JP5234318B2 - Automatic frequency control method and apparatus - Google Patents

Description

  The present invention relates to a radio communication system that multiplex-transmits a known signal (also referred to as a reference signal or a pilot signal) together with a data signal, and more particularly to an automatic frequency control method and apparatus on the receiving side.

  In recent years, the development of communication technology has been remarkable, and a system for communicating large amounts of data at high speed not only in wired communication but also in wireless communication is being realized. In particular, with the widespread use of mobile terminals such as mobile phones, research and development of next-generation communication methods that enable multimedia data such as moving images and voices to be used on mobile terminals are being conducted.

  What is attracting attention as a next-generation communication method is, for example, an Orthogonal Frequency Division Multiplex (OFDM) communication method employed in LTE (Long Term Evolution) of 3GPP (3rd Generation Partnership Project). The OFDM communication system is a communication system based on multi-carrier transmission, and can modulate / demodulate a plurality of carrier waves at once. However, in OFDM, if there is a frequency error in the carrier between the transmitter and the receiver, the orthogonality between the subcarriers is lost and inter-carrier interference occurs, so usually automatic frequency control that corrects the carrier frequency error. (AFC: Automatic Frequency Control) is adopted.

  In order for the AFC to function with high accuracy, it is necessary to accurately detect the frequency error of the carrier wave, and various methods have been proposed. For example, Patent Document 1 (Japanese Patent Laid-Open No. 2000-228656) proposes a method that uses a correlation based on a pilot signal (hereinafter referred to as a reference signal) that is regularly transmitted during an effective symbol period. Specifically, using a reference signal distributed in one frame, a complex correlation value between the received signal and the reference signal sequence held by the receiving device is calculated, and the phase component of the correlation peak value The frequency error is calculated based on

JP 2000-228656 A 3GPP TS 36.211 V8.0.0 (2007-09)

  However, in Patent Document 1 described above, since correlation is always performed using reference signals that are distributed and arranged in one frame and always have the same time interval, a frequency detection error occurs in an environment with a lot of noise and interference. The possibility increases and the control accuracy of AFC is lowered.

  An object of the present invention is to provide an automatic frequency control method and apparatus that enables accurate AFC control even in an environment with a lot of noise and interference, and a wireless communication apparatus using the same.

An automatic frequency control apparatus according to the present invention is an automatic frequency control apparatus in a wireless communication system that transmits a plurality of different reference signals spread in a frequency direction and / or a time direction , and one reference in one subcarrier. Correlation calculation means for calculating a correlation value using a channel estimation value by signal interpolation and a channel estimation value not by interpolation of the other reference signal, and frequency control based on a plurality of correlation values obtained by the correlation calculation means And a frequency control means for generating a signal.

  According to the present invention, accurate AFC control is possible even in an environment with a lot of noise and interference.

  Hereinafter, an automatic frequency control method and apparatus according to an embodiment of the present invention will be described in detail by taking an LTE system discussed in 3GPP as an example. In particular, the reference signal configuration of the LTE system described in 3GPP TS 36.211 (Non-Patent Document 1) is taken as an example, but the present invention is not limited to this.

1. Correlation Calculation FIG. 1 is a diagram schematically showing an arrangement example of reference signals used in one embodiment of the present invention. The vertical axis represents time, and the horizontal axis represents the frequency represented by subcarrier number k. Here, an arbitrary slot #n and a part of # (n + 1) following it are shown.

“R ₁ ” in the figure represents the arrangement of the first reference signals inserted at the beginning of each slot at intervals of 6 subcarriers. “R ₂ ” represents the arrangement of the second reference signals inserted at intervals of 6 subcarriers in the middle of each slot (after the time T ₁ from the beginning), but the frequency is 3 subcarriers with respect to the first reference signal. Arranged to be shifted in the direction. The interval T ₂ between the second reference signal in slot #n and the first reference signal in the subsequent slot # (n + 1) has a length obtained by subtracting time T ₁ from one slot time.

