CN1825773A - Method for code group identification and frame synchronization - Google Patents
Method for code group identification and frame synchronization Download PDFInfo
- Publication number
- CN1825773A CN1825773A CNA2006100094292A CN200610009429A CN1825773A CN 1825773 A CN1825773 A CN 1825773A CN A2006100094292 A CNA2006100094292 A CN A2006100094292A CN 200610009429 A CN200610009429 A CN 200610009429A CN 1825773 A CN1825773 A CN 1825773A
- Authority
- CN
- China
- Prior art keywords
- hard decision
- symbol
- decision symbol
- coded
- decoder
- 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.)
- Granted
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 for number of patent application is 03110683.8, the applying date is on April 22nd, 2003, denomination of invention is divided an application for the application of " identification of code group group and synchronous method and the device of edge circle ".
Technical field
The present invention relates to be used in direct sequence CDMA (direct-sequence code division multipleaccess, DS-CDMA) coding group identification in the communication system and the synchronous method of frame boundaries, for example broadband (wide-band) cdma system and third generation partnership plan (3
RdGeneration partnershipproject, 3GPP) system.
Background technology
At present, the mobile telephone system (cellular system) of direct sequence (direct-sequence) CDMA roughly can be classified into two kinds.A kind of is the accurate synchronous synchro system in minizone, base (inter-cell), and another kind is the nonsynchronous asynchronous system in minizone, base.For synchro system, the conversion (hand-off) of mobile phone between the sub-district, base will be very quick, because the zero-time of the identification code (identical code) of contiguous sub-district, base only is different fixing side-play amounts with the sub-district, base that is connected instantly.But in order to reach synchronous purpose, each sub-district, base also needs the equipment of many costlinesses, such as global positioning system (global positioning system, GPS) and accurately sequential oscillator (oscillator).The sequential oscillator provides the clock of a base station, and GPS provide clock to the time foundation.Except apparatus expensive, for a sub-district, base that is difficult to receive gps signal, for example in basement or intensive residential area, asynchronous system also has the difficulty in the enforcement.
As for asynchronous system, for example European wideband CDMA system that 3GPP proposed, each base station has adopted two synchronizing channels (synchronization channels).Via acquiring the synchronous code that is transmitted in two synchronizing channels, mobile terminal (mobile phone for example) can be set up good binding (link) and can not produce the problem of broken string when base station is changed.These two synchronizing channels be respectively main synchronizing channel (primary synchronization channel, PSCH) and secondary synchronization channel (secondarysynchronization channel, SSCH).In asynchronous system, PSCH is the general channels of all base stations, and (primary synchronizationcode PSC), is denoted as C to send once Primary Synchronisation Code by each time slot (slot)
p, constitute.The length of each PSC is 256 chips (chip), and PSC is when sending, and the time slot of down link (downlink) channel will be sent synchronously.
Contain among the SSCH one by 15 secondary synchronization code (secondary synchronization code SSC), is denoted as C
Ssc i, the identification ordered series of numbers that is constituted.The identification ordered series of numbers is constantly sent with cycle one time frame (time frame), and wherein, each SSC is by one group 16 quadratures (orthogonal) sign indicating number CS
1～CS
16Selected, and the length of each yard is 256 chips.One by one parallel of the SSC of identification in the ordered series of numbers and side by side being sent by base station with PSC.The identification number dependent of dead military hero also is the foundation of a base station down link scrambler (downlink scrambling code) in one of them of 64 different code group groups (code group).In other words, to be a length constituted by the ordered series of numbers of 15 orthogonal codes each code group group.Sign indicating number in 64 code group groups puts in order the position as shown in Figure 9.64 specific ordered series of numbers be for the ordered series of numbers that cyclic shift produced that makes them all be unique and design.That is to say, if the number of times of cyclic shift between 0 to 14, produced 960 (=64*15) individual ordered series of numbers is all differing from each other.Utilize this specific character, the zero-time of an employed code group group of base station and time frame can be identified in the mobile terminal.
