JP3246646B2 - Code division multiple access communication receiver - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a code division multiple access communication (CDMA) used between a base station and a mobile station in a mobile communication so as to obtain a path diversity effect. Related to a receiving device.

[0002]

2. Description of the Related Art FIG. 6 shows a conventional code division multiple access communication system (hereinafter referred to as a spread spectrum communication system).
The spread spectrum transmitter 404 performs, for example, BPSK modulation on the information signal, directly performs spread spectrum on the modulated output signal using a spread code, converts the spread output signal into a high frequency band signal, and transmits the signal from the antenna 401. The transmission signal is received by the two antennas 402 and 403 of the spread spectrum receiver 400, and these reception signals are converted to signals of an intermediate frequency band as necessary, though not shown, and then are correlated by the correlators 405 and 407. Each is despread by the same spreading code as that of the transmitter 404. As these correlators 405 and 407, a matched filter or a sliding correlator is used.

The outputs of the correlators 405 and 407 are detected by detectors 406 and 408, respectively, and the rake (RAK) is detected.
E) Supply to receiver 412. As the detectors 406 and 408, a synchronous detector, a delay detector and the like are used. RA
In the KE receiver 412, the respective detection outputs of the detectors 406 and 408 are supplied to delay lines 410 and 411 with taps, respectively, and are sequentially delayed from each tap by one chip period Tc of the spread code of the correlators 405 and 407. Delayed signals are obtained, and these delayed signals are combined by the combining circuit 409 and output. Hereinafter, the number m of taps of each of the delay lines 410 and 411 with taps is referred to as a window width m. The above-mentioned combination includes maximum ratio combining in which weighting is performed in proportion to the amplitude level of the delay signal and combining, and equal gain combining in which weighting is not performed (weights are all set to 1). When the correlators 405 and 407 are matched filters, the delay time between taps thereof is made equal to the delay time between taps of the delay lines 410 and 411.

[0004] Not all of the delayed wave signals received by the antennas 402 and 403 are input to the synthesis circuit 412, but signals of a number corresponding to the number of taps of the delay lines 410 and 411 with taps are input in the order of reception input. Is done. For example, when the number of taps of the tapped delay lines 410 and 411 in the signals received by the antennas 402 and 403 are each 6 taps, the combining circuit 412 outputs, as shown in FIG. Correlation detection levels a _{1 to} a ₆ and correlation detection levels b ₁ to b during despread detection output
₆ signals (6 signals in total) are input.

[0005] As described above, a statistically independent delayed wave (hereinafter, referred to as a path) subjected to fading due to multiple wave propagation.
Is received by a plurality of antennas (hereinafter, referred to as branches), and the number of paths to be combined is increased to improve communication quality. It is effective to improve the communication quality by using the path diversity and the space diversity together.

[0006]

However, there has been a case where the communication quality cannot be always obtained as expected by the combined use of the path diversity and the space diversity. In order to elucidate this point, various experimental studies were performed as described below. FIG. 8 is a diagram showing a delay wave characteristic curve (shape of a group of delay waves) when the delay time difference of the delay wave with respect to the main wave is plotted on the horizontal axis and the relative level of the delay wave with respect to the main wave is plotted on the vertical axis. is there. In an actual multi-wave propagation environment, an incoming delayed wave has a spread as shown in FIG. 8, and the shape of this characteristic curve is an exponential function type. A delay spread τ is a quantity that quantitatively represents the spread of a delayed wave in multipath propagation.
s Is used. Delay spread τs Is the square root of the second moment around the average propagation delay time difference, where each incoming radio wave is weighted by the received power, and is expressed by the following equation. (Reference: descriptive personal communication technology Ohm p.70) τs = √ (Σ ( τi-τ A) 2 P (τi) / ΣP (τi)) τ A = ΣτiP (τi) / ΣP (τi) (1)ま で is from i = 1 to N, where τi: the propagation delay time of the i-th delayed wave, P (τ
i): Power of the i-th delayed wave.

