JP5633271B2 - COMMUNICATION DEVICE, COMMUNICATION METHOD, AND COMMUNICATION SYSTEM - Google Patents

Description

  The present invention relates to a communication apparatus and a communication method for receiving, for example, transmission signals transmitted from a plurality of antennas using a plurality of antennas, and a communication system using such a communication apparatus.

Conventionally, there has been a great demand for high-speed data transfer in wireless communication. Therefore, standardization of next-generation data communication standards such as Long Term Evolution (LTE) is underway. In high-speed data communication standards such as LTE, Multiple Input Multiple Output (MIMO) technology that allows two communication devices to both increase and decrease the bandwidth by sending and receiving signals in parallel using multiple antennas Has been.
In a communication system employing MIMO technology, a transmission apparatus having a plurality of antennas divides one or more data streams into a plurality of transmission signals. The transmission apparatus transmits each transmission signal via a separate antenna. On the other hand, a receiving device having a plurality of antennas receives a transmission signal from a transmitting device at each antenna. Then, the receiving device separates transmission signals transmitted simultaneously from signals received by the respective antennas. As a method for separating the transmission signal, for example, a least mean square error (MMSE) and a maximum likelihood detection (Maximum Likelihood Detection, MLD) are known. A receiver using MLD compares a set of received signals actually received with a set of received signals estimated from a set of transmission signal candidates that may be transmitted. The transmission signal candidate corresponding to the most probable group is the signal that is actually transmitted. In order to compare the received signal set estimated with respect to the set of transmission signal candidates and the actual received signal set, the receiving apparatus using the MLD is a square Euclidean distance between the two received signal sets Calculate metrics such as
MLD can obtain superior reception characteristics as compared with linear separation methods such as MMSE. However, the amount of calculation performed by the MLD to separate the transmission signal from the reception signal is larger than the amount of calculation performed by the linear separation method such as MMSE to separate the transmission signal. In particular, as the number of transmission signals transmitted simultaneously and the number of values that can be taken in the modulation scheme used for transmission increase, the number of metric calculations in MLD increases exponentially.
Therefore, MLD with a reduced amount of calculation has been proposed (see, for example, Patent Documents 1 to 3 and Non-Patent Document 1).

ただし、Nは2以上の整数である。なお、送信信号の数は、受信信号の数、すなわち、受信装置が有するアンテナの数と同一でなくてもよい。同時に送信される送信信号の数よりも、受信装置が有するアンテナの数が多ければ、MIMO技術による信号伝送が可能となる。また行列Hは、送信信号と受信信号の対応関係を記述する有効チャネル行列を表す。（１）式では、簡単化のために、送信信号に加わるノイズは省略されている。QRM-MLD法では、有効チャネル行列Hがユニタリー行列Qと三角行列Rに分解され、次式のように表される。

（１）式の両辺に、左側からユニタリー行列Qのエルミート共役Q^Hを乗じることにより、次式が得られる。

なお、ベクトルz=(z₀, z₁, ..., z_N-1)は、受信信号ベクトルYと行列Q^Hとの積により得られる受信信号ベクトルのユニタリ変換ベクトルである。（３）式に示されるように、ユニタリ変換ベクトルzの各要素に関連する送信信号の数は異なる。例えば、信号z_N-1に対しては、送信信号x_N-1のみが関連する。一方、信号z₀に対しては、N個の送信信号x₀〜x_N-1が関連する。 For example, Non-Patent Document 1 discloses a technique for controlling the number of symbol replica candidates using reliability information of symbol replica candidates in the QRM-MLD method in which the number of metric calculations is reduced by combining QR decomposition and the M algorithm. Is disclosed.
First, the QRM-MLD method will be described. N transmission signals (x ₀ , x ₁ , ..., x _N-1 ) transmitted simultaneously from the transmission device and N reception signals (y ₀ , y ₁ ,. .., y _N-1 ) is expressed by the following equation.

However, N is an integer of 2 or more. Note that the number of transmission signals may not be the same as the number of reception signals, that is, the number of antennas included in the reception apparatus. If the number of antennas included in the receiving device is larger than the number of transmission signals transmitted at the same time, signal transmission by the MIMO technique becomes possible. The matrix H represents an effective channel matrix that describes the correspondence between the transmission signal and the reception signal. In equation (1), noise added to the transmission signal is omitted for the sake of simplicity. In the QRM-MLD method, the effective channel matrix H is decomposed into a unitary matrix Q and a triangular matrix R, and is expressed as follows.

By multiplying both sides of the equation (1) by the Hermitian conjugate Q ^H of the unitary matrix Q from the left side, the following equation is obtained.

Note that the vector z = (z ₀ , z ₁ ,..., Z _N−1 ) is a unitary transformation vector of the received signal vector obtained by the product of the received signal vector Y and the matrix Q ^H. As shown in Equation (3), the number of transmission signals related to each element of the unitary transformation vector z is different. For example, for a signal z _N-1, only relevant transmission signals x _N-1. On the other hand, N transmission signals x _{0 to} x _N−1 are related to the signal z ₀ .

The receiving apparatus inputs a value of z ₀ to z _N−1 obtained from the actually received received signal and a set of symbol replicas corresponding to the transmitted signal that may be transmitted to the vector X in Equation (3). The square Euclidean distance with the value of z ₀ ^′ to z _N−1 ^′ obtained by doing is calculated as a metric. This symbol replica is a signal temporarily set by the receiving apparatus. In MLD, the receiving apparatus estimates that a set of symbol replicas that minimizes the sum of metrics obtained for all of z _{0 to} z _N−1 is an actually transmitted transmission signal.

The reception apparatus using the QRM-MLD method calculates metrics in order from the signal z _{0 to} z _N−1 having the smallest number of related transmission signals. For example, in Equation (3), the number of transmission signals related to the signal z _N-1 is minimized. Therefore, the receiving apparatus calculates a metric for the signal z _N-1 in the first stage, and calculates a metric for the signal z _N-2 with the _second smallest number of related transmission signals in the next stage. . At that time, the receiving apparatus calculates metrics only for M symbol replicas in order from the smallest metric calculated for the signal z _{N-m + 1} of the previous stage in the m-th stage. For example, assume that M = 3 and the transmission signals x _{0 to} x _N−1 are modulated by 64QAM. In this case, there are 64 symbol replicas for each transmission signal. And since the signal z _N-1 is associated with only the transmission signal x _N-1, the receiving apparatus 64 symbol replica c _{N-1, 0} to c corresponding to the transmitted signal x _N-1 in the first stage Calculate the metric for _N-1,63 . As a result, the symbol replicas corresponding to the three metrics are c _{N-1, a} , c _{N-1, b} , c _{N-1, c} (where 0 ≦ a, b, c Assume that ≦ 63 and a ≠ b ≠ c). In this case, when the receiving apparatus calculates a metric for the signal z _N-2 in the second stage, three symbol replicas c _{N−1, a} , c _{N−1, b} are transmitted for the transmission signal x _N−1. , C _{N-1, c} only. Then, the receiving apparatus transmits one of the three symbol replicas c _{N−1, a} , c _{N−1, b} , c _{N−1, c} and 64 symbol replicas c that the transmission signal x _N−2 can take. _A metric is calculated for each pair with any of _N-2,0 to _cN-2,63 .
As described above, the reception apparatus using the QRM-MLD method does not have to calculate the metrics for all the symbol replica pairs, and thus the amount of calculation can be reduced.

In an example of a method for further reducing the number of metric operations than the QRM-MLD method, the receiving device subtracts the signal components of surviving symbol replica candidates from the unitary transformed signals z _{0 to} z _{N-1 at} each stage of calculating the metric. Find the remaining received signal. Based on the residual received signal, the receiving apparatus ranks each symbol replica candidate of the transmission signal for calculating the branch metric in the order in which the branch metric is expected to be small. Then, the receiving apparatus calculates the cumulative value of the branch metrics in order from the highest ranked symbol replica candidate, and survives only the symbol replica candidates whose cumulative value of branch metrics is smaller than a predetermined threshold. In addition, ranking for symbol replica candidates is performed, for example, by determining a quadrant to which a residual received signal belongs.
The predetermined threshold is set so that the larger the estimated noise power contained in the received signal is, the larger the value is such that the combination of more symbol replica candidates becomes the candidate for the transmitted signal as the signal reception environment is worse. The For example, when the transmission signal is multiplexed by Orthogonal Frequency-Division Multiplexing (OFDM), the noise power used to determine the predetermined threshold is the error in the channel estimation value between subcarriers. Estimated from the average value of power.

  Moreover, in order to perform error correction decoding processing such as turbo code decoding or Viterbi decoding from the separated transmission signal, the likelihood of the symbol replica candidate combination other than the maximum likelihood symbol replica candidate combination is also used. There is. However, the transmission signal separation method that does not calculate the metric for all symbol replica candidate combinations as described above cannot directly calculate the likelihood for symbol replica candidate combinations other than the most likely symbol replica candidate combinations. There is. Thus, methods have been proposed for calculating likelihoods for combinations of symbol replica candidates whose metrics are not calculated (for example, Patent Document 4 and Non-Patent Document 2).

JP 2008-301188 A JP 2008-182332 A JP 2006-270430 A JP 2008-124843 A

Kawai, Higuchi, Maeda, Sawahashi, “Adaptive Survival Symbol Replica Candidate Selection Method Based on Reliability Information Using Each Stage Independent Candidate Control in OFDM MIMO Multiplexing Using QRM-MLD”, IEICE Technical Report RCS2004-221, 2004 November, pp. 61-66 Seki, Ito, Shikakura, Kawai, Maeda, Higuchi, “Sensitive information generation method for soft decision turbo decoding suitable for symbol replica candidate reduction QR decomposition-MLD in OFCDM MIMO multiplexing”, IEICE Technical Report RCS2003-314, 2004 March, pp. 67-72

An arithmetic unit that separates transmission signals by improving the MLD method with a smaller amount of calculation than the MLD method or the MLD method is hereinafter referred to as an MLD arithmetic unit for convenience. In particular, the number of metric calculations performed by the MLD calculator that implements the transmission signal separation method that combines the QRM-MLD method described above and determination of the metric cumulative value using a threshold based on noise power varies depending on the noise power. To do.
For example, it is assumed that a plurality of transmission signal sets are multiplexed by OFDM, and one transmission signal set is superimposed on one subcarrier. Here, under a multipath environment, reception quality varies for each subcarrier due to the influence of frequency selective fading. In the above known technique, the threshold for the cumulative value of branch metrics is determined according to the noise power. Therefore, if the reception quality of each subcarrier is different from each other, there is a possibility that the number of times the metric is calculated by the MLD calculator differs for each subcarrier, that is, for each set of transmission signals.
For example, for a set of transmission signals superimposed on subcarriers whose threshold is set to a sufficiently large value due to high noise power, the cumulative value of branch metrics does not exceed the threshold. For this reason, the MLD computing unit calculates a metric for all the survival candidates set in advance for each stage. That is, the total number of metric calculations for such subcarriers is maximized.
On the other hand, for a set of transmission signals superimposed on subcarriers with a low threshold due to low noise power, the cumulative value of branch metrics may exceed the threshold at any stage. In such a case, the number of symbol replicas for which the metric is calculated is smaller than the preset number of surviving candidates. Therefore, the total number of metric calculations for such subcarriers is less than the maximum value of the total metric calculation count.

In order for the MLD calculator to separate transmitted signals within a predetermined time regardless of noise power, the MLD calculator is designed so that the amount of calculation corresponding to the maximum value of the total number of metric calculations can be executed within the predetermined time. The Therefore, the MLD calculator is redundant for subcarriers whose metric calculation count is less than the maximum value. In particular, when a single MLD computing unit separates transmission signals for a plurality of subcarriers, the circuit scale of the MLD computing unit is the number of subcarriers for which the MLD computing unit performs signal separation processing, and the total number of metric computations. The amount of computation multiplied by the maximum value can be executed within a predetermined time. Therefore, there is a possibility that the MLD computing unit becomes redundant as the number of subcarriers n increases.
When the circuit scale of an MLD computing unit increases, it becomes difficult to mount the MLD computing unit on a mobile station device such as a mobile phone. In addition, the larger the scale of the arithmetic circuit, the more power is consumed in the arithmetic circuit. Therefore, it is preferable that the circuit scale of the MLD arithmetic unit is smaller for a receiving device driven by a battery such as a portable terminal. Therefore, it is not preferable that the MLD arithmetic unit becomes redundant.

  Therefore, the present specification provides a communication device and a communication method capable of suppressing an increase in circuit scale of an MLD computing unit that separates a plurality of transmission signals based on a plurality of reception signals received via a plurality of antennas. With the goal.

  According to one embodiment, a communication device is provided. This communication device is connected to any one of a plurality of antennas and a plurality of antennas, and receives a plurality of reception signals respectively received from a transmission device having a plurality of antennas via the connected antennas. And a selection unit that selects a reception signal set corresponding to the transmission signal set from signals received by the plurality of reception units, and each transmission signal included in the transmission signal set is separated from the reception signal set And an arithmetic unit. The arithmetic unit adjusts the number of transmission signal group candidates according to the reception state of the reception signal group, and transmits the candidate having the highest posterior probability of being a reception signal group among the transmission signal group candidates. A transmission signal estimation unit that estimates as a set of signals, an output unit that outputs a predetermined signal used for decoding the set of transmission signals, and a calculation amount by the transmission signal estimation unit are counted. When the calculation amount reaches a predetermined threshold before completing the estimation of the set, the output unit outputs a predetermined signal, while the transmission signal estimation unit estimates the transmission signal before the calculation amount reaches the predetermined threshold. And a control unit that causes the transmission signal estimation unit to output the estimated transmission signal set when the unit finishes the estimation of the transmission signal set.

