Weighted decoding method and circuits for comma-free reed-solomon codes

Description

FIELD OF THE INVENTION

[0001] The invention generally relates to a weighted decoding method and circuits for Comma-Free Reed-Solomon codes that applies to a cell search in a wideband code division multiple access system, and more particularly relates to a weighted decoding method that is applied to the decoding circuit that is based on a systolic array and a decoding circuit that is based on a folding systolic array.

BACKGROUND OF THE INVENTION

[0002] As regards the third generation partnership project (hereinafter referred as the 3GPP) wideband code division multiple access (hereinafter referred as W-CDMA) system, the cell search process employed by a cellular phone has to execute a series of detections and testing procedures of synchronization codes in order to synchronize the cellular phone with its best base station. In general, the cell search algorithm of the 3GPP W-CDMA is divided into three ordered steps, namely 1) slot synchronization, 2) frame synchronization and code-group identification, and 3) scrambling-code determination. The first step involves using a primary synchronization code (hereinafter referred as PSC) to achieve slot synchronization. The second step involves using both the secondary synchronization codes (SSCs) and the Comma-Free Reed-Solomon (hereinafter referred as CFRS) codes to achieve frame synchronization and code-group identification. The third step involves using all the possible scrambling codes of the identified code group to execute a de-scrambling procedure for scrambling-code determination. Before achieving the synchronization, the cellular phone couldn't begin to decode various channel messages broadcasted by the base station or measure various channel effects.

[0003] In this invention, we are concerned with that the second step of the synchronization procedure involves decoding the CFRS code so as to determine the frame boundary and code group. The CFRS code is a combination of Comma-Free code and Reed-Solomon code and thus it has the features of both Comma-Free code and Reed-Solomon code. The Comma-Free code has a feature that no new Comma-Free code could be created by combining any two Comma-Free codes. It is capable of both error detection and synchronization simultaneously. The synchronization capability of Comma-Free code is used by the W-CDMA to achieve frame synchronization. Generally, a Comma-Free code decoding circuit is composed of simple correlators. However, in the 3GPP W-CDMA, the Comma-Free code is transmitted intermittently as opposed to the continual transmission in other applications. Therefore, an ordinary Comma-Free code decoder is not applicable to the synchronization procedure of the 3GPP W-CDMA.

[0004] The CFRS code is also a (15, 3) Reed-Solomon (hereinafter referred as RS) code with a powerful error correction feature that is essential to the 3GPP W-CDMA. As regards the design of a RS decoder, it has been discussed in detail in many articles, and the most widely used decoding method could be summed up in four steps: 1) syndrome calculation of the received codeword, 2) error locator polynomial and error evaluator computations, 3) error location identification, and 4) error value calculation. However, in the 3GPP W-CDMA, the set of CFRS codes is composed of 64 special RS codes and thus an ordinary RS decoder is unsuitable.

[0005] The 3GPP W-CDMA uses 64 CFRS codes to represent 64 code groups. Each codeword consists of 15 symbols. Every code group includes eight scrambling codes. Every base station treats one of the eight scrambling codes of the code group to which it belongs as its scrambling code for differentiating itself from the other stations. To be connected through a certain base station, a cellular phone has to achieve scrambling code synchronization with the base station first. As described above, the synchronization process has to comprise the decoding procedure of CFRS code with a view to executing frame boundary detection and code-group determination.

[0006] The symbols of CFRS code are chosen from the elements of Galois Field (GF) (16). Among the 16 secondary synchronization codes, the nth synchronization code is transmitted via a secondary synchronization channel represents that the nth element of GF (16) is regarded as the code symbol. Symbols are selected to form 64 CFRS codes to represent 64 code groups, each of which consists of 15 symbols. To transmit CFRS codes, a base station sends identical codes in each frame. However, prior to frame synchronization, the initial position for the receipt of CFRS codes is not necessarily occupied by the first code symbol, but can be one of the 15 code symbols. Since the same CFRS code is transmitted in each frame, it is possible to receive 15 code symbols continuously even before a frame boundary is determined. The consecutive received 15 code symbols result in a cyclic-shift version of a CFRS codeword. The aim of decoding is to identify the received codeword is which one of the 64 CFRS codes in order to determine the code group, and to detect the cyclic-shift time cyclic-shift so as to determine the frame boundary.