When the shape of the spread of the delayed wave is an exponential function type, the level decreases as the delay time difference from the main wave increases. If the power of the signal component of the delayed wave is expressed as S and the thermal noise power of the receiver is expressed as N, the average S / N of the main wave is maximum since N is constant, and the delayed wave S having a large delay time difference with respect to the main wave is obtained. Is reduced, the average S / N of the delayed wave is reduced. Furthermore, in mobile communication, signal power fluctuates over several tens of dB due to fading, so that signal power and receiver noise power may be reversed.
FIG. 9A shows the error rate characteristics with respect to the reception level when the delay spread τs is constant and the window width m is a parameter. It can be seen that the error rate characteristics change depending on the window width m. FIG. 9B is a diagram in which the minimum reception level for obtaining the required quality in FIG. 9A is obtained by using the window width m as a parameter. There is a window width m _W that minimizes this minimum reception level. When the delay spread τs is small, the optimum value of the window width m is small, and when the delay spread τs is large, the optimum value of the window width m is large. The reason for the existence of the optimum value is that if the window width m is too large, the noise power is combined to the correlator output when the noise power is larger than the signal power, and the communication quality deteriorates. This is because communication quality deteriorates because effective delayed wave signal power cannot be synthesized. The conventional spread spectrum communication system has a problem that a maximum diversity effect cannot be expected in response to various propagation characteristics, particularly delay characteristics, because the window width is constant.

Conventionally, a delay signal from a tap position for a continuous window width m of the delay line has been supplied to the synthesis circuit. For this reason, for example, when the window width m = 3 and the signal shown in FIG. 7 is input, the signals a ₁ , a ₂ , a ₃ and the signals b ₁ , b ₂ , b ₃ are conventionally supplied to the combining circuit. Is done. However, a person from a signal a ₂ of a ₄ large level, a ₄ is not available, conventionally not effectively utilizing the input signal at this point, could not sufficiently expected much diversity effect. In order to improve this point, conventionally, the number of delayed wave signals supplied to the synthesis circuit
When the number of delay wave signals supplied from 0 and 411 is 4, that is, when the window width m = 2, the window width m is set to, for example, 3, and the delay line 41
0,411, a total of six delayed wave signals a ₁ -a
In some cases, four are supplied in order of increasing level from a ₃ , b _{1 to} b ₃ , and in this example, b ₁ , b ₂ , a ₁ , and b ₃ are supplied to the synthesis circuit. However, even in this case, since the number of delayed wave signals supplied to the synthesis circuit is constant, if the delay spread τs is small, for example, if the number of supplied delayed wave signals is too large, the communication quality may deteriorate. Will be easily understood.

SUMMARY OF THE INVENTION In view of the above problems, it is an object of the present invention to provide a receiver capable of detecting a spread of a delay wave and optimally controlling a window width in accordance with the detected value to effectively extract a diversity effect. And The present invention further provides a receiver capable of optimally controlling the window width according to the spread of the delay wave and extracting and synthesizing a signal having a large reception level irrespective of the order of reception input to effectively extract the diversity effect. The purpose is to:

[0010]

According to the first aspect of the present invention, signals received from n (n is an integer of 1 or more) antennas are despread by n correlators using the same spreading code. The n despread outputs are detected by n detectors, respectively, and the n detected outputs are sequentially delayed by one chip period of the spreading code by n delay means (m is 2 In the code division multiple access communication receiving apparatus that obtains the delayed signals of the above (integer) and combines the delayed signals by the combining circuit, at least one of the n correlators is sequentially shifted by at least one 1-chip cycle (k Level detecting means for detecting the level of an output satisfying ≧ m), calculating means for calculating the spread of the delayed wave using the k levels detected by the level detecting means, and calculating means for calculating the spread of the delayed wave. Depending on the value of the spread of the delayed wave ; And a window width changing means for changing the number m of delayed signals taken out from the n-number of each delay unit.

