CN100367682C - Method for code group identification and frame synchronization - Google Patents
Method for code group identification and frame synchronization Download PDFInfo
- Publication number
- CN100367682C CN100367682C CNB2006100094292A CN200610009429A CN100367682C CN 100367682 C CN100367682 C CN 100367682C CN B2006100094292 A CNB2006100094292 A CN B2006100094292A CN 200610009429 A CN200610009429 A CN 200610009429A CN 100367682 C CN100367682 C CN 100367682C
- Authority
- CN
- China
- Prior art keywords
- symbols
- hard decision
- symbol
- sequence
- code
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Abstract
A method and apparatus having a modified Reed-Solomon decoder is used for finding a specific code group used by a base station and the frame timing synchronization with the base station. The modified Reed-Solomon decoder uses a standard Reed-Solomon decoder and some reliability measurements computed from the received code word symbols. If the reliability of a received symbol is too low, this symbol is considered as erasure. By selecting code word symbols with higher reliabilities and erasing code word symbols with lower reliabilities, the symbol error probability is reduced and the performance is improved. Several modified Reed-Solomon decoders and a few decoding strategies are introduced in order to decode the received code word sequences with a power- and memory-effective method.
Description
The application is a divisional application with patent application number 03110683.8, application date 2003, 4/22, entitled "method and apparatus for code group identification and frame boundary synchronization".
Technical Field
The present invention relates to a method of code group identification and frame boundary synchronization for use in a direct-sequence code division multiple access (DS-CDMA) communication system, such as a wideband (wide-band) CDMA system and a third generation partnership project (3) rd generationnpartnershirp project,3 GPP).
Background
Currently, mobile telephone systems (cellular systems) of direct-sequence CDMA can be classified roughly into two types. One is a synchronous system in which base cells are precisely synchronized, and the other is an asynchronous system in which base cells are not synchronized. For a synchronous system, the hand-off of the handset between base cells will be very fast because the start time of the identification code (identity code) of the neighboring base cell is only a different fixed offset from the base cell that is currently connected. However, each base cell also requires a lot of expensive equipment for synchronization purposes, such as a Global Positioning System (GPS) and an accurate timing oscillator (oscillator). The time sequence oscillator provides a clock of a base station, and the GPS provides a clock time-pair basis. In addition to the expensive equipment, asynchronous systems have difficulty implementing a base cell that has difficulty receiving GPS signals, such as in basements or dense residential areas.
As for asynchronous systems, such as wideband CDMA systems proposed by 3GPP in europe, two synchronous channels (synchronization channels) are used for each base station. By obtaining the synchronization codes transmitted in the two synchronization channels, the mobile station (e.g., mobile phone) can establish a good link and will not cause disconnection problem during base station switching. These two synchronization channels are a Primary Synchronization Channel (PSCH) and a Secondary Synchronization Channel (SSCH), respectively. In asynchronous systems, the PSCH is a channel common to all base stations, and is a primary synchronization code (Primary synchronization code) sent once per slot (slot)code, PSC), denoted C p And (c) a step of forming. Each PSC has a length of 256 chips (chips), and when a PSC is transmitted, a time slot of a downlink (downlink) channel is synchronously transmitted.
The SSCH contains a Secondary Synchronization Code (SSC) of 15 numbers, denoted C ssc i And forming an identification number sequence. Identify the sequence in weeksSending out a time frame (timeframe) continuously, wherein each SSC is composed of a set of 16 orthogonal Codes (CS) 1 ～CS 16 Selected and each code is 256 chips long. SSCs in the identification number sequence are sent by the base station in parallel and simultaneously with the PSC. The identification number sequence belongs to one of 64 different code groups (codegroup), and is also a basis for a base station downlink scrambling code (downlink scrambling code). In other words, each code group is composed of a sequence of 15 orthogonal codes. The code arrangement order positions in the 64 code groups are shown in fig. 9. The 64 specific sequences are designed so that the sequences generated by their cyclic shifts are all unique. That is, as long as the number of cyclic shifts is between 0 and 14, the resulting 960 (= 64 × 15) number series are all different from each other. By using this feature, the mobile station can identify the code group used by a base station and the start time of the time frame.
When a mobile station of a wideband CDMA system proposed by 3GPP performs a base station cell search, the mobile station first finds a base station providing the strongest signal, and then finds the downlink scrambling code and time frame synchronization of the found base station. Such a base cell search can be represented by the following three steps:
the first step is as follows: time slot synchronization
When searching for the base cell, the mobile station first finds out the PSC with the largest signal energy and its timing in the PSCH. This is typically done by a filter corresponding to the PSCH. Since the PSC transmitted at the PSCH is the same for each base station, the PSC with the highest signal energy corresponds to the base station that provides the strongest signal. The slot length and the starting time of the base station with the strongest signal can be determined by the PSC timing. Therefore, the mobile station can automatically adjust its internal clock (clock) to synchronize with the slot of the strongest signal base station.
The second step: code group identification and frame boundary synchronization
After the slot synchronization, the mobile station can go to the SSCH to collect SSCs, and then find out which code group of 64 code groups the identification number sequence from the base station found in the first step belongs to, and from which time a time frame should start, i.e. frame boundary synchronization. An intuitive implementation is to receive the SSCs in 15 slots consecutively in a time frame length to form a sequence and then compare the received sequence with 960 sequences possible from 64 code groups and 15 cyclic shifts. Because the 960 number series are all unique, the identical number series can be found therein. Therefore, it can know which code group the found base station sends, and can know that the cyclic shift of several time slots is the time frame starting time of the found base station, so as to achieve the purpose of time frame synchronization.