According to another embodiment, a predetermined number of transmission signals include a plurality of antennas and a modulation unit that superimposes a set of a predetermined number of transmission signals on any of a plurality of subcarriers having different frequency bands. Each of the transmission signals included in the set is transmitted from a different antenna among the plurality of antennas, the plurality of antennas, and any one of the plurality of antennas, via the connected antennas A plurality of receiving units that respectively receive a predetermined number of transmission signal sets, and a selection unit that selects a predetermined number of received signal sets corresponding to the predetermined number of transmission signal sets from signals received by the plurality of receiving units. And a reception device having a computing unit that separates each transmission signal included in the corresponding transmission signal set from each of the predetermined number of reception signal sets.
Then, the arithmetic unit of the receiving apparatus adjusts the number of transmission signal set candidates according to the reception state of the reception signal set, and among the transmission signal set candidates, the posterior probability of being the reception signal set is the highest. A transmission signal estimation unit that estimates a high candidate as a set of transmission signals, an output unit that outputs a predetermined signal used for decoding the transmission signal set, and a calculation amount by the transmission signal estimation unit, When the calculation amount reaches a predetermined threshold before the transmission signal set estimation ends, the output unit outputs a predetermined signal, while the transmission signal estimation unit calculates the calculation amount before the predetermined threshold. When the transmission signal estimation unit finishes estimating the transmission signal set, the control unit causes the transmission signal estimation unit to output the estimated transmission signal set.
On the other hand, the transmission apparatus applies cyclic delay diversity to at least one of the plurality of subcarriers when frequency selective fading does not occur for a predetermined number of sets of transmission signals.

The objects and advantages of the invention will be realized and attained by means of the elements and combinations particularly pointed out in the appended claims.
It should be understood that both the foregoing general description and the following detailed description are exemplary and explanatory and are not restrictive of the invention as claimed.

  The communication apparatus and the communication method disclosed in this specification can suppress an increase in the circuit scale of the MLD arithmetic unit.

It is a schematic block diagram of the communication system which has a receiver by one Embodiment. It is a schematic block diagram of the stream separation part which the receiving apparatus by 1st Embodiment has. It is a schematic block diagram of an MLD calculator. (A) ~ (c) are diagrams each showing the estimate of the transmitted signal x _0, the relationship between the quadrant estimate belongs. It is an operation | movement flowchart of a transmission signal separation process. It is an operation | movement flowchart of a transmission signal separation process. It is a graph which shows the simulation result of a block error rate in case coefficient α = 0.8. It is a schematic block diagram of the stream separation part of the receiver by 2nd Embodiment. (A) is a figure which shows an example of the received signal quality calculated | required about each subcarrier, (b) is a figure which shows an example of the subcarrier allocated to each MLD computing unit. It is an operation | movement flowchart of the transmission signal separation process by 2nd Embodiment. It is a schematic block diagram of the MLD calculator by 3rd Embodiment. It is a schematic block diagram of the communication system by 4th Embodiment. It is a schematic block diagram of the base station apparatus with which the receiver which concerns on any embodiment was integrated. It is a schematic block diagram of the mobile station apparatus with which the receiver which concerns on any embodiment was integrated.

Hereinafter, a receiving apparatus according to an embodiment will be described with reference to the drawings.
This receiving apparatus receives a plurality of transmission signals transmitted from a transmitting apparatus having a plurality of antennas via a plurality of antennas by using the MIMO technology. The receiving apparatus separates the transmission signal from the reception signal received by each of the plurality of antennas. In order to separate the transmission signal, the MLD computing unit included in this receiving apparatus determines the transmission signal based on the transmission signal separation method that combines the QRM-MLD method and determination of the cumulative metric value using a threshold value based on noise power. To separate. Then, when the amount of calculation during execution of the transmission signal separation process reaches a predetermined threshold, the MLD calculator stops the estimation of the transmission signal and outputs a predetermined signal value. As a result, the receiving apparatus suppresses the circuit scale of the MLD computing unit to the circuit scale of the computation amount corresponding to the threshold value.

  In the present embodiment, a plurality of sets of transmission signals are multiplexed in the OFDM format. However, a plurality of sets of transmission signals may be multiplexed according to other multiplexing schemes that may have different reception states for each channel or subcarrier. Such multiplexing schemes include, for example, frequency division multiplexing or single carrier frequency division multiplexing (SC-FDM).

FIG. 1 is a schematic configuration diagram of a communication system 1 according to the first embodiment. The communication system 1 includes a transmission device 2 having two antennas 21-1 and 21-2 and a reception device 3 having two antennas 31-1 and 31-2. And the transmitter 2 radiates | emits the radio signal on which the transmission signal was superimposed simultaneously from each antenna 21-1, 21-2. On the other hand, the receiving device 3 receives the radio signal radiated from the transmitting device 2 via the antennas 31-1 and 31-2. In addition, the signal received with each antenna 31-1, 31-2 is called a received signal below. The receiving device 3 obtains a transmission signal based on each received signal.
In this embodiment, the number of antennas included in the transmission apparatus 2 is merely an example, and the number of antennas included in the transmission apparatus 2 is an arbitrary number of two or more that can be physically mounted on the transmission apparatus 2. Can be set to Similarly, the number of antennas included in the reception device 3 is merely an example, and the number of antennas included in the reception device 3 is set to an arbitrary number of two or more that can be physically mounted on the reception device 3. Can do. Furthermore, the number of antennas included in the transmission device 2 may be different from the number of antennas included in the reception device 3.

  The transmission apparatus 2 includes two antennas 21-1, 21-2, a code word generation unit 22, an encoding unit 23, a modulation unit 24, two transmission units 25-1, 25-2, and a control unit. 26. The code word generation unit 22, the encoding unit 23, the modulation unit 24, the transmission units 25-1, 25-2, and the control unit 26 may be mounted on the transmission device 2 as separate circuits, and these circuits are integrated. The transmitter 2 may be mounted as a single integrated circuit.

The codeword generation unit 22 divides the data to be transmitted into codewords having a length of a transport size block (Transport Block Size, TBS) determined by the control unit 26. The codeword can be, for example, MAC-PDU data according to the specifications of the media access control (MAC) layer and the protocol data unit (PDU) layer. Further, the codeword generation unit 22 refers to the number of streams for the codeword encoded by the encoding unit 23 determined by the control unit 26 and assigns the generated codeword to the encoding unit 23. The number of streams represents the number of data transmitted simultaneously for one codeword.
The codeword generation unit 22 passes the generated codeword to the encoding unit 23.

  The encoding unit 23 performs an error correction encoding process such as a convolutional encoding process or a turbo encoding process on the codeword received from the codeword generating unit 22. Furthermore, the encoding unit 23 generates a data stream by dividing the codeword encoded by the number of streams acquired from the control unit 26. Then, the encoding unit 23 outputs the encoded codeword to the modulation unit 24 in units of data streams.

The modulation unit 24 modulates the data stream received from the encoding unit 23 according to the modulation mode MOD determined by the control unit 26. The modulation unit 24 superimposes the modulated data stream on a subcarrier together with various control signals and pilot signals received from the control unit 26 as a transmission signal. At that time, the modulation unit 24 superimposes a set of one transmission signal transmitted simultaneously from each antenna on subcarriers having the same frequency band. Then, the modulation unit 24 multiplexes a plurality of subcarriers transmitted from the same antenna according to OFDM, that is, by performing inverse Fourier transform.
Furthermore, the modulation unit 24 outputs the multiplexed transmission signal to one of the transmission units 25-1 and 25-2 according to the precoding matrix determined by the control unit 26.

  Each of the transmission units 25-1 and 25-2 superimposes the multiplexed transmission signal input from the modulation unit 24 on a carrier wave having a radio frequency. The transmission units 25-1 and 25-2 have high power amplifiers. Then, the transmission units 25-1 and 25-2 amplify the intensity of the multiplexed transmission signal to a desired level. The transmission units 25-1 and 25-2 are connected to the antennas 21-1 and 21-2, respectively, and transmit multiplexed transmission signals as radio signals via the connected antennas.

The control unit 26 controls the entire transmission device 2. For example, the control unit 26 refers to the feedback information received from the receiving device 3 and determines the length of the codeword, the modulation mode, the number of streams of each codeword, and the precoding matrix. Then, the control unit 26 notifies information related to the length of the codeword, the modulation mode, and the like to each related unit of the transmission device 2.
Note that the feedback information includes, for example, a CQI (Channel Quality Indicator) value indicating the quality of the signal received by the receiving device 3. The control unit 26 can demodulate and decode a signal from the receiving device 3 received from a receiving unit (not shown) connected to one of the antennas, and extract feedback information from the decoded signal.

  Next, the receiving device 3 according to the first embodiment will be described. The reception device 3 includes two antennas 31-1 and 3-2, two reception units 32-1 and 32-2, a demodulation unit 33, a decoding unit 37, and a data integration unit 38. The receiving units 32-1 and 32-2, the demodulating unit 33, the decoding unit 37, and the data integrating unit 38 may be mounted on the receiving device 3 as separate circuits, and as an integrated circuit in which these circuits are integrated, The receiver 3 may be mounted.

The receiving units 32-1 and 32-2 are connected to the antennas 31-1 and 31-2, respectively. The receiving units 32-1 and 32-2 receive the radio signals transmitted from the antennas 21-1 and 21-2 of the transmission device 2 via the antennas 31-1 and 31-2, respectively.
Each of the reception units 32-1 and 32-2 has a low noise amplifier, and amplifies the received radio signal by the low noise amplifier. Then, the reception units 32-1 and 32-2 generate a baseband signal having a baseband frequency by superimposing a signal having an intermediate frequency on the amplified radio signal, and output the baseband signal to the demodulation unit 33. To do.

  The demodulator 33 separates the transmission signals transmitted from the antennas of the transmission device 2 from the baseband signals received from the reception units 32-1 and 32-2. For this purpose, the demodulation unit 33 includes a Fourier transform unit 34, a channel estimation unit 35, and a stream separation unit 36.

The Fourier transform unit 34 is an example of a selection unit, and reproduces a signal for each subcarrier by performing Fourier transform on each baseband signal received from the reception units 32-1 and 32-2. And the Fourier-transform part 34 selects the signal of the frequency band corresponded to the subcarrier with which the transmission signal was superimposed among the signals reproduced | regenerated from the baseband signal received from each receiving part as a received signal, respectively. Therefore, the Fourier transform unit 34 obtains a set of reception signals for each subcarrier on which the transmission signal is superimposed. The Fourier transform unit 34 passes the received signal set to the channel estimation unit 35 and the stream separation unit 36.
Note that the demodulation unit 33 has the same number of Fourier transform units as the number of antennas included in the reception device, and each Fourier transform unit performs Fourier transform on a baseband signal generated from radio signals received by different antennas. Also good.

ここで、x₀、x₁は、それぞれ、アンテナ２１−１、２１−２から送信された送信信号を表す。送信信号ベクトルXは、各送信信号x₀、x₁を要素とするベクトルである。またy₀、y₁は、それぞれ、アンテナ３１−１、３１−２を介して受信された受信信号を表す。受信信号ベクトルYは、各受信信号y₀、y₁を要素とするベクトルである。行列Hはチャネル行列を表し、その要素h_ijは、例えば、パイロット信号に対するチャネルインパルス応答として求められる。さらに、ベクトルnはノイズベクトルを表し、その要素n₀、n₁は、それぞれ、アンテナ３１−１、３１−２で受信された受信信号に含まれるノイズ成分を表す。なお、チャネルの順序は上記の例に限定されない。例えば、送信信号x₀、x₁はそれぞれ、アンテナ２１−２、２１−１から送信された送信信号であってもよく、また、y₀、y₁は、それぞれ、アンテナ３１−２、３１−１を介して受信された受信信号であってもよい。
チャネル推定部３５は、チャネル行列をストリーム分離部３６へ渡す。
なお、復調部３３は、多重化されているサブキャリアの数と同数のチャネル推定部を有し、各チャネル推定部がそれぞれ互いに異なるサブキャリアのチャネル行列を生成してもよい。
ストリーム分離部３６は、サブキャリアごとに、チャネル行列に基づいて受信信号の組から送信信号の組に含まれる各送信信号を分離する。なお、ストリーム分離部３６の詳細については後述する。 The channel estimation unit 35 obtains a channel matrix that represents the relationship between the transmission signal and the reception signal for each subcarrier. For example, the channel estimation unit 35 sets a channel impulse response value of a signal known to the transmission device 2 and the reception device 3 such as a pilot signal included in a signal transmitted from each antenna of the transmission device 2 for each subcarrier. Ask. The channel estimation unit 35 uses the channel impulse response value as each element of the channel matrix. In this case, the relationship between the transmission signal and the reception signal is expressed by the following equation using a channel matrix.

