Method and apparatus for cell search for W-CDMA with non-ideal sampling

Abstract

The invention discloses a cell search method for a CDMA system for a three-stage cell search procedure. The cell search method have the steps of matching of incoming signals by a chip matched filter; over-sampling the incoming signals N times against a chip rate by a sampling device; transmitting the N over-samplers into a serial test unit, selecting a best sample among N over-samples by the serial test unit controlled by a first stage detector; and transmitting the best sample of the N over-samples into a second stage detector and a third stage detector for a trial. The best sample of the N over-samples has a maximum sampling value of the N over-samples. The invention significantly reduces the effects of non-ideal sampling so as to accomplish fast cell search.

Description

BACKGROUND OF THE INVENTION

[0001] 1. Field of the Invention

[0002] The present invention generally relates to a cell search method and apparatus for the code division multiple access (CDMA) system, and more particularly to a cell search method and apparatus for the wideband-code division multiple access (W-CDMA) system to reduce the effects of non-ideal sampling.

[0003] 2. Description of the Related Art

[0004] CDMA cellular systems based on code division multiple access (CDMA) using a direct sequence spread spectrum (DSSS) technology greatly increases the channel capacity. These systems are receiving much attention in the recent work on ground mobile communication systems. In general, bandwidth efficiency of a CDMA system is much better than that of other multiple access systems (FDMA, TDMA) because of the universal frequency reuse property. Moreover, the cell planning is also easy in these systems. Hence, a CDMA cellular system can be a promising system in the future.

[0005] Third generation partnership project (3GPP) wideband code division multiple access/frequency division duplex system (W-CDMA/FDD) has been adopted as one of the standards for the IMT-2000 third generation system. In CDMA cellular systems, the procedure used by a user equipment (UE) to search for the best cell is referred to as “cell search”. Fast cell search is very important in order to reduce the UE switched-on delay (initial search), to increase standby time (idle mode search) and to maintain good link quality during handover (active mode search).

[0006] U.S. Pat. No. 6,038,250 issued to Shou et al., entitled “Initial Synchronization Method And Receiver for DS-CDMA Inter Base Station Asynchronous Cellular System”, discloses that cells are searched at a high speed using an initial synchronization method and a receiver for a DS-CDMA inter base station asynchronous cellular system. A base band received signal is input to a matched filter and is correlated with a spread code supplied from a spread code generator. A signal electric power calculator calculates the electric power of the correlation output of the matched filter, and outputs the result to a long code synchronization timing determiner, a threshold value calculator, and a long code identifier. During the initial cell search, the spread code generator outputs a short code that is common to the control channel of each of the base stations. After the long code synchronization timing has been determined, each of the segments of the N chips which constituting a portion of the synthesized spread code sequence is sequentially replaced and output.

[0007] U.S. Pat. No. 6,185,244 issued to Nystrom et al., entitled “Cell searching in a CDMA communications system” discloses a coding scheme for more effectively acquiring a long code and frame timing during a cell search in a CDMA communications system. A code set of length M Q-ary code words including symbols from a set of Q short codes is defined with certain properties. The primary property to be satisfied is that no cyclic shift of a code word yields a valid code word. The other properties to be satisfied are that there is a one-to-one mapping between a long code message and a valid code word, and a decoder should be able to find both the random shift (thereby implicitly finding the frame timing) and the transmitted code word (i.e., its associated long code indication message) in the presence of interference and noise, with some degree of accuracy and reasonable complexity.