The third step: scrambling code identification
After the two steps are completed through the two synchronization channels, the mobile station then uses a common pilot channel (CPICH) to identify the primary scrambling code (primary scrambling code) by comparing SSCs of the code group found in the second step. Then, a Primary Common Control Physical Channel (PCCPCH) can be found. Then, the information of the system and the base station can be read.
The second step of the base cell search procedure is the subject of the present invention. The SSCH is used to determine frame boundary synchronization. A frame of 15 SSCH symbols forms a code sequence, and the 64 different code sequences are taken from a codebook (codebook). The same code sequence is repeated for each frame in a base station. Any code phase shift (code phase shift) of the 64 different code sequences is selected to be different from all code phase shifts of all other code sequences. With these properties, the frame boundary can be detected by identifying the correct starting phase of the SSCH symbol sequence. To meet the above properties and to maximize the minimum distance between different coding sequences, a (15,3) reed solomon code in the galois field (GF (16)) has been proposed. Regarding this error correction code, a standard reed solomon decoder of the reed solomon code can be found in textbooks (15,3) and can correct up to 6 symbol errors. However, the number of symbol errors may often exceed 6 due to frequency errors (frequency error), channel repetition, channel noise, or other reasons. Therefore, a standard reed-solomon decoder often cannot recover a valid code sequence.
In the journal "IEEEJournaronSelecttedDareasincommunications vol.18, no.8 August2000", yi-PingEricWang proposes another approach. The method proposed by Wang is: after achieving slot synchronization, the receiver operates to correlate the received signal from the SCH with all 16S-SCH sequences and accumulate N in accordance with the 64 Reed-Solomon code sequences used t SSCH correlation for a period of time, there are 15 hypothetical frame boundaries per symbol sequence. There are a total of 960 hypotheses. At the end, the hypothesis with the largest cumulative amount is selected as the candidate for the frame boundary-code group pair, and the next stage is given for scrambling code identification.
The method proposed by Wang has a good implementation but requires a lot of memory and computational effort. The present invention provides a power and memory efficient method using a standard reed solomon decoder incorporating a reliability measure.
Disclosure of Invention
The present invention overcomes the above-mentioned disadvantages of the conventional frame boundary synchronization and code group identification, and it is a primary object of the present invention to provide a method and apparatus for frame boundary synchronization and code group identification with low power and low memory requirements. Accordingly, the apparatus of the present invention comprises a correlator bank (correlator bank) provided with a plurality of correlators (correlators), a hard decision and reliability measurement unit (hard decision and reliability measurement unit), a code sequence identifier (code sequence identifier), a frame boundary finder (frame boundary finder), and a code group identification unit (code group identification unit).
Each time a signal is received, the signal is sent to the correlator bank to identify the correlation between the signal and the 16 orthogonal code symbols. The hard decision and reliability measurement unit then picks a hard decision symbol with the highest correlation and calculates the reliability using a function of the correlation of the 16 correlations.
It is another object of the present invention to provide a modified reed solomon decoder in a code sequence identifier. In a preferred embodiment of the present invention, the modified Reed-Solomon decoder uses a threshold to determine whether the encoded symbol should be erased based on a measure of the reliability of the hard-decision symbol. When the number of valid symbols is greater than or equal to a threshold value between 3 and 15, the entire code sequence is sent to a standard Reed-Solomon error and erasure decoder for decoding.
In another embodiment, the modified Reed-Solomon decoder compares the number of erasures in an encoded sequence to a threshold, the threshold being an integer between 0 and 12. If the number of erasers is not greater than the threshold, the encoded sequence is sent to a standard error and erasure decoder. If the decoder does not decode a valid code sequence, the k code symbols with the lowest reliability are erased and the new code sequence is sent again to the standard error and erasure decoder.
It is still another object of the present invention to utilize at least two frames of encoded symbols to reduce the probability of symbol errors and improve the performance of the encoded sequence identifier. In this way, a symbol and reliability update unit (symbol and reliability update unit) is added to the code sequence identifier. Since the code symbols are transmitted periodically, after a frame of codes is received and recorded, the next received code symbol should ideally be equal to the first code symbol in the recorded frame. The next code symbol and its reliability are used to replace the corresponding code symbol in the recorded frame. When more than one frame of symbols is received, a decision is made whether to replace the hard decision symbols with additional symbols based on their reliability.
Another decoding strategy that uses more than one code sequence is to first receive two code sequences and then generate a new code sequence by comparing the two code sequences. And if the corresponding code symbols in the two received code sequences are not equal, erasing the code symbol. This new encoded sequence is then sent to the standard error and erasure decoder.
Another decoding strategy involves the use of hard decisions in voting (voting) on multiple frames of the encoded symbol. After receiving several code sequences, their hard decision symbol values are recorded. A new encoded sequence is generated by using the corresponding encoded symbols of the plurality of tickets over the plurality of frames. This new encoded sequence is then sent to the standard error and erasure decoder.