When the base cell search was carried out in the mobile terminal of the wideband CDMA system that proposes as 3GPP, the mobile terminal can be looked for a base station that peak signal can be provided earlier, and then go for out that found base station the down link scrambler and the time frame synchronization.Such base cell search can following three steps be represented:
The first step: slot synchronization
When searching the sub-district, base, the mobile terminal can be found out PSC and its sequential of signal energy maximum earlier at PSCH.This action generally is to finish by a filter that corresponds to PSCH.Just because of each base station is all identical at the PSC that PSCH sent, so the PSC of signal energy maximum corresponds to the base station that peak signal can be provided.And just can determine the slot length and the zero-time of the base station of peak signal by the sequential of PSC.Therefore, the mobile terminal just can adjust automatically clock internal (clock) with the slot synchronization of the base station of peak signal.
Second step: the identification of code group group is synchronous with frame boundaries
After the slot synchronization, the mobile terminal just can be arrived and collect SSC among the SSCH, so just, can find out the identification ordered series of numbers that base station sent that the first step finds is which the code group group that belongs to 64 code group groups on earth, and time frame should from when, and just frame boundaries is synchronous.A kind of implementation method of intuition be when receiving one in the frame length SSC in continuous 15 time slots then this is received ordered series of numbers and compares with 15 issuable 960 ordered series of numbers of cyclic shift to form an ordered series of numbers by 64 code group groups.Because these 960 ordered series of numbers all are unique, institute is so that can be in wherein finding out identical ordered series of numbers.Like this, can know that just the base station that is found is to send which code group group, also can know the time frame time started that should the several time slots of cyclic shift be only this base station that finds, the purpose of frame synchronization when reaching.
The 3rd step: scrambler identification
After finishing above-mentioned two steps by two synchronizing channels, the mobile terminal is then by a Common Pilot Channel (common pilot channel, CPICH), utilize second SSC that goes on foot the code group group that is found to contrast one by one, just can confirm main scrambler (primary scrambling code).Then, (primary common control physical channel PCCPCH) just can find Primary Common Control Physical Channel.Then, the information of system and base station can read.
Second step of base cell search process is a theme of the present invention.SSCH is with deciding frame boundaries synchronous.A frame of 15 SSCH symbols has formed a coded sequence, and these 64 different coded sequences are taken from code book (codebook).Each frame in a base station repeats identical coded sequence.It is that all code phases that are different from all other coded sequences move that 64 arbitrary code phases of different coded sequences of this that is selected move (code phase shift).Had these character, the border of frame just can be via the correct start-phase that identifies the SSCH symbol sebolic addressing, and is detected.In order to satisfy above-mentioned character and the minimum range between the different coded sequences to be maximized, therefore (15,3) reed-solomon code of character used in proper names and in rendering some foreign names Lip river Wa Chang (GF (16)) lining has been proposed.About this error correcting code, can in textbook, find the standard reed solomon decoder of (15,3) reed-solomon code, and can correct to 6 symbol errors.Yet, because the noise of changeable, the channel of frequency errors (frequency error), channel, or other reason, the number of symbol error may be usually above 6.So the standard reed solomon decoder usually can't be replied an effectively sign indicating number sequence.
In the periodical of " IEEE Journal on Selected Areas in Communications vol.18, no.8August 2000 ", Yi-Ping Eric Wang has proposed another kind of method.The method that Wang proposed is: reach the period synchronously after, the running of receiving terminal at first receives that with SCH signal is relevant with 16 all S-SCH sequences, according to 64 reed solomon product code sequences that are used, is accumulated in N then
tThe SSCH correlation of individual period, each character string has the frame boundaries of 15 hypothesis.So have 960 hypothesis.When finishing, the hypothesis with largest cumulative amount is chosen as the candidate that frame boundaries-the code group group is right, and gives the next stage as sign indicating number identification usefulness at random.
The method that Wang proposed has preferable implementation result, but needs a large amount of internal memory and evaluation work.Utilization of the present invention combines the standard reed solomon decoder of reliability measurements (reliability measurement), proposes the profitable method of a kind of power and internal memory.