[0007] cyclic-shift

[0008] According to the above description, there are 64 CFRS codes and 15 cyclic-shift versions of each codeword. Thus there can be a total of 960 (64×15) combinations of cyclic-shift codeword. The determination of these 960 cyclic-shift hypotheses requires a lot of complicated calculations, and more importantly, timely accomplishment of these calculations in order to avoid delay in the synchronization procedure that may cause more serious problems. Hence, decoding speed has a direct impact on real-time synchronization. A “fast” CFRS decoder is an indispensable component for the cell search algorithm.

[0009] The decoding methods discussed in the existing literature regarding the mathematical calculations of synchronization merely involve direct comparison of these 960 combinations, thus there is no other better decoding method, nor is there any effective architecture for hardware implementation. Therefore, it is a good idea to use the direct decoding method based on the 960 combinations to develop decoding hardware architecture that works more effectively, as well as to put forward a decoding method that is more accurate.

SUMMARY OF THE INVENTION

[0010] In view of the aforesaid technical problems, the invention provides a kind of speedy decoding circuit architecture that applies to the CFRS code of the 3GPP W-CDMA, not only to support various kinds of cell search algorithms, but also to meet the demand for frequent decoding in multi-candidate cell search algorithms.

[0011] As regards the cell search procedure of the 3GPP W-CDMA, there are many options for the execution strategies. For example, serial execution of the three steps of the synchronization procedure, simultaneous execution of the synchronization procedure in a pipelined manner, or selecting several slot boundary candidates in step 1 and simultaneously executing steps 2 & 3 by means of the various slot boundary candidates are all possible and feasible. Each algorithm requires different CFRS decoding speed. Serial cell search does not require fast decoding. Pipelined cell search requires a decoder with high decoding speed. The multi-candidate method requires a decoder with extremely high decoding speed because of the need for frequent decoding. The invention provides the means to solve the aforesaid problems, i.e. the decoding circuits that work at high or low decoding speeds. Whenever a high decoding speed is required, it could meet the need of the cell search algorithm. Whenever a low decoding speed is acceptable, it executes decoding with the minimal number of components so as to reduce power consumption.

[0012] Another object of the invention is to provide a weighted CFRS decoding method wherein 960 possible combinations of CFRS codes are decoded by means of weighted correlation after 15 code symbols have been completely received.

[0013] As regards the decoding method put forward by the invention, its circuit is based on the concept of systolic array architecture. This architecture is characterized by speedy decoding and thus it meets various demands of cell search algorithms. In addition, with the folding technology of the systolic array architecture, the weighted decoding method put forward by the invention reduces the area of its array without jeopardizing its speedy decoding feature.

[0014] To gain further insight into the characteristics and the implementation of the invention, illustration and detailed explanation of the preferred embodiment are provided below:

BRIEF DESCRIPTION OF THE DRAWINGS

[0015]
FIG. 1 depicts the comparison of the efficiency of CFRS decoding at different decoding lengths;

[0016]
FIG. 2 shows the coherent decoding process (step 2) of the synchronization procedure; the reference value for coherent detection is the decoding result of the same chip of the primary synchronization decoder;

[0017]
FIG. 3 shows a CFRS decoder based on SA architecture, wherein xi denotes the received code symbol and yj,i denotes the result of a correlating comparison;

[0018]
FIG. 4 depicts the structure of IPG;

[0019]
FIG. 5 depicts the circuit of WPE;

[0020]
FIG. 6 depicts the circuit of BPE;

[0021] FIGS. 7A-7C shows the table of 64 sets of CFRS codes and the table of its segments after folding;

[0022]
FIG. 8 shows a CFRS decoder based on the folding SA architecture; the number of times of folding is 2; the 64×15 array is reduced to a 16×15 array, but the decoding duration is still 94 cycles;

[0023]
FIG. 9 depicts the circuit of a refresh mechanism; and

[0024]
FIG. 10 compares the performance of weighted decoding with that of non-weighted decoding.

DETAILED EXPLANATION OF THE INVENTION

[0025] As shown in FIG. 1, the 960 combinations of CFRS codes can be decoded better with a complete set of 15 code symbols rather than with part of the code symbols. Hence, in the invention, after 15 code symbols have been received, a frame boundary is determined by making reference to a direct comparison of the 15 code symbols.

[0026] The decoding method adopted by the invention can be denoted in the following way:

\begin{matrix} (i, j) = Arg {\max {X_{i} \otimes H_{j}} \begin{matrix} i = 0 \sim 14 \\ j = 0 \sim 63 \end{matrix}} & (1) \end{matrix}

[0027] where i denotes the index for the 15 cyclic-shift versions, j denotes the index for the 64 CFRS codes, Xi denotes the result of i time(s) of rightward cyclic-shift of the CFRS codes received, while X0 denotes the result of arbitrary cyclic-shift of the received CFRS codes in the beginning; and, Xi={x1,x2,x3, . . . ,x15},xkεGF(16),1≦k≦15. Hj denotes one of the 64 CFRS codes, and Hj={h1,h2,h3, . . . ,h15},hkεGF(16),1≦k≦15. Finally, Xi{circle over (x)}Hj refers to making a correlating comparison.