It is yet another object of the present invention to provide a method of frame boundary synchronization. By looking at the 64 code sequences in fig. 9, the present invention finds that the first code symbol in a frame must have the smallest symbol value. If the smallest symbol value is unique, the symbol is the beginning of the frame (head of the frame). If the smallest symbol value occurs twice, the next symbol after the start of the frame must have a smaller value than the next symbol after the other found smallest symbol.
It is another object of the present invention to provide a method for identifying memory benefits in a code group of a coded sequence. The 64 code groups for efficient coding of comma-free reed-solomon codes also have this property: the code sequences in each group are uniquely identifiable by the first 3 code symbols. By storing the first 3 code symbols of each of the 64 code groups, the code number (code number) of the received code sequence can be identified.
The invention relates to a method for generating a correct coding sequence, which is characterized by comprising the following steps:
(a) Receiving a frame of encoded symbols;
(b) Forming a hard decision symbol sequence from the frame of the encoded symbols, each hard decision symbol having a corresponding reliability value;
(c) Sending the hard decision symbol sequence and the corresponding reliability value to the modified decoder;
(d) Generating a correct code sequence using the modified decoder, and ending the method if the modified decoder successfully decodes;
(e) If the total number of the received hard decision symbols is greater than the limit, ending the method; (f) receiving a number of additional encoded symbols; (g) Updating the hard decision symbol sequence and the corresponding reliability value according to the additional coding symbol; and
(h) Returning to the step (c).
The method of generating a correct code sequence according to the present invention, wherein the step (g) of using an additional code symbol having a position m is according to the following steps: (g1) Calculating new hard decision symbols and new corresponding reliability values for the additional coded symbols; (g2) Identifying a combined hard decision symbol having a received position equal to the remainder of m divided by n, n being the number of symbols in the sequence of hard decision symbols; (g3) If the new hard decision symbol is constantly equal to the combined hard decision symbol, increasing the corresponding reliability value of the combined hard decision symbol; (g4) If the new hard decision symbol is not identical to the combined hard decision symbol and the new corresponding reliability value is less than the combined hard decision symbol corresponding reliability value, decreasing the combined hard decision symbol corresponding reliability value; and (g 5) if the new hard decision symbol is not identical to the combined hard decision symbol and the new corresponding reliability value is greater than the corresponding reliability value of the combined hard decision symbol, replacing the combined hard decision symbol with the new hard decision symbol and replacing the corresponding reliability value with a newly reduced corresponding reliability value.
The method of generating a correct encoded sequence of the present invention wherein the modified decoder in step (d) performs the steps of: (i) Receiving a new symbol from the hard decision symbol sequence; (ii) If the reliability value corresponding to the new symbol is larger than the reliability critical value, the new symbol is determined to be an effective symbol, otherwise, the new symbol is determined to be an ineffective symbol; (iii) recording the total number of valid symbols; (iv) If the total number of recorded valid symbols is greater than or equal to the threshold value of the total number of valid symbols, sending the coded sequence to an error and erasure decoder according to the valid symbols, otherwise executing step (vi); (v) If the error and erasure decoder decodes successfully, then the correct code sequence is generated and the step (d) is ended; and (vi) if all symbols of the hard decision symbol sequence are received, ending step (d), otherwise returning to step (i).
The method for generating a correct code sequence according to the present invention, wherein the hard decision symbol sequence has at most 15 hard decision symbols, and the threshold value of the total number of valid symbols is an integer between 3 and 15 and is a function of the number of received hard decision symbols.
In the method of generating a correct code sequence according to the present invention, the modified decoder in step (d) performs the following steps: (i) Receiving a sequence of hard-decision symbols comprising valid symbols and invalid symbols; (ii) If the total number of invalid symbols in the hard decision symbol sequence is not greater than the critical value of the total number of invalid symbols, sending the coded sequence to an error and erasure decoder according to the valid symbols, otherwise ending the step (d); (iii) If the error and erasure decoder successfully decodes, then generating the correct code sequence and ending the step (d); (iv) Selecting k symbols from the current valid symbols of the hard decision symbol sequence, wherein the k symbols have k lowest reliability values in the current valid symbols of the hard decision symbol sequence; (v) Deeming the k symbols to be invalid and determining a total number of invalid symbols from the sequence of hard-decision symbols; and (vi) if the total number of invalid symbols of the hard decision symbol sequence determined in step v is not greater than the threshold value of the total number of invalid symbols, returning to step (ii), otherwise, ending the step (d).
The method for generating a correct code sequence according to the present invention, wherein the hard decision symbol sequence has at most 15 hard decision symbols, and the threshold value of the total number of invalid symbols is an integer between 0 and 12.
Another method of generating the correct coding sequence according to the invention comprises the following steps: (a) receiving first and second frames of encoded symbols; (b) Forming first and second hard decision symbol sequences from the first and second frames of the encoded symbols; (c) Comparing each hard decision symbol in the first sequence with each corresponding hard decision symbol in the second sequence, if the corresponding hard decision symbols are identical, determining that the hard decision symbols are valid symbols, otherwise, determining that the hard decision symbols are invalid symbols; (d) If the total number of invalid hard decision symbols is less than the critical value of the total number of invalid hard decision symbols, sending the coded sequence to an error and erasure decoder according to the valid hard decision symbols determined in the step (c), otherwise, ending the method; and (d) if the error and erasure decoder successfully decodes, then generating the correct code sequence and ending the method.
