JPH11205864A - Method for searching long code in asynchronous cellular system between ds-cdma stations - Google Patents

Abstract

PROBLEM TO BE SOLVED: To allow a method to search long codes with high accuracy. SOLUTION: In the identification processing of a long code, a correlation arithmetic operation is applied between a prescribed chip of a first composite code and a received signal sample and its output power is calculated (S14). The above processing is repeatedly executed for k-sets of different chips of the same composite code (S16), and the mean value is compared with a prescribed threshold (S19). When the mean value is smaller than the prescribed threshold, the above processing is executed for second and succeeding composite codes (S20). Furthermore, peak outputs of a multi-path are summed, and the above processing may be executed regarding the sum.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] The present invention relates to a DS-CDMA.
(Direct Sequence-Code Division Multiple Acces
s) The present invention relates to a long code search method in an asynchronous cellular system between base stations.

[0002]

2. Description of the Related Art With the development of land mobile communication in recent years, direct spreading (D / D) capable of greatly increasing the channel capacity has been proposed.
A CDMA cellular system using a code division multiple access (CDMA) system using S) type spread spectrum (SS) has attracted attention. Generally, in the CDMA system, since there is mutual interference with other stations, another multiple access system (FDM)
A, TDMA), the frequency use efficiency is degraded. However, in the cellular system, since the spatial frequency reuse efficiency (the cell repetition rate of the same frequency) contributes to the overall frequency utilization efficiency, the CDMA system that is resistant to interference and has a high cell repetition rate is also a promising system. .

[0003] Such DS-CDMA cellular systems are classified into two systems: an inter-base-station synchronization system that strictly synchronizes the time between all base stations, and an inter-base-station asynchronous system that does not. The inter-base-station synchronization system realizes inter-base-station synchronization using another system such as GPS. In each base station, the same long code is used with a different delay for each base station. The initial cell search only needs to synchronize the timing of the long code. Further, the peripheral cell search at the time of handover can be performed at higher speed because the mobile station is notified of the code delay information of the peripheral base station from the base station to which the mobile station belongs.

On the other hand, in the asynchronous system between base stations, the spreading code used in each base station is changed to identify the base station. Therefore, the mobile station needs to identify the spreading code in the initial cell search. Becomes Also, in the peripheral cell search at the time of handover, the number of spread codes to be identified can be limited by obtaining information on the spread codes used in the peripheral base station from the base station to which the cell belongs. However, in either case, the search time is longer than in the case of the inter-base station synchronization system, and the time required for cell search becomes enormous when a long code is used as a spreading code. However, the asynchronous system between base stations has an advantage that other systems such as GPS are not required.

A cell search method has been proposed which can solve the problem of the asynchronous system between base stations and can perform initial synchronization at high speed (Kenichi Higuchi, Mamoru Sawahashi, Fumiyuki Adachi, "DS-CDMA base station"). Two-Step Fast Initial Synchronization Method for Long Code in Inter-Asynchronous Cellular System "IEICE Technical Report,
CS-96, RCS96-12 (1996-05)).
In the proposed initial synchronization method, first, the short code common to each cell is despread using a matched filter to detect the timing of the long code, and then each cell is matched using a matched filter or a sliding correlator. This is to specify a unique long code.

Hereinafter, the proposed initial synchronization method will be described. In the proposed asynchronous cellular system between base stations, each of the base stations BS1 to BSN has a different long code LC # 1, LC # 2,.
#N and short code SC for identifying each channel
Transmission is performed with a mobile station using symbols spread twice using # 0 to SC # M. Here, the short codes SC # 0 to SC # M are common to each cell, and a common short code SC # 0 is assigned to a control channel for each cell.

The above-described proposed two-stage high-speed initial synchronization method will be described in detail with reference to FIG. The upper part of the figure shows an example of a received signal in the mobile device, and shows received signals of the control channels transmitted from the base stations BS _k , BS _k-1 and BS _k-2 respectively. As shown in the drawing, each control channel has a symbol (hatched portion in the drawing) spread by only the short code SC # 0 assigned to the control channel common to each base station in one long code cycle. . This is realized by not performing long code spreading in a fixed cycle.
The other symbol positions are double-spread by the long code LC # i and the short code SC # 0 which are different for each base station. As a result, even if the timing of the long code between cells is synchronously received by the mobile station, the control code can be demodulated. As described above, the control channels transmitted from the respective base stations such as BS _{k to} BS _k-2 are multiplexed asynchronously and received by the mobile station.

The mobile station performs a cell search in the following two-stage configuration. First, in the first stage,
As shown in the figure, the mobile device detects the correlation between the received signal sample and the replica of the short code SC # 0 for the control channel using the matched filter. As described above, each control channel in the received signal has a symbol (hatched portion in the figure) spread with the short code SC # 0 common to each base station at the cycle of the long code. Therefore, when the correlation is detected using the short code symbol replica during one long code period, as shown in the figure, the position corresponding to the reception timing of the symbol spread by the short code # 0 in each control channel is obtained. , A correlation peak is detected. On mobile devices,
The timing at which the largest correlation peak is detected is determined to be the long code synchronization timing of the control channel of the base station desired to be connected.

Next, the process proceeds to the second stage.
In order to identify the base station, identification of a long code spreading a control channel detecting the long code synchronization timing is performed using one sliding correlator. For this reason, in the initial cell search, the long code groups LC # 1 to LC #
N, a long code LC # i is sequentially selected from among N, a composite code of the selected long code LC # i and the common short code SC # 0 is generated, and the synthesized code is synchronized with the synchronization timing obtained in the first stage. Then, correlation detection is performed. In the peripheral cell search at the time of handover, a composite code of a long code LC # i + short code SC # 0 is similarly sequentially generated from a long code group of peripheral cells notified from the currently connected base station, Correlation detection is performed on the synchronization timing. In this way,
Long code LC #i until the correlation detection value exceeds the threshold
Is changed, and the long code LC # k exceeding the threshold value is determined to be the long code of the reception control channel, and the cell search ends. Thereby, the base station can be identified.