[0008] U.S. Pat. No. 6,289,007 issued to Kim et al., entitled “Method for Acquiring A Cell Site Station in Asynchronous CDMA Cellular Communication Systems”, discloses that a group code and a cell code are multiplexed and then used as a pilot code for discriminating a base station in asynchronous cellular CDMA (Code Division Multiple Access) communication systems. Using the multiplexed code, interferences are reduced in case of using two pilot codes. A method for acquiring a cell site station in asynchronous CDMA (Code Division Multiple Access) communication systems including a base station controller, a plurality of mobile stations and base stations, and discriminating the base stations by using different sequences, the method including the steps of: a) assigning a group code of the cell as a pilot code of an inphase channel of the base stations; b) assigning a cell code of the cell as a pilot code of a quadrature channel of the base stations; and c) multiplexing the pilot codes of inphase channel and the quadrature channel, and generating an inphase/quadrature pilot code.

[0009] Now referring to FIG. 1, it will be helpful to understand the simplified frame structure of the 3GPP W-CDMA/FDD system. First, in 3GPP W-CDMA/FDD system, the cell search is typically carried out in three stages by employing two peculiarly designed synchronization channels and a common pilot channel. In the first stage 110, a primary synchronization channel (PSCH) 111 is used for slot synchronization. The primary synchronization channel (PSCH) 111 consists of a primary synchronization code which is denoted acp, where “a” (=±1) depends on the existence of the transmit diversity at the base station (BS). In the second stage 120, the secondary synchronization channel (SSCH) 121 is used for frame/code group identification. The secondary synchronization channel (SSCH) 121 consists of the secondary synchronization code, which is denoted by acs, and the coefficient of cs is similar to that in PSCH. In the third stage 130, a common pilot channel (CPICH) 131 is used for determination of the downlink scrambling code. As shown, the 10-ms frame consists of 15 slots, and the system uses the chip rate of 3.84 Mchips/sec. Eventually, there are 38400 chips in a frame and 2560 chips in a slot. In addition, PSCH and SSCH are 256-chip long and only transmitted at the beginning of the slot boundaries.

[0010] Conventional cell search processes for the 3GPP W-CDMA/FDD system can be divided into two broad categories: the serial search and the pipeline search processes. The serial search process needs to go through all the three stages of synchronization one by one before a new three-stage attempt can commence. This usually takes much more time than the pipeline search, where the three-stage processes can be performed concurrently so as to reduce the search time, of course, at the expense of larger complexity and computation power.

[0011] Now referring to FIG. 2, it is a simplified diagram for the conventional three-stage serial cell search processes. For convenience, a complete three-stage cell search procedure will be referred as a trial, and trials are repeated again and again until the search succeeds. Trials are not overlapped in the serial search, that is, only one stage (a block 211, a block 212 and a block 213) works at a time, which implies lower power consumption, but at the expense of a longer search time.

[0012] On the other hand, referring to FIG. 3, it is a simplified diagram for the conventional three-stage pipeline search processes. In the pipeline search processes, three stages are performed concurrently, which means the trials are overlapped. Namely, a block 311, a block 321 and a block 331 are in the same trial. Obviously, in the pipeline search processes, more trials are possible for a fixed amount of time, and hence result in a faster search. Of course, they require more power consuming. Note that no extra hardware is needed for the pipeline search processes, as compared with the serial search. For simplicity, we assume 10 (ms) is required for each stage, and then (K+2)×10 (ms) of the total processing time is necessary for a successful search which is terminated in the Kth trial.

[0013] However, a common fundamental assumption were usually made in the above prior art on the cell search. That is, the sampling at the output of the chip-matched filter is perfect (ideal sampling). In practice, the sampling at the output of the chip-matched filter is not perfect (non-ideal sampling). The non-ideal sampling may reduce the signal to interference ratio and increase the cell search time. The performance of cell search is degraded significantly due to the imperfection.

SUMMARY OF THE INVENTION

[0014] The primary object of the present invention is to provide a cell search method for the CDMA system, especially for W-CDMA system, in order to to reduce the effects of non-ideal sampling and to accomplish fast cell search.

[0015] Another object of the present invention is to provide a new cell search apparatus for the CDMA system, especially for W-CDMA system, to realize the cell search under the effects of non-ideal sampling without increasing the hardware complexity.