Assuming such a reference signal configuration, in the automatic frequency control method according to the present embodiment, subcarriers to which no reference signal is mapped use a channel estimation value interpolated from the channel estimation value of an adjacent reference signal. To perform correlation calculation. That is, when performing the interpolation process in the frequency direction, the first channel estimation value of the reference signal R ₁ of the channel estimation value T _R1 obtained by interpolation calculation using the second reference signal R ₂ channel estimates And the channel estimation value of T _R2 obtained by the interpolation calculation using. Similarly, when interpolation processing is performed in the time direction, the channel estimation value of T _R1 obtained by interpolation using the channel estimation value of the second reference signal R2 and the channel estimation value of the first reference signal R1 are obtained. The channel estimation value of _TR2 obtained by the used interpolation calculation is used.

In this way, in the same subcarrier, the correlation between the channel estimation value obtained by interpolation of one reference signal and the channel estimation value not obtained by interpolation of the other reference signal can be obtained. That is, not only one slot which is the time interval of each reference signal but also correlations at different time intervals T ₁ and T ₂ can be obtained, and a correlation value with higher reliability can be obtained.

2. Circuit Configuration Next, a circuit configuration for obtaining a frequency control signal for automatic frequency control (AFC) using the reference signal by interpolation described above will be described. Here, the signal arrangement shown in FIG. 1 is assumed.

  FIG. 2 is a block diagram showing a main part of the AFC apparatus according to the present embodiment. The AFC device 10 according to the present embodiment receives the channel estimation value of the reference signal from the channel estimation unit 11 and outputs a frequency control signal for controlling the oscillation frequency of the oscillator. The channel estimation unit 11 interpolates channel estimation values of subcarriers to which no reference signal is mapped from channel estimation values of mapped reference signals, and calculates channel estimation values of all subcarriers # 1 to #m as complex correlation calculation units. 12 is output. The complex correlation calculation unit 12 obtains a complex correlation value using these channel estimation values, and the frequency control unit 13 generates a frequency control signal based on the complex correlation value.

The channel estimation unit 11 uses the channel estimation values h _Ri (n, k) and h _Ri (n + 1, k) of the reference signal Ri mapped from the multicarrier received signal and the channel estimation value h _Rj (n, k) obtained by interpolation. k) is output to the complex correlation calculation unit 12. Here, n is a slot number and k is a subcarrier number. If attention is paid to subcarrier #k in the reference signal arrangement example shown in FIG. 1, h _Ri (n, k) and h _Ri (n + 1, k) are subcarriers in the first block of slots #n and # (n + 1), respectively. This is a channel estimation value of the reference signal R ₁ mapped to #k. H _Rj (n, k) is a channel estimation value obtained by interpolation of subcarrier #k in the fifth block of slot #n.

The complex correlation calculation unit 12 uses the channel estimation value h _Ri (n, k) of the reference signal Ri in the slot #n, the channel estimation value h _Rj (n, k) obtained by interpolation of the reference signal Rj, and the slot #. Complex correlation values C _Ri-Rj (n, k), C _Rj-Ri (n, k), and C _Ri-Ri (with the channel estimation value h _Ri (n + 1, k) of the reference signal Ri at (n + 1) At least one of n, k) is calculated. Complex correlation values C _Ri-Rj (n, k), C _Rj-Ri (n, k) and C _Ri-Ri (n, k) are calculated for all subcarriers to which one reference signal is mapped. Alternatively, a combination of complex correlation values calculated for each subcarrier may be determined.

Here, C _Ri-Rj (n, k) is a channel estimation value h _Rj (n, k) obtained by interpolation between the channel estimation value h _Ri (n, k) of the reference signal Ri in the slot #n and the reference signal Rj. k) and a complex correlation value. C _Rj−Ri (n, k) is the channel estimation value h _Rj (n, k) obtained by interpolation of the reference signal Rj and the channel estimation value h _Ri (n + 1, k) of the reference signal Ri in slot # (n + 1). ) And a complex correlation value. Further, C _Ri-Ri (n, k) is a channel estimation value h _Ri (n, k) of the reference signal Ri in the slot #n and a channel estimation value h _Ri (n + 1, n) of the reference signal Ri in the slot # (n + 1). k) and a complex correlation value.

The frequency control unit 13 is based on the complex correlation values C _Ri-Rj (n, k), C _Rj-Ri (n, k), and C _Ri-Ri (n, k) calculated over all subcarriers. Then, a frequency control signal for AFC is generated. Details will be described later.