As described above, by separating the timing synchronization of the long code and the identification of the long code, the cell search can be performed at a high speed. In a typical asynchronous cellular system between base stations, it is necessary to perform (corresponding to the number of spreading codes x the number of phases of spreading codes) correlations to perform a cell search. According to (number of spreading codes + number of phases of spreading codes)
Only about twice correlation detection is required.

Further, the present applicant has proposed a method capable of executing initial synchronization in an asynchronous cellular system between base stations at a higher speed (Japanese Patent Application No. 9-1960).
issue). In this method, a long code is identified at high speed by detecting a correlation with a predetermined intercept of the synthesized code. As a result, a long code can be identified more quickly than in the case of using the above-described sliding correlator, and high-speed initial synchronization can be performed.

Referring to FIG. 14, the proposed method will be described. In FIG. 14, 1 is a complex type matched filter, 2 is a spreading code generator, 3 is a power calculator,
Reference numeral 4 is a long code synchronization timing determination unit, 5 is a threshold value calculation unit, and 6 is a long code identification unit. The matched filter 1 includes a sampling clock CL as shown in FIG.
The shift register 11 has a number of stages (128 stages in this example) equal to the number of chips of a section of a composite code, which will be described later, and spreads the sections of the composite code to be loaded. A code register 12, a plurality of (in this example, 128) multipliers 13 for respectively multiplying data of corresponding stages of the shift register 11 and the spread code register 12, and It comprises an adder 14 for calculating the sum of outputs. The matched filter 1 includes a charge coupled device (CCD).
e), a filter using a SAW (Surface Acoustic Wave) filter, or a filter using a digital IC circuit can be used. However, the use of an analog matched filter proposed by the present applicant is only for calculation speed and power consumption. This is preferable from the viewpoint of calculation accuracy.

[0013] The spread spectrum transmission signal from the base station is input from the receiving antenna to the high-frequency receiving unit, converted into an intermediate frequency signal, multiplied by the output of the intermediate frequency oscillator, and multiplied by the low-pass filter. Signal. This baseband received signal is input to the matched filter 1 and is correlated with the spread code replica supplied from the spread code generator 2. The power calculator 3 calculates the power of the correlation output of the matched filter 1 and outputs the calculated power to the long code synchronization timing determiner 4, threshold calculator 5, and long code identifier 6. Here, the matched filter 1 is a complex type matched filter, and outputs correlation outputs of an in-phase component (I component) and a quadrature component (Q component). The power calculator 3 calculates the absolute value (square) (I ² + Q ² ).

The spreading code generator 2 is controlled by a long code synchronization timing determiner 4 and a long code identifier 6. As described above, at the time of the initial cell search, the spreading code generation unit 2 outputs the common short code SC # 0 to the control channel of each base station. The cells, that is, the segments of the N chips of the combined code #i of the long code LC # i and the short code SC # 0 unique to each base station are output while being sequentially replaced. In addition, in the peripheral cell search before handover, similarly to the above-described initial cell search, a common control code from the control channel of each base station is received, and based on this, the long code synchronization timing of each base station is determined. Is determined. After the long code synchronization timing of the handover destination base station is determined, a plurality of long codes LC # i to be searched are searched based on information on the long codes of the neighboring cells received from the control channel of the base station to which the handover belongs. And short code S
Each piece of the partial N chips of the composite code of C # 0 is output while being sequentially replaced.

The long code synchronization timing determination section 4
In the case of the initial cell search, the short code SC # 0 is loaded from the spreading code generation unit 2 into the PN code register 12 in the matched filter 1, and the average power of the maximum correlation value (averaging over a predetermined number of long code periods) The selected timing is the timing at which the output power value is output, and the timing is output to the spreading code generator 2 and the threshold calculator 5 as the long code synchronization timing.
The threshold calculation unit 5 calculates a threshold to be output to the long code identification unit 6 based on the power of the maximum correlation value at the time of the long code synchronization timing. In the case of a neighboring cell search before handover, the short code SC # 0 is similarly loaded from the spreading code generator 2 into the matched filter 1, and the maximum correlation value of the maximum correlation value is excluded except for the base station currently communicating. The timing at which the average power is output is selected, and this timing is output as the long code synchronization timing of the handover destination base station to the spreading code generation unit 2 and the like. Are loaded into the matched filter 1.

After detecting the long code synchronization timing, the long code identification unit 6 replaces the above-mentioned intercepts of the respective composite codes and sequentially loads them, and compares the output of the signal power calculation unit 3 with a predetermined threshold value. If the threshold value is exceeded, the code number of the long code corresponding to the combined code loaded in the spreading code generation unit 2 at this time is determined to be the long code of the base station to be received. The output of the matched filter 1 and the output of the signal power calculation unit 3 are output to a reception data processing unit (not shown) as necessary. For example, the output of the matched filter 1 can be output to a rake combining circuit, or the output of the signal power calculation unit 3 can be output to a multipath detection unit to perform path diversity reception.

The proposed method will be described in more detail with reference to the flow charts of FIGS. As described above, this initial synchronization method has a configuration including two stages. As shown in FIG. 15A, a long code timing detection process in step S100 and a long code identification process in step S200 are performed. Consists of FIG. 15B shows the result of the step S.
100 is a flowchart of a long code timing detection process. First, in step S101, the common short code SC # 0 is stored in the PN code register 12 of the matched filter 1 from the spreading code generation unit 2.
To load. As a result, the matched filter 1 outputs a correlation output between the baseband received signal sample and the common short code SC # 0.

Next, proceeding to step S102, the power calculator 3 calculates the power of the correlation output from the matched filter 1. Then, the process proceeds to step S103, where the power value exceeding a predetermined threshold value and its timing (corresponding time) are stored, and an average value thereof is calculated over a plurality of cycles. Then, the maximum value is selected from the average values, and the chip synchronization and the long code synchronization timing are determined from the timing. The above is the long code timing detection processing in step S100.