[0028] In order to ensure the accuracy of the CFRS decoding results, the invention puts forward a weighted decoding method wherein decoding is carried out by means of weighted correlation after 15 code symbols have been completely received. FIG. 2 depicts this weighted correlation decoding method.

[0029] First of all, the determination of the secondary synchronization code (SSC) is carried out according to a coherent detection. The detection value of the primary synchronization code is regarded as a reference value. After coherently combining the detection values of 16 SSCs with that of the primary synchronization code (PSC), find the greatest value from the 16 results. The greatest value represents which CFRS code symbol transmitted. FIG. 2 depicts the coherent combination. With the SSC decoder 901 and the SSC decoder 902, the received SSC is decoded by means of 16 the secondary synchronization codes 903. The primary synchronization code result is multiplied by the decoding result of the two SSC decoders 901 & 902 through the multipliers 904 & 905. After the coherent combination, those results are added up by the adder 906 so as to acquire the coherent detection result of the 16 SSCs.

[0030] The CFRS codes determination element 907 determines the CFRS code symbol. After 15 code symbols have been received, the CFRS decoder 908 makes comparisons according to the 64×15 CFRS codes book 909, and outputs the maximum value of this cyclic-shift instance. This maximum decoding value, which is acquired by means of using coherent detection results as weights of symbols, is the so-called “weight” of this code symbol. While making a weighted correlating comparison, it is necessary to compare the code symbols first. If the code symbols are identical, then the weights of the code symbols are added to the weighted decoding results. Finally, all possible weighted decoding results are compared.

[0031] As regards the weighted correlating comparison method put forward by the invention, its decoding method may be denoted in the following way:

\begin{matrix} (i, j) = Arg {\max {X_{i} \cdot H_{j}} \begin{matrix} i = 0 \sim 14 \\ j = 0 \sim 63 \end{matrix}} & (2) \end{matrix}

[0032] where i denotes the index for the 15 cyclic-shift versions, j denotes the index for the 64 CFRS codes, Xi denotes the result of i time(s) of rightward cyclic-shift of the CFRS codes received, while X0 denotes the result of arbitrary cyclic-shift of the CFRS codes received in the beginning; and, Xi={x1,x2,x3, . . . ,x15},xkεGF(16),1≦k≦15. Hj denotes one of the 64 CFRS codes, and Hj={h1,h2,h3, . . . ,h15},hkεGF(16),1≦k≦15. Finally, Xi{circle over (x)}Hj refers to making a weighted correlating comparison that may be denoted in the following way:

\begin{matrix} X_{i} \cdot H_{j} = \sum_{k = 1}^{15} w_{k} (x_{k} \otimes h_{k}) & (3) \end{matrix}

[0033] where k denotes the index of the code symbol in a correlating comparison, wk denotes the weight of the code symbol, xk. As mentioned above, wk is the maximum detections value of SSC. xk{circle over (x)}hk refers to making a correlating comparison between xk and hk. If they are found to be identical, then the result is 1. If they are found to be different, then the result is 0. The judgment may be converted into the logical gate of exclusive-OR, XOR, as far as hardware is concerned.

[0034] The aforesaid weighted decoding method is carried out by executing the weighted correlating comparison denoted by equation (3), and finding the greatest weighted correlation value from 960 ones of its kind. The cyclic-shift index i, and the index of the 64 sets of CFRS codes saved beforehand, j, of this value is the decoding result. The index of the 64 CFRS codes saved beforehand, j, determines the code group, while the cyclic-shift index i determines the frame boundary, so as to complete step 2 of the synchronization procedure of 3GPP W-CDMA.

[0035] According to equation (2), the decoding method put forward herein is to carry out rightward cyclic-shift again with respect to the arbitrary cyclic-shift result of the received CFRS codes. Only 14 times of cyclic-shift are required—much less than the 64×14 times of cyclic-shift performed for each of the 64 sets of CFRS codes saved beforehand. As a result, power consumption for the execution of a circuit is reduced.