According to the second aspect of the present invention, received signals from n (n is an integer of 1 or more) antennas are detected by n detectors, and the n detected outputs are converted to n correlators. Are despread with the same spreading code, and m (m is an integer of 2 or more) delayed signals are obtained by sequentially delaying the n despread outputs by one chip period of the spreading code by n delay means. In a code division multiple access communication receiver for synthesizing these delayed signals by a synthesizing circuit, at least one of the n correlators is sequentially shifted by at least one chip period and output k (an integer satisfying k ≧ m). Level detecting means for detecting the level of the delay wave, calculating means for calculating the spread of the delayed wave based on the detection level from the level detecting means, and n delays are calculated by the value of the spread of the delayed wave calculated by the calculating means. Of the delayed signal extracted from the means ; And a window width changing means for changing the m.

According to the third aspect of the present invention, the first or second aspect is provided.
In the invention, j signals (integers satisfying 2 ≦ j ≦ n × m) are selected from the m delay signals from the n delay means, that is, n × m delay signals in descending order of their level. Selection means for supplying the data to the synthesis circuit.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows an embodiment of the present invention in which the number of antennas is n = 2 and the number of delay line taps is m = 6. It is attached. In the following description, it is assumed that the output number k extracted from the correlators 405 and 407 for level detection is equal to the tap number m of the delay lines 410 and 411. Generally, k ≧ m. The signals received by the antennas 402 and 403 are BPF
(Bandpass filter) Correlator 40 through 103 and 104
5,407. The outputs of the correlators 405 and 407 are output from the A / D converters 107 and 10 of the RAKE receiving circuit 412.
8 are sampled and converted into digital values at every one-chip cycle Tc. That is, the levels of the outputs of the correlators 405 and 407 which are sequentially shifted by one-chip cycle are detected. Is stored in the memory 123. 6 taps t11, t12 of the tapped delay line 410 in memory 123, ... correlation detection level a ₆ of the delay signals corresponding to t16, a ₅ ... 6 taps of a ₁ and tapped delay line 411 t21, t22 ,…
t26 to the correlation detection level b ₆ of the delay signals obtained _{respectively,} b _5, ... b ₁ are stored. The delay spread calculation unit 112 calculates the delay spread τs from the values of these detection levels by using the equation (1). The delay time τi in equation (1) corresponds to the delay time between taps of the tapped delay lines 410 and 411 (chip cycle: an integral multiple of Tc).
The delay spread τs can be directly obtained by using the equation (1). However, as shown in FIG. 2A, when the shape of the group of the delay waves is an exponential function type, the level of the delay wave is the delay time τ
It decreases at a constant slope with i. Assuming that this slope is q, q = − (10 iTc /
τs) log (e) (dB / μs) (i: integer, Tc:
The chip period, iTc = 1 μs, and the unit of τs is given by μs). Using the plurality of levels stored in the memory 123, the slope q is obtained by the least squares approximation method. From this, the inverse calculation is performed to calculate the delay spread τs. In any method for determining the delay spread τs, the A / D
It is only necessary to use the level detected by one of the converters 107, 108. The spread of the delayed waves reaching the antennas 402 and 403 is almost equal. The delay spread τs may be calculated based on the detection levels obtained from both the A / D converters 107 and 108, and the average thereof may be obtained. According to the delay spread τs obtained in this way, the memory 1
The delay spread τ obtained from the relation table between the delay spread τs and the optimum window width m shown in FIG.
Determine the optimal window width m for s. This result is input to the switch selection means 114, and the delay line 410 with a tap and
Switches S11 to S11 connected in series with each of the taps 411
S16 and S21 to S26 are turned on.
That is, when the obtained optimum window width m is, for example, 3, FIG.
, Taps t14 to t1 for the branch 1
6, the signal at branch t2 for tap t24
Switch S1 so that the signal from t26 to t26 is output.
4 to S16 and S24 to S26 are turned on, and the switch S1 is turned on.
1 to S13 and S21 to S23 are OFF. That is, assuming that the obtained optimum window width is m, the signal at taps t (16− (m−1)) to t16 for branch 1 and the taps t (26− (m−1)) to t2 for branch 2
6 so that the switch S (16-
(M-1)) to S16, S (26- (m-1)) to S2
6 is turned ON. The number of taps (positions) of the window width m corresponding to the delay spread thus obtained is extracted and supplied to the synthesis circuit 409. At this time, taps t11 and t2 are set so that the maximum ratio combination of the detection signals is performed at the timing.
Before 1, timing adjustment delay units 117 and 118 for adjusting the time by the arithmetic processing are provided. FIG. 2
In the table B, the relationship shown in FIG. 9 is obtained for each delay spread value, and the result is created in advance.