3. As described above, since the AFC control value is obtained by using a plurality of complex correlation values between different reference signals spread in the frequency direction and / or the time direction, the frequency of the carrier wave can be obtained even in an environment with a lot of noise or interference. It is possible to suppress a decrease in accuracy of the AFC device that corrects the error.

4). Wireless Communication Device FIG. 3 is a block diagram showing a configuration of a receiving unit of a wireless communication device according to an embodiment of the present invention. A wireless communication apparatus according to this embodiment includes an antenna 101, an orthogonal detector 102, an analog / digital (A / D) converter 103, a guard interval removal unit 104, a fast Fourier transform (FFT) unit 105, and channel estimation. Section 106, channel demodulation section 107, channel decoding section 108, AFC section 109, and oscillator (TCXO: Temperature Compensated Xtal Oscillator) 110.

  In addition, the AFC unit 109 includes a complex correlation calculation unit 121, a threshold value determination unit 122, an average calculation unit 123, an Arctan calculation unit 124, a Δf conversion unit 125, a loop filter unit 126, and a TCXO control voltage calculation unit 127. These circuits may be configured by hardware, but equivalent functions can be realized by executing the respective programs on a program control processor such as a CPU.

  In FIG. 3, the RF signal received by the antenna 101 is first subjected to quadrature detection by the quadrature detector 102 in accordance with the oscillation frequency from the TCXO 110, and the received signal obtained thereby is digitally converted by the A / D converter 103. . Subsequently, the guard interval removing unit 104 removes the guard interval from the received signal and outputs it to the FFT unit 105. The FFT unit 105 converts the received signal in the time domain into the frequency domain by Fourier transform. As is well known, since the frequency domain component is converted into a time domain signal wave by inverse Fourier transform on the transmission side, the frequency component is separated from the time domain signal wave by the Fourier transform of the FFT unit 105, and the user The symbol of the data modulation signal can be extracted. Each frequency component obtained by the FFT unit 105 is used for channel estimation by the channel estimation unit 106.

  The channel estimation unit 106 estimates the influence received by the radio channel by dividing the received signal by a known reference signal. For subcarriers to which no reference signal is mapped, as described above, channel estimation values for all subcarriers can be calculated by interpolating based on channel estimation values of adjacent reference signals.

  The channel demodulation unit 107 performs demodulation using the channel estimation value, and the channel demodulated signal is decoded by the channel decoding unit 108 to obtain user data.

  The channel estimation value is input not only to the channel demodulation unit 107 but also to the AFC unit 109. The AFC unit 109 estimates a frequency error from the input channel estimation value, and outputs this to the oscillator 110 as a frequency control signal. The TCXO 110 corrects the oscillation frequency according to the frequency control signal and maintains the radio frequency received by the quadrature detector 102.

5. AFC operation Normally, the basic operation of AFC is as follows. A complex correlation calculation unit calculates a complex correlation value between the channel estimation values. The Arctan calculation unit converts the complex correlation value into a phase change amount, and the Δf conversion unit converts the phase change amount into a frequency error Δf. Further, the frequency error Δf is averaged over a plurality of slots by the loop filter, and the TCXO control voltage calculation unit calculates a correction value that makes the frequency error zero, and the control voltage corresponding to the correction value is frequency controlled. The signal is output to the TCXO 110 as a signal to adjust the oscillation frequency.

  On the other hand, the AFC unit 109 according to the present embodiment obtains an AFC control value using a complex correlation value between different reference signals spread in the frequency direction and the time direction. Further, the threshold value determination unit 122 provides a level threshold value and / or a phase difference threshold value, selects only effective correlation values and outputs them to the average calculation unit 123. Correlation values with low reliability are averaged by the average calculation unit 123. Do not include in the calculation. Thereby, an accurate AFC control value can be obtained even in an environment where the signal-to-noise ratio SNR is poor. Hereinafter, the complex correlation calculation unit 121, the threshold determination unit 122, and the average calculation unit 123, which are characteristic configurations in the present embodiment, will be described in detail.

First, the complex correlation calculation unit 121 interpolates the channel estimation value h _R1 (n, k) of the first reference signal R ₁ and the interpolation for the subcarrier #k to which the first reference signal R ₁ of the slot #n is mapped. The complex correlation value C _R1-R2 (n, k) with the channel estimation value h _R2 (n, k) of the second reference signal R ₂ obtained by the above is calculated by the following equation (1).