[0036] The aforesaid weighted decoding method is summed up as follows:

[0037] 1. Preceding Procedure

[0038] First of all, the determination and assumption of SSC is carried out according to a coherent decoding strategy. The decoding value of the primary synchronization code is regarded as a reference value. Acquire 16 results by coherently combining the decoding value of 16 SSC with that of the primary synchronization code (PSC).

[0039] Then, find the greatest value from the 16 sets of results. The position of the greatest value among the 16 results represents the position of the transmitted CFRS code symbol among the elements of GF (16). The greatest value acquired after the coherent detection is regarded as the weight of this code symbol.

[0040] 2. Weighted Decoding Procedure

[0041] First of all, make a correlating comparison between the received CFRS codes and the 960 combinations. If the code symbols are identical, the weights of these code symbols are added to the weighted correlating comparison results. If they are different, nothing should be done.

[0042] Find the greatest correlating value from all the 960 correlating comparison results. The position of the code of the greatest result among the 64 sets of CFRS codes indicates the intended code group. The number of times of its cyclic-shift represents the frame boundary.

[0043] To implement the aforesaid decoding method, SA architecture and folding SA architecture may be adopted. Explanation is provided below:

[0044] I. The Decoding Circuit Based on SA Architecture

[0045] As shown in FIG. 3, the decoding circuit comprises several components, namely an Input Pattern Generator (IPG) 10, a 64×15 Weighted Processing Element Array (WPEA) 20, and a 64×1 Boundary Processing Element Array (BPEA) 30.

[0046] The main function of the IPG 10 is to generate 15 cyclic-shift versions from the arbitrarily cyclic-shift CFRS codes, X, wherein Xi,0≦i≦14, then output the cyclic-shift versions to the WPEA 20 in a skewed-form required by the SA architecture. The output comprises all 15 -cyclic-shift versions, and, as shown in FIG. 4, its circuit structure comprises a shift register 11, a Johnson counter 13, and a masking combinational circuit 12. Its process is described below.

[0047] First of all, a code received, X, is loaded to the shift register 11. The shift register 11 then generates its 15 cyclic-shift versions separately. The Johnson counter 13 generates the required skewed mask. With the masking combinational circuit 12, the 15 cyclic-shift versions are turned into the required 15 cyclic-shift versions in skewed-form generated by the skewed mask, and the latter is regarded as the input to the underlying WPEA 20 that is based on SA architecture.

[0048] The WPEA 20 is a kind of SA architecture composed of 64×15 WPE. The structure of each WPE is shown in FIG. 5, and the primary function of the WPE is to make a weighted correlating comparison. IPG 10 generates all the 15 cyclic-shift versions that are then compared with 64 CFRS codes.

[0049] There are four registers in each WPE, namely an H register 202, X register 201, Y result register 203 and W register 206. The H register 202 is for storing the code symbols, hk, of CFRS code beforehand, or, in other words, the 64×15 CFRS code symbol table 40 shown in FIG. 7A is put in the H register 202 of every WPE of the 64×15 WPEA 20 by mapping. Since there are 64 CFRS codes, and each set consists of 15 code symbols, it is processed by a 64×15 WPEA 20. The X register 201 is for storing the code symbols, Xk, sent by the overhead WPE. Given the comparison combinational circuit 204 shown in FIG. 5, when the two code symbols stored in the X register 201 and the H register 202, respectively, are identical, and given the switch function of the choice element 207, the weight wk of xk saved temporarily in the W register is added to the result of the weighted correlating comparison. The sum is then sent to the WPE on the right. In the meanwhile, the received code symbol xk that has been compared as well as its weight wk are sent to the underlying WPE.

[0050] The weighted computation is as follows. The SSC decoding result of the first half of step 2 of the synchronization procedure is treated as the input to a weight, that is, wk. Make a correlating comparison between the received code symbol and the code symbol that is saved in advance. If they are identical, that is, xk=hk, then conduct a switch by means of the choice element 207, by adding the weight wk of xk saved temporarily in the W register 206 to the result of the weighted correlating comparison.

[0051] The 64×15 WPEA 20 connects to a set of 64×1 BPEA 30 that lies on its right. Each row of the WPEA 20 is in charge of making the correlating comparison with a—CFRS codes. The last WPE output at the right end of each row is the result of correlating comparison, yj,i. A BPE compares this comparison results, yj,i. After comparing the 15 results of the same code, each BPE begins to compare the results of individual rows, as shown in FIG. 6.