Next, the long code identification processing in step S200 will be described with reference to the flowchart in FIG. Here, first, a long code number i is set to an initial value 1 (step S201), and a synthesized code #i (a synthesized code of the long code LC # i and the common short code SC # 0) is set in the matched filter 1. A predetermined number N chips (in this example, the first chip to the twelfth
A section of 8 chips (128 chips) is loaded (step S202). As a result, in the matched filter 1, a correlation operation between the intercept of the composite code # 1 and a received signal sample is executed. Then, the process proceeds to step S203, and it is determined whether or not the power value of the correlation output output from the power calculation unit 3 is larger than a predetermined threshold value calculated by the threshold value calculation unit 5. If the result of this determination is YES, the process proceeds to step S206,
The long code number i is determined to be the corresponding long code.

On the other hand, if the decision result in the step S203 is NO, the process advances to a step S204 to decide whether or not the determined long code number i is the last long code. As a result, when it is not the last long code, the process proceeds to step S205, the long code number i is updated to i + 1, and the process returns to step S202.
The correlation process is performed on the composite code corresponding to the updated long code i. When the last long code is determined, it is determined that the detection of the long code synchronization timing in step S100 is an error.
Returning to step S100, long code synchronization timing detection is performed again.

Here, the correlation process between the intercept of each synthesized code and the received baseband signal will be described in detail with reference to FIG. In FIG. 17, the horizontal axis is a time axis, and (a) is a baseband received signal sample input to the shift register 11 of the matched filter 1;
(B) shows an intercept of each of N chips (in this example, 128 chips) of each of the composite codes # 1 to # 512 used as reference codes in the correlation processing. As shown in this figure, for the composite code # 1 of the first long code LC # 1 and the common short code SC # 0, the first to 128th chips are referred to by
Loaded into the N code register 12. As described above, the received baseband signal is sampled for each sampling clock CL from the long code synchronization timing and sequentially input shifted to the shift register 11, and the contents of each stage of the shift register 11 and the PN code storage are stored. The contents of each stage of the register 12 and the 128 multipliers 13
Are respectively multiplied. When the baseband received signal samples for 128 chips are stored in the shift register 1, the sum of the outputs of the multipliers 13 from the adder 14, that is, the composite code # 1
And the correlation output between the received signal sample and the received signal sample.

Next, at the timing when the baseband reception signal is newly inputted by M chips (M is an integer) and shifted M times, the shift register 11 stores the (M + 1) th to (M + 128) th of the baseband reception signal. Sample is stored. At this time, the PN code register 12 stores the (M + 1) th chip of the second composite code # 2 to
Load the (M + 128) th chip. Thereby, the adder 14 outputs 128 of the second synthesized code # 2.
Correlation output between the intercept of the chip and the received signal sample (1
A partial correlation output for 28 chips) is output.

Then, until the power value of the partial correlation output exceeds the threshold value and the determination result of the step S203 (FIG. 103) becomes YES, the third synthesized code is sequentially provided at intervals of M chips. Each partial correlation is calculated in the order of # 3 and the fourth synthesized code # 4. The value of M may theoretically be set to M = 1, but it is preferable that M is set to about M = 4 in consideration of the accuracy of the chip synchronization, the variation of the correlation peak, and the like. It is.

As described above, according to the proposed method, the fact that the preceding received sample is stored in the shift register in the matched filter is used, so that each of the synthesized codes serving as reference codes is used. The intercepts can be sequentially switched and the partial correlations thereof can be detected simply by placing a small interval of M chips, and it becomes possible to identify a long code at a very high speed. For example, as described above, when the number of intercept chips is 128 chips, the value of M is 4, and the total number of long codes is 512, at most 128+ (512-1) × 4 = 2172 chips Search can be performed once for the long code of.

[0025]

According to the above-described initial synchronization method, the cell search can be executed at a high speed, but it cannot be said that sufficient consideration has been given to the influence of noise. Further, as indicated by the dotted line in FIG. 14, a multipath actually occurs. However, in the method described above, it cannot be said that sufficient consideration is given to this multipath. Therefore,
It is desired to identify a long code with higher accuracy.

Therefore, the present invention provides an asynchronous CDM between base stations.
It is an object of the present invention to provide a high-speed and high-accuracy long code search method and a receiver in an A cellular system.

[0027]

In order to achieve the above object, a long code search method in an asynchronous cellular system between DS-CDMA base stations according to the present invention uses a spreading code sequence including a long code unique to each cell. A DS-CDMA base station that searches for a long code of a received signal with a part of the long code or a code based on the partial code (hereinafter, these are referred to as “intercepts”) after detecting the synchronization timing of the long code. A long code search method in an inter-cell asynchronous cellular system,
(1) sequentially shifting the position of the partial code in the long code to generate a predetermined number of the segments; (2)
A correlation operation between these intercepts and the received signal is performed, and (3)
Calculating power based on the output of the correlation operation; (4) calculating an average value of power corresponding to the predetermined number of intercepts of the long code; and (5) calculating an average value of the power until the average value exceeds a predetermined threshold. (6) When the average value exceeds a predetermined threshold, the long code corresponding to the intercept is specified, and thereby the cell corresponding to the long code is specified.

[0028] In another long code search method in the asynchronous cellular system between DS-CDMA base stations of the present invention, a synchronization timing of the long code is detected by using a spread code sequence including a long code unique to each cell. Later, the DS-C searches for a long code of the received signal by using a part of the long code or a code based on the partial code (hereinafter, these are referred to as “intercepts”).
A long code search method in an asynchronous cellular system between DMA base stations, wherein (1) a predetermined number of segments are generated by sequentially shifting positions of the partial codes in the long code, and (2) the segments and the reception are generated. A correlation operation with the signal is performed, (3) an average value of correlation outputs corresponding to the predetermined number of intercepts of the long code is calculated, (4) power is calculated based on the average value, and (5) Change the long code until the power exceeds a predetermined threshold,
(6) When the power exceeds a predetermined threshold, a long code corresponding to the intercept is specified, and thereby a cell corresponding to the long code is specified.