[0016] In order to achieve the above objects, the present invention provides a cell search method for the CDMA system, especially for W-CDMA system, to reduce the effects of non-ideal sampling by using a new cell search procedure under the three-stage cell search procedure. In the first stage, a PSCH is used for slot synchronization; in the second stage, a SSCH is used for frame/code group identification after slot synchronization; and in the third stage, a CPICH is used for the determination of the downlink scrambling code. The cell search method according to the present invention comprises the steps of: matching incoming signals by a chip matched filter; over-sampling the incoming signals N times against a chip rate by a sampling device; transmitting the N over-samples into a serial test unit by a down sampler; selecting a best sample among N over-samples by the serial test unit controlled by a first stage detector in the first stage; and transmitting the best sample of the N over-samples into a second stage detector and a third stage detector to accomplish a trial. The trial of the cell search method further comprises the steps of achieving a slot synchronization in the first stage by a first stage detector; achieving a code group and frame synchronization in the second stage by a second stage detector; selecting a scrambling code in the third stage by the third stage detector; testing the scrambling code against a threshold η0 by a comparator; determining the scrambling code by a first decision device after the comparator; wherein if the threshold is exceeded, then the selected scrambling code goes for a synchronization verification unit, otherwise the trial is considered to be failure, and a new trial will be restarted without penalty; and determining the scrambling code by a second decision device after the synchronization verification unit; wherein if the scrambling code passes the second decision device, then the trial is succeed, otherwise the a new trial will be restarted with a penalty time TP (ms) and; a initial sampling point at the matched filter is assumed uniformly distributed.

[0017] One feature of the method according to the present invention is that the best sample of the N over-samples (Y1(k), Y2(k), . . . YN(k)) has a maximum sampling value of the N over-samples in the absence of the interference.

[0018] The present invention also provides a cell search apparatus for the CDMA system, especially for W-CDMA system, dealing with the three-stage cell search processes, wherein in the first stage, a PSCH is used for slot synchronization; in the second stage, a SSCH is used for frame/code group identification based on the slot boundary reported from the first stage; and in the third stage, a CPICH is used for determining the downlink scrambling code. The cell search apparatus according to the present invention comprises a chip matched filter, used for detecting incoming signals; a sampling device connected to the chip matched filter, used for over-sampling the incoming signals by N times against a chip rate; a down-sample device connected to the sampling device, used for transmitting the N over-samplers; a serial test unit connected to the sampling device and controlled by a first stage detector, used for selecting a best sample of the N over-samples; the first stage detector connected to the serial test unit, used for slot synchronization; a second stage detector connected to the serial test unit, used for frame boundary and code group synchronization; a third stage detector connected to the serial test unit, used for determining a scrambling code; and a identification unit connected to the third stage detector, used for determining the trial succeed or not.

[0019] The identification unit further comprises a comparator connected to the third stage detector, used for testing the output of the third stage detector against a threshold η0; a first decision device connected to the comparator, used for determining the scrambling code; wherein if the threshold is exceeded, then the selected scrambling code goes for a synchronization verification unit, otherwise the trial is considered to be failure, and a new trial will be restarted without penalty; and the synchronization verification unit connected to the first decision device, used for verifying the scrambling code; and a second decision device connected to the synchronization verification unit, used for determining the acceptance of the scrambling code; wherein if the scrambling code passes the second decision device, then the trial is succeed, otherwise the a new trial will be restarted with a penalty time TP (ms) and ; a initial sampling point at the matched filter is assumed uniformly distributed.

[0020] One feature of the apparatus according to the present invention is that the best sample of the N over-samples (Y1(k), Y2(k), . . . YN(k)) has a maximum sampling value of the N over-samples in the absence of the interference.

[0021] Other objects, advantages, and novel features of the invention will become more apparent from the following detailed description when taken in conjunction with the accompanying drawing.