The invention also provides a method for generating a correct coding sequence, which comprises the following steps: receiving a frame of a plurality of coded symbols; (b) Forming a plurality of hard decision symbol sequences, each sequence corresponding to one of all received frames of coded symbols; (c) Forming a new sequence of hard decision symbols, the value of each hard decision symbol in the new sequence being determined by the number of votes taken for the corresponding hard decision symbol in the plurality of sequences; (d) sending the new sequence to an error and erasure decoder; and (e) if the error and erasure decoder successfully decodes, then generating the correct code sequence and ending the method.
The present invention further provides a method for generating a correct coding sequence, further comprising the steps of: (f) Receiving a new frame of code symbols if the error and erasure decoder fails to successfully decode a valid code sequence; and (g) returning to step (b).
The foregoing and other objects and advantages of the invention will be apparent from the following detailed description of the embodiments, taken in conjunction with the accompanying drawings and the appended claims.
Drawings
For further explanation of the technical contents of the present invention, the following detailed description is given with reference to the accompanying drawings and examples, in which:
fig. 1 is a system block diagram of an apparatus for code group identification synchronization with frame boundaries in accordance with the present invention.
FIG. 2 illustrates a flow diagram of a preferred embodiment of a modified Reed-Solomon decoder, called a "threshold erase decoder".
FIG. 3 illustrates a flow diagram of another preferred embodiment of a modified Reed-Solomon decoder, called a "segmented erasure decoder".
Fig. 4 illustrates the sign and reliability update procedure.
Fig. 5 illustrates a decoding strategy when more than one frame is received.
Fig. 6 illustrates a decoding strategy for multiple frames with hard decisions only.
Fig. 7 illustrates a decoding strategy for multiple frames with votes.
Fig. 8 illustrates a frame boundary finder.
Fig. 9 encodes a sequence diagram.
Detailed Description
The main idea of the invention is to use the standard error-and-erasure decoder in conjunction with reliability measurement to perform code group identification and frame boundary synchronization in the UMTS WCDMA system. Fig. 1 is a system block diagram of an apparatus for code group identification synchronization with frame boundaries in accordance with the present invention. The apparatus comprises a correlator bank comprising a plurality of correlators 101, a hard decision and reliability measurement unit 102, a code sequence identifier 103, a frame boundary finder 104, and a code group identification unit 105.
As is well known, each of the SSCs corresponds to a valid code in the (15,3) code. Typically, the error rate of hard-decision symbols (hard-decision symbols) after passing through the 16Walsh code correlator is too high for most case 1 standard decoders to successfully recover a valid code. However, decoders still have many advantages, such as smaller memory requirements and lower computational complexity (computational complexity).
According to the invention, each time a signal is received, it is sent to a correlator bank comprising 16 correlators 101, in order to identify its code symbols CS orthogonal to 16 01 ，CS 02 …, and CS 16 The correlation between them. At time m, the output correlation values from the 16 correlators are { r } 01 m ，r 02 m ，…，r 16 m }. Hard decision symbol value R at time m m From { CS 01 ，CS 02 ，…，CS 16 Select one of the correlations with the highest correlation. Thus, the reliability measure can be defined as 16 correlation values r 01 m ，r 02 m ，…，r 16 m Is used to measure the hard decision symbol value R m The reliability of (2). For example, the reliability measure L m Can be defined as
Or is
As shown in FIG. 1, the hard decision and reliability measurement unit 102 receives the packetThe correlation values from the correlators 101 are hard-decided to pick out a symbol R m . Then, according to the above-mentioned formula, the selected symbol R is selected m Calculating its reliability measure L m 。
Since each of the 64 code groups is a valid code in the (15,3) code, after recovering a valid code for the reed solomon decoder, the present invention selects the code symbol with the highest reliability measure and erases the remaining code symbols. Based on the properties of the (15,3) code, if all 15 code symbols are received, a maximum of 12 code symbols with lower reliability measures can be erased.
According to the invention, the code sequence identifier 103 comprises a modified decoder 111. A preferred embodiment of the modified decoder 111 is a "threshold-erasure decoder" in which a threshold σ is used based on the reliability measure calculated in the hard decision and reliability measurement unit 102 r To determine whether the coded symbols should be erased.
Fig. 2 illustrates a flow diagram of an implementation of a threshold erase decoder as a modified decoder 111. When a new code symbol is received, its hard-decision symbol value and corresponding reliability value are recorded. If the reliability is less than the threshold value sigma r The received encoded symbol is considered to be an invalid symbol and erased, and the corresponding reliability value is set to- ∞. Otherwise, if the reliability value is larger than sigma r Its hard decision symbol value is recorded and the number of Valid Symbols (VSN) is incremented by 1.
Assume another critical value σ v Is an integer between 3 and 15 and is a function of the number of Received Symbols (RSN). When VSN is larger than or equal to sigma v The entire code sequence is then sent to the standard error and erasure decoder. If the decoding process fails and the RSN is less than 15, another new code symbol is received. If the reliability value of the newly received character is greater thanσ r The new code sequence containing the newly received code symbols is then forwarded to the standard error and erasure decoder. When the standard error and erasure decoder returns a valid code, the entire decoding process is left and ends.