Here, x ₀ and x ₁ represent transmission signals transmitted from the antennas 21-1 and 21-2, respectively. The transmission signal vector X is a vector having the transmission signals x ₀ and x ₁ as elements. Y ₀ and y ₁ represent received signals received via the antennas 31-1 and 31-2, respectively. The received signal vector Y is a vector having the received signals y ₀ and y ₁ as elements. The matrix H represents a channel matrix, and its element h _ij is obtained as a channel impulse response to a pilot signal, for example. Further, the vector n represents a noise vector, and its elements n ₀ and n ₁ represent noise components included in the received signals received by the antennas 31-1 and 31-2, respectively. Note that the channel order is not limited to the above example. For example, the transmission signals x ₀ and x ₁ may be transmission signals transmitted from the antennas 21-2 and 21-1, respectively, and y ₀ and y ₁ may be the antennas 31-2 and 31-, respectively. 1 may be a received signal received via 1.
The channel estimation unit 35 passes the channel matrix to the stream separation unit 36.
Note that the demodulation unit 33 may have the same number of channel estimation units as the number of multiplexed subcarriers, and each channel estimation unit may generate different channel matrices of subcarriers.
The stream separation unit 36 separates each transmission signal included in the transmission signal set from the reception signal set based on the channel matrix for each subcarrier. Details of the stream separator 36 will be described later.

The demodulator 33 demodulates the separated transmission signal in accordance with the modulation method applied to the transmission signal, and passes the demodulated signal to the decoding unit 37.
The demodulator 33 calculates feedback information such as a CQI value to be fed back to the transmission apparatus 2 and sends the feedback information to a transmitter (not shown). The transmitter generates a radio signal by orthogonally modulating a signal having feedback information and then superimposing the signal on a carrier wave having a radio frequency. The transmission unit then radiates a radio signal carrying feedback information via any antenna.

  The decoding unit 37 generates an encoded codeword by combining the data streams received from the demodulation unit 33. Then, the decoding unit 37 performs an error correction decoding process on the encoded code word. The decoding unit 37 outputs the codeword subjected to the decoding process to the data integration unit 38.

  The data integration unit 38 reproduces the original data by combining the code words received from the decoding unit 37 in a predetermined order. Then, the data integration unit 38 outputs the original data to other components (not shown) of the reception device 3.

Details of the stream separation unit 36 will be described below.
FIG. 2 is a schematic configuration diagram of the stream separation unit 36. The stream separation unit 36 includes four MLD calculators 40-1 to 40-4. Each MLD computing unit 40-1 to 40-4 separates a transmission signal from a set of reception signals superimposed on the same subcarrier among the reception signals received by the antennas 31-1 and 31-2. . Then, the MLD computing units 40-1 to 40-4 output the separated transmission signals to the decoding unit 37.
The number of MLD calculators included in the stream separation unit 36 is not limited to four. It is preferable that the number of MLD calculators included in the stream separator 36 is set so that the number of subcarriers in which each MLD calculator separates transmission signals for one frame is equal. For example, if the number of subcarriers multiplexed by OFDM is 8, the stream separation unit 36 preferably has 8, 4, 2, or 1 MLD computing units.

FIG. 3 is a schematic configuration diagram of the MLD computing unit 40-1. Since each MLD computing unit 40-1 to 40-4 has the same configuration, detailed description of the MLD computing units 40-2 to 40-4 will be omitted.
The MLD computing unit 40-1 separates each transmission signal included in the transmission signal set from the reception signal set. For this purpose, the MLD computing unit 40-1 includes a transmission signal estimation unit 41, a log likelihood ratio calculation unit 42, an alternative value output unit 43, a control unit 44, and a selector 45.
Each of these units included in the MLD computing unit 40-1 may be a separate circuit. Alternatively, each of these units included in the MLD computing unit 40-1 may be a single arithmetic circuit that realizes the function of each unit.

The transmission signal estimation unit 41 adjusts the number of transmission signal group candidates according to the reception state of the reception signal group, and among the transmission signal group candidates, the posterior probability of being the reception signal group is the highest. A high candidate is estimated as a set of transmission signals. Therefore, in the present embodiment, the transmission signal estimation unit 41 uses a transmission signal separation method that combines the QRM-MLD method and the determination of the cumulative metric value using a threshold value based on the noise power included in the reception signal.
The transmission signal estimation unit 41 includes a QR decomposition unit 411, a ranking determination unit 412, a metric calculation unit 413, and a determination unit 414.

そしてQR分解部４１１は、（４）式の両辺に、それぞれ左側からユニタリ行列のエルミート共役Q^Hを乗じる。これにより、QR分解部４１１は、受信信号ベクトルYをユニタリ変換したユニタリ変換ベクトルzを求める。ユニタリ変換ベクトルzと、上三角行列R及び送信信号ベクトルXとの関係は次式で表される。

ここで、ユニタリ変換信号z₀、z₁は、それぞれユニタリ変換ベクトルzの要素であり、何れも、受信信号y₀、y₁に関する成分を有する。
なお、QR分解部４１１は、QR分解を行うために、例えば、ギブンス回転、ハウスホルダー変換あるいはグラム・シュミット分解などを用いることができる。
QR分解部４１１は、ユニタリ変換ベクトルz及び上三角行列Rをランキング決定部４１２及びメトリック計算部４１３へ渡す。 The QR decomposition unit 411 performs QR decomposition on the channel matrix H received from the channel estimation unit 35 into a unitary matrix Q and an upper triangular matrix R. The channel matrix H is expressed as follows using a unitary matrix Q and an upper triangular matrix R.

The QR decomposition unit 411, (4) to both sides of the equation is multiplied by the Hermitian conjugate Q ^H from the left, respectively unitary matrices. Accordingly, the QR decomposition unit 411 obtains a unitary transformation vector z obtained by unitary transformation of the received signal vector Y. The relationship between the unitary transformation vector z, the upper triangular matrix R, and the transmission signal vector X is expressed by the following equation.

Here, the unitary transformation signals z ₀ and z ₁ are elements of the unitary transformation vector z, and both have components related to the reception signals y ₀ and y ₁ .
Note that the QR decomposition unit 411 can use, for example, Givens rotation, Householder conversion, or Gram-Schmidt decomposition to perform QR decomposition.
The QR decomposition unit 411 passes the unitary transformation vector z and the upper triangular matrix R to the ranking determination unit 412 and the metric calculation unit 413.

The ranking determination unit 412 ranks the symbol replicas for each transmission signal. At that time, the ranking determination unit 412 pays attention to the unitary conversion signals that are elements of the unitary conversion vector z in order from the smallest number of related transmission signals. A stage is set for each unitary conversion signal. The ranking determination unit 412 obtains an estimated value of one transmission signal for each stage, and ranks the symbol replicas in order from the estimated value by quadrant determination.
Note that the symbol replica is generated by the receiving device 3 with a value that the transmission signal can take in the modulation scheme applied to the transmission signal.

In the present embodiment, referring to the equation (6), the unitary conversion signal z ₁ includes only components related to the transmission signal x ₁ . Therefore, the ranking determination unit 412 focuses on the unitary conversion signal z ₁ in the first stage. Then, the ranking determination unit 412 sets (z ₁ / r ₁₁ ) as the estimated value u ₁ of the symbol x ₁ .
Ranking determining unit 412 to determine the rank of each symbol replica for the transmission signal x _1, in the coordinate system of the I and Q signals indicating the signal values that can take the transmitted signal x _1, estimates of the transmitted signal x ₁ determining the quadrant value u ₁ belongs. Then, the ranking determination unit 412 uses the center position of the quadrant to which the estimated value u ₁ belongs as an origin, and performs, for example, at least ((1/2) log ₂ m ₁ ) quadrant determination, thereby being close to the estimated value u ₁ . In turn, the rank for each symbol replica is determined. Note that m ₁ is the number of values that the modulation scheme applied to the transmission signal x ₁ can take. For example, if the modulation scheme is QPSK, m ₁ = 2 and the modulation scheme is 64QAM. In this case, m ₁ = 8.

The quadrant determination process will be described with reference to FIGS. 4 (a) to 4 (c). As an example, the modulation scheme applied to the transmitted signal x ₁ is 64QAM. 4A to 4C, the horizontal axis represents the I signal component, and the vertical axis represents the Q signal component. The points 401, respectively, a signal point corresponding to the signal values that can take the transmitted signal x _1. For example, the point 401a represents a signal value corresponding to the symbol “101111”. An asterisk 410 represents the estimated value u ₁ . In this example, both the I signal component and the Q signal component of the estimated value u ₁ are negative values.

As shown in FIG. 4A, the estimated value u ₁ is located in the fourth quadrant 420 because both the I signal component and the Q signal component are negative.
Therefore, next, the ranking determining unit 412 uses the estimated value when the signal value (−4 / √42, −4 / √42) corresponding to the center of the fourth quadrant to which the estimated value u ₁ belongs is set as the origin O ′. u Determine the quadrant to which ₁ belongs. That is, the ranking determination unit 412 obtains an estimated value u ′ ₁ by subtracting the signal value (−4 / √42, −4 / √42) corresponding to the center of the fourth quadrant from the estimated value u ₁ , It is determined whether the I signal component and the Q signal component of the estimated value u ′ ₁ are positive or negative. As shown in FIG. 4B, in this example, the estimated value u ₁ belongs to the first quadrant 430 with the origin O ′ as a reference.

Next, the ranking determination unit 412 uses the signal value (−2 / √42, −2 / √42) corresponding to the center of the first quadrant 430 to which the estimated value u ₁ belongs, relative to the origin O ′ as the origin. O ″, the quadrant to which the estimated value u ₁ belongs is determined. That is, the ranking determination unit 412 determines the signal value (−2 / √42, − corresponding to the center of the first quadrant 430 from the estimated value u ₁ . 2 / √42) "seek _1, the estimated value u" estimated value u by subtracting the determined I and Q signal components of ₁ or respectively positive or negative. Then illustrated in FIG. 4 (c) As shown, the estimated value u ₁ belongs to the second quadrant 440 with reference to the origin O ″. The second quadrant 440 corresponds to one symbol “110001”. Therefore, the signal point corresponding to the symbol “110001” is closest to the estimated value u ₁ . Therefore, the ranking unit 412 sets the symbol replica of “110001” to the highest rank.

Similarly, the ranking determination unit 412 repeats the process of determining the quadrant to which the estimated value u ₁ belongs, thereby obtaining the second to Mth symbol replicas with respect to the estimated value u ₁ . M is the number of surviving symbol replicas in each stage, and is set to 3, for example.

ここでr₀₀、r₀₁は、それぞれ上三角行列Rの要素である。またc_1i(i=1,2,..,M)は、最初のステージにおいて生き残った送信信号x₁のシンボルレプリカである。
ランキング決定部４１２は、ステージごとに、注目する送信信号のシンボルレプリカのランクをメトリック計算部４１３へ通知する。 When the second and subsequent stages are the stage of interest, the ranking determination unit 412 excludes the component of the transmission signal selected up to the previous stage from the unitary conversion signal of interest of the stage of interest. Thereby, the ranking determination unit 412 can obtain an estimated value included in one transmission signal of interest included in the transmission signal vector X. Then, the ranking determining unit 412 ranks each symbol replica for the transmission signal of interest by executing quadrant determination processing for the transmission signal of interest.
For example, the ranking determining unit 412 pays attention to the unitary conversion signal z ₀ in the second stage when referring to the equation (6). Unitary transformation signal z ₀ has components of both the transmitted signal x ₀ and x _1. Therefore, the ranking determination unit 412 obtains an estimated value u _0i (i = 1, 2,... M) of the transmission signal x ₀ according to the following equation.

Here, r ₀₀ and r ₀₁ are elements of the upper triangular matrix R, respectively. C _1i (i = 1, 2,..., M) is a symbol replica of the transmission signal x ₁ that survives in the first stage.
The ranking determination unit 412 notifies the metric calculation unit 413 of the rank of the symbol replica of the transmission signal of interest for each stage.

ここで(c_0j,c_1j)(j=1,2,...,k、ただしk≦M)は、それぞれ、送信信号x₀、x₁の候補として生き残ったシンボルレプリカである。
あるいは、メトリック計算部４１３は、メトリックd_1j及びd_2jとして、次式のように、ユニタリ変換信号z₀、z₁と、シンボルレプリカの組(c_0j,c_1j)を（６）式の右辺の第１項に入力することにより得られた推定受信信号とのマンハッタン距離を算出してもよい。

なお、関数Re(P)は、変数Pの実数成分を出力する関数であり、関数Im(P)は、変数Pの虚数成分を出力する関数である。またメトリック計算部４１３は、ユニタリ変換信号z₀、z₁と、生き残ったシンボルレプリカの組(c_0j,c_1j)間の距離を表す他の指標をメトリックとして計算してもよい。
メトリック計算部４１３は、メトリックを１回計算する度に、メトリックを計算したことを表す演算実行信号を制御部４４へ通知する。 The metric calculation unit 413 selects, for each stage, a surviving symbol replica set that minimizes the total metric value in the stage immediately preceding that stage. Then, the metric calculation unit 413 calculates metrics in order from the higher rank for the symbol replica of the transmission signal of interest in the stage. Then, the metric calculation unit 413 adds a symbol replica whose metric has been newly calculated to the selected symbol replica set. In addition, the metric calculation unit 413 updates the metric total value by adding the metric calculated at the stage of interest to the total metric value calculated for the selected symbol replica pair.
The metric is an index of the probability that the transmission signal is actually transmitted. In the present embodiment, the metric d _1j in the first stage and the metric d _2j in the second stage are each expressed by the following equations using the square Euclidean distance.

Here, (c _0j , c _1j ) (j = 1, 2,..., K, where k ≦ M) are symbol replicas that survived as candidates for the transmission signals x ₀ and x ₁ , respectively.
Alternatively, the metric calculation unit 413 uses the unitary transformation signals z ₀ and z ₁ and the symbol replica set (c _0j , c _1j ) as the metrics d _1j and d _2j as shown in the following equation, and the right side of the equation (6). The Manhattan distance with the estimated received signal obtained by inputting to the first term may be calculated.