BRIEF DESCRIPTION OF THE DRAWINGS

[0022] FIG. 1 shows a simplified frame structure for the 3GPP WCDMA/FDD system.

[0023] FIG. 2 shows a traditional serial cell search method for the 3GPP WCDMA/FDD system. (In the case, 10 (ms) processing time for each stage is assumed)

[0024] FIG. 3 shows a traditional pipelined cell search method for the 3GPP WCDMA/FDD system. (In the case, 10 (ms) processing time for each stage is assumed)

[0025] FIG. 4 shows a new cell search method using serial test in first stage (STS1) method according to the present invention.

[0026] FIG. 5 shows a flowchart of the new cell search method using serial test in first stage (STS1) method according to the present invention.

[0027] FIG. 6 is the architecture of the cell search system using serial test in first stage (STS1) method according to the present invention.

[0028] FIG. 7 shows a signal model used in the present invention.

[0029] FIG. 8 shows the drawing of comparing the different cell search method with the effects of non-ideal sampling.

DESCRIPTION OF THE PREFERRED EMBODIMENT

[0030] The main idea of STS1 method is that the selection of the best sample has a maximum sampling values of the N over-samples. And, the selection of the best sample in the first stage improves the performance which non-ideal sampling is taken into account.

[0031] Now referring to FIG. 4, it shows the operation of the STS1 method according to the embodiment of the present invention. According to the present invention, 512 scrambling codes are used to differentiate different cells in the downlink and are reused all over the system. The codes are divided into 64 groups with 8 codes in each group. Each code is 38400-chip long and hence extends over a frame. Since the cell sites are not synchronized, the codes always begin their new period at the frame boundaries. As shown in FIG. 4, the method deals with three-stage processes. The first stage 410, having a primary synchronization channel (PSCH), is used for slot synchronization. By using the same PSCH for each cell and by transmitting PSCH at the slot boundaries only, slot synchronization can be easily achieved by synchronization to PSCH. Furthermore, a generalized hierarchical Golay sequence is employed as the primary synchronization code (PSC) for easy implementation. The second stage 420, having a SSCH, is used for frame/code group identification after slot synchronization. Frame synchronization and code group identification can be achieved by detecting the secondary synchronization channel (SSCH), which is spread by one of the 16 orthogonal spreading codes, called the secondary synchronization codes (SSCs). The SSCs are orthogonal to PSC in order to reduce mutual interference. In addition, to facilitate fast frame/code group identification, the SSCH is farther encoded into a set of 64 code words by using a (15,3) comma-free Reed Solomon code (CFRS), with each codeword in the set representing a code group. Because of the nice property of comma free, the frame synchronization is accomplished once the code group is identified. The third stage 430, having a common pilot channel (CPICH), is used for the determination of the downlink scrambling code. After the code group is identified, the scrambling code can be determined easily by selecting one of the 8 codes in the group by using CPICH.

[0032] FIG. 5 shows a flowchart of the new cell search method using the serial test in first stage (STS1) method according to the embodiment of the present invention. To describe the flowchart, FIG. 6 is also needed. FIG. 6 is the architecture of the cell search apparatus according to the present invention. As shown, the architecture is used to accommodate the serial test in first stage (STS1) method. A cell search apparatus 200 for the CDMA system, especially for W-CDMA system, comprises a chip matched filter 210, a sampling device 220 connected to the matched filter 210, a down-sample device 230 connected to the sampling device 220, a serial test unit 245 connected to the down-sample device 230, a first stage detector 240 connected to the serial test unit 245, a second stage detector 250 connected to the serial test unit 245, a third stage detector 260 connected to the serial test unit 245, a identification unit 280 connected to the third stage detector 260. The identification unit 280 further comprises a comparator 270 connected to the third stage detector 260, a first decision device 275 connected to the comparator 270, a synchronization verification unit 290 connected to the first decision device 275, and a second decision device 295 connected to the synchronization verification unit 290.