Similarly, the channel estimation value h _R2 (n, k) of the second reference signal R ₂ obtained by interpolation and the channel estimation value h _R1 (n +) of the first reference signal R _{1 in} the next slot # (n + 1). 1, k) complex correlation value C _R2-R1 with (n, k), and the first reference of the first reference signal R ₁ of the channel estimate h _R1 (n, k) and the next slot # (n + 1) A complex correlation value C _R1-R1 (n, k) with the channel estimation value h _R1 (n + 1, k) of the signal R ₁ is calculated. At this time, C _R2-R1 (n, k) and C _R1-R1 (n, k) are different from C _R1-R2 (n, k) because the time interval between correlated channel estimates is C _{R1 Corrections} are made by the following equations (2) and (3) so that complex correlation values are obtained at the same time interval as _-R2 (n, k).

In the equations (2) and (3), T ₁ is the time between the first reference signal R ₁ and the second reference signal R ₂ as described above, and T ₂ is the second reference signal R ₂ and the next time. Is the time between the first reference signal R ₁ of the slots.

Next, in order to reduce the influence of noise, the three complex correlation values C _R1-R2 (n, k), C _R2-R1 (n, k), C _R1 obtained by the equations (1) to (3) are used. _At least one of _−R1 (n, k) is subjected to threshold determination regarding the level and / or phase difference by the threshold determination unit 122 and then averaged by the average calculation unit 123. The threshold determination process of the threshold determination unit 122 is executed in order to increase the reliability of the finally obtained complex correlation value (details will be described later).

The average calculation unit 123 performs the above operation for all the subcarriers to which the first reference signal R ₁ is mapped, and calculates the average between the subcarriers as an overall complex correlation value C. The Arctan calculation unit To 124.

  The Arctan calculation unit 124 converts the complex correlation value C into a phase change amount, and the Δf conversion unit 125 converts it into a frequency error Δf. Further, the frequency error Δf is averaged over a plurality of slots by the loop filter 126 and output to the TCXO control voltage calculation unit 127. The TCXO control voltage calculation unit 127 calculates a correction value that sets the frequency error Δf to 0, and generates control voltage data corresponding to the correction value. The control voltage data is analog-converted by a digital / analog converter (not shown) and output to the TCXO 110 as a frequency control signal. The oscillation frequency thus controlled in frequency is output from the TCXO 110 to the quadrature detector 102.

6). Threshold Determination FIG. 4A is a complex plan view for explaining the level threshold determination of the threshold determination unit in the present embodiment, and FIG. 4B is the level and phase difference threshold determination of the threshold determination unit in the present embodiment. It is a complex top view for demonstrating.

  As shown in FIG. 4A, the threshold value determination unit 122 is provided with a level threshold value in advance, and the average calculation unit 123 determines that the complex correlation value level calculated by the complex correlation calculation unit 121 is lower than the level threshold value. Exclude from processing. For example, if the complex correlation calculation unit 121 calculates three complex correlation values C1, C2, and C3, and C1 and C2 are higher than the level threshold value and only C3 is low, the correlation value C3 having a lower level is converted to noise or interference. It is regarded as a correlation value with low reliability that has been degraded by the influence of the above, etc., and is not included in the average. As described above, only the correlation values C1 and C2 having high levels can be selected and used for the averaging process, and a highly reliable AFC control value can be obtained even in an environment where the SNR is poor.

  In addition, the method of determining whether or not the correlation value has a low reliability is not limited to the level threshold determination, but can also determine that the reliability is low even if the phase component is significantly different from other correlation values. Therefore, as shown in FIG. 4B, it is more desirable to use the level threshold value and the phase difference threshold value as threshold judgment criteria.