The function Re (P) is a function that outputs the real component of the variable P, and the function Im (P) is a function that outputs the imaginary component of the variable P. Further, the metric calculation unit 413 may calculate another index representing the distance between the unitary conversion signals z ₀ and z ₁ and the surviving symbol replica pair (c _0j , c _1j ) as a metric.
Each time the metric calculation unit 413 calculates the metric once, the metric calculation unit 413 notifies the control unit 44 of an operation execution signal indicating that the metric has been calculated.

In the present embodiment, in the first stage, the cumulative value of the metric is d _1j itself. In the second stage, the cumulative value of the metric is the sum of d _1j and d _2j . Each time the metric calculation unit 413 calculates the cumulative value of the metric, the metric calculation unit 413 notifies the determination unit 414 of the total value and the corresponding symbol replica set. Each time the metric calculation unit 413 calculates the total value of the metric, the metric calculation unit 413 stores the total value and the corresponding symbol replica set in a built-in memory.

The determination unit 414 determines whether or not the cumulative value of the metric of each symbol replica set exceeds a predetermined threshold at each stage. The determination unit 414 determines whether the number of symbol replicas for which the metric has been calculated has reached the number of survivors set in advance for each stage. If the cumulative value is equal to or less than a predetermined threshold and the number of symbol replicas for which the metric has been calculated does not reach the surviving number, the determination unit 414 determines the metric for the symbol replica having the next rank with respect to the metric calculation unit 413. To calculate.
On the other hand, when the cumulative value exceeds the predetermined threshold value, the determination unit 414 causes the ranking determination unit 412 to perform symbol replica ranking for the transmission signal corresponding to the stage next to the stage of interest. If the number of symbol replicas for which a metric has been calculated for the stage of interest has reached a preset number of survivors, the determination unit 414 ends the calculation of the metric at that stage. Then, the determination unit 414 causes the ranking determination unit 412 to perform symbol replica ranking for the transmission signal corresponding to the stage next to the stage of interest.

ここでE_m,minは、現時点でのステージmにおけるメトリックの累計値の最小値である。Xは所定の係数であり、予め設定される。またσ²は、推定された雑音電力である。雑音電力σ²は、例えば次式で算出される。

ここでP_pich/dtchは、1サブキャリア当たりのパイロットチャネルの送信電力とデータチャネルの送信電力との比である。またξ_p,q,kは、送信装置２のアンテナpと受信装置３のアンテナq間のサブキャリアkに対する複素フェージング変動であり、例えば、フーリエ変換部３４から得られるパイロットチャネルとすればよい。またN_t、N_rは、それぞれ、送信信号が放射される送信装置２のアンテナ数及び送信信号を受信する受信装置３のアンテナ数である。またN_subは、サブキャリア数である。 The predetermined threshold is set for each stage based on the noise power of the received signal, for example. For example, the predetermined threshold delta _m at stage m, as described in Non-Patent Document 1, is calculated by the following equation.

Here, E _{m, min} is the minimum value of the cumulative value of the metric in stage m at the present time. X is a predetermined coefficient and is set in advance. Σ ² is the estimated noise power. The noise power σ ² is calculated by the following equation, for example.

Here, P _{pich / dtch} is a ratio between the transmission power of the pilot channel and the transmission power of the data channel per subcarrier. Further, ξ _{p, q, k} is a complex fading fluctuation with respect to the subcarrier k between the antenna p of the transmission device 2 and the antenna q of the reception device 3, and may be a pilot channel obtained from the Fourier transform unit 34, for example. N _t and N _r are the number of antennas of the transmission device 2 that radiates the transmission signal and the number of antennas of the reception device 3 that receives the transmission signal, respectively. N _sub is the number of subcarriers.

If the stage of interest is the last stage, the surviving symbol replica set is a candidate set of transmission signals. Therefore, the determination unit 414 obtains the minimum value of the cumulative value of the metrics calculated so far. Then, the determination unit 414 estimates that the set of symbol replicas corresponding to the minimum metric cumulative value is the set of transmission signals that are actually transmitted. In the present embodiment, a symbol replica set (c _0min , c _1min ) estimated to be a set (x ₀ , x ₁ ) of transmission signals actually transmitted is referred to as a maximum likelihood symbol set. Symbols (x ₀ ^(ML) , x ₁ ^(ML) ) corresponding to symbol replicas included in the maximum likelihood symbol set are referred to as maximum likelihood symbols.
The determination unit 414 outputs the minimum value of the metric cumulative value and the maximum likelihood symbol set (c _0min , c _1min ) to the log likelihood ratio calculation unit 42. The determination unit 414 also outputs to the log-likelihood ratio calculation unit 42 each symbol replica set for which the metric cumulative value has been calculated up to the final stage and the corresponding metric cumulative value.

  When the transmission signal estimation unit 41 receives a stop instruction signal from the control unit 44 before calculating the metric for the last stage, the transmission signal estimation unit 41 stops the transmission signal estimation process. In this case, the determination unit 414 outputs nothing to the log likelihood ratio calculation unit 42.

ここで、d_r,min(b_n=1)、d_r,min(b_n=0)は、それぞれ、最尤シンボルx_r ^(ML)(r=0,1)の左からn番目のビットが'1'、'0'のときのメトリックの累計値である。d_minは、最尤シンボルセット(c_0min,c_1min)について算出されたメトリックの累計値を表す。関数bit(x_r ^(ML),n)は、最尤シンボルx_r ^(ML)の左からn番目のビット値を表す。また関数invbit(x_r ^(ML),n)は、最尤シンボルx_r ^(ML)の左からn番目のビットの反転値を表す。そして関数min_a=b(d)は、a=bの条件を満たすメトリックの累計値dのうちの最小値を出力する関数である。さらに、d(c_0j ^(survived), c_1j ^(survived))(j=0,1,...,k-1)は、最終ステージまで生き残ったシンボルレプリカの組(c_0j ^(survived), c_1j ^(survived))について算出されたメトリックの累計値を表す。
また対数尤度比算出部４２は、次式に従ってビットLLR_r(n)を算出してもよい。

対数尤度比算出部４２は、推定された送信信号と、各ビットLLR_r(n)及び対応する反転ビットシンボルとをセレクタ４５へ出力する。 The log-likelihood ratio calculation unit 42 calculates a log-likelihood ratio (LLR) for each bit of the estimated transmission signal used in error correction decoding processing by soft decision. The log likelihood ratio calculation unit 42 calculates a bit LLR _r (n) that is an LLR for the nth bit from the left of the transmission signal x _r (r = 0, 1) according to the following equation.

Here, d _{r, min} (b _n = 1) and d _{r, min} (b _n = 0) are the nth bit from the left of the maximum likelihood symbol x _r ^(ML) (r = 0, 1), respectively. Is the cumulative value of the metric when is '1' or '0'. d _min represents a cumulative value of metrics calculated for the maximum likelihood symbol set (c _0min , c _1min ). The function bit (x _r ^(ML) , n) represents the nth bit value from the left of the maximum likelihood symbol x _r ^(ML) . The function invbit (x _r ^(ML) , n) represents the inverted value of the nth bit from the left of the maximum likelihood symbol x _r ^(ML) . The function min _{a = b} (d) is a function that outputs the minimum value of the cumulative values d of the metrics that satisfy the condition a = b. Furthermore, d (c _0j ^(survived) , c _1j ^(survived) ) (j = 0,1, ..., k-1) is a set of symbol replicas (c _0j ^(survived) , c _1j ^(survived) ) represents the cumulative value of the metric calculated.
In addition, the log likelihood ratio calculation unit 42 may calculate the bit LLR _r (n) according to the following equation.

The log likelihood ratio calculation unit 42 outputs the estimated transmission signal, each bit LLR _r (n), and the corresponding inverted bit symbol to the selector 45.

Note that the transmission signal estimation unit 41 calculates the metric cumulative value only for the symbol replica pair that survived to the final stage, so the metric cumulative value for the symbol replica pair including the inverted bit symbol is calculated. There may not be. In this case, the likelihood for the symbol replica including the inverted bit symbol is not directly calculated.
Therefore, in such a case, the log likelihood ratio calculation unit 42 calculates the likelihood of the symbol replica whose metric cumulative value has not been calculated, for example, as follows.
The log likelihood ratio calculation unit 42 averages the least square Euclidean distance between the received signal estimated from the set of maximum likelihood symbols and the actual received signal set for a plurality of symbols. The log-likelihood ratio calculation unit 42 sets a value obtained by multiplying the average value by a predetermined coefficient as the likelihood of a set of symbol replicas for which a cumulative metric value has not been calculated. The predetermined coefficient can be set to 3, for example.
Alternatively, the log-likelihood ratio calculation unit 42 may calculate the likelihood of a set of symbol replicas for which the cumulative value of the metric has not been calculated according to another likelihood calculation method disclosed in Non-Patent Document 2. .

The alternative value output unit 43 is an example of an output unit, and includes, for example, a nonvolatile semiconductor memory circuit, and a log likelihood ratio LLR of each bit as a predetermined value used for error correction decoding of a set of transmission signals. _r Store the preset value of (n). The LLR _r (n) of each bit has a value indicating that the reliability of the preset value of the transmission signal is low, for example, a value of about 0 or ± 0.5. When the substitute value output unit 43 receives the output instruction signal from the control unit 44, the substitute value output unit 43 outputs the preset value of LLR _r (n) of each bit to the selector 45.

  The control unit 44 includes, for example, a processor and a nonvolatile or volatile semiconductor memory circuit. And the control part 44 counts the frequency | count of the calculation of the metric calculated in the transmission signal estimation part 41 for every flame | frame. In the present embodiment, the control unit 44 updates the metric calculation count by incrementing the metric calculation count by 1 each time an arithmetic execution signal is received from the metric calculation unit 413 of the transmission signal estimation unit 41. Then, the control unit 44 stores the updated metric calculation count in a built-in memory circuit.

Further, each time the metric calculation count is updated, the control unit 44 determines whether or not the updated metric calculation count has reached a predetermined calculation amount threshold value corresponding to the allowable maximum calculation amount of the MLD calculator 40-1. To do. If the updated metric calculation count does not reach the calculation amount threshold, the control unit 44 causes the transmission signal estimation unit 41 to continue the transmission signal estimation process.
On the other hand, when the updated metric calculation count reaches the calculation amount threshold value, the control unit 44 transmits a stop instruction signal that instructs the transmission signal estimation unit 41 to stop the transmission signal estimation process. Further, the control unit 44 transmits an output instruction signal to the alternative value output unit 43. Further, the control unit 44 transmits a control signal to the selector 45 so that the preset value of each bit LLR _r (n) received from the alternative value output unit 43 is output.

The allowable maximum calculation amount of the MLD calculator 40-1 is the maximum value of the calculation amount that the MLD calculator 40-1 can finish a predetermined number of transmission signal separation processes within a predetermined time. This predetermined time is determined in advance according to, for example, the signal processing speed required of the receiving device 3. The larger the maximum allowable calculation amount, the more the MLD calculator 40-1 executes many calculations simultaneously, so the circuit scale of the MLD calculator 40-1 increases.
For example, the calculation amount when the transmission signal estimation unit 41 calculates a metric for each of the number of surviving candidate symbol replicas set in advance at each stage is the maximum calculation amount MAXA. In this case, the allowable maximum calculation amount of the MLD calculator 40-1 is the maximum calculation amount MAXA, the number ns of subcarriers on which the MLD calculator 40-1 performs transmission signal separation processing, and the coefficient α (where 0 <α < It is a value (MAXA × ns × α) multiplied by 1). Therefore, the calculation amount threshold is set to (MAXA × ns × α), for example. The coefficient α is determined in advance according to the communication environment in which the receiving device 3 is assumed to be used. When the reception quality of each subcarrier is different under a multipath environment, the amount of computation of transmission signal separation processing varies for each subcarrier. For this reason, when the delay dispersion is large, the signal power increases depending on the subcarrier, so that the transmission signal is separated even with a relatively small number of metric calculations. Therefore, the coefficient α can be set to a relatively small value (for example, 0.6 to 0.7).
On the other hand, when the delay dispersion is small, the signal power may be small for all subcarriers. In such a case, in the transmission signal estimation process, the number of symbol replicas whose metrics are calculated at each stage increases. Therefore, if the allowable maximum calculation amount is small, there is a high possibility that the MLD calculator 40-1 cannot separate transmission signals. Therefore, as the delay dispersion assumed is smaller, the coefficient α is set to a value closer to 1.

When the control signal is not received from the control unit 44, the selector 45 outputs the set of transmission signals, the inverted bit symbol, and each bit LLR _r (n) received from the log likelihood ratio calculation unit 42. On the other hand, when receiving a control signal from the control unit 44, the selector 45 outputs each bit LLR _r (n) received from the alternative value output unit 43.

5 and 6 are operation flowcharts of the transmission signal separation process executed by the MLD computing unit 40-1. This transmission signal separation process is executed for each received signal corresponding to one symbol for each subcarrier.
The QR decomposition unit 411 performs QR decomposition on the channel matrix H into a unitary matrix Q and an upper triangular matrix R (step S101). The QR decomposition unit 411 generates a unitary transformation vector z by multiplying the Hermitian conjugate Q ^H of the unitary matrix Q to the reception signal vector Y (step S102). The QR decomposition unit 411 outputs the unitary transformation vector z to the ranking determination unit 412 and the metric calculation unit 413.