[0033] Now referring to FIG. 5 and FIG. 6. In step 600, the chip matched filter 210 in front of the cell search apparatus 200 matches the incoming signals 190. In step 610, the sampling device 220 over-samples the incoming signal 190 N times against the chip rate and the N over-samples in each chip duration labeled by Y1(k), Y2(k), . . . YN(k) are fed into the down-sample device 230. In step 615, the down-sample device 230 transmits the N over-samples into the serial test unit 245 and the serial test unit 245 controlled by the first stage detector 240 selects a best sample of the N over samples. In step 620, the serial test unit 245 transmits the best sample into the second stage detector 250 and the third stage detector 260. The block 411 at the first stage, the block 421 at the second stage and the block 431 at the third stage (see FIG. 4 ) test the same selected best sample in which the best sample has the maximum sampling value among the N over-samples so as to reduce the adverse effect of non-ideal sampling. Once the trial fails, the next trial will be tested all over again by the random selected sample at hand. The block at the first stage, the block at the second stage and the block at the third stage (see FIG. 4 ) test another the same sample. The trial will still process unless the scrambling code is accepted by the synchronization verification unit processes. Furthermore, the selecting of the sampling points is carried out serially and cell search time of (N+2)×10 (ms) is required for each trial, and, of course, (K+N+1)×10 (ms) is required for each search if the cell search is succeeded at Kth trial.

[0034] In step 630, the first stage detector 240 is used for slot synchronization. A non-coherent type matched filter, in which the matched filter is divided into several small segments and the outputs of each segment are combined with their absolute values, is partitioned into four (4) segments and used as the first stage detector 240 for slot synchronization in the first stage. The first stage detector uses a generalized hierarchical Golay sequence as a primary synchronization code (PSC) in a primary synchronization channel (PSCH). Non-coherent accumulation over fifteen (15) slots is found appropriate to additive white Gaussian noise. More than one slot boundary may be selected in the first stage to go through the next stages for a better performance.

[0035] In step 640, the code group and frame synchronization can be accomplished in the second stage, after slot synchronization. The second stage detector 250 uses sixteen (16) matched filters for the detection of the sixteen (16) SSCs. Coherent accumulation results from the channel estimation coming out from the first stage. After collecting fifteen (15) decisions, they are correlated with the sixty-four (64) CFRS code words, each with fifteen (15) possible cyclic shift positions. This results in 960 correlation values. And finally, the code group and cyclic shift position associated with the maximum value are identified as the desirable code group and frame boundary, respectively. In step 650, in the third stage, a scheme is employed to detect the desirable scrambling code out of the eight (8) codes according to the code group identified in the second stage. Basically, in the third stage, the third stage detector 260 selects a scrambling code in the third stage. The third stage detector 260 is actively correlated with eight (8) possible scrambling codes and votes for the possible candidate by selecting the maximum value out of the eight (8) codes once in a 256-chip (symbol) duration. Finally, after 150 symbols (one frame), the maximum ballot is transmitted to a identification unit 280 connected to the third stage detector 260 for determining the trial succeed or not. The maximum ballot is tested against the threshold η0. The threshold is set up with the constant false alarm rate. In step 660, the output 261 of the third stage detector 260 is tested against a threshold η0 271 by the comparator 270. The first decision device 275 after the comparator 270 determines the scrambling code true or false. If the threshold is not exceeded, the trial is considered to be failure, and a new trial will be restarted without penalty. And, if the threshold is exceeded, then the selected scrambling code goes for the synchronization verification unit 290.

[0036] In step 670, the second decision device 295 after the synchronization verification unit 290 verifies the selected scrambling code, wherein if the scrambling code passes the second decision device 295, then the trial is succeed, otherwise the a new trial will be restarted with a penalty time TP (ms) and ; a initial sampling point at the matched filter is assumed uniformly distributed. We assume that the false alarm in the verification unit processes is negligible. Thus, the penalty time paid for the false alarm in the second decision device 295 is TP (ms).