Fig. 3 illustrates another preferred embodiment of a modified decoder 111, called a "segment erasure decoder" (eraseKbyKdecoder). A coded sequence containing 15 coded symbols is fed to a "segmented erasure decoder". Recording hard decision symbol values (R) 0 ，R 1 ，R 2 ，…，R 14 ) And its corresponding reliability (L) 0 ，L 1 ，L 2 ，…，L 14 ). Note that: some coded symbols are considered erasures (erasures) and their reliability is set to- ∞.
Calculating the total number of erasures e 0 And a critical value sigma between 0 and 12 e For comparison. If total number of erasures e 0 Not more than sigma e The encoded sequence is sent to a standard error and erasure decoder. If the decoding process fails, first (L) 0 ，L 1 ，L 2 ，…，L 14 ) In a descending order ((L) (0) ，L (1) ，L (2) ，…，L (14) ) Wherein L is (i) Correspond to (i) . At this time, there is e 0 One code symbol is erased, thus, R (0) ，R (1) ，…， R (e0-1) Is erased and L (0) ＝L (1) ＝…＝L (e0-1) And = - ∞. At L (e0) ，L (e0+1) ，…，L (14) The lowest k reliabilities corresponding to symbols that are not erased are L (e0) ，L (e0+1) ，…，L (e0+k-1) 。
Then, the corresponding reliability is L (e0) ，L (e0+1) ，…，L (e0+k-1) K code symbols R (e0) ， R (e0+1) ，…，R (e0+k-1) Erased, where k is a positive integer and is the current e 0 Function of value, that is to say, inK is variable during each erase. Thus, the number of erased bits becomes e 0 + k. Comparing the current erased number (e) 0 + k) and threshold σ e If the number of erased bits is not greater than σ e The new encoded sequence (containing k erasures) is again sent to the decoder. When the standard error and erasure decoder sends a valid code sequence or the number of erasures exceeds the threshold value sigma e The entire erasure-comparison-decoding process is then finished.
To further reduce the symbol error probability and improve the efficiency of operation of the code sequence identifier 103, the present invention may use frames of not only one code symbol. Accordingly, a symbol and reliability update unit 112 may be added to the code sequence identifier 103, as shown in FIG. 1. A method of updating hard-decision symbol values and reliability measures when more than 15 symbols are received, and a decoding procedure using more than one frame, will be discussed below.
Since the 15 code symbols are transmitted periodically, if an encoded sequence of 15 code symbols cannot be decoded, it is not necessary to discard the encoded sequence. In other words, new coded symbols can be received and used to update the hard-decision symbol values and their corresponding reliability measures. Fig. 4 illustrates one embodiment of this method of updating hard decision symbol values and reliability measurements.
Assume that the code sequence of the previously received 15 code symbols is (R) 0 ，R 1 ，R 2 ，…，R 14 ) And its corresponding reliability is (L) 0 ，L 1 ，L 2 ，…，L 14 ). Since the 15 code symbols are transmitted periodically, ideally,
R i ＝R imod15 i＝15，16，17，...。
after receiving a frame of coded symbols, the total RSN value is 15. When the 16 th coded symbol is received, a hard decision measurement R' (or R) 15 ) And its corresponding reliability L' (or L) 15 ) Is recorded. Ideally, R' should be equal to R 0 . However, due to noise and other reasons, the hard decision measurement R 0 And R 15 May not be equal. If these two hard decision measurements (R) 0 And R 15 ) If equal, the reliability L is updated by increasing a certain value 0 . The increased value is a measure of the original reliabilityL 0 And the currently received reliability L 15 The function of interest. For example, the new reliability may be represented by the sum of these two reliabilities:
L 0(after updating) ＝L 0 +L 15 。
however, if these two hard decision measurements (R) 0 And R 15 ) Not equal, the symbol and corresponding reliability must be represented by its corresponding reliability (L) 0 And L 15 ) And comparing and updating according to the compared result. The updated hard decision measurement is set to have a symbol value corresponding to a greater reliability value.
Also, the updated reliability should be reduced. The decrease of the reliability is also L 0 And L 15 As a function of (c). For example, in the case of a liquid,
L 0(after updating) ＝max(L 0 ，L 15 )-min(L 0 ，L 15 )
expressed by the same symbol, when the 17 th code symbol (R) is received 16 ) When the RSN value is equal to 16 16 Should be equal to R 1 The hard decision symbol value and reliability modification procedure may then be applied to R 1 And R 16 And so on, and so forth.
By updating R i And R i mod 15 The hard decision measurement and reliability method of (2) can then be introduced into the decoding process. Figure 5 illustrates a decoding strategy when more than one frame (containing 15 symbols) is received. When 15 coded symbolsR＝(R 0 ，R 1 ，R 2 ，…，R 14 ) When the original code sequence of (a) cannot be decoded, σ N A new code symbol (R) 15 ，R 16 ，R 17 ，R 18 ) Is received. Applying hard decision measurements and reliability update procedures to R 0 And R 15 ，R 1 And R 16 ，R 2 And R 17 And R 3 And R 18 A new set of coding sequences is obtained. It is worth mentioning that even though the hard decision measurements may not be changed, their corresponding reliabilities may be different.