Delay line 410 with tap for branch 1
The delayed signal output from taps t11 to t16 of A is
The weighting factors W11 to W16 for the delayed signals calculated by the maximum ratio combining weighting factor calculation means 113 based on the levels calculated by the / D converter 107 are multiplied by the multipliers M11 to M16, respectively. However, since the optimum window width m is determined, multipliers M (16− (m−1)) to M10 are used. The delay signal output from the taps t21 to t26 of the tapped delay line 411 for the branch 2 is A /
Based on the level calculated by the D converter 108, the weighting coefficients W21 to W26 for the delayed signal calculated by the maximum ratio combining weighting coefficient calculating means 113 are multiplied by the multipliers M21 to M26, respectively. However, since the optimum window width m is determined, multipliers M (26- (m-1)) to M20 are used. The signal multiplied by the weight for maximum ratio combining is input to the combining circuit 409 and combined to output a demodulated signal.

The above principle is based on n antennas × delay line taps.
J (multiple paths satisfying 2 ≦ j ≦ n × m)
The same applies to the case of selecting the path of (number). sand
That is, for example, when n = 2, the optimum window width m is set as described above.
After the determination, the detection level a in the memory 123₁~ A
m, b₁J are selected in ascending order from ~ bm, and j
Are stored in the memory 125. Figure
Reference numeral 4 denotes a case where j = 4 and m = 3. In this case m
= 3, taps t14 to t16 and t24 to t26 are selected
But the delay signal a from these taps₁~ A_Three, B₁
~ B_ThreeThe level is b ₁> A₁> B_Two> B_Three> A_Three> A_Two
, B ₁, A₁,
b_Two, B_ThreeAre selected and combined at the maximum ratio and output.
It is.

As shown by a dotted line in FIG.
407 are output to the A / D converters 107 and 1 respectively.
08 to perform level detection. Further, depending on the processing procedure on the transmission side, as shown in FIG.
02, 403 to the band-pass filter 10
3 and 104, respectively, and first the detectors 406 and 408
And detects the signal in the intermediate frequency band.
The baseband signal may be obtained by despreading in steps 5 and 407, respectively, and supplied to the tapped delay lines 410 and 411. In this case, the data signal is spread on the transmitting side, and the spread signal is used to receive a signal output by, for example, BPSK modulating a carrier signal with the spread output. The level detection by the A / D converters 107 and 108 is performed by the correlators 405 and
407 output. In addition, parts corresponding to those in FIG. 1 are denoted by the same reference numerals, and redundant description is omitted.