[0037] Extensive computer simulations are used to evaluate the performance of new cell search algorithms comparing to the traditional method. The emphasis will be on the effects of non-ideal sampling. The simulation results are under the conditions: N=2, the maximum Doppler shift is 185.2 Hz (100-km/hr), the processing time for each stage detection is 10 (ms), TP=250 (ms) and η0 is set with a false alarm of 10−5. In addition, the transmit powers of the physical channels are distributed as follows. First, PSCH and SSCH have the same power, and the power ratio of CPICH and SCH (PSCH+SSCH) is unity. Second, the power of CPICH is 10% to the total transmit power. In other words, during cell search, 80% of the transmit power is contributed to the intra-cell interference. And finally, a geometry factor G=(PI+Ppsc+Pssc+Ppc)/Px□ is used to model the location of UE in a cell. The higher the G, the closer the UE from the desired base station. Cumulative distribution function (CDF) of search time is the performance index used for evaluating different search algorithms.

[0038] Referring to FIG. 7, it depicts a signal model used in the present invention with the effects of non-ideal sampling. Using a base-band representation, the received signal r(t) is given by 1$r\left(t\right)=\sum _{i=-\infty }^{\infty }\left\{\left[\sqrt{P_{\mathrm{psc}}}{c}_{\mathrm{psc}}\left(i\right)+\sqrt{P_{\mathrm{ssc}}}{c}_{\mathrm{ssc}}\left(i\right)+\sqrt{P_{\mathrm{pc}}}{c}_{\mathrm{pc}}\left(i\right)\right]h\left(t-{\mathrm{iT}}_e-\tau \right)+\sqrt{P_l}{n}_l\left(t\right)\right\}g\left(t\right)+\sqrt{P_x}{n}_x\left(t\right)$

[0039] where Ppsc, Cpsc, Pssc, Cssc and Ppc, Cpc are the power and spreading codes of the PSCH, SSCH and CPICH, respectively, g(t) is the complex-valued Rayleigh fading gain, h(t) is a square root raised cosine shaping function with roll-off factor 0.22, Tc is the chip duration τ is the initial random delay and is modeled as a random variable with uniform distribution over (−0.5Tc, 0.5Tc). In addition, PI and PX are the power of the intra-cell interference nI(t) and inter-cell interference nX(t), respectively, where nI(t) and nX(t) are modeled as zero mean additive white Gaussian noise with unity variance. Three observations on the modeling of above are worthy mentioning here. First, for simplicity, only flat fading channels are considered, and only the channels relevant to the cell search are treated explicitly; all other channels are included either in the interference terms nI(t) or nX(t). Second, τ denotes the effects of non-ideal sampling, which have been neglected in prior art. Third, it is assumed that the effects of non-ideal sampling and the carrier drift are due to the same source of frequency instability of oscillators.

[0040] FIG. 8 shows the effects of non-ideal sampling, where “ideal” is used to denote the ideal sampling. As can be seen, non-ideal sampling results in significant performance degradation, especially for the low signal noise ratio SNR case (a small G)—a case of practical interest. In the figure, 90% of occasions are finished in 110 (ms) for all the algorithms for G=6 (dB). On the other hand, STS1 accomplishes 90% of the searches in 300 (ms), but it takes more than 450 (ms) for the conventional algorithm under G=0 (dB). Clearly, STS1 is the one of choice under the condition of non-ideal sampling. Note that at the very low percentile of the CDF, the conventional algorithm performs better; this can be attributed to that in very few occasions the sampling in the conventional algorithm happens to be very close the optimum one.

[0041] Accordingly, the cell search method for the CDMA system of the present invention significantly reduces the effects of non-ideal sampling in the CDMA system so as to accomplish fast cell search. The cell search apparatus for the CDMA system of the present invention can realize the cell search under the effects of non-ideal sampling without increasing the hardware complexity. It should be appreciated that the method and apparatus according to the present invention can also be applied to the mobile equipment and wireless PDA system.