The ranking determination unit 412 calculates, as an estimated value of the transmission signal of interest, a residual component obtained by canceling the component of the symbol replica set that survived from the unitary conversion signal of the reception signal corresponding to the stage of interest to the previous stage. (Step S103). Each stage corresponds to one unitary conversion signal included in the unitary conversion vector z. A unitary conversion signal having a smaller number of related transmission signals is first noted.
The ranking determination unit 412 ranks the symbol replicas in the order close to the estimated value of the transmission signal of interest by quadrant determination (step S104). Then, the ranking determination unit 412 notifies the metric calculation unit 413 of the rank of each symbol replica.

The metric calculation unit 413 selects, for the stage of interest, a symbol replica set that minimizes the metric accumulated value in the previous stage. In addition, the metric calculation unit 413 sets symbol replicas to be focused on in the rank order for the stage to be focused on (step S105). The metric calculation unit 413 calculates a metric for the symbol replica of interest. Then, the metric calculation unit 413 adds the calculated metric to the metric accumulated value obtained for the selected symbol replica pair, and updates the metric accumulated value (step S106).
The metric calculation unit 413 stores the metric accumulated value and the symbol replica set in a built-in memory, and notifies the determination unit 414 of the metric accumulated value and the symbol replica set. The metric calculation unit 413 notifies the control unit 44 of an operation execution signal indicating that the metric has been calculated.

Upon receiving the calculation execution signal, the control unit 44 updates the metric calculation count by incrementing it by 1 (step S107). Then, the control unit 44 determines whether or not the updated metric calculation count has reached a predetermined calculation amount threshold value corresponding to the allowable maximum calculation amount of the MLD calculator 40-1 (step S108).
If the updated metric calculation count has reached the calculation threshold (step S108—Yes), the control unit 44 stops the transmission signal estimation by the transmission signal estimation unit 41 (step S109). Further, the control unit 44 causes the alternative value output unit 43 to output a preset value of the log likelihood ratio of each bit (step S110). Then, the MLD computing unit 40-1 ends the transmission signal separation process.

On the other hand, when the updated metric calculation count is less than the calculation amount threshold (No in step S108), the control unit 44 continues the transmission signal estimation by the transmission signal estimation unit 41 as shown in FIG. Then, the determination unit 414 determines whether or not the cumulative metric value is greater than or equal to a threshold value determined based on the noise power of the received signal (step S111).
If the cumulative value of the metric is less than the threshold (step S111-No), the determination unit 414 determines whether the number of symbol replicas selected in rank order in the stage of interest has reached a predetermined number of survivors (step) S112). If the number of selected symbol replicas has not reached the predetermined number of survivors (No in step S112), the determination unit 414 instructs the metric calculation unit 413 to select a symbol replica of the next rank. Then, the MLD computing unit 40-1 executes the processing after step S105.

On the other hand, when the cumulative value of the metric is equal to or greater than the threshold (step S111—Yes), the determination unit 414 determines whether the stage of interest is the final stage (step S113). Alternatively, also when the number of selected symbol replicas has reached a predetermined number of survivors (step S112—Yes), the determination unit 414 determines whether or not the stage of interest is the final stage (step S113).
If the stage of interest is not the final stage (step S113—No), the determination unit 414 moves to the next stage (step S114). The determination unit 414 notifies the ranking determination unit 412 to perform the next stage process. Then, the MLD computing unit 40-1 executes the processing after step S103.
On the other hand, if the stage of interest is the final stage (step S113—Yes), the determination unit 414 obtains the minimum value of the cumulative value of metrics (step S115). Then, the determination unit 414 estimates the set of symbol replicas corresponding to the minimum value of the cumulative metric value as the set of transmission signals that are actually transmitted (step S116). The determination unit 414 outputs the minimum value of the cumulative metric value and the estimated transmission signal to the log likelihood ratio calculation unit 42.

The log likelihood ratio calculation unit 42 calculates the log likelihood ratio between the maximum likelihood symbol and the inverted bit symbol for each bit included in the symbol corresponding to the estimated transmission signal (step S117). Then, the log likelihood ratio calculation unit 42 outputs the set of estimated transmission signals, the log likelihood ratio and the inverted bit symbol used for calculating the log likelihood ratio to the decoding unit 37 via the selector 45. .
Thereafter, the MLD computing unit 40-1 ends the transmission signal separation process.

FIG. 7 shows a graph in which the block error rate by the transmission signal separation process according to the first embodiment and the block error rate by the transmission signal separation process disclosed in Non-Patent Document 1 are calculated by simulation. In this simulation, it is assumed that transmission signals are radiated from the four antennas included in the transmission device, and the reception device receives transmission signals radiated from the four antennas. The transmission signal modulation method is 16QAM, and a 6-pass exponential attenuation model is assumed as the signal propagation environment.
Further, in the transmission signal separation processing according to the first embodiment, the coefficient α for determining the threshold value that defines the calculation amount of the MLD calculator is set to 0.8.

In FIG. 7, the horizontal axis represents the SN ratio of the received signal, and the vertical axis represents the block error rate. A graph 701 represents a block error rate when the coding rate is 3/4 and the transmission signal separation process disclosed in Non-Patent Document 1 is used. A graph 702 represents a block error rate when the coding rate is 3/4 and the transmission signal separation processing according to the first embodiment is used. Further, a graph 711 represents a block error rate when the coding rate is 8/9 and the transmission signal separation process disclosed in Non-Patent Document 1 is used. A graph 712 represents the block error rate when the coding rate is 8/9 and the transmission signal separation processing according to the first embodiment is used.
As shown in FIG. 7, the block error rate according to the transmission signal separation process of the first embodiment and the non-patent document 1 are disclosed for both cases where the coding rate is 3/4 or 8/9. There is no significant difference between the block error rate by the transmission signal separation process.
Therefore, the circuit scale of the MLD computing unit that performs the transmission signal separation process according to the first embodiment is larger than the circuit scale of the computing unit that performs the transmission signal separation process disclosed in Non-Patent Document 1 by a coefficient α = 0.8. It can be seen that it can be reduced by the amount of computation equivalent to 20%.

  As described above, the MLD computing unit monitors the amount of computation when the transmission signal separation processing is executed, and stops the transmission signal separation processing when the amount of computation exceeds the maximum allowable computation amount of the MLD computing unit. Then, a log likelihood ratio set in advance is output. Therefore, the amount of computation executed by this MLD computing unit can be smaller than the maximum amount of computation by the transmission signal separation method combining the QRM-MLD method and the determination of the metric cumulative value using a threshold based on noise power. As a result, the circuit scale of the MLD computing unit can be made smaller than the circuit scale of the computing unit that realizes the transmission signal separation method that combines the QRM-MLD method and the determination of the metric cumulative value using a threshold based on noise power. Therefore, the circuit scale of the receiving apparatus according to the first embodiment can be reduced.

Next, a receiving apparatus according to the second embodiment will be described.
The receiving apparatus according to the second embodiment has MLD computing units that are smaller than the number of subcarriers. This receiving apparatus estimates the amount of calculation required for transmission signal separation processing for each subcarrier according to the signal reception state of each subcarrier. The receiving apparatus allocates subcarriers to each MLD computing unit so that variation in the amount of computation of the transmission signal separation processing by each MLD computing unit is reduced.
The receiving apparatus according to the second embodiment differs from the receiving apparatus according to the first embodiment only in the stream separation unit. Therefore, hereinafter, the stream separation unit will be described.
In this embodiment, it is assumed that the number of subcarriers multiplexed by OFDM is 8.

FIG. 8 is a schematic configuration diagram of the stream separation unit 50 of the receiving device according to the second embodiment. The stream separation unit 50 includes a scheduler 51, four MLD calculators 40-1 to 40-4, and a rearrangement unit 52.
Each of these units included in the stream separation unit 50 may be a separate arithmetic circuit. Alternatively, each of these units included in the stream separation unit 50 may be one arithmetic circuit that realizes the function of each unit.
In FIG. 8, each part of the stream separation unit 50 is assigned the same reference number as the reference number of the corresponding component of the stream separation unit 36 according to the first embodiment shown in FIG. 2. The stream separation unit 50 is different from the stream separation unit 36 according to the first embodiment in that it includes a scheduler 51 and a rearrangement unit 52.

ここでMは、送信信号が放射されるアンテナ数かつ送信信号を受信するアンテナ数である。またγ_sは、サブキャリアsにおけるSNRである。そしてHは、M×M個の要素を持つチャネル行列であり、H^Hは、チャネル行列Hのエルミート共役である。そしてIは、M×M個の要素を持つ単位行列であり、det(A)は、行列Aの行列式を表す。 The scheduler 51 determines subcarriers to be assigned to the MLD calculators 40-1 to 40-4 based on the received signal quality for each subcarrier. For this purpose, the scheduler 51 includes a reception quality measurement unit 511 and a computing unit allocation unit 512.
Reception quality measuring section 511 measures the reception signal quality of each subcarrier. In the present embodiment, the reception quality measurement unit 511 obtains the capacity for each subcarrier as the reception signal quality.
For this purpose, the received signal quality measurement unit 511, for example, based on a pilot channel signal superimposed on each subcarrier received from the Fourier transform unit 34, a signal-to-noise ratio (Signal to Noise Ratio, SNR) for each subcarrier. ) The reception quality measurement unit 511 receives the channel matrix from the channel estimation unit 35. Received signal quality measurement section 511 calculates capacity C _s (s = 1, 2,..., N _sub , where N _sub is the number of subcarriers) according to the following equation.

Here, M is the number of antennas from which the transmission signal is radiated and the number of antennas that receive the transmission signal. Γ _s is the SNR in subcarrier s. H is a channel matrix having M × M elements, and H ^H is a Hermitian conjugate of the channel matrix H. I is a unit matrix having M × M elements, and det (A) represents a determinant of the matrix A.

The received signal quality measurement unit 511 may calculate a signal power to interference plus noise ratio (SINR) for each subcarrier instead of the capacity as the received signal quality.
Received signal quality measurement section 511 measures the received signal quality of each subcarrier for each frame. Alternatively, the received signal quality measurement unit 511 may measure the received signal quality of each subcarrier at a rate of once every predetermined number of frames. The predetermined number is set to about 2 to 10, for example.
Received signal quality measurement section 511 notifies received signal quality of each subcarrier to allocation section 512.

The assigning unit 512 assigns the sub-carriers assigned to the MLD computing units 40-1 to 40-4 so that the variation in the amount of computation of the transmission signal separation processing by each MLD computing unit is reduced based on the received signal quality for each subcarrier. Determine your career.
As is clear from the equations (10) and (11), the smaller the signal power, that is, the lower the received signal quality, the greater the threshold Δm that defines the amount of transmission signal separation processing performed by the MLD calculator. Therefore, the lower the received signal quality, the higher the possibility that the amount of computation of the transmission signal separation process will increase.
Therefore, for example, the assigning unit 512 ranks the subcarriers according to the received signal quality corresponding to the subcarriers. Then, allocation section 512 determines subcarriers to be allocated to each MLD calculator so that the sum of the subcarrier ranks allocated to each MLD calculator is as equal as possible.

FIG. 9A is a diagram showing an example of received signal quality obtained for each subcarrier, and FIG. 9B is a diagram showing an example of subcarriers assigned to each MLD computing unit.
In FIG. 9A, the numerical value attached to the vertical axis is the identification number of the subcarrier, and the horizontal axis is the capacity. Bar graphs 901 to 908 represent the capacities of subcarriers 1 to 8, respectively. In this example, the capacity increases in the order of subcarriers 5, 7, 6, 3, 1, 4, 2, and 8.

In FIG. 9B, numerical values 1 to 4 attached to the vertical axis are identification numbers of the MLD computing units, and correspond to the MLD computing units 40-1 to 40-4, respectively. The horizontal axis represents the amount of calculation. The bar graphs 911 to 914 represent the amount of computation of the transmission signal separation processing in each MLD computing unit. Further, numbers 1 to 8 given in the bar graphs 911 to 914 are identification numbers of subcarriers.
In this example, the assigning unit 512 assigns the four subcarriers 5, 7, 6, and 3 to the MLD computing units 40-1 to 40-4, respectively, in descending order of capacity. Then, the assigning unit 512 assigns the remaining subcarriers to the MLD computing units 40-1 to 40-4, respectively, in descending order of capacity.
For this reason, one of the two subcarriers allocated to each of the MLD calculators 40-1 to 40-4 requires a small amount of calculation. The possibility that the amount exceeds the maximum allowable calculation amount of the MLD calculator is low.

  Allocation section 512 outputs the subcarrier channel matrix and the received signal allocated to each MLD computing unit 40-1 to 40-4. The allocation unit 512 notifies the rearrangement unit 52 of allocation information including a set of the identification number of each MLD arithmetic unit and the identification number of the subcarrier allocated to the MLD arithmetic unit.

Each MLD computing unit 40-1 to 40-4 separates the transmission signal of the subcarrier assigned by the scheduler 51. Each of the MLD computing units 40-1 to 40-4 outputs a set of subcarrier transmission signals, an inverted bit symbol for each bit, and a log likelihood ratio to the rearrangement unit 52.
In the present embodiment, the circuit scale of each of the MLD computing units 40-1 to 40-4 is such that when the two transmission signal separation processes each have a calculation amount equal to or less than the maximum allowable calculation amount, the transmission is performed twice. The signal separation process is set to be completed within a predetermined time required for the receiving device 3.

  Based on the allocation information, the rearrangement unit 52 sets the transmission signal, the inverted bit symbol, and the log likelihood ratio of each subcarrier received from each MLD computing unit 40-1 to 40-4 to be the same as the order of the transmission signals. Rearrange as follows. The rearrangement unit 52 then outputs the rearranged transmission signal set, the inverted bit symbol, and the log likelihood ratio to the decoding unit 37.