[0052] The I comparator 301 is for comparing the results of the same row. A result that is greater is stored in the maximum value and cyclic-shift index register 306 first. After the maximum result of the same row has been generated, the J comparator 302 begins to compare the maximum value of the existing row with that of the overhead row, selects the greater one, saves it and its code group index j and cyclic-shift index i in the maximum value and group & cyclic-shift index register 308, and sends them to the underlying BPE. The multiplexer 305 makes reference to the result of I comparator 301, and saves the greater result value, yj,i, and its cyclic-shift index i in the maximum value and cyclic-shift index register 306. In the event that the new yj,i value is greater than the yj,max value, that is, when the multiplexer 305 chooses line 1, then its cyclic-shift index i is provided by the cyclic-shift index 303. The multiplexer 307 makes reference to the result of J comparator 302, and saves the greater result value, yj,i, its cyclic-shift index i and code group index j, etc., in the maximum value and group & cyclic-shift index register 308. In the event that the result of the existing row, yj,max, is greater than the yj−1,max value of the overhead row, then its group index is provided by the code group index 304.

[0053] The output of the BPE at the bottom (that is, the output result of the maximum value and group & cyclic-shift index register 308) is the result of decoding. The comparison result of the greatest correlation is the desired decoding result. The code group index j of this result denotes that the CFRS code received, X, is the jth code of all the possible 64 CFRS codes, while the cyclic-shift index i denotes that the CFRS code received, X, results from i times of cyclic-shift of the original code symbol. According to the “Background of the Invention” section of this document, in the 3GPP W-CDMA, the code group index, j, denotes a code group, while the cyclic-shift index, i, denotes a frame boundary. This is the end of step 2 of the cell search.

[0054] Please refer to FIG. 3 again. The whole decoding procedure is: load the CFRS codes received to IPG 10; the IPG 10 generates 15 cyclic-shift versions in order; input the 15 cyclic-shift versions in a skewed-form to the 64×15 WPEA 20; the WPEA 20 makes correlating comparisons between the 15 cyclic-shift versions and 64 codes saved beforehand; a comparison is made on each row with respect to one set of possible CFRS codes; as regards the comparison result of a row, the BPE of the row identifies the most probable cyclic-shift position; search vertically, that is, from the top to the bottom, and find the greatest result of cyclic-shift position of individual rows; the output of the BPE at the bottom is the decoding result.

[0055] The decoding duration of this decoding circuit is 94 cycles. From the point of view of a system, the chip rate is 3.84 MHz. In general, the minimum frequency designed by a circuit should be 3.84 MHz. In other words, the designed circuit only spends time for a maximum 94 chips on decoding. Thus, there is sufficient time to prepare for the following step of determining scrambling codes in an ordinary synchronization procedure. Even if decoding has to be executed ten or twenty times in multi-candidate cell search algorithms, this decoder is able to finish decoding before the end of the slot wherein the 15th secondary synchronization code is decoded. Thus it does not delay the execution of step 3 of the cell search procedure. The total time spent: 20 (number of times of decoding)*94(the duration of decoding)+256(receipt of the secondary synchronization code)=2136<2560 (slot length).

[0056] II. The Folding Technology and its Application to the Decoding Circuit Based on SA Architecture

[0057] As regards the aforesaid decoding circuit based on SA architecture, with the folding technology put forward by the invention, the SA of the WPEA 20 is folded so that its size is reduced from 64×15 to 32×15 and thus its array area is reduced. As regards the code symbols that have to be saved beforehand, the 64×15 CFRS code symbol table 40 (shown in FIG. 7A) is also folded and thus turned into 32×15, as shown in FIG. 7B, comprising an upper half segment 41 and a lower half segment 42. Similarly, the 64×1 BPEA 30 is also folded and thus turned into 32×1. Despite the folding, a decoding speed of 94 cycles is maintained. The folding process does impact the decoding speed, depending on the number of times of folding. Explanation is herein given on the folding technology in a case where, for example, folding is carried out twice successively. As shown in FIG. 8, after the WPEA 20 has been folded twice, its original 64×15 systolic array is diminished and turned into a 16×15 systolic array. The CFRS code symbol table that has to be saved beforehand for the WPEA 20 is, by the same token, turned into four segments, namely the first segment 43, the second segment 44, the third segment 45 and the fourth segment 46, as shown in FIG. 7C.

[0058] The maximum number of successive folds without jeopardizing decoding speed is two. FIG. 8 shows the architecture that results from the folding, wherein the IPG 10 and each WPE remain identical to their appearance shown in FIGS. 4, 5, respectively, thus they need not be modified. However, the BPE has to be slightly modified so that its architecture is capable of feedback, as shown in FIG. 6.