  As shown in FIG. 4B, the threshold value determination unit 122 is provided with a level threshold value and a phase difference threshold value in advance, and the level of the complex correlation value calculated by the complex correlation calculation unit 121 is lower than the level threshold value. Alternatively, those whose phase is away from the phase difference threshold range determined by another correlation value and the phase difference threshold are excluded from the processing of the average calculation unit 123. For example, the complex correlation calculation unit 121 calculates three complex correlation values C1, C2, and C3, all of which are higher than the level threshold, but only C3 is separated from the phase difference threshold range including the phases of C1 and C2. Then, this correlation value C3 is regarded as a correlation value with low reliability that has deteriorated due to the influence of noise or interference, and is not included in the average. As described above, only the correlation values C1 and C2 having a high level and within the predetermined phase difference threshold range are selected and used for the averaging process, thereby obtaining a highly reliable AFC control value even in an environment with a poor SNR. be able to.

  Note that the threshold determination unit 122 may use other methods as long as it is a selection unit that can select only effective correlation values by excluding those that are considered to have low reliability from a plurality of correlation values.

7). Combination of Complex Correlation Calculations As described above, the complex correlation calculation unit 121 performs three types of complex correlation values C _R1-R2 (n, k) and C _R2-R1 (n, k) according to the equations (1) to (3). ), C _R1-R1 (n, k) can be calculated, and these three types of complex correlation values can be obtained for all reference signals, and the average can be used as the average complex correlation value of each subcarrier. .

FIG. 5 is a reference signal arrangement diagram for explaining a first example of a combination of complex correlation operations in the present embodiment. Here, with respect to the first reference signal channel estimate h _R1 of R ₁ mapped for each channel estimate h _R1 3 kinds of complex correlation values _{C R1-R2 (n, k} ), C R2-R1 (n , k) and C _R1-R1 (n, k) are calculated to obtain an average.

  Other combinations of complex correlation operations are possible. For example, instead of using all the complex correlation values, one (or several) of three types of complex correlation values is selected for each subcarrier, and the average is used as the average complex correlation value of the subcarrier. Also good.

FIG. 6 is a reference signal arrangement diagram for explaining a second example of the combination of complex correlation operations in the present embodiment. Here, with respect to the first reference signal channel estimate h _R1 of R ₁ mapped for each channel estimate h _R1 3 kinds of complex correlation values _{C R1-R2 (n, k} ), C R2-R1 (n , k) and C _R1-R1 (n, k), only C _R1-R2 (n, k) and C _R1-R1 (n, k) are alternately calculated to obtain an average. However, it is not limited to this pattern, Arbitrary combinations can be used.

In this embodiment, three types of complex correlation values C _R1-R2 (n, k), C _R2-R1 (n, k), and C _R1-R1 (n, k) are used. It is not limited. The number may be less than three or more. For example, the second reference signal channel estimate _{R 2 h R2 (n, k} ) and the complex with the next slot # (n + 1) of the second reference signal channel estimate _{R 2 h R2 (n + 1} , k) Correlation values C _R2-R2 (n, k) and the like can be added as other combinations.

Furthermore, in the above-described embodiment, the complex correlation value is calculated by paying attention only to the subcarrier to which the first reference signal R ₁ is mapped. However, the present invention is not necessarily limited to this. The subcarriers to which the second reference signal R ₂ is mapped may be noted, or complex correlation values may be calculated for all subcarriers using values interpolated in the frequency direction and included in the average.

8). Effect According to the present embodiment, the AFC control value is obtained by using a plurality of complex correlation values between different reference signals spread in the frequency direction and the time direction. A decrease in accuracy of the AFC circuit that corrects the frequency error can be suppressed.

  In addition, a complex correlation value level threshold and / or phase difference threshold is provided, and a correlation value having a low level or a correlation value whose phase is far from that of other correlation values that are affected by noise, interference, or the like is low in reliability. Therefore, the accuracy of control by the AFC circuit can be further increased by not including it in the average.

  In the above embodiment, the SISO (Single Input Single Output) system has been described. However, the present invention is not limited to this. It can also be applied to systems using multiple antennas such as MIMO (Multi Input Multi Output), receive diversity, and transmit diversity. In this case, after obtaining the correlation value for each branch, an average between the branches may be taken.

  Further, the configuration of the reference signal targeted by the present invention is not limited to that shown in FIGS. Similarly, in other reference signal configurations, it is only necessary to obtain the AFC control value using a plurality of complex correlation values between a plurality of reference signals having different time differences. Similarly, the LTE discussed in 3GPP has been described as an example, but the present invention is not necessarily limited thereto. The present invention can also be used in systems using other OFDM transmission systems and automatic frequency control apparatuses in other wireless communication systems.