FIG. 10 is an operation flowchart of the stream separation process executed by the stream separation unit 50.
The received signal quality measurement unit 511 of the scheduler 51 measures the received signal quality of each subcarrier (step S201). Received signal quality measurement section 511 notifies received signal quality of each subcarrier to allocation section 512 of scheduler 51. The assigning unit 512 assigns subcarriers to each MLD computing unit based on the received signal quality of each subcarrier so that the variation in the computation amount of each MLD computing unit is reduced (step S202). Allocation section 512 outputs the subcarrier channel matrix and the received signal allocated to each MLD computing unit 40-1 to 40-4. The allocation unit 512 notifies the rearrangement unit 52 of allocation information including a set of the identification number of each MLD arithmetic unit and the identification number of the subcarrier allocated to the MLD arithmetic unit.

Each of the MLD computing units 40-1 to 40-4 performs transmission signal separation processing on the assigned subcarriers (step S203). Refer to FIGS. 5 and 6 for details of the operation flow of the transmission signal separation processing.
Each MLD computing unit 40-1 to 40-4 outputs a set of transmission signals for each subcarrier, an inverted bit symbol, and a log likelihood ratio to the rearrangement unit 52.
Based on the allocation information, the rearrangement unit 52 converts the transmission signal set, the inverted bit symbol, and the log likelihood ratio of each subcarrier received from each MLD computing unit 40-1 to 40-4 to the original transmission signal. Rearrangement is made to be the same as the order (step S204). The rearrangement unit 52 then outputs the rearranged transmission signal set, the inverted bit symbol, and the log likelihood ratio to the decoding unit 37.
Then, the stream separation unit 50 ends the stream separation process.

  The receiving apparatus according to the second embodiment determines subcarriers to be assigned to each MLD computing unit so that variation in the amount of computation of transmission signal separation processing is reduced. Therefore, this receiving apparatus can reduce the probability that the amount of calculation of the transmission signal separation process exceeds the maximum allowable amount of calculation of the MLD calculator.

Next, a receiving apparatus according to the third embodiment will be described.
The MLD computing unit included in the receiving apparatus according to the third embodiment has a computation smaller than the computation amount of the QRM-MLD method when the computation amount of the transmission signal separation processing by maximum likelihood estimation exceeds the allowable maximum computation amount of the MLD computation unit. The transmission signal is separated using a quantity of transmission signal separation processing.

FIG. 11 is a schematic configuration diagram of an MLD calculator included in the receiving apparatus according to the third embodiment. When the receiving apparatus has a plurality of MLD calculators, one or more arbitrary MLD calculators among the plurality of MLD calculators are MLD calculators according to the third embodiment, and other MLD calculators. The computing unit may be the MLD computing unit according to the first embodiment.
The MLD computing unit 60 according to the third embodiment includes a transmission signal estimation unit 41, a log likelihood ratio calculation unit 42, an alternative value output unit 43, a control unit 44, and a selector 45. Further, the alternative value output unit 43 includes a simple transmission signal estimation unit 46 and a second log likelihood ratio calculation unit 47.
Each of these units included in the MLD computing unit 60 may be a separate circuit. Alternatively, each unit included in the MLD computing unit 60 may be a single circuit that realizes the function of each unit.
In FIG. 11, the same reference numerals as those of the corresponding components of the MLD calculator 40-1 according to the first embodiment shown in FIG. The MLD calculator 60 is different from the MLD calculator 40-1 according to the first embodiment with respect to the configuration of the alternative value output unit 43.
Therefore, only the point regarding the alternative value output unit 43 will be described below.

In this embodiment, the control unit 44 activates the simple transmission signal estimation unit 46 of the alternative value output unit 43 when it is determined that the metric calculation count of the transmission signal separation processing has reached a predetermined threshold.
The simple transmission signal estimation unit 46 is an example of a second transmission signal estimation unit. Then, the simple transmission signal estimation unit 46 separates the transmission signal from the reception signal received from the Fourier transform unit 34 according to a transmission signal separation method having a calculation amount smaller than the calculation amount of the transmission signal separation processing by the transmission signal estimation unit 41. Note that the simple transmission signal estimation unit 46 can use, for example, the MMSE method, the Vertical-Bell Laboratories layered space-time (V-BLAST) method, or the zero forcing method as the transmission signal separation method.
The simple transmission signal estimation unit 46 outputs the obtained transmission signal set to the second log likelihood ratio calculation unit 47 of the alternative value output unit 43.

The second log-likelihood ratio calculation unit 47 determines the log likelihood between the transmission signal and the inverted bit symbol for each bit of the symbol of the received transmission signal according to the transmission signal separation method adopted by the simple transmission signal estimation unit 46. Calculate the degree ratio. Then, the second log likelihood ratio calculation unit 47 outputs the set of transmission signals, the log likelihood ratio, and the inverted bit symbol to the selector 45. For example, various known methods can be used as a log likelihood ratio calculation method when a set of transmission signals is obtained by the MMES method, the V-BLAST method, the zero forcing method, or the like.
The operation flow of the transmission signal separation process executed by the MLD computing unit 60 in this embodiment is compared with the operation flow of the transmission signal separation process according to the first embodiment shown in FIGS. 5 and 6 in step S110. Only the processing is different. In this embodiment, instead of the alternative value output unit 43 outputting a preset value of the log likelihood ratio of each bit, the alternative value output unit 43 includes a simple transmission signal estimation unit 46 and a second log likelihood ratio calculation unit. The transmission signal obtained by 47, the log likelihood ratio, etc. are output.

  In this embodiment, the threshold for the calculation amount used in the control unit 44 is the same as that in the first embodiment, corresponding to the calculation amount by the simple transmission signal estimation unit 46 and the second log likelihood ratio calculation unit 47. It is preferable that the threshold is set lower than the threshold for the calculation amount. This is because the calculation amount when the transmission signal estimation unit 41 cannot separate the transmission signals is larger by the calculation amount of the alternative value output unit 43 than the calculation amount by the MLD calculator according to the first embodiment.

  In the receiving apparatus according to the third embodiment, even when the MLD computing unit cannot separate transmission signals according to the maximum likelihood estimation method, the transmission signals are separated according to a simple transmission signal separation method. Therefore, this receiving apparatus can separate the transmission signals with a certain accuracy.

Next, a communication system according to the fourth embodiment will be described.
The receiving device included in the communication system according to the fourth embodiment obtains the degree of correlation of signals between subcarriers and notifies the transmitting device of the degree of correlation. On the other hand, if the transmission device included in the communication system determines that the communication environment has almost no frequency selective fading based on the degree of correlation, the transmission apparatus performs cyclic delay diversity (CDD) on the transmission signal. Apply. As a result, this communication system causes a variation in the signal quality of the received signal for each subcarrier, thereby reducing the possibility that the amount of computation in the MLD computing unit of the receiving device exceeds the maximum allowable computation amount of the MLD computing unit. Let

FIG. 12 is a schematic configuration diagram of a communication system 10 according to the fourth embodiment. The communication system 10 includes a transmission device 12 having two antennas 21-1 and 21-2 and a reception device 13 having two antennas 31-1 and 31-2. And the transmission apparatus 12 radiates | emits the radio signal on which the transmission signal was superimposed simultaneously from each antenna 21-1, 21-2. On the other hand, the reception device 13 receives the radio signal radiated from the transmission device 12 via the antennas 31-1 and 31-2.
In FIG. 12, the same reference numerals as those of the corresponding components of the transmitting apparatus 2 and the receiving apparatus 3 according to the first embodiment shown in FIG. It was attached. The receiving device 13 is different from the receiving device 3 according to the first embodiment in that it includes a correlation degree calculation unit 70 and a transmission unit 71. The transmission device 12 is different from the transmission device 2 according to the first embodiment in that it includes a reception unit 72.
Therefore, hereinafter, only the points relating to the correlation degree calculation unit 70, the transmission unit 71, the reception unit 72, and the application of CDD will be described.

The correlation calculation unit 70 of the receiving device 13 receives the pilot channel signal of each subcarrier from the Fourier transform unit 34 for each frame. Correlation degree calculation unit 70 then calculates the degree of correlation between signals of pilot channels in the same frame between adjacent subcarriers. Correlation degree calculation unit 70 generates a feedback signal including each correlation degree, and passes the feedback signal to transmission unit 71.
The transmitter 71 generates a radio signal by orthogonally modulating the feedback signal and then superimposing it on a carrier wave having a radio frequency. The transmitter 71 is connected to the antenna 31-2 via a high frequency switch (not shown) or a duplexer (not shown). The transmitter 71 then radiates a radio signal carrying the feedback signal via the antenna 31-2.

The transmission device 12 receives a radio signal from the reception device 13 via the antenna 21-2. The radio signal is sent to the receiving unit 72 connected to the antenna 21-2 via a high frequency switch (not shown) or a duplexer (not shown).
The receiving unit 72 superimposes a signal having an intermediate frequency on the received radio signal and converts the frequency of the received signal into a baseband signal having a baseband frequency, and then outputs the baseband signal to the control unit 26.

The control unit 26 demodulates the baseband signal received from the receiving unit 72 and extracts a feedback signal from the demodulated signal.
The control unit 26 compares each correlation level included in the feedback signal with a predetermined correlation level threshold value. When any correlation degree is higher than the correlation degree threshold, the control unit 26 determines to apply CDD to the transmission signal. Then, the control unit 26 transmits a control signal instructing the modulation unit 24 to apply CDD. On the other hand, when all the correlation degrees are below the correlation threshold, the control unit 26 determines that CDD is not applied to the transmission signal. Then, the control unit 26 transmits a control signal that instructs the modulation unit 24 not to apply CDD.
Note that the correlation degree threshold is set to, for example, the minimum value of the correlation degree when the communication path between the transmission apparatus 12 and the reception apparatus 13 is a single path and frequency selective fading does not occur.

When receiving a control signal instructing to apply CDD from the control unit 26, the modulation unit 24 sets a different cyclic shift amount for each subcarrier. In the present embodiment, it is preferable that the variation in the received signal quality for each subcarrier is large because it is possible to avoid an increase in the amount of computation of transmission signal separation processing for all subcarriers. In particular, the number of MLD calculators included in the receiving apparatus may be smaller than the number of multiplexed subcarriers. In this case, the reception device 13 adjusts the subcarriers assigned to each MLD calculator as in the second embodiment, so that the calculation amount of the transmission signal separation processing exceeds the allowable maximum calculation amount of the MLD calculator. Probability can be reduced.
Therefore, it is preferable that the modulation unit 24 sets the cyclic shift amount so that the phase difference in the frequency domain between the subcarriers becomes large. For example, the modulation unit 24 sets the cyclic shift amount for each subcarrier to e ^{−j2π (k−1) / 2} . Note that k is a subcarrier identification number (k = 1, 2,..., N _sub ). In this case, every time the subcarrier identification number k increases, the phase rotates by -180 °. Therefore, in the time domain, the component for each subcarrier of the signal of each frame is cyclically shifted by half the data length excluding the guide interval in the frame.

  The communication system according to the fourth embodiment forcibly causes frequency selective fading by applying CDD to a transmission signal when a multipath environment is not established, and makes the received signal quality different for each subcarrier. be able to. Therefore, this communication system reduces the possibility that the amount of computation in the MLD computing unit of the receiver exceeds the allowable maximum computing amount of the MLD computing unit by causing variations in the signal quality of the received signal for each subcarrier. be able to.

In each of the above-described embodiments, the metric calculation unit of the MLD computing unit determines the number of multiplications and the number of divisions included in the calculated metric instead of transmitting an operation execution signal to the control unit every time a metric is calculated. May be notified. In this case, the control unit obtains the sum of the number of multiplications and the number of divisions every time it receives the number of multiplications and the number of divisions in each transmission signal separation process. Then, when the sum exceeds a predetermined threshold, the control unit stops the transmission signal separation process. The predetermined threshold is a value obtained by multiplying the sum of the number of multiplications and the number of divisions when the metric is calculated for the number of surviving symbol replicas set at all stages by the coefficient α (0 <α <1). Is set.
Further, in each of the above embodiments, when the decoding unit of the receiving apparatus performs error correction decoding processing by hard decision, the log likelihood ratio calculation unit and the second log likelihood ratio calculation unit may be omitted.

Further, when the receiving apparatus has a plurality of MLD calculators and a scheduler, the allowable maximum calculation amount of any one of the plurality of MLD calculators is larger than the allowable maximum calculation amount of other MLD calculators. Each MLD computing unit may be designed. In this case, the scheduler assigns the subcarrier having the lowest received signal quality to an MLD calculator having a maximum allowable calculation amount larger than that of other MLD calculators. Thereby, the receiving apparatus can increase the possibility that the transmission signal can be separated by maximum likelihood estimation for all subcarriers.
Alternatively, any one of the plurality of MLD calculators may be the MLD calculator according to the third embodiment. In this case, the scheduler assigns the subcarrier having the lowest received signal quality to the MLD computing unit according to the third embodiment. Thereby, the receiving apparatus can increase the possibility that the transmission signal can be separated by the maximum likelihood estimation, and can separate the transmission signal with a certain degree of reliability even when the transmission signal cannot be separated by the maximum likelihood estimation.

  Further, by having both the constituent elements of the transmitting device 2 or 12 and the constituent elements of the receiving device 3 or 13, a communication device that can perform both transmission and reception of signals by MIMO technology is formed. In this case, the antennas 21-1 and 21-2 of the transmission device 2 or 12 and the antennas 31-1 and 31-2 of the reception device 3 or 13 are replaced with a common set of antennas. Each of the common set of antennas is connected to one of the transmission units 25-1 and 25-2 included in the transmission device 2 or 12 and the reception units 32-1 and 32-included in the reception device 3 or 13 via a duplexer. 2 is connected.