Novel coding sequencesR' and New reliability sequencesL' is sent to the modified decoder. If the new coding sequenceR' cannot be decoded again, in addition to σ N A new code symbol may be received to obtain another new set of code sequencesR'' and New reliability sequencesL'' wherein σ N May be any positive integer and may vary with each update procedure.R'' andL' is again sent to the modified decoder shown in fig. 2 and 3, or a combination of both. When the modified decoder returns a valid code sequence, the entire decoder process is complete. To avoid infinite loops (endlessloop) due to low signal-to-noise ratio (SIN) or other factors, the present invention limits the total RSN value to terminate the loop. When the total number of received symbols exceeds a predetermined numberAt integer values MAX-RSN, the current coding sequence is discarded.
Fig. 6 illustrates another decoding strategy for more than one coded sequence with only hard decisions. First, two code sequences are receivedR 1 AndR 2 . With hard decision measurements of respectivelyR 1 ＝(R 1 0 ，R 1 1 ，R 1 2 ，…， R 1 14 ) AndR 2 ＝(R 2 0 ，R 2 1 ，R 2 2 ，…，R 2 14 ). Comparison of R 1 j And R 2 j J =0,1,2, …,14. If a hard decision measurement (R) 1 j And R 2 j ) Is not equal to R j It is set to erase. Comparison ofR 1 AndR 2 after 15 symbols in the sequence, if the total number of erased e is 0 Less than a critical value σ e Containing e 0 An erased code sequenceR＝(R 0 ， R 1 ，R 2 ，…，R 14 ) Is sent to a standard error and erasure decoder, where σ e Is an integer from 1 to 13.
If the code sequenceR' cannot be successfully decoded, the two encoded sequences may be discarded or an attempt made to use other decoding strategies. In another aspect, another code sequence comprising 15 code symbolsR 3 May continue to be received. By comparing the encoded sequences using the procedure described aboveR 3 And the previous coding sequenceR', can be atR' A new result code sequence is recorded. If it isRTotal number of erasures declared in e 0 Less than sigma e Containing e 0 An erased code sequenceR' is sent to a standard error and erasure interpreter. Note, σ e The value may be lowered. When the standard error and erasure decoder returns a valid code, or receives a number of code sequences equal to the maximum number of code sequences allowed, the entire process ends.
Fig. 7 illustrates another alternative hard-decision decoding strategy for voting. Initially, σ is received s A code sequence and hard decision measurements thereofR 1 ＝(R 1 0 ，R 1 1 ，R 1 2 ，…，R 1 14 )，R 2 ＝(R 2 0 ，R 2 1 ，R 2 2 ，…， R 2 14 )，…，R σs ＝(R σs 0 ，R σs 1 ，R σs 2 ，…，R σs 14 ) Is recorded. For each coded symbol, its hard-decision measurement R j J =0,1,2, …,14, set as the set { R } 1 j ，R 2 j ，R 3 j ，…，R σs j Most tickets (majorityvote). And the coding sequence of the result is recorded inR＝(R 0 ，R 1 ，R 2 ，…，R 14 ) And sent to a standard error and erasure interpreter. If the decoding process fails, a new code sequence may be received and a plurality of tickets may be employed. The resulting encoded sequence is then decoded. When the standard error and erasure decoder returns a valid code or the number of received code sequences is equal to the maximum number of allowed code sequences, the entire decoding strategy is complete.
As shown in fig. 1, after the code sequence identifier 103 identifies the correct code sequence, the frame boundary of the code sequence is found by the frame boundary finder. Fig. 8 illustrates a method of finding frame boundaries after de-encoding to recover a valid encoded sequence.
Referring to fig. 9, the 64 code sequences have valid code bits of the comma-free reed-solomon code. That is, all codes have no internal reproducibility (internal repetition), and in each of the 15 code symbols, the first code symbol thereof has the smallest symbol value and the smallest symbol value is found at most twice in the code sequence. If the smallest symbol value is unique, this symbol is the beginning of the frame. If the minimum symbol value is found twice, the next symbol after the beginning of the frame must have a smaller value than the next symbol of the minimum symbol found in the other time periods. For example, if the smallest symbol value is found in the number of periods =0 and the number of periods = j, the symbol in the number of periods =1 must have a smaller symbol value than the symbol in the number of periods = j + 1. For example, the encoding sequence in group 0 is (1,1,2, 8,9, 10, 15,8, 10, 16,2,7, 15,7, 16), where the smallest sign value is 1. The minimum sign value is found twice in the number of periods =0 and the number of periods =1, respectively. Comparing the next symbol of the two symbol values, i.e., the number of slots =1 and the number of slots =2, the symbol after the start of the frame, i.e., the number of slots =1, can be found to have a smaller symbol value. For another example, the code sequence in the group 63 is (9, 12, 10, 15, 13, 14,9, 14, 15, 11, 11, 13, 12, 16, 10), where the smallest symbol value is 9. The minimum symbol value is found twice in the number of time slots =0 and the number of time slots = 6. Comparing two symbol values of the next symbol, that is, the number of slots =1 and the number of slots =7, the symbol after the start of the frame, that is, the number of slots =1, can be found to have a smaller symbol value.
As described above, the valid code sequence recovered by the decoder may be a periodic shift (cyclic shift) of the original code sequence. The frame boundaries can be determined by finding the smallest two symbol values in the encoded sequence. If the minimum two symbol values are not equal, the head index of the frame boundary is the position of the minimum symbol value. If the smallest two symbol values are equal, the head position can be determined by comparing the next symbol of the two symbol values. Based on the above properties, it is not difficult to find the start position of the frame boundary.