[0059] After two times of successive folding, the size of the WPEA 20 is reduced from 64×15 to 16×15, while the size of the BPEA 30 is also reduced from 64×1 to 16×1, thus the areas of both arrays decrease to one-fourth of their original sizes. The reduction in array area is achieved at the cost of the addition of a “refresh mechanism” for refreshing the code symbol, hk, stored beforehand in the H register of the WPE. In other words, after two times of successive folding, the 64×15 CFRS code symbol table 40 originally stored in advance in the WPEA 20 (shown in FIG. 7A) is divided into four segments 43, 44, 45 and 46, as shown in FIG. 7C. The first time only one segment is placed in the H register of the WPEA 20 shown in FIG. 8. The second time one segment is saved first. Then, after the previous segment has been decoded, the code symbol of the next segment is refreshed by the refresh mechanism and saved in the WPE.

[0060] A refresh cluster is shown in FIG. 8, and a code symbol is refreshed cluster by cluster. In other words, each time only one cluster is refreshed. The formation of clusters depends on the execution condition of SA. After each WPE has made a correlating comparison with the last code symbol it receives, it should immediately make a correlating comparison for the code symbol of the next segment in the next cycle. Hence, it is necessary to refresh the code symbol saved beforehand in individual WPEs, that is, to refresh the content value in the H register. Since the rows must be refreshed, and the WPEs are separated from each other by one position. All the WPEs that have to be refreshed in the same cycle are collectively known as a refresh cluster. Clusters are refreshed in order, so that all the code symbols that are saved beforehand in 16 rows and are required to be refreshed can be refreshed simultaneously. There is a total of 15 refresh clusters, namely the first refresh cluster, the second refresh cluster, . . . the 15th refresh cluster. They are refreshed in order, or, in other words, every cluster is refreshed every 15 cycles.

[0061]
FIG. 9 shows the refresh mechanism that comprises 15 refresh counters 501, a read circuit 502 of Read-Only Memory (ROM) 60, a refresh register 503 and a cluster allocation circuit 504. The 15 refresh counters 501 figure the time taken to refresh individual refresh clusters. Once the refresh time runs out, the refresh counters 501 trigger the read circuit 502 of Read-Only Memory (ROM) 60. The read circuit 502 then reads the code symbols of the next segment to be saved beforehand, and places the code symbols in the refresh register 503. The cluster allocation circuit 504 sends the new code symbols to the corresponding H register of the WPE (for saving CFRS code symbol beforehand). This is the way the refresh procedure is executed. The CFRS Read-Only Memory (ROM) 60 shown in FIG. 9 is the ROM that saves the 64×15 CFRS code symbol table 40, as shown in FIG. 7A.

[0062] In addition, as shown in FIG. 8, the code symbol received and sent downward by the 15th row of the WPEA 20 has to be sent to the first row as feedback because, in the meantime, second segment decoding has to be started in the first row. The direct feedback exempts IPG 10 from generating skewed input repeatedly whenever a new segment is to be decoded. As a result, power consumption is reduced.

[0063] Please refer to FIG. 8. The decoding process of the folding architecture is as follows: IPG 10 generates the 15 types of cyclic-shift results of the received CFRS codes in order, and outputs these cyclic-shift results in a skewed form to the 16×15 WPEA 20. In the meantime, the refresh mechanism loads and refreshes the code symbol of the first segment 43 shown in FIG. 7C according to the sequence of the refresh clusters. The WPEA 20 makes a correlating comparison between these 15 cyclic-shift versions and the refreshed segment code symbols. A correlating comparison is made on a set of possible CFRS codes for each row. As regards the correlating comparison result of a row, the BPE of the row finds the most probable result of cyclic-shift, searches vertically, that is, from top to bottom, and find the most probable result of cyclic-shift of individual rows. The output of the BPE at the bottom is the decoding result for the first segment. This result has to be sent to the first BPE as feedback, and be compared with the result of the next segment.

[0064] After the refresh procedure executed by the refresh mechanism on the first cluster and the lapse of the following 15 cycles, the correlating comparison between the code symbol 43 of the first segment and the cyclic-shift of the received code symbol is completed. The code symbol 43 of the first segment has to be refreshed and turned into the code symbol 44 of the second segment, so that the third segment 45 and the fourth segment 46 can be refreshed by analogy. The process does not end unless and until the correlating comparison of the fourth segment is completed. The output of the BPE at the bottom is the decoding result.