In this case, in particular, when the time division multiplexing method is adopted as the signal multiplexing method, the channel of the signal transmitted from one communication device to the other communication device, and one communication device is the other. Correlation between channels of signals received from the communication device is high. Therefore, when the correlation between the subcarriers of the signal received by one communication device from the other communication device is high, between the subcarriers included in the signal transmitted from the one communication device to the other communication device. It is estimated that the degree of correlation is also high.
Therefore, in a modification of the fourth embodiment, the control unit of one communication device calculates the correlation between adjacent subcarriers of the signal received from the other communication device, and the correlation is the correlation threshold. CDD may be applied when exceeding.

  Next, a mobile station and a base station apparatus of a mobile communication system that employs a receiving apparatus or a communication apparatus according to any of the above embodiments will be described.

  FIG. 13 is a schematic configuration diagram of a base station apparatus in which the transmission apparatus and the reception apparatus described above are incorporated. The base station apparatus 100 includes a line termination unit 101, a baseband processing unit 102, a call control unit 103, a plurality of communication units 104-1 to 104-n, and a plurality of antennas 105-1 to 105-n. . Note that n is a natural number of 2 or more. The baseband processing unit 102, the call control unit 103, and the communication units 104-1 to 104-n may be separate circuits, or each of these units is a single integrated circuit in which these circuits are integrated. It may be.

  The line termination unit 101 has a communication interface for connecting to the core network. The line termination unit 101 terminates the core network to which the higher-level device is connected. Then, the line terminating unit 101 receives a downlink signal transmitted to the mobile station apparatus from the core network, and outputs the downlink signal to the baseband processing unit 102. On the other hand, the line termination unit 101 receives the uplink signal received from the mobile station apparatus from the baseband processing unit 102, and outputs the uplink signal to the core network.

  The baseband processing unit 102 implements the functions of the codeword generation unit 22, the encoding unit 23, the modulation unit 24, and the control unit 26 of the transmission device 2 or 12 in the above embodiment. Furthermore, the baseband processing unit 102 implements the functions of the demodulation unit 33, the decoding unit 37, and the data integration unit 38 of the receiving device 3 or 13 in each of the above embodiments.

  Baseband processing section 102 determines the number of streams of each codeword, MOD, TBS, and precoding matrix based on feedback information received from the mobile station apparatus. Also, the baseband processing unit 102 divides the downlink signal received from the core network into codewords having the length of the transport size block TBS. In addition, the baseband processing unit 102 performs error correction coding processing on the codeword. Further, the baseband processing unit 102 generates a data stream by dividing the codeword encoded with the determined number of streams. Then, the baseband processing unit 102 orthogonally modulates each data stream according to the modulation mode MOD to generate a transmission signal that is a downlink signal. Furthermore, baseband processing section 102 multiplexes a set of a plurality of transmission signals according to the OFDM scheme, and outputs the multiplexed transmission signals to any of antennas 105-1 to 105-n according to a precoding matrix.

The baseband processing unit 102 receives uplink signals received from the antennas 105-1 to 105-n from the communication units 104-1 to 104-n. Then, the baseband processing unit 102 separates the transmission signal transmitted from each antenna of the mobile station apparatus from the uplink signal according to each of the above embodiments or modifications thereof. Then, the baseband processing unit 102 generates an encoded codeword by combining the separated transmission signals. The baseband processing unit 102 performs error correction decoding processing on the encoded codeword. The baseband processing unit 102 combines the decoded codewords to reproduce the original uplink signal. Then, the baseband processing unit 102 outputs the uplink signal to the core network via the line termination unit 101.
Further, the baseband processing unit 102 calculates feedback information such as a CQI value, a RANK value, and a precoding vector to be notified to the mobile station apparatus, and uses the feedback information as one of the communication units 104-1 to 104-n. It transmits to a mobile station apparatus via one.

  The call control unit 103 executes call control processing such as calling, response, disconnection, and handover between a mobile station device such as a portable terminal that communicates via the base station device 100 and the base station device 100. Then, the call control unit 103 instructs the baseband processing unit 102 to start or end the operation according to the result of the call control processing.

Each of the communication units 104-1 to 104-n includes one of the transmission units of the transmission device 2 and one of the reception units of the reception device 3 according to the above-described embodiment. The transmission units and reception units included in the communication units 104-1 to 104-n are connected to the antennas 105-1 to 105-n via duplexers (not shown), respectively. Communication units 104-1 to 104-n amplify the downlink signal received from baseband processing unit 102, and transmit the amplified downlink signal via antennas 105-1 to 105-n.
Communication units 104-1 to 104-n receive uplink signals transmitted from mobile station apparatuses via antennas 105-1 to 105-n. Then, the communication units 104-1 to 104-n amplify the received uplink signal and pass it to the baseband processing unit 102.

  FIG. 14 is a schematic configuration diagram of a mobile station apparatus in which the transmission apparatus and the reception apparatus described above are incorporated. The mobile station apparatus 200 includes a control unit 201, a baseband processing unit 202, a call control unit 203, a plurality of communication units 204-1 to 204-n, and a plurality of antennas 205-1 to 205-n. Note that n is a natural number of 2 or more. The control unit 201, the baseband processing unit 202, the call control unit 203, and the communication units 204-1 to 204-n may be separate circuits, or these units are integrated with each other. One integrated circuit may be used.

  The control unit 201 controls the entire mobile station device 200. The control unit 201 executes various application programs that operate on the mobile station apparatus 200. For this purpose, the control unit 201 includes a processor, a nonvolatile memory, and a volatile memory. When an application for communication such as telephone or data communication is activated by a user operation via an operation unit (not shown) such as a keypad included in the mobile station apparatus 200, the control unit 201 calls according to the application. The control unit 203 is operated. Then, the control unit 201 performs information source encoding processing on the data requested to be transmitted by the application or the voice signal acquired from the microphone (not shown) included in the mobile station device 200. Then, the control unit 201 passes signals obtained as a result of these processes to the baseband processing unit 202 as uplink signals. In addition, when the control unit 201 receives a downlink signal from the baseband processing unit 202, the control unit 201 acquires a speech signal or data by executing decoding processing of an information source code. And the control part 201 passes an audio | voice signal to the speaker (not shown) which the mobile station apparatus 200 has. Further, the control unit 201 displays the acquired data on a display (not shown) included in the mobile station device 200.

  The baseband processing unit 202 implements the functions of the codeword generation unit 22, the encoding unit 23, the modulation unit 24, and the control unit 26 of the transmission device 2 or 12 in the above embodiment. Furthermore, the baseband processing unit 202 implements the functions of the demodulation unit 33, the decoding unit 37, and the data integration unit 38 of the receiving device 3 or 13 in each of the above embodiments.

  Baseband processing section 202 determines the number of streams of each codeword, MOD, TBS, and precoding matrix based on feedback information received from the base station apparatus. In addition, the baseband processing unit 202 divides the uplink signal into codewords having the length of the transport size block TBS. The baseband processing unit 202 performs error correction coding processing on the codeword. Further, the baseband processing unit 202 generates a data stream by dividing the codeword encoded with the determined number of streams. Then, the baseband processing unit 202 generates a transmission signal by performing orthogonal modulation on each data stream in accordance with the modulation mode MOD. Furthermore, the baseband processing unit 202 multiplexes a plurality of sets of transmission signals according to, for example, the SC-FDM scheme. The baseband processing unit 202 outputs the multiplexed transmission signal to any of the communication units 204-1 to 204-n according to the precoding matrix.

The baseband processing unit 202 receives downlink signals from the communication units 204-1 to 204-n. Then, the baseband processing unit 202 separates the transmission signal transmitted from each antenna of the base station apparatus from the downlink signal according to each of the above embodiments or modifications thereof. Then, the baseband processing unit 202 generates an encoded codeword by combining the separated transmission signals. The baseband processing unit 202 performs error correction decoding processing on the encoded codeword. The baseband processing unit 202 combines the codewords that have been subjected to the decoding process, and restores the original downlink signal. Then, the baseband processing unit 202 outputs the downlink signal to the control unit 201.
The baseband processing unit 202 calculates feedback information such as a CQI value, a RANK value, and a precoding vector to be notified to the base station apparatus, and uses the feedback information as one of the antennas 205-1 to 205-n. To the base station apparatus via

  The call control unit 203 executes call control processing such as calling, response, disconnection, and handover between the mobile station apparatus 200 and the base station apparatus. Then, the call control unit 203 instructs the baseband processing unit 202 to start or end the operation according to the result of the call control processing.

Each of the communication units 204-1 to 204-n includes one of the transmission units of the transmission device 2 and one of the reception units of the reception device 3 according to the above-described embodiment. The transmitting unit and the receiving unit included in each of the communication units 204-1 to 204-n are connected to the antennas 205-1 to 205-n via duplexers (not shown), respectively. Then, the communication units 204-1 to 204-n amplify the uplink signal received from the baseband processing unit 202, and transmit the amplified uplink signal via the antennas 205-1 to 205-n.
The communication units 204-1 to 204-n receive downlink signals transmitted from the base station apparatuses via the antennas 205-1 to 205-n. Then, the communication units 204-1 to 204-n amplify the received downlink signal and pass it to the baseband processing unit 202.

  All examples and specific terms listed herein are intended for instructional purposes to help the reader understand the concepts contributed by the inventor to the present invention and the promotion of the technology. It should be construed that it is not limited to the construction of any example herein, such specific examples and conditions, with respect to showing the superiority and inferiority of the present invention. Although embodiments of the present invention have been described in detail, it should be understood that various changes, substitutions and modifications can be made thereto without departing from the spirit and scope of the present invention.

The following supplementary notes are further disclosed regarding the embodiment described above and its modifications.
(Appendix 1)
Multiple antennas,
A plurality of receiving units connected to any one of the plurality of antennas, each receiving a set of transmission signals transmitted from a transmitting device having the plurality of antennas via the connected antennas;
A selection unit that selects a set of reception signals corresponding to the set of transmission signals from signals received by the plurality of reception units;
An arithmetic unit that separates each transmission signal included in the transmission signal set from the reception signal set;
The computing unit is
The number of candidates for the set of transmission signals is adjusted according to the reception state of the set of received signals, and among the candidates for the set of transmission signals, the candidate with the highest posterior probability of becoming the set of received signals is selected as the candidate A transmission signal estimation unit that estimates a set of transmission signals;
An output unit for outputting a predetermined signal used for decoding the set of transmission signals;
The amount of calculation performed by the transmission signal estimation unit is counted, and the predetermined signal is output to the output unit when the calculation amount reaches a predetermined threshold before the transmission signal estimation unit finishes estimating the set of transmission signals. On the other hand, when the transmission signal estimation unit finishes estimating the set of transmission signals before the calculation amount reaches the predetermined threshold, the transmission signal estimation unit A control unit for outputting a set;
A communication device.
(Appendix 2)
The communication device according to appendix 1, wherein the predetermined threshold is an allowable maximum calculation amount for the arithmetic unit to separate the set of transmission signals within a predetermined time.
(Appendix 3)
A first predetermined number of sets of the transmission signals are multiplexed according to a predetermined multiplexing method, and the second predetermined number of the arithmetic units less than the first predetermined number are included;
The reception signal quality is obtained for each of the first predetermined number of sets of the transmission signals, and based on the reception signal quality, the variation in the amount of calculation by the second predetermined number of arithmetic units is suppressed. The communication apparatus according to appendix 1 or 2, further comprising a scheduler that assigns each set of transmission signals to any one of the second predetermined number of computing units.
(Appendix 4)
The scheduler ranks each of the sets of transmission signals according to the received signal quality, and the sum of the ranks of the sets of transmission signals assigned to each of the second predetermined number of calculators is equal. The communication apparatus according to appendix 3, wherein each set of transmission signals is assigned to any one of the second predetermined number of computing units.
(Appendix 5)
The communication apparatus according to appendix 3 or 4, wherein the predetermined multiplexing scheme is an orthogonal frequency division multiplexing scheme.
(Appendix 6)
The output unit is
A second transmission signal estimator that estimates the set of transmission signals from the set of received signals with a calculation amount smaller than the calculation amount of the transmission signal estimation unit;
The output unit outputs the set of transmission signals estimated by the second transmission signal estimation unit as the predetermined signal when the calculation amount of the transmission signal estimation unit reaches the predetermined threshold. The communication device according to any one of appendices 1 to 5.
(Appendix 7)
further,
A decoding unit for decoding the set of transmission signals and reproducing an uplink signal;
A line termination unit for outputting the uplink signal to the core network;
The communication device according to any one of appendices 1 to 6, which is a base station device having
(Appendix 8)
further,
The communication apparatus according to any one of appendices 1 to 6, which is a mobile station apparatus having a decoding unit that decodes the set of transmission signals and reproduces a downlink signal.
(Appendix 9)
A plurality of antennas, and a modulation unit that superimposes a set of a predetermined number of transmission signals on any of a plurality of subcarriers having different frequency bands, and transmitting signals included in the set of the predetermined number of transmission signals Each transmitting from a different antenna of the plurality of antennas,
A plurality of antennas, a plurality of receiving units that are connected to any one of the plurality of antennas and receive the predetermined number of sets of transmission signals via the connected antennas, and received by the plurality of receiving units A selection unit that selects the predetermined number of received signal pairs corresponding to the predetermined number of transmission signal sets from the received signals, and the transmission signal set corresponding to each of the predetermined number of received signal sets. A communication system having a receiving device having an arithmetic unit for separating each transmission signal included,
The arithmetic unit of the receiving device is:
The first transmission of the predetermined number of transmission signal sets corresponding to the first reception signal set according to the reception state of the first reception signal set of the predetermined number of reception signal sets The number of candidates for the signal set is adjusted, and among the candidates for the first set of transmission signals, the candidate with the highest posterior probability that becomes the first set of received signals is selected as the set of first transmission signals. A transmission signal estimator that estimates as
An output unit for outputting a predetermined signal used for decoding the first set of transmission signals;
The amount of calculation by the transmission signal estimation unit is counted, and when the calculation amount reaches a predetermined threshold before the transmission signal estimation unit finishes estimating the first transmission signal set, the output unit On the other hand, when the transmission signal estimation unit finishes estimating the first transmission signal set before the calculation amount reaches the predetermined threshold, the transmission signal estimation unit A controller that outputs the set of estimated transmission signals;
Have
The transmission device applies a cyclic delay diversity to at least one of the plurality of subcarriers when frequency selective fading does not occur for the predetermined number of sets of transmission signals.
(Appendix 10)
The receiving device is:
A correlation calculating unit that calculates a correlation between subcarriers adjacent to each other in a frequency band of the plurality of subcarriers;
A transmission unit that transmits the degree of correlation to the transmission device;
When the correlation degree is equal to or greater than a minimum value when no frequency selective fading occurs for the predetermined number of transmission signal sets, the frequency selective fading does not occur for the predetermined number of transmission signal sets. The communication system which judges that.
(Appendix 11)
A set of transmission signals transmitted from a transmission device having a plurality of antennas is received via each of the plurality of antennas, and each transmission included in the set of transmission signals from a set of received signals corresponding to the set of transmission signals. A communication method by a communication device for separating signals,
Adjusting the number of transmission signal set candidates according to the reception state of the reception signal set;
Among the candidates for the set of transmission signals, the candidate having the highest posterior probability that becomes the set of the reception signals is estimated as the set of transmission signals,
Count the amount of computation related to the estimation,
When the calculation amount reaches a predetermined threshold before finishing the estimation of the transmission signal set, a predetermined signal used for decoding the transmission signal set is output, while the calculation amount is a predetermined threshold value. When the estimation of the set of transmission signals is finished before reaching the value, the estimated set of transmission signals is output.
A communication method.