In the above description, a BPSK modulation signal is assumed for the sake of simplicity, but another modulation signal, for example, a QPSK modulation signal may be used. Also in this case, the detector 406,
A synchronous detector or a delay detector is used as 408. Both are quadrature detectors, and two synchronous component outputs (I output) and quadrature component outputs (Q output) are obtained. , A delay line with a tap is used. That is, there are two systems, I and Q, respectively. When despreading is performed after detection, correlators 405 and 407 are also provided with two systems of I and Q, respectively. If different spreading codes are used on the transmitting side for the I and Q components, the receiving apparatus also uses I and Q of the detection output.
The Q component is despread with the same different spreading code as the transmitting side. The different sets of spreading codes are the same for each antenna system (branch). Furthermore, in the above description, a delay line with taps is used to obtain a delay signal sequentially shifted by one chip cycle. However, other means such as using a memory and delaying reading by one chip cycle for writing may be used. . Further, the number n of antennas is not limited to two, and may be three or more. As n increases, both effects of space diversity and path diversity increase. At n−1, by changing the number of window widths m, that is, the number of paths according to the spread of the delay wave, the minimum reception level for obtaining the required quality changes, and an appropriate window width m may be set.
The number j of the delay signals extracted from the n × m delay signals and supplied to the combining circuit 409 may be two or more.

In FIG. 1, switches S11 to S16,
When omitting steps S21 to S26 and turning off the switch,
The weight of the serial multiplier may be set to 0. Further, the present invention may be applied not only to the maximum ratio combining but also to equal gain combining. In this case, the weight coefficients W11 to W1 of the tap coefficients are set.
6, W21 to W26 are set to 1 or each of the multipliers M11 to M
16, M21 to M26 may be omitted. In the above description, the spread bandwidth B of the received signal is assumed to be constant. However, when a signal having a different spread bandwidth B is received, a normalized delay spread obtained by multiplying the delay spread τs by the spread bandwidth B and an optimum window width are obtained. May be stored in advance, and the optimum window width may be determined by obtaining the normalized delay spread Bτs from the detection level of the received signal. The correlators 405, 407 described above,
The detectors 406 and 408 can be configured by digital signal processing.

In the above description, the table shown in FIG. 2B is prepared in advance, and an appropriate window width m is obtained from the calculated delay spread τs. Not limited to this arrangement, a prerequisite for generation of tables shown in FIG. 2B, for example, for various positive normalized delay spread τs for each delay line number n as shown in FIG. 10A, bit error rate 10 ^- Lowest E to get ³
The b / No characteristic is stored in a table using the window width m as a parameter, or the lowest Eb / No characteristic that can obtain an error rate of 10 ⁻³ with respect to various normalization delay spreads as shown in FIG. With the window width number m as a parameter,
The number n of various delay lines and the number P of various delay signals supplied to the synthesis circuit 409 are stored in a table in advance, and the table corresponding to FIG. 10A is referred to based on the calculated delay spread τs. When a line parallel to the vertical axis is drawn upward from the corresponding position of the delay spread above, a window width m, which is a parameter value of a curve that intersects first, is obtained, and this window width m is used. Alternatively, for a table of various values P supplied to the synthesis circuit 409, Eb / N of a curve that first intersects from below with a vertical line passing through the calculated delay spread τs
It is also possible to use the minimum Eb / No P and the parameter m in the minimum of o. Also, given branch number n,
A reference table may be determined from the number of supplied delay signals P and the parameter value m of the curve that intersects first when the calculated delay spread is raised from the position on the horizontal axis of the calculated delay spread.

[0020]

As described above, according to the present invention, by optimally controlling the window width with respect to the delay spread, the S / N of the synthesized signal is always maximized regardless of the magnitude of the delay spread of the delay wave. It can be. According to the present invention, space diversity and path diversity can be used more effectively than before, and communication quality can be improved.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

2A is a diagram illustrating an inclination of attenuation in the spread of a delay wave, and FIG. 2B is a diagram illustrating an optimum window width with respect to a delay spread;

FIG. 3 is a diagram showing an example of delay line tap output.

FIG. 4 is a diagram showing delay line tap outputs when a plurality of paths are selected.

FIG. 5 is a block diagram showing an embodiment of the invention of claim 2;

FIG. 6 is a block diagram of a conventional code division multiple access receiver.