After determining the frame boundary, the apparatus of the present invention utilizes the code group identification unit 105 to identify the code group. Referring to fig. 9, it can be observed that the code sequences in each group can be uniquely identified by the first three code symbols. With this property, only the first 3 columns in FIG. 9 need to be compared. The number of code groups can be confirmed by comparing the first three code symbols. Therefore, the code group identification unit greatly reduces the requirement of the memory.
It is worth mentioning that fig. 2 and 3 illustrate the "critical value-erasure decoder" and the "segmented erasure decoder" of the modified decoder 111 of the present invention, respectively. Various changes may be made to these decoders. For example, both "threshold-erase decoders" and "segmented-erase decoders" may be combined if desired.
However, the above description is only a preferred embodiment of the present invention, and the scope of the present invention should not be limited thereby. All equivalent changes and modifications made according to the claims of the present invention should still fall within the scope of the patent of the present invention.
Claims (9)
1. A method of generating a correct coding sequence, comprising the steps of:
(a) Receiving a frame of encoded symbols;
(b) Forming a sequence of hard decision symbols from the frame of coded symbols, each hard decision symbol having a corresponding reliability value;
(c) Sending the hard decision symbol sequence and the corresponding reliability value to the modified decoder;
(d) Generating a correct code sequence using the modified decoder, and ending the method if the modified decoder successfully decodes;
(e) If the total number of the received hard decision symbols is greater than the limit, ending the method;
(f) Receiving a number of additional coded symbols;
(g) Updating the hard decision symbol sequence and the corresponding reliability value according to the additional coding symbol; and
(h) Returning to the step (c).
2. The method of generating a correct code sequence according to claim 1, wherein the step (g) of using an additional code symbol at position m is according to the steps of:
(g1) Calculating new hard decision symbols and new corresponding reliability values for the additional coded symbols;
(g2) Identifying a combined hard decision symbol having a received position equal to the remainder of m divided by n, n being the number of symbols in the sequence of hard decision symbols;
(g3) If the new hard decision symbol is constantly equal to the combined hard decision symbol, increasing the corresponding reliability value of the combined hard decision symbol;
(g4) If the new hard decision symbol is not identical to the combined hard decision symbol and the new corresponding reliability value is less than the corresponding reliability value of the combined hard decision symbol, reducing the corresponding reliability value of the combined hard decision symbol; and
(g5) If the new hard decision symbol is not identical to the combined hard decision symbol and the new corresponding reliability value is greater than the corresponding reliability value of the combined hard decision symbol, replacing the combined hard decision symbol with the new hard decision symbol and replacing the corresponding reliability value with a newly reduced corresponding reliability value.
3. The method for generating correct code sequences according to claim 1, wherein the modified decoder in step (d) performs the steps of:
(i) Receiving a new symbol from the hard decision symbol sequence;
(ii) If the reliability value corresponding to the new symbol is larger than the reliability critical value, the new symbol is determined to be an effective symbol, otherwise, the new symbol is determined to be an ineffective symbol;
(iii) Recording the total number of valid symbols;
(iv) If the total number of the recorded valid symbols is greater than or equal to the threshold value of the total number of the valid symbols, sending the coded sequence to an error and erasure decoder according to the valid symbols, otherwise, executing the step (vi);
(v) If the error and erasure decoder successfully decodes, then generating the correct code sequence and ending the step (d); and
(vi) If all the symbols of the hard decision symbol sequence are received, ending the step (d), otherwise returning to the step (i).
4. A method of generating a correct code sequence as claimed in claim 3, wherein the hard decision symbol sequence has at most 15 hard decision symbols, and the threshold value for the total number of valid symbols is an integer between 3 and 15 and is a function of the number of received hard decision symbols.
5. The method of claim 1 wherein the modified decoder in step (d) performs the steps of:
(i) Receiving hard decision symbols of a sequence, including valid symbols and invalid symbols;
(ii) If the total number of invalid symbols of the hard decision symbol sequence is not greater than the threshold value of the total number of invalid symbols, sending the coded sequence to an error and erasure decoder according to the valid symbols, otherwise ending the step (d);
(iii) If the error and erasure decoder decodes successfully, then the correct code sequence is generated and the step (d) is ended;
(iv) Selecting k symbols from the current valid symbols of the hard-decision symbol sequence, wherein the k symbols have k lowest reliability values in the current valid symbols of the hard-decision symbol sequence;
(v) Identifying the k symbols as invalid and determining the total number of invalid symbols from the sequence of hard-decision symbols; and
(vi) If the total number of invalid symbols of the hard decision symbol sequence determined in step v is not greater than the threshold value of the total number of invalid symbols, returning to step (ii), otherwise ending the step (d).
6. The method for generating correct code sequence according to claim 5, wherein the hard decision symbol sequence has at most 15 hard decision symbols, and the threshold value for the total number of invalid symbols is an integer between 0 and 12.