[0065] The process of mathematical calculation for the decoding of folding architecture is similar to that of the architecture previously put forward, except that the components in charge of execution are different. The 16×15 WPEA 20 and the 15×1 BPEA 30 shown in FIG. 7 are repeatedly used. The concept of refreshing the codes saved beforehand and the concept of the refresh mechanism are incorporated into the design. As a result, the area of a decoder is reduced, but the decoding speed of the decoder is not. The decoding time required by the decoder remains the same, that is, 94 cycles.

[0066] The CFRS decoding method put forward by the invention involves weighted correlating comparison. As shown in FIG. 10, given a weighted method, the effect of a decoding circuit can be increased by approximately 1˜0.5-dB.

[0067] While the invention has been described by way of example and in terms of a preferred embodiment, it is to be understood that the invention is not limited thereto. It should be understood that the detailed description and specific examples, while indicating preferred embodiments of the invention, are given by way of illustration only, since various changes and modifications within the spirit and scope of the invention will become apparent to those skilled in the art from this detailed description. Therefore the scope of protection for the invention should conform to the claims attached below.

Claims

1. A weighted decoding method for Comma-Free Reed-Solomon (CFRS) codes applied to a wideband code division multiple access (W-CDMA) system, which utilizes the detection of primary synchronization code (PSC) generated in a first step of the synchronization procedure of said system to decode said CFRS codes which comprise 64 code groups and each code group contains 15 code symbols; comprising the steps of: determining the secondary synchronization code: coherently combine 16 secondary synchronization codes with said primary synchronization code; calculating the greatest value of said 16 coherent secondary synchronization code detection results and treat it as the weighted value of said code symbol of said CFRS codes; making a weighted correlating comparison, which is to make a correlating comparison on said code symbols of said CFRS codes; when said code symbols of said CFRS codes are identical, then said weighted value is added to said code symbols of said CFRS codes; and performing 15 times of cyclic-shift for said CFRS codes received and making a correlating comparison to acquire a cyclic-shift index for determining a frame boundary.
2. The weighted decoding method of claim 1, wherein said cyclic-shift index is acquired by the equation:
3. A weighted decoding circuit for Comma-Free Reed-Solomon codes for receiving and decoding a Comma-Free Reed-Solomon (CFRS) code comprising: an input pattern generator, which receives the arbitrarily cyclic-shift CFRS codes for generating 15 cyclic-shift versions of CFRS codes and outputs to a systolic array in a skewed form; a weighted processing element array composed of 64×15 weighted processing elements, for receiving the skewed-form CFRS codes to make weighted correlating comparisons and output a weight correlating comparison result; and a boundary processing element array composed of 64×1 boundary processing elements, for calculating the greatest said correlating comparison result of a row of said systolic array to find greatest results of individual rows and set the acquired index as a decoding result.
4. The weighted decoding circuit of claim 3, wherein said input pattern generator is composed of a shift register, a Johnson counter and a masking combinational circuit; said shift register receives the 15 symbols of said received CFRS code and their corresponding weights, cyclically shifts them and outputs them as 15 cyclic-shift versions of said CFRS code and weights; said Johnson counter sends a mask counting signal to said combinational circuit; said masking combinational circuit outputs the skewed-form CFRS codes according to said mask counting signal.
5. The weighted decoding circuit of claim 3, wherein each of said weighted processing element is composed of a first register, a second register, a third register, a fourth register, a (XNOR-AND) combinational circuit, an accumulator and a choice element; said first register stores the symbols of said CFRS codes beforehand; said second register stores the received code symbols of said CFRS codes so as to facilitate their downward transmission; said third register stores said correlating comparison results; said fourth register stores a weighted value; said (XNOR-AND) combinational circuit compares two code symbols in said first register and said second register, respectively, to see whether they are identical; said choice element decides whether said weighted value is to be added to the final result according to said correlating comparison result; said accumulator adds said weighted value to a final result.
6. The weighted decoding circuit of claim 5, wherein said weighted value is acquired by means of the primary synchronization code generated in the first step of the synchronization procedure of the system of wideband code division multiple access, and said weighted value is acquired by the steps of: determining the secondary synchronization code: coherently combine 16 secondary synchronization codes with said primary synchronization code; and calculating the greatest value of said 16 secondary synchronization codes and treat it as said weighted value.
7. The weighted decoding circuit of claim 3, wherein each of said boundary processing element is composed of a first comparator, a second comparator and a combinational circuit; said first comparator compares the comparison results of the same said row, while said second comparator compares the comparison results of individual rows; said combinational circuit stores the temporary results of said first comparator and those of said second comparator, and sends the comparison result of said row to said second comparator for comparison after said first comparator has made a comparison with respect to said row.