FIG. 7 is a diagram showing a delay line tap output.

FIG. 8 is a diagram showing the spread of a delayed wave.

9A is a diagram illustrating an error rate with respect to a reception level using a window width as a parameter, and FIG. 9B is a diagram illustrating a reception level that obtains required quality with respect to a window width using a delay spread as a parameter.

FIG. 10A is a characteristic curve diagram showing the lowest Eb / No value with respect to delay spread using the window width m as a parameter when all delay signals of the determined window width m are supplied to the synthesis circuit;
Is a delay spread τs having as parameters the number of branches (number of delay lines) n and the window width m for each number of delay signals P supplied to the combining circuit 409, and the lowest E which can obtain the required bit error rate.
It is a figure which shows the example of a relationship with b / No.

Continuation of front page (56) References JP-A-7-74687 (JP, A) JP-A-7-74685 (JP, A) JP-A-9-162779 (JP, A) JP-A-8-265217 (JP, A) JP-A-8-65206 (JP, A) JP-A-8-65205 (JP, A) JP-A-8-116303 (JP, A) JP-A 1-188037 (JP, A) Yoshio Karasawa Et al., Space-Path Hybrid Diversity System for Base Station Reception in CDMA Mobile Communication, IEICE Technical Report SAT93-12, Japan, May 21, 1993, p. 41-47 Akihiro Higashi et al., Adaptive RAKE Diversity (ARD) Reception System, IEICE Spring Conference Proceedings B-301, Japan, p. 2-301 Hideaki Okamoto et al., DS-CDMA Reception Considering Delay Spread Diversity Correction Method and Required Ed / No, IEICE General Conference Proceedings B-393, Japan, March 11, 1996, p. 393 Hideaki Okamoto et al., A study on delay profile and number of paths in microcells, Proceedings of the IEICE Communication Society Conference B-20, Japan, August 30, 1996, p. 20 (58) Field surveyed (Int.Cl. ⁷ , DB name) H04J 13/00-13/06 H04B 1/69-1/713 H04B 7/08 H04B 7/005

Claims (3)

(57) [Claims] 1. A reception signal from n antennas (n is an integer of 1 or more) is despread by n correlators with the same spreading code, and the n despread outputs are respectively detected by n detections. A detector (m is an integer of 2 or more) is obtained by sequentially delaying the n detection outputs by n delay means by one chip period of the spreading code, respectively. In a code division multiple access communication receiving apparatus for combining by a combining circuit, at least one of the n correlators is sequentially shifted by one chip period and k output levels (an integer satisfying k ≧ m) are output levels. A level detecting means for detecting, a calculating means for calculating the spread of the delayed wave based on a detection level from the level detecting means, and a value of the spread of the delayed wave calculated by the calculating means, from the n delay means, Number of delayed signals to extract m
And a window width changing means for changing the width of the code division multiple access communication. 2. A signal received from n (n is an integer of 1 or more) antennas is detected by n detectors, and the n detection outputs are respectively detected by n correlators by the same spreading code. The despreading is performed, and the n despread outputs are sequentially delayed by one chip period of the spreading code by n delay means, respectively (m is 2
A delay signal of the above integer), and combining these delayed signals by a combining circuit, in a code division multiple access communication receiving apparatus, wherein at least one of the n correlators is sequentially shifted by one chip period from the k correlators. Level detecting means for detecting an output level of (an integer satisfying k ≧ m); calculating means for calculating the spread of the delayed wave based on the detection level from the level detecting means; and a delayed wave calculated by the calculating means The number m of delayed signals extracted from each of the n delay means is determined by the value of the spread of
And a window width changing means for changing the width of the code division multiple access communication. 3. The n delay means select j (integer satisfying 2 ≦ j ≦ nxm) from n × m delay signals composed of the m delay signals in descending order of their level. 3. The receiving device for code division multiple access communication according to claim 1, further comprising a selecting unit for supplying the signal to the combining circuit.