Further, still another DS-CDM of the present invention
The long code search method in the asynchronous cellular system between A base stations uses a spreading code sequence including a long code unique to each cell, and after detecting the synchronization timing of the long code, a part of the long code or a partial code of the long code. Search for a long code of a received signal by a code based on
A long code search method in an asynchronous cellular system between S-CDMA base stations, comprising: (1) shifting a position of the partial code in the long code to define a plurality of partial codes; A predetermined number of the intercepts are generated by cyclically shifting a code based on a target code, (2) a correlation operation is performed between these intercepts and the received signal, and (3) power is calculated based on an output of the correlation operation. And (4) selecting a power exceeding a predetermined threshold from these powers, (5) calculating an average value of the selected powers, and (6) calculating the average until the average value exceeds a predetermined threshold. (7) specifying a long code corresponding to the intercept when the average value exceeds a predetermined threshold, thereby specifying a cell corresponding to the long code.

Furthermore, still another DS-C of the present invention
The long code search method in the asynchronous cellular system between the DMA base stations uses a spreading code sequence including a long code unique to each cell and a short code for a common control channel common to each cell, and uses the synchronization timing of the long code. A long code search in an asynchronous cellular system between DS-CDMA base stations for searching for a long code of a received signal by a part of the long code or a code based on this partial code (hereinafter, these are referred to as "intercepts") after detection. (1) performing a correlation operation between the short code and the received signal, and cyclically integrating the power of the correlation output over a plurality of symbols;
From the cyclic integration result, a maximum value of the power and a timing corresponding to a power peak within a predetermined time from the maximum value are selected, and (3) a position of the partial code in the long code is shifted to obtain a plurality of partial codes. Define the code,
(4) cyclically shifting a code based on each of the defined partial codes to generate a predetermined number of the segments;
A correlation operation between these intercepts and the received signal is performed, and (5)
Calculating power based on the output of the correlation operation; (6) calculating an average value of the power at the selected timing among the powers; and (7) calculating the long code until the average value exceeds a predetermined threshold. (8) When the average value exceeds a predetermined threshold, a long code corresponding to the intercept is specified, and thereby a cell corresponding to the long code is specified.

Furthermore, still another DS-C of the present invention
The long code search method in the asynchronous cellular system between DMA base stations uses a spreading code sequence including a long code unique to each cell, and after detecting the synchronization timing of the long code, a part of the long code or a partial code of the long code. (Hereinafter referred to as “intercept”)
A long code search method in an asynchronous cellular system between DS-CDMA base stations, which searches for a long code of a received signal by: (1) shifting a position of the partial code in the long code to a plurality of partial codes; A predetermined number of the segments are generated by cyclically shifting a code based on each of the defined partial codes, and (2) performing a correlation operation between these segments and the received signal, ) Calculate the power from the correlation operation output, (4) select a power exceeding a predetermined threshold from these powers, and (5) correspond to the selected power while cyclically shifting the partial codes. Rake synthesis by fading correction of the correlation calculation output of the I and Q components
(6) Calculate the average value of these rake synthesis results, and (7)
Calculating power based on the average value; (8) changing the long code until the power exceeds a predetermined threshold;
(9) When the power exceeds a predetermined threshold, a long code corresponding to the intercept is specified, and thereby a cell corresponding to the long code is specified.

Further, still another DS-C of the present invention
The long code search method in the asynchronous cellular system between the DMA base stations uses a spreading code sequence including a long code unique to each cell and a short code for a common control channel common to each cell, and uses the synchronization timing of the long code. A long code search in an asynchronous cellular system between DS-CDMA base stations that searches for a long code of a received signal by a part of the long code or a code based on the partial code (hereinafter, these are referred to as "intercepts") after the detection. (1) performing a correlation operation between the short code and the received signal, and cyclically integrating the power of the correlation output over a plurality of symbols;
(2) A maximum value of the power and a timing corresponding to a power peak within a predetermined time from the maximum value are selected from the cyclic integration result. (3) The position of the partial code in the long code is shifted to a plurality of positions. And a predetermined number of the segments are generated by cyclically shifting a code based on each of the defined partial codes, and (4) performing a correlation operation between these segments and the received signal, (5) While cyclically shifting each of the partial codes, the I and Q components of the correlation operation output are fading-corrected at the selected timing and rake-combined. (6) The average value of these rake-combined results is calculated. (7) calculating power based on the average value; (8) changing the long code until the power exceeds a predetermined threshold; and (9) determining that the power has exceeded a predetermined threshold. To identify the long code corresponding to the sections, whereby it is intended to identify the cell corresponding to the long code. Furthermore, the intercept is a part of a composite code of a short code and a long code corresponding to each communication channel.

[0033]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The long code search method according to the present invention will be described below. The long code search method according to the present invention is basically the same as the proposed method described with reference to FIGS. This is performed on the configuration, and the description overlapping with the items described with reference to FIGS. 14 to 16 will be omitted.

First, a first embodiment of the long code search method of the present invention will be described with reference to FIG.
In this embodiment, in order to prevent the influence of noise, an average value of the power of the correlation output for a plurality of continuous symbols is calculated, and the determination is made based on the average value.