8. A folding weighted Comma-Free Reed-Solomon (CFRS) decoding circuit, comprising: an input pattern generator, which receives the arbitrarily cyclic-shift CFRS codes for generating 15 cyclic-shift versions of CFRS codes and outputs to a systolic array in a skewed form; a plurality of weighted processing element arrays folded from a systolic array composed of 64×15 weighted processing elements at least one time, for receiving the skewed-form CFRS codes—to make weighted correlating comparisons and output a weighted correlating comparison result; a plurality of boundary processing element arrays folded from a systolic array composed of 64×1 boundary processing elements at least one time, for calculating a greatest said correlating comparison result of a row of said skewed-form CFRS codes of said systolic array to find greatest results of individual rows and set the acquired index as a decoding result; and a refresh mechanism for refreshing the code symbol saved in advance in each said processing element.
9. The decoding circuit of claim 8, wherein the preferably times of folding the systolic array architecture formed by said 64×15 processing elements and the systolic array architecture of the 64×1 boundary processing elements is two.
10. The decoding circuit of claim 8, wherein said input pattern generator is composed of a shift register, a Johnson counter and a masking combinational circuit; said shift register receives the 15 symbols of said CFRS codes and their corresponding weighted values, cyclically shifts them and outputs them as 15 cyclic-shift versions of said CFRS codes and weighted values; said Johnson counter sends a masking signal to said combinational circuit; said masking combinational circuit outputs the skewed-form CFRS codes according to said masking signal.
11. The decoding circuit of claim 8, wherein each of said weighted processing element is composed of a first register, a second register, a third register, a fourth register, a (XNOR-AND) combinational circuit, an accumulator and a weight adder; said first register stores the symbols of said CFRS codes beforehand; said second register stores the received code symbols of said CFRS codes so as to facilitate their downward transmission; said third register stores said correlating comparison results; said fourth register stores a weighted value; said (XNOR-AND) combinational circuit compares the two code symbols in said first register and said second register, respectively, to see whether they are identical or not; said weighted value adder decides whether said weighted value is to be added to the final result, according to said correlating comparison result; said accumulator adds the correlating comparison result to the final result.
12. The decoding circuit of claim 11, wherein said weighted value is acquired by means of the detection of primary synchronization code (PSC) generated in the first step of the synchronization procedure of the system of wideband code division multiple access (W-CDMA), and said weighted value is acquired by the steps of: determining the secondary synchronization code: coherently combine 16 sets of secondary synchronization codes with said primary synchronization code; and calculating the greatest value of said 16 sets of secondary synchronization codes and treat it as said weighted value.
13. The decoding circuit of claim 8, wherein each of said weighted processing element is composed of a first register, a second register, a third register, a fourth register, a (XNOR-AND) combinational circuit, an accumulator and a choice element; said first register stores the symbols of said CFRS codes beforehand; said second register stores the received code symbols of said CFRS codes so as to facilitate their downward transmission; said third register stores said correlating comparison results; said fourth register stores a weighted value; said (XNOR-AND) combinational circuit compares the two code symbols in said first register and said second register, respectively, to see whether they are identical or not; said choice element decides whether said weighted value is to be added to the final result, according to said correlating comparison result; said accumulator adds said weighted value to a final result.
14. The decoding circuit of claim 8, wherein each of said boundary processing element is composed of a first comparator, a second comparator and a combinational circuit; said first comparator compares the comparison results of the same said row, while said second comparator compares the comparison results of individual rows; said combinational circuit stores the temporary results of said first comparator and those of said second comparator, and sends the comparison result of said row to said second comparator for comparison after said first comparator has made a comparison with respect to said row.
15. The decoding circuit of claim 8, wherein said refresh mechanism is composed of a plurality of counters, a read circuit of Read-Only Memory (ROM), a refresh register and a cluster allocation circuit; said plurality of counters figure out the time taken to refresh individual refresh clusters in a duration of figuring out 15 refresh clusters; once the refresh time runs out, said read circuit of ROM read the code symbol of the next segment to be saved beforehand, and place the code symbol in said refresh register; said cluster allocation circuit then sends the new code symbol to said first register of said corresponding processing element for saving said CFRS code symbol beforehand to complete the refresh procedure; the formation of said 15 refresh clusters depends on the execution condition of said systolic array.

Weighted decoding method and circuits for comma-free reed-solomon codes

Information

Publication Number

Date Filed

Date Published

Inventors

CPC

US Classifications

International Classifications

Abstract

Description

Claims