[0042] Although the invention has been explained in relation to its preferred embodiment, it is not used to limit the invention. It is to be understood that many other possible modifications and variations can be made by those skilled in the art without departing from the spirit and scope of the invention as hereinafter claimed.

Claims

1. A cell search method for a CDMA system for a three-stage cell search procedure, comprising the steps of: matching incoming signals by a chip matched filter; over-sampling the incoming signals N times against a chip rate by a sampling device; transmitting the N over-samples into a serial test unit by a down sampler; selecting a best sample among N over-samples by the serial test unit controlled by a first stage detector in the first stage; and transmitting the best sample of the N over-samples into a second stage detector and a third stage detector to accomplish a trial.

2. A cell search method as claimed in claim 1, wherein the trial of the cell search method further comprises the steps of: achieving a slot synchronization in the first stage by a first stage detector; achieving a code group and frame synchronization in the second stage by a second stage detector; selecting a scrambling code in the third stage by the third stage detector; testing the scrambling code against a threshold η0 by a comparator; determining the scrambling code by a first decision device after the comparator; wherein if the threshold is exceeded, then the selected scrambling code goes for a synchronization verification unit, otherwise the trial is considered to be failure, and a new trial will be restarted without penalty; and determining the scrambling code by a second decision device after the synchronization verification unit; wherein if the scrambling code passes the second decision device, then the trial is succeed, otherwise the a new trial will be restarted with a penalty time TP (ms) and ; a initial sampling point at the matched filter is assumed uniformly distributed.

3. A cell search method as claimed in claim 1, wherein the best sample of the N over-samples (Y1(k), Y2(k), . . . YN(k)) has a maximum sampling value of the N over-samples in the absence of the interference.

4. A cell search method as claimed in claim 2, wherein achieving the slot synchronization further comprises the step of: the first stage detector uses a generalized hierarchical Golay sequence as a primary synchronization code (PSC) in a primary synchronization channel (PSCH).

5. A cell search method as claimed in claim 2, wherein achieving a code group and frame synchronization further comprises the step of: the second stage detector uses 16 secondary synchronization codes (SSCs) orthogonal to the PSC in a secondary synchronization channel (SSCH).

6. A cell search method as claimed in claim 5, wherein the SSCH is further encoded into a set of 64 code words by using a (15,3) comma-free Reed Solomon code (CFRS), with each codeword in the set representing a code group, to identify a frame boundary and code group.

7. A cell search method as claimed in claim 2, wherein the scrambling code is determined by selecting one of the 8 codes by in a common pilot channel (CPICH) after the code group and frame boundary is identified.

8. A cell search method as claimed in claim 2, wherein the selecting of the sampling points is carried out serially and cell search time of (N+2)×10 (ms) is required for each trial, and, of course, (K+N+1)×10 (ms) is required for each search if the cell search is succeeded at Kth trial.

9. A cell search method as claimed in claim 2, wherein the first stage detector uses a non-coherent combined matched filter for slot synchronization in the first stage.

10. A cell search method as claimed in claim 4, wherein the first stage detector selects more than one slot boundary candidate of the primary synchronization code (PSC) for the slot synchronization, although only one candidate will be considered.

11. A cell search method as claimed in claim 5, wherein the second stage detector uses 16 matched filters for the detection of the secondary synchronization codes (SSCs) for the code group and frame synchronization.

12. A cell search method as claimed in claim 6, wherein the code group and frame boundary use a code group and cyclic shift position associated with the maximum value in the second stage, respectively.

13. A cell search method as claimed in claim 2, wherein the threshold η0 is preset according to the constant false alarm rate.

14. A cell search method as claimed in claim 1, wherein the cell search method is used for wideband code division multiple access, frequency division duplex system (W-CDMA/FDD).