FIG. 1 is a functional block diagram of the long code identification unit 6 when the first embodiment is applied. In FIG. 1, reference numeral 31 denotes the power calculator 3
This is an analog-to-digital converter that receives the output from FIG. 14 and converts it into digital data. The output of the analog-to-digital converter 31 is input to an averaging unit 32, which calculates the average value of the correlation signal power for a plurality of consecutive symbols. That is,
The averaging section 32 holds the correlation signal power converted into the digital data for a plurality of symbols, adds these, divides them, and calculates the average value. The output of the averaging unit 32 is input to the determination unit 33, where the output is compared with the threshold value calculated by the threshold value calculation unit 5 (FIG. 14). in this way,
The influence of noise can be prevented by calculating the average value of the partial correlation outputs for a plurality of times and comparing the average value with the threshold value. Note that the number of chips of the intercept of the composite code for detecting the partial correlation has a length equal to the number of chips corresponding to one symbol. The averaging unit 3
2 and the determination unit 33 can also be realized using an analog operation circuit. In this case, the analog-to-digital converter 31 can be omitted.

The long code identification processing in the first embodiment will be described with reference to the flowchart of FIG. 2 and the timing chart of FIG. Here, the long code identification processing of FIG. 2 is executed after the long code timing detection processing shown in FIGS. 15A and 15B. In FIG. 3, the horizontal axis indicates the time axis, (1) indicates received sample data,
(2) is the sampling clock CL. As described above, the baseband received signal is sampled for each sampling clock, and the received sample data of (1) is stored in the shift register 11 in the matched filter 1.
2 shows the received signal sample stored in the last stage (the 128th stage). (3) is an intercept (PN intercept) of the synthesized code stored in the PN code register 12 in the matched filter 1, and (4) is an intercept of the received signal sample output from the power calculator 3 and the PN intercept. (5) is the average output of the correlation signal power output from the averaging unit 32, and (6) is the determination unit 3
3 shows the output. It should be noted that, for simplicity, in this timing chart and the timing chart described later, M in FIG. 17 is set to 1 and the delay time required for processing in each unit is ignored. .

When the long code identification process is started, first, in step S11, the number J of the composite code to be determined is set to the initial value 1. Next, in step S12, the number k of the intercept of the composite code #J is set to the initial value 1. Then, in step S13, the intercept k of the combined code #J is obtained from the spreading code generation unit 2 (FIG. 14) in the matched filter 1.
Loaded into the N code register 12. In the example shown in FIG. 3, during the period T1, the first intercept (#
1.1), that is, the first to 128th chips of the composite code # 1 have been loaded. Thereby, in the matched filter 1, the correlation between the received signal sample and the intercept # 1.1 is calculated, and the power is output from the power calculator 3 (FIG. 14). Next, step S1
At 5, the power value is stored and held together with the timing data.

Next, the process proceeds to step S16, where it is determined whether or not the number k of the intercept of the composite code is equal to the expected final value. In the example shown, k =
It is set to 3. In the period T1, the result of this determination is NO, so the process proceeds to step S17, where k is increased by 1, and the process returns to step S13. As a result, this time, the segment 2 (# 1.2) which is the second to 129th chips of the composite code # 1 is stored in the PN code register 1
2 is loaded. Thereby, the correlation between the received signal sample stored in the shift register 11 and the intercept 2 is obtained in the period T2, and the power is output (step S14). Then, the average data and the timing information are held (step S15). Then, it is determined again whether or not k has reached the scheduled maximum value. In this way, the correlation process is performed while sequentially moving the intercept of the combined code until k = 3.

Then, k = 3, and step S1 is performed.
When the result of the determination in step 6 is YES, the process proceeds to step S18, and the average value of the correlation signal power corresponding to the k symbols for the combined code held in step S15 is calculated. Then, in step S19,
It is determined whether or not the average value is equal to or greater than a predetermined threshold.
If the determination result is YES, the long code corresponding to the combined code is determined to be the long code of the base station, and the long code identification processing ends.

On the other hand, if the result of the determination in step S19 is N
In the case of O, the process proceeds to step S20, and it is determined whether or not the determined synthesized code is the last long code to be determined. When the determination result is YES, it is determined that there is an error in the detection of the long code synchronization timing in the step S100, and the step 100 (FIG.
Return to 5 (a)). If the result of this determination is NO, the number J of the composite code is increased by one, and the process returns to step S12 to repeat the above-described processing. For example,
As shown in FIG. 3, assuming that J = 2, the first first segment # 2.4 of the composite code # 2 (this is the fourth to 131st chips of the composite code # 2) The PN
It is loaded into the sign register 12 and is correlated with the received signal samples.

As described above, in this embodiment,
For each intercept of the synthesized code, a correlation between the received signal sample and a plurality of consecutive intercepts is obtained, and the long code is identified based on the average value of the signal power. Therefore, the effects of noise and fading can be averaged, and erroneous determination can be reduced. In the above example, the average value of the power of the three correlation outputs is calculated. However, the present invention is not limited to this, and the average value of the power of an arbitrary number of correlation outputs can be calculated.

Next, a second embodiment of the present invention will be described with reference to a functional block diagram of FIG. 4, a flowchart of FIG. 5, and a timing chart of FIG.
As shown in FIG. 4, this embodiment is suitable for application when the channel (control channel) for identifying the long code is not modulated, and is suitable for the first embodiment. It is more resistant to noise and fading than it is.

As shown in FIG. 4, in this embodiment, the output of the matched filter 1 and its power output are directly input to the A / D converter 41. And
The output of the A / D converter 41 is input to the averaging unit 42,
As in the case described above, the correlation outputs for a plurality of continuous segments shifted by one chip for one combined code are held, and when the power output exceeds a predetermined threshold, the average value of the correlation outputs is represented by I, It is calculated for each of the Q components. Then, the average value is input to the power calculation unit 43 to calculate the power (I ² + Q ² ), and the determination unit 44 compares the average value with the above-described threshold to determine the long code. Note that the averaging unit 42, the power calculation unit 43, and the determination unit 44 can also be configured using an analog operation circuit. In this case, the analog-to-digital converter 41 can be omitted.

The processing in the second embodiment shown in the flow chart of FIG. 5 is different from the flow chart of FIG. 2 in that the processing in step S34 is simply a correlation processing, and The difference is that a power calculation process in step S39 is added after the output process. The other processes are the same as those in FIG.
Since this is the same as the flowchart shown in FIG.
Detailed description thereof will be omitted. Also, the timing chart of FIG. 6 is the same as that of FIG. 3 except that the power output of (5) is output after calculating the average output of (4). Description is omitted.