  The present invention is applicable to an automatic frequency control device of an information terminal such as a mobile phone, PHS (Personal Handyphone System), PDA (Personal Data Assistance, Personal Digital Assistants).

It is a figure which shows typically the example of arrangement | positioning of the reference signal used in one Embodiment of this invention. It is a block diagram which shows the principal part of the AFC apparatus by this embodiment. It is a block diagram which shows the structure of the receiver of the radio | wireless communication apparatus by one Example of this invention. (A) is a complex plan view for explaining the level threshold judgment of the threshold judgment unit in the present embodiment, and (B) is a complex for explaining the level and phase difference threshold judgment of the threshold judgment unit in the present embodiment. It is a top view. It is a reference signal arrangement | positioning figure for demonstrating the 1st example of the combination of the complex correlation calculation in a present Example. It is a reference signal arrangement | positioning figure for demonstrating the 2nd example of the combination of the complex correlation calculation in a present Example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Automatic frequency control (AFC) apparatus 11 Channel estimation part 12 Complex correlation calculation part 13 Frequency control part 121 Complex correlation calculation part 122 Threshold determination part 123 Average calculation part 124 Arctan calculation part 125 Δf conversion part 126 Loop filter 127 TCXO control voltage calculation Part

Claims (11)

An automatic frequency control apparatus in a radio communication system of a system for transmitting a plurality of different reference signals spread in a frequency direction and / or a time direction,
Correlation calculation means for calculating a correlation value using a channel estimation value by interpolation of one reference signal and a channel estimation value not by interpolation of the other reference signal in the same subcarrier ;
Frequency control means for generating a frequency control signal based on a plurality of correlation values obtained by the correlation calculation means;
An automatic frequency control device comprising:   The frequency control means includes selection means for selecting an effective correlation value from the plurality of correlation values with a predetermined threshold as a reference, and generates the frequency control signal based on the effective correlation value. The automatic frequency control apparatus according to claim 1.   The automatic frequency control device according to claim 1, wherein the frequency control unit generates the frequency control signal based on an average of the correlation values. An automatic frequency control method in a wireless communication system that transmits a plurality of different reference signals spread in a frequency direction and / or a time direction,
In the same subcarrier, a correlation value is calculated using a channel estimation value by interpolation of one reference signal and a channel estimation value not by interpolation of the other reference signal ,
Generate a frequency control signal based on the calculated correlation values,
Controlling the reception frequency according to the frequency control signal ;
An automatic frequency control method characterized by the above. Selecting an effective correlation value based on a predetermined threshold value from the plurality of correlation values;
Generating the frequency control signal based on the effective correlation value;
The automatic frequency control method according to claim 4.   6. The automatic frequency control method according to claim 4, wherein the frequency control signal is generated based on an average of the correlation values. A wireless communication apparatus in a multicarrier transmission system of a system for transmitting a plurality of different reference signals spread in a frequency direction and / or a time direction,
Oscillation means for generating an oscillation frequency for reception;
Channel estimation means for estimating a channel estimation value of each subcarrier based on the received reference signal;
Correlation calculation means for calculating a correlation value using a channel estimation value by interpolation of one reference signal and a channel estimation value not by interpolation of the other reference signal in the same subcarrier ;
A frequency control means for generating a frequency control signal based on a plurality of correlation values and controlling the oscillation frequency of the oscillation means;
A wireless communication apparatus comprising:   The frequency control means includes selection means for selecting an effective correlation value from the plurality of correlation values with a predetermined threshold as a reference, and generates the frequency control signal based on the effective correlation value. The wireless communication apparatus according to claim 7.   The radio communication apparatus according to claim 7 or 8, wherein the frequency control means generates the frequency control signal based on an average of the correlation values.   A wireless communication system comprising the wireless communication device according to claim 7. An automatic frequency control apparatus in a wireless communication system of a system for transmitting a plurality of different reference signals spread in a frequency direction and / or a time direction, comprising:
In the same subcarrier, a correlation value is calculated using a channel estimation value obtained by interpolation of one reference signal and a channel estimation value not obtained by interpolation of the other reference signal ,
Generating a frequency control signal based on a plurality of correlation values;
A program characterized by functioning as follows.

Images