DESCRIPTION OF SYMBOLS 1,10 Communication system 2,12 Transmitting device 21-1, 21-2 Antenna 22 Codeword generation part 23 Encoding part 24 Modulation part 25-1, 25-2 Transmission part 26 Control part 3, 13 Receiving apparatus 31-1 , 31-2 antenna 32-1, 32-2 receiving unit 33 demodulating unit 34 Fourier transform unit 35 channel estimating unit 36, 50 stream separating unit 37 decoding unit 38 data integrating unit 40-1 to 40-4, 60 MLD computing unit 41 transmission signal estimation unit 411 QR decomposition unit 412 ranking determination unit 413 metric calculation unit 414 determination unit 42 log likelihood ratio calculation unit 43 alternative value output unit 44 control unit 45 selector 46 simple transmission signal estimation unit 47 second log likelihood ratio 47 Calculation unit 51 Scheduler 511 Reception quality measurement unit 512 Allocation unit 52 Rearrangement unit 70 Correlation degree calculation unit 71 Transmission unit 72 receiving unit 100 base station apparatus 101 line terminating unit 102 baseband processing unit 103 call control unit 104-1 to 104-n communication unit 105-1 to 105-n antenna 200 mobile station apparatus 201 control unit 202 baseband processing unit 203 Call control unit 204-1 to 204-n Communication unit 205-1 to 205-n Antenna

Claims (7)

Multiple antennas,
A first predetermined number of transmissions connected to any one of the plurality of antennas, transmitted from a transmission device having the plurality of antennas via the connected antennas , and multiplexed according to a predetermined multiplexing scheme A plurality of receivers each receiving a set of signals;
A selection unit that selects a set of received signals corresponding to each of the first predetermined number of sets of transmission signals from signals received by the plurality of receiving units;
A second predetermined number of computing units less than the first predetermined number for separating each transmission signal included in the corresponding set of transmission signals from the set of received signals ;
The reception signal quality is obtained for each of the first predetermined number of sets of the transmission signals, and based on the reception signal quality, the variation in the amount of calculation by the second predetermined number of arithmetic units is suppressed. A scheduler that assigns each of the first predetermined number of sets of transmission signals to any one of the second predetermined number of computing units ;
Each of the second predetermined number of computing units is:
Assigned to adjust the number of sets of candidates of the transmission signals allocated according to a set of reception state of the received signal, and of the set of candidates of the allocated the transmission signal, the allocated the A transmission signal estimation unit that estimates a candidate having the highest posterior probability as a set of received signals as the set of the assigned transmission signals;
An output unit for outputting a predetermined signal used for decoding the set of the assigned transmission signals;
The calculation amount by the transmission signal estimation unit is counted, and when the calculation amount reaches a predetermined threshold before the transmission signal estimation unit finishes estimating the set of the assigned transmission signals, the output unit When the predetermined signal is output, and when the transmission signal estimation unit finishes estimating the set of the allocated transmission signals before the calculation amount reaches a predetermined threshold, the transmission signal estimation unit A control unit for outputting the set of estimated transmission signals to
A communication device. Multiple antennas,
A plurality of receiving units connected to any one of the plurality of antennas, each receiving a set of transmission signals transmitted from a transmitting device having the plurality of antennas via the connected antennas;
A selection unit that selects a set of reception signals corresponding to the set of transmission signals from signals received by the plurality of reception units;
An arithmetic unit that separates each transmission signal included in the transmission signal set from the reception signal set;
The computing unit is
The number of candidates for the set of transmission signals is adjusted according to the reception state of the set of received signals, and among the candidates for the set of transmission signals, the candidate with the highest posterior probability of becoming the set of received signals is selected as the candidate A transmission signal estimation unit that estimates a set of transmission signals;
An output unit for outputting a predetermined signal used for decoding the set of transmission signals;
The amount of calculation performed by the transmission signal estimation unit is counted, and the predetermined signal is output to the output unit when the calculation amount reaches a predetermined threshold before the transmission signal estimation unit finishes estimating the set of transmission signals. On the other hand, when the transmission signal estimation unit finishes estimating the transmission signal set before the calculation amount reaches a predetermined threshold, the transmission signal estimation unit causes the transmission signal estimation unit to A control unit for outputting
Have
The output unit is
A second transmission signal estimator that estimates the set of transmission signals from the set of received signals with a calculation amount smaller than the calculation amount of the transmission signal estimation unit;
The output unit outputs the set of transmission signals estimated by the second transmission signal estimation unit as the predetermined signal when the calculation amount of the transmission signal estimation unit reaches the predetermined threshold . communication equipment. further,
A decoding unit that decodes the plurality of transmission signals and reproduces an uplink signal;
A line termination unit for outputting the uplink signal to the core network;
The communication apparatus according to claim 1 or 2 is a base station apparatus having a. A plurality of antennas, and a modulator for superimposing the one of the plurality of sub-carriers having mutually different frequency bands set the first predetermined number of transmission signals, said set of first predetermined number of transmission signal Transmitting devices that transmit each of the transmission signals included in the plurality of antennas from different antennas, and
A plurality of antennas, a plurality of receiving units connected to any one of the plurality of antennas and receiving the first predetermined number of sets of transmission signals via the connected antennas; and the plurality of receiving units from the signal received by the parts, the first selection unit for selecting a set of received signals that correspond to each set of a predetermined number of transmission signals, said set of first predetermined number of the received signal A second predetermined number of computing units that are smaller than the first predetermined number and separate the transmission signals included in the corresponding transmission signal set from each of the first predetermined number of the transmission signals. A set of the first predetermined number of the transmission signals is obtained so as to obtain reception signal quality for each of the sets, and based on the received signal quality, a variation in the amount of calculation by the second predetermined number of calculators is suppressed. Each of the second predetermined number A communication system comprising a receiver and a scheduler to assign to one of the arithmetic unit,
Each of the second predetermined number of computing units of the receiving device is:
Said set of first predetermined number of transmission signals corresponding to the set of the first of said assigned receive signal according to a set of the reception states of a predetermined number of received signal the allocated ones of the set of received signals Adjusting the number of candidates for the set of assigned transmission signals, and among the candidates for the set of assigned transmission signals, the candidate having the highest posterior probability of becoming the assigned received signal set is A transmission signal estimation unit that estimates a set of assigned transmission signals;
An output unit for outputting a predetermined signal used for decoding the set of assigned transmission signals;
The amount of calculation by the transmission signal estimation unit is counted, and when the calculation amount reaches a predetermined threshold before the transmission signal estimation unit finishes estimating the assigned transmission signal set, the output unit On the other hand, if the transmission signal estimation unit finishes estimating the set of transmission signals allocated before the calculation amount reaches the predetermined threshold, the transmission signal estimation unit A controller that outputs the set of estimated transmission signals;
Have
The communication device applies cyclic delay diversity to at least one of the plurality of subcarriers when frequency selective fading does not occur for the first predetermined number of sets of transmission signals. A first predetermined number of sets of transmission signals transmitted from a transmission apparatus having a plurality of antennas and multiplexed according to a predetermined multiplexing scheme are respectively received via a plurality of antennas, and the first predetermined number for each set of number transmission signal, a communication method by a communication device for separating the respective transmission signals included from a set of corresponding received signals to a set of the transmission signal,
The reception signal quality is obtained for each of the first predetermined number of sets of transmission signals, and the amount of calculation by the second predetermined number of arithmetic units smaller than the first predetermined number is varied based on the received signal quality. And assigning each of the first predetermined number of sets of transmission signals to any of the second predetermined number of computing units,
Each of the second predetermined number of computing units is:
Assigned to adjust the number of sets of candidates of the transmission signals allocated according to a set of reception state of the received signal,
Of the set of candidates of the allocated the transmission signal, estimates the combination become posterior probability is highest candidates of the allocated the received signal as the set of the allocated the transmission signal,
Count the amount of computation related to the estimation,
Outputting a predetermined signal used for decoding the allocated transmission signal set when the computation amount reaches a predetermined threshold before the estimation of the allocated transmission signal set is completed, When the estimation of the assigned transmission signal set is completed before the calculation amount reaches a predetermined threshold, the estimated transmission signal set is output.
A communication method. A plurality of antennas, and a modulation unit that superimposes a set of a predetermined number of transmission signals on any of a plurality of subcarriers having different frequency bands, and transmitting signals included in the set of the predetermined number of transmission signals Each transmitting from a different antenna of the plurality of antennas,
A plurality of antennas, a plurality of receiving units that are connected to any one of the plurality of antennas and receive the predetermined number of sets of transmission signals via the connected antennas, and received by the plurality of receiving units A selection unit that selects the predetermined number of received signal pairs corresponding to the predetermined number of transmission signal sets from the received signals, and the transmission signal set corresponding to each of the predetermined number of received signal sets. A communication system having a receiving device having an arithmetic unit for separating each transmission signal included,
The arithmetic unit of the receiving device is:
The first transmission of the predetermined number of transmission signal sets corresponding to the first reception signal set according to the reception state of the first reception signal set of the predetermined number of reception signal sets The number of candidates for the signal set is adjusted, and among the candidates for the first set of transmission signals, the candidate with the highest posterior probability that becomes the first set of received signals is selected as the set of first transmission signals. A transmission signal estimator that estimates as
An output unit for outputting a predetermined signal used for decoding the first set of transmission signals;
The amount of calculation by the transmission signal estimation unit is counted, and when the calculation amount reaches a predetermined threshold before the transmission signal estimation unit finishes estimating the first transmission signal set, the output unit On the other hand, when the transmission signal estimation unit finishes estimating the first transmission signal set before the calculation amount reaches the predetermined threshold, the transmission signal estimation unit A controller that outputs the set of estimated transmission signals;
Have
The transmitter applies cyclic delay diversity to at least one of the plurality of subcarriers when frequency selective fading does not occur for the predetermined number of sets of transmission signals;
The output unit of the receiving device is:
A second transmission signal estimator that estimates the set of transmission signals from the set of received signals with a calculation amount smaller than the calculation amount of the transmission signal estimation unit;
The output unit outputs the set of transmission signals estimated by the second transmission signal estimation unit as the predetermined signal when the calculation amount of the transmission signal estimation unit reaches the predetermined threshold. Communications system. A set of transmission signals transmitted from a transmission device having a plurality of antennas is received via each of the plurality of antennas, and each transmission included in the set of transmission signals from a set of received signals corresponding to the set of transmission signals. A communication method by a communication device for separating signals,
Adjusting the number of transmission signal set candidates according to the reception state of the reception signal set;
Among the candidates for the set of transmission signals, the candidate having the highest posterior probability that becomes the set of the reception signals is estimated as the set of transmission signals,
Count the amount of computation related to the estimation,
When the calculation amount reaches a predetermined threshold before finishing the estimation of the transmission signal set, a predetermined signal used for decoding the transmission signal set is output, while the calculation amount is a predetermined threshold value. When the estimation of the set of transmission signals is finished before reaching the value, the estimated set of transmission signals is output.
Including
The output of the predetermined signal means that the transmission signal is transmitted from the set of received signals with a smaller calculation amount than the calculation amount for estimating a candidate having the highest posterior probability as the set of received signals as the set of transmission signals. Estimating a set of signals and outputting the estimated set of transmission signals as the predetermined signal;
Communication method.

Images