As described above, according to the second embodiment, since the power is calculated from the average value of the correlation output (I component and Q component) from the matched filter 1, the positive and negative noise components are calculated. Are mutually canceled, and a greater noise reduction effect than in the case of the first embodiment can be expected. However, when data modulation is performed on the control channel, the average output becomes 0. Therefore, this embodiment is suitable when the control channel is not modulated.

Next, a third embodiment of the present invention will be described with reference to a functional block diagram of FIG. 7, a flowchart of FIG. 8, and a time chart of FIG. In this embodiment, multipath is taken into account, and the correlation signal powers of the received signals of a plurality of paths are added, and the determination is made based on the average value thereof. Therefore, the received signal power can be used more effectively than in the case of using a single-path received signal as in the first and second embodiments described above, and the effects of noise and fading can be further reduced. be able to.

In FIG. 7, reference numeral 51 denotes the power calculator 3
An A / D converter 52 receives the output of FIG. 14 and converts it into a digital signal. The A / D converter 52 holds correlation signal power data corresponding to each path from the A / D converter 51 and averages them. An averaging unit 53 for calculating a value is a determining unit for comparing the output of the averaging unit 52 with the threshold data from the threshold calculating unit 5 (FIG. 14).

FIG. 8 is a flowchart for explaining the operation of the third embodiment. Here, as in the case described above, J indicates the number of the composite code to be determined, that is, the long code, and k indicates the number of a plurality of symbols for calculating the average value of the correlation output for one composite code. Further, i is the number of chips for setting a delay period for detecting a multipath.

In the timing chart of FIG. 9, the horizontal axis represents time, (1) is received sample data, (2) is a sampling clock CL, and (3) is a PN code register in the shift register 1. 12 shows an intercept of a composite code stored in No. 12. The hatched portion in (3) indicates that a peak was detected at this timing. (4) is the output of the power calculator 3 (FIG. 14), (5) is the average output of the multipath correlation output for one intercept, and (6) is the multipath of a plurality of intercepts for one composite code. The average value of the correlation output, (7) indicates the judgment output. In addition,
FIG. 9 shows a timing chart when k is 2 and i is 6. In this case, a multipath signal having a delay of up to six chip times can be used.

When the long code identification process in this embodiment is started, first, in step S51,
The number J of the composite code to be determined is initialized to 1. Then, in steps S52 and S53, k and i are set to the initial value 1 respectively. Thus, in step S54, the first combined code # 1 is stored in the PN code register 12 of the matched filter 1.
The first section (# 1.1-0), that is, the first to 128th chips of the composite code # 1 are loaded (FIG. 9).
Period T1). Then, the correlation between the received signal sample and the intercept (# 1.1-0) is calculated in the matched filter 1, and the signal power is output from the power calculator 3 (S55). When the signal power is equal to or higher than the predetermined threshold, the signal power is held together with the timing information (step S56), and the correlation power of each path is added (step S57).

The timing at which the correlation calculation output is sufficiently large can be detected in advance by the short code of the common control channel. In this case, the threshold processing in step S56 is not required. That is, when detecting the long code synchronization timing, the power of the correlation output with the short code SC # 0 common to the control channel of each base station is cyclically integrated over a plurality of symbols, that is, the correlation peak for a plurality of symbols is provided. Are integrated, and the timing corresponding to the maximum value and the power peak within a predetermined time from the maximum value is detected. Then, the output of the power calculator 3 at the selected timing is added in the step S57.

Next, the routine proceeds to step S58, where it is determined whether or not the value of i has reached the maximum value. If the determination result is NO, the process proceeds to step S59, i is increased by 1, and in step S60, the intercept stored in the PN code register 12 is cyclically shifted by one stage. Thus, the PN code register 12 has
The 128th chip and the 1st to 127th chips (# 1.1-1) of the composite code # 1 in which the data of the 1st to 128th chips of the composite code # 1 are cyclically shifted by one chip are stored. (Period T2 in FIG. 9). Then, returning to step S55, the correlation between the received signal sample and the one-chip cyclically shifted composite code # 1.1-1 is output. At this time, as shown by hatching in FIG. 9, when the signal power of the correlation output becomes a predetermined value or more (peak), the power value is held together with the timing data in step S56. Thereafter, this process is repeated until the value of i reaches the set maximum value, and the correlation power of each path is added (step S57).

If the decision result in the step S58 is YES, the process proceeds to a step S61, in which it is determined whether or not k has reached the maximum value. If the result of this determination is NO, the operation proceeds to step S62, the value of k is increased by one, and the operation returns to step S53. Then, a new intercept k (in FIG. 9, T
7 (that is, the seventh chip to the 134th chip in the composite code # 1) is loaded into the PN code register 12 and a correlation process with the received signal sample is executed. Thereafter, in the same manner as described above, the intercept stored in the PN code register 12 is cyclically shifted until the value of i reaches the maximum value to detect a correlation, and the peak correlation power is added. The above processing is repeated until the value of k reaches the maximum value. If the determination result in step S61 is YES, the process proceeds to step S63, where the correlation power of the plurality of paths in each intercept calculated by the above processing is calculated. The average of the sum is calculated. And step S
Proceeding to 64, it is determined whether the average value is equal to or greater than a predetermined threshold.

If the decision result in the step S64 is NO, the process proceeds to a step S65, and similarly to the above case, J
Is determined to have reached the maximum value. If the value has not reached the maximum value, the value of J is updated in step S66, and the process proceeds to step S52. On the other hand, if the decision result in the step S64 is YES, it is determined that there is an error in the long code synchronization timing detection, and the process returns to the step S100 ((a) in FIG. 15). On the other hand, if the decision result in the step S64 is YES, the process proceeds to a step S67 to decide that the long code is the long code of the base station.