7. A method of generating a correct coding sequence, comprising the steps of:
(a) Receiving first and second frames of encoded symbols;
(b) Forming first and second hard decision symbol sequences from the first and second frames of the encoded symbols;
(c) Comparing each hard decision symbol in the first sequence with each corresponding hard decision symbol in the second sequence, if the corresponding hard decision symbols are identical, determining that the hard decision symbol is an effective symbol, otherwise, determining that the hard decision symbol is an ineffective symbol;
(d) If the total number of invalid hard decision symbols is less than the critical value of the total number of invalid hard decision symbols, sending the coded sequence to an error and erasure decoder according to the valid hard decision symbols determined in the step (c), otherwise, ending the method; and
(d) If the error and erasure decoder decodes successfully, the correct code sequence is generated and the method is ended.
8. A method of generating a correct coding sequence, comprising the steps of:
(a) Receiving a frame of a plurality of coded symbols;
(b) Forming a plurality of hard decision symbol sequences, each sequence corresponding to one of all received frames of coded symbols;
(c) Forming a new sequence of hard decision symbols, the value of each hard decision symbol in the new sequence being determined by the number of votes taken for the corresponding hard decision symbol in the plurality of sequences;
(d) Sending the new sequence to an error and erase decoder; and
(e) If the error and erasure decoder decodes successfully, the correct code sequence is generated and the method is ended.
9. The method of generating a correct coding sequence according to claim 8, further comprising the steps of:
(f) Receiving a new frame of code symbols if the error and erasure decoder fails to successfully decode a valid code sequence; and
(g) Returning to the step (b).
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US41253202P | 2002-09-19 | 2002-09-19 | |
| US60/412,532 | 2002-09-19 | ||
| US10/316,319 | 2002-12-10 |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB031106838A Division CN1275407C (en) | 2002-09-19 | 2003-04-22 | Method and apparatus for code group identification and frame synchronization |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1825773A CN1825773A (en) | 2006-08-30 |
| CN100367682C true CN100367682C (en) | 2008-02-06 |
Family
ID=36936243
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2006100094292A Expired - Fee Related CN100367682C (en) | 2002-09-19 | 2003-04-22 | Method for code group identification and frame synchronization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100367682C (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008205689A (en) * | 2007-02-19 | 2008-09-04 | Sony Corp | Communication apparatus, communicating method, and computer program |
| CN101656541B (en) * | 2009-09-15 | 2012-10-03 | 中兴通讯股份有限公司 | Coding method and device of RS codes |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1344117A (en) * | 2000-09-09 | 2002-04-10 | 三星电子株式会社 | Equipment and method for search base station in asynchronous mobile communicating system |
| US20020041579A1 (en) * | 2000-09-19 | 2002-04-11 | Motohiro Tanno | Cell search method for mobile station in mobile communication system and mobile station |
| US20020064211A1 (en) * | 2000-08-30 | 2002-05-30 | Po-Tsun Chen | Method and apparatus for code group identification and frame synchronization in DS/CDMA systems |
-
2003
- 2003-04-22 CN CNB2006100094292A patent/CN100367682C/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020064211A1 (en) * | 2000-08-30 | 2002-05-30 | Po-Tsun Chen | Method and apparatus for code group identification and frame synchronization in DS/CDMA systems |
| CN1344117A (en) * | 2000-09-09 | 2002-04-10 | 三星电子株式会社 | Equipment and method for search base station in asynchronous mobile communicating system |
| US20020041579A1 (en) * | 2000-09-19 | 2002-04-11 | Motohiro Tanno | Cell search method for mobile station in mobile communication system and mobile station |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1825773A (en) | 2006-08-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20080244364A1 (en) | Method And Apparatus For Code Group Identification And Frame Synchronization By Use Of Reed-Solomon Decoder And Reliability Measurement For UMTS W-CDMA | |
| US10674462B2 (en) | Facilitating synchronization between a base station and a user equipment | |
| JP3427056B2 (en) | Cell search in CDMA communication system | |
| CN100568848C (en) | The apparatus and method of receiving packet data control channel in the mobile communication system | |
| US6571099B1 (en) | Cell searching method in asynchronous wideband code division multiple access system | |
| MXPA01013040A (en) | Method and apparatus for fast wcdma acquisition. | |
| WO2000016512A1 (en) | Rate detection in radio communication systems | |
| US7929509B2 (en) | Reduced cell acquisition time | |
| US6892072B2 (en) | Method of optimizing the search for cells in a mobile telecommunication network | |
| US20050018646A1 (en) | Multistage PN code acquisition circuit and method | |
| CN100367682C (en) | Method for code group identification and frame synchronization | |
| US6628634B2 (en) | Fast decoding of long codes | |
| US7366141B2 (en) | Cell search method and apparatus in a WCDMA system | |
| US7254163B2 (en) | Method for WCDMA frame synchronization and related device | |
| EP1817860B1 (en) | Tfci decoding apparatus and method | |
| JP2001346239A (en) | Mobile wireless communication equipment and cell search method | |
| Sriram et al. | Fast acquisition method for DS-CDMA systems employing asynchronous base stations | |
| Sriram et al. | An analysis of the 3-stage search process in W-CDMA | |
| JP2000138615A (en) | Method for capturing synchronization and device therefor | |
| Shieh et al. | Soft-decision Reed-Solomon decoding strategies for code group identification and frame synchronization in UMTS WCDMA | |
| EP1161111A1 (en) | Mobile communication terminal and method of communication | |
| Liao et al. | New Code Synchronization Algorithm for the Secondary Cell-search Stage in WCDMA |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20080206 Termination date: 20150422 |
|
| EXPY | Termination of patent right or utility model |