15. A cell search method as claimed in claim 1, wherein the cell search method is used for the mobile equipment and wireless PDA system.

16. A cell search apparatus for a CDMA system for a three-stage (a first stage, a second stage and a third stage ) cell search procedure, comprising: a chip matched filter, used for matching incoming signals; a sampling device connected to the chip matched filter, used for over-sampling the incoming signals by N times against a chip rate; a down-sample device connected to the sampling device, used for transmitting the N-over samplers; a serial test unit connected to the sampling device and controlled by a first stage detector, used for selecting a best sample of the N over-samples; the first stage detector connected to the serial test unit, used for slot synchronization; a second stage detector connected to the serial test unit, used for frame boundary and code group synchronization; a third stage detector connected to the serial test unit, used for determining a scrambling code; and a identification unit connected to the third stage detector, used for determining the trial succeed or not.

17. A cell search apparatus as claimed in claim 16, wherein the identification unit further comprises: a comparator connected to the third stage detector, used for testing the output of the third stage detector against a threshold η0; a first decision device connected to the comparator, used for determining the scrambling code; wherein if the threshold is exceeded, then the selected scrambling code goes for a synchronization verification unit, otherwise the trial is considered to be failure, and a new trial will be restarted without penalty; and the synchronization verification unit connected to the first decision device, used for verifying the scrambling code; and a second decision device connected to the synchronization verification unit, used for determining the acceptance of the scrambling code; wherein if the scrambling code passes the second decision device, then the trial is succeed, otherwise the a new trial will be restarted with a penalty time TP (ms) and ; a initial sampling point at the matched filter is assumed uniformlly distributed.

18. A cell search apparatus as claimed in claim 16, wherein the best sample of the N over-samples (Y1(k), Y2(k), . . . YN(k)) has a maximum sampling value of the N over-samples in the absence of the interference.

19. A cell search apparatus as claimed in claim 17, wherein the first stage detector uses a generalized hierarchical Golay sequence as a primary synchronization code (PSC) in a primary synchronization channel (PSCH).

20. A cell search apparatus as claimed in claim 17, wherein the second stage detector uses 16 secondary synchronization codes (SSCs) orthogonal to the PSC in a secondary synchronization channel (SSCH).

21. A cell search apparatus as claimed in claim 20, wherein the SSCH is further encoded into a set of 64 code words by using a (15,3) comma-free Reed Solomon (CFRS) code, with each codeword in the set representing a code group, to identify frame boundary/code group.

22. A cell search apparatus as claimed in claim 17, wherein the scrambling code can be determined by selecting one of the 8 codes by in a common pilot channel (CPICH) after the code group and frame boundary is identified.

23. A cell search apparatus as claimed in claim 16, wherein the selecting of the sampling points is carried out serially and cell search time of (N+2)×10 (ms) is required for each trial, and, of course, (K+N+1)×10 (ms) is required for each search if the cell search is succeeded at Kth trial.

24. A cell search apparatus as claimed in claim 17, wherein the first stage detector uses a non-coherent combined matched filter for slot synchronization in the first stage.

25. A cell search apparatus as claimed in claim 19, wherein the first stage detector selects more than one slot boundary candidate of the primary synchronization code (PSC) for the slot synchronization, although only one candidate will be considered.

26. A cell search apparatus as claimed in claim 20, wherein the second stage detector uses 16 matched filter for the detection of the SSCs.

27. A cell search apparatus as claimed in claim 21, wherein the code group and frame boundary use a code group and cyclic shift position associated with the maximum value in the second stage, respectively.

28. A cell search apparatus as claimed in claim 17, wherein the threshold η0 is preset according to the constant false alarm rate.

29. A cell search apparatus as claimed in claim 16, wherein the cell search method is used for wideband code division multiple access, frequency division duplex system (W-CDMA/FDD).

30. A cell search apparatus as claimed in claim 16, wherein the cell search apparatus is used for the mobile equipment and wireless PDA system.