As described above, according to this embodiment, the sum of the correlation powers of a plurality of paths is added in addition to using the average value of a plurality of symbols as in the first embodiment. Used. Therefore, the effects of noise and fading can be eliminated more than in the first and second embodiments using a single-path received signal.

Next, a fourth embodiment of the present invention will be described with reference to a functional block diagram of a long code identification unit in FIG. 10, a flowchart in FIG. 11, and a timing chart in FIG. This embodiment uses a multipath as in the third embodiment described above. However, instead of simply adding the power of the correlation output of each path, the correlation of each of the I and Q components is used. After fading correction of the output, rake combining is performed, and the rake combining result is averaged for a plurality of symbols, and then power is calculated. This makes it possible to determine a long code with higher precision than in the third embodiment. This embodiment is suitable for application when the control channel is unmodulated.

In FIG. 10, reference numeral 61 denotes an A / D converter to which the output of the matched filter 1 is directly inputted.
Is a fading correction unit that performs fading correction on the output of the A / D converter. Here, an unmodulated control channel is used for long code identification, and the phase error due to fading is estimated by averaging the I and Q components of the matched filter output over a plurality of symbols. Therefore, the fading phase error of the output of the matched filter 1 is corrected by the fading correction unit 62. A correction coefficient is calculated based on the phase error thus obtained, and the correction coefficient is multiplied by the correlation peak of each of the I and Q components to perform fading correction. 63 is a rake combining and averaging unit for rake combining the phase-corrected correlation output of each path output from the fading correction unit 62 and calculating an average value of the rake combined outputs corresponding to a plurality of symbols;
Is the power of the output of the rake combining and averaging unit 63 (I ² +
Q ² ), a power calculator, 65 is the threshold calculator 5
The determination unit compares the threshold value supplied from FIG. 14 with the output of the power calculation unit 64.

In the flowchart of FIG. 11, steps up to step S74 are the same as those of the third embodiment shown in FIG. 8 described above. Then, in the fourth embodiment, in step S75, the power of the correlation output from the matched filter 1 exceeding the predetermined threshold is held, and the above-described processing is performed by the loop of steps S75 to S78. Similarly, multipath correlation output is detected. As in the case described above, it is possible to detect in advance the timing at which the correlation operation output is sufficiently large by the short code of the common control channel, and in this case, the threshold processing in step S75 becomes unnecessary. .

FIG. 12 shows the peak of the correlation output with respect to the first combined code # 1.1 in the periods T2, T3 and T6. Then, fading correction of each path detected in step S79 is performed. This correction is performed using a different correction coefficient for each pass. Then, in step S80, rake combining of the correlation output of each path subjected to fading correction is performed. As described above, this processing is executed for the correlation output for a plurality of k pieces of the same composite code (step S82), and the average value is calculated in step S83. In the example shown in FIG. 10, for the first combined code # 1, the average value of the rake combined output with respect to intercept # 1.1 and intercept # 1.7 is calculated. Then, the process proceeds to a step S84, in which the power of the average value is calculated and compared with the threshold value.

As described above, according to this embodiment, the average value of the result of rake combining by performing the fading correction of each path is calculated, and the long code is identified using the power value. Thus, it is possible to perform identification with higher accuracy than in the case of the above-described first to third embodiments.

In the embodiment described above, the intercept is described as an intercept of a predetermined number of chips of a composite code of a long code and a specific short code common to each cell. There is no limit. A part of a predetermined number of chips of a long code unique to each cell may be used as the segment itself, or the segment, that is, a composite code of a part of the long code and a specific short code common to each cell may be added with “+1”. Or a composite code modulated by multiplying by “−1”.

[0062]

As described above, according to the long code search method of the present invention, long codes can be identified while eliminating the effects of noise and fading.
High-precision long code search can be performed.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of a long code identification unit for describing a first embodiment of a long code search method according to the present invention.

FIG. 2 is a flowchart for explaining the operation of the first embodiment in the long code search method of the present invention.

FIG. 3 is an example of a timing chart in the first embodiment of the long code search method of the present invention.

FIG. 4 is a functional block diagram of a long code identification unit for describing a second embodiment of the long code search method according to the present invention.

FIG. 5 is a flowchart for explaining the operation of the second embodiment in the long code search method of the present invention.

FIG. 6 is an example of a timing chart in a second embodiment of the long code search method of the present invention.

FIG. 7 is a functional block diagram of a long code identification unit for describing a third embodiment of the long code search method according to the present invention.

FIG. 8 is a flowchart for explaining the operation of the long code search method according to the third embodiment of the present invention.

FIG. 9 is an example of a timing chart in a third embodiment of the long code search method of the present invention.

FIG. 10 is a functional block diagram of a long code identification unit for describing a fourth embodiment of the long code search method according to the present invention.

FIG. 11 is a flowchart illustrating an operation of a long code search method according to a fourth embodiment of the present invention.

FIG. 12 is an example of a timing chart according to a fourth embodiment of the long code search method of the present invention.

FIG. 13 is a diagram for explaining a conventional two-stage initial synchronization method.

FIG. 14 is a block diagram showing a functional configuration in a proposed initial synchronization method.

FIG. 15 is a flowchart for explaining the operation of the proposed initial synchronization method.

FIG. 16 is a flowchart for explaining the operation of the proposed initial synchronization method.

FIG. 17 is a diagram for explaining the operation of the proposed initial synchronization method.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 matched filter 2 spreading code generator 3, 43, 64 power calculator 4 long code synchronization timing determiner 5 threshold calculator 6 long code identifier 11 shift register 12 PN code register 13 multiplier 14 adder 21, 31, 41 , 51, 61 A / D conversion unit 22, 33, 44, 53, 65 Judgment unit 32, 42, 52 Average processing unit 62 Fading correction unit 63 Rake combining, average processing unit

Claims (7)

[Claims] 1. After detecting the synchronization timing of the long code using a spreading code sequence including a long code unique to each cell, a part of the long code or a code based on this partial code (hereinafter, these are referred to as “ A long code search method in an asynchronous cellular system between DS-CDMA base stations, which searches for a long code of a received signal by "intercept". (1) The position of the partial code in the long code is sequentially shifted. (2) performing a correlation operation between the intercept and the received signal; (3) calculating power based on an output of the correlation operation; (4) the predetermined number of the long codes (5) change the long code until the average value exceeds a predetermined threshold value; A) identifying a long code corresponding to the intercept when the average value exceeds a predetermined threshold value, thereby identifying a cell corresponding to the long code; Long code search method in. 2. After detecting a synchronization timing of the long code using a spreading code sequence including a long code unique to each cell, a part of the long code or a code based on the partial code (hereinafter, these are referred to as “ A long code search method in an asynchronous cellular system between DS-CDMA base stations, which searches for a long code of a received signal by "intercept". (1) The position of the partial code in the long code is sequentially shifted. (2) performing a correlation operation between these sections and the received signal; (3) calculating an average value of correlation outputs corresponding to the predetermined number of sections of the long code; ) Calculate the power based on this average value, (5) change the long code until this power exceeds a predetermined threshold, (6) When the power exceeds a predetermined threshold, a long code corresponding to the intercept is specified, and thereby a cell corresponding to the long code is specified.
A long code search method in an asynchronous cellular system between S-CDMA base stations. 3. After detecting the synchronization timing of the long code using a spreading code sequence including a long code unique to each cell, a part of the long code or a code based on the partial code (hereinafter, these are referred to as “ A long code search method in the DS-CDMA base station asynchronous cellular system, wherein the long code of the received signal is searched for by using the "intercept". , And a predetermined number of the segments are generated by cyclically shifting a code based on each of the defined partial codes. (2) Performing a correlation operation between these segments and the received signal, (3) Calculate the power based on the output of the correlation operation, and (4) Select the power exceeding a predetermined threshold from these powers (5) calculating an average value of the selected powers; (6) changing the long code until the average value exceeds a predetermined threshold; (7) when the average value exceeds a predetermined threshold A long code search method in an asynchronous cellular system between DS-CDMA base stations, wherein a long code corresponding to the intercept is specified, and thereby a cell corresponding to the long code is specified. 4. A part of the long code after detecting a synchronization timing of the long code using a spreading code sequence including a long code unique to each cell and a short code for a common control channel common to each cell. Alternatively, a long code search method in a DS-CDMA base station asynchronous cellular system, in which a long code of a received signal is searched for using a code based on the partial code (hereinafter, these are referred to as “intercepts”), A correlation operation between the short code and the received signal is performed, and the power of the correlation output is cyclically integrated over a plurality of symbols. (2) The maximum value of the power from the cyclic integration result and the power peak within a predetermined time from the maximum value (3) The position of the partial code in the long code is shifted. And a plurality of partial codes are defined, and a predetermined number of the segments are generated by cyclically shifting a code based on the defined partial codes. (4) Correlation between these segments and the received signal (5) Calculate the power based on the output of the correlation calculation, (6) calculate the average value of the power at the selected timing among these powers, and (7) calculate the average value of the power at a predetermined value. (8) specifying a long code corresponding to the intercept when the average value exceeds a predetermined threshold, and thereby specifying a cell corresponding to the long code. A long code search method in a DS-CDMA base station asynchronous cellular system. 5. After detecting a synchronization timing of the long code using a spreading code sequence including a long code unique to each cell, a part of the long code or a code based on the partial code (hereinafter, these are referred to as “ A long code search method in the DS-CDMA base station asynchronous cellular system, wherein the long code of the received signal is searched for by using the "intercept". , And a predetermined number of the segments are generated by cyclically shifting a code based on each of the defined partial codes. (2) Performing a correlation operation between these segments and the received signal, (3) Power is calculated from the correlation calculation output, (4) Power exceeding a predetermined threshold is selected from these powers, ) While the respective partial codes are cyclically shifted, the correlation calculation outputs of the I and Q components corresponding to the selected powers are subjected to fading correction and rake-combined. (7) calculating power based on the average value; (8) changing the long code until the power exceeds a predetermined threshold; and (9) the intercept when the power exceeds a predetermined threshold. Characterized in that a long code corresponding to the long code is identified, and thereby a cell corresponding to the long code is identified.
A long code search method in an asynchronous cellular system between S-CDMA base stations. 6. A part of the long code after detecting a synchronization timing of the long code using a spreading code sequence including a long code unique to each cell and a short code for a common control channel common to each cell. Alternatively, a long code search method in a DS-CDMA base station asynchronous cellular system, in which a long code of a received signal is searched for using a code based on the partial code (hereinafter, these are referred to as “intercepts”), A correlation operation between the short code and the received signal is performed, and the power of the correlation output is cyclically integrated over a plurality of symbols. (2) The maximum value of the power from the cyclic integration result and the power within a predetermined time from the maximum value (3) The position of the partial code in the long code is selected. A plurality of partial codes, and a predetermined number of the segments are generated by cyclically shifting a code based on each of the defined partial codes. (4) Correlation between the segments and the received signal (5) performing rake synthesis by fading-correcting the I and Q components of the correlation calculation output at the selected timing while cyclically shifting each of the partial codes; and (6) averaging these rake synthesis results. (7) Calculate the power based on this average value; (8) Change the long code until this power exceeds a predetermined threshold; (9) The power exceeds the predetermined threshold Sometimes identifying a long code corresponding to the intercept, thereby identifying a cell corresponding to the long code.
A long code search method in an asynchronous cellular system between S-CDMA base stations. 7. The method according to claim 1, wherein the intercept is a part of a composite code of a short code and a long code corresponding to each communication channel.
3. The long code search method in the asynchronous cellular system between DS-CDMA base stations according to the item [1].