Summary of the invention
The present invention overcomes the shortcoming that above-mentioned conventional frame border is synchronous and the code group group is discerned, and one of its main purpose is for providing a kind of and the low-power of code group group identification and the method and apparatus of low memory requirements synchronous for frame boundaries.According to this, device of the present invention comprises a correlator (correlator bank), have several correlators (correlators), hard decision and reliability measuring unit (hard decision and reliabilitymeasurement unit), a coded sequence identifier (code sequence identifier), a frame boundaries finder (frame boundary finder) and a code group group recognition unit (code groupidentification unit).
Whenever receiving a signal, this signal just is sent in this correlator, with the correlation between the coded identification of discerning this signal and 16 quadratures.Hard decision and reliability measuring unit are picked out a hard decision symbol with high correlation then, and utilize one to go out reliability with 16 relevant function calculation of correlation.
Another object of the present invention provides the reed solomon decoder of the modification in a kind of coded sequence identifier.In preferred embodiment of the present invention, this reed solomon decoder of revising is according to the reliability measurements of hard decision symbol, and whether this is wiped free of to use a critical value to decide coded identification.When the number of effective symbol during more than or equal to a critical value between 3 and 15, the reed-solomon mistake that entire coded sequence is just delivered to standard is lost decoder and is decoded with wiping.
In another embodiment, this reed solomon decoder of revising is number and a critical value of eraser in a coded sequence relatively, and this critical value is the integer between 0 and 12.If eraser's number is not more than critical value, then coded sequence is just delivered to the mistake of standard and is wiped decoder.If decoder is not replied an efficient coding sequence, k the coded identification with lowest reliable degree just is wiped free of, and new coded sequence is sent to the mistake of this standard once again and wipes decoder.
Another purpose of the present invention is to utilize at least two frames of coded identification, goes to reduce the probability of symbol error, and improves the implementation result of coded sequence identifier.According to this, add a symbol and reliability and change unit (symbol and reliability update unit) in the coded sequence identifier.Because coded identification is periodically to be transmitted, after receiving and having write down a frame of coding, first coded identification in the frame that the coded identification that the next one is received should equal to be recorded in theory.This next coded identification and its reliability are used for changing corresponding coded identification in the frame that is recorded.When the symbol received more than a frame, according to their reliability, whether decision removes to change the hard decision symbol with extra symbol.
Another uses decoding policy more than a coded sequence at first receiving two coded sequences, by these two coded sequences of comparison, produces a new coded sequence then.If the corresponding codes symbol in two coded sequences receiving is unequal, then wipe this coded identification.This new coded sequence is sent to the mistake of standard then and wipes decoder.
Another decoding policy is included in a plurality of frames of coded identification, uses hard decision with ballot formula (voting).After receiving several coded sequences, write down their hard decision value of symbol.Corresponding codes symbol via adopting majority vote in a plurality of frames produces a new coded sequence.This new coded sequence is sent to the mistake of standard then and wipes decoder.
Another purpose of the present invention provides the synchronous method of a kind of frame boundaries.Look into 64 coded sequences in Fig. 9 via sight, the present invention finds that first coded identification in a frame must have minimum value of symbol.If minimum value of symbol is unique, this symbol is exactly the beginning (head of the frame) of this frame.If minimum value of symbol has occurred twice, then the next symbol after the frame beginning must have than the less value of next symbol after the minimum symbol of finding out at other.
Another object of the present invention provides a kind of profitable method of internal memory of code group group of recognition coding sequence.64 coding groups that exempt from the efficient coding of comma reed-solomon code also have this characteristic: the coded sequence in each group can be by 3 initial unique identifications of coded identification.By 3 the initial coded identifications that store each the coding group in 64 coding groups, the sign indicating number number (code number) of paid-in coded sequence just can be identified.
A kind of method that produces correct coded sequence of the present invention is characterized in that, comprises the following step:
(a) frame of received code symbol;
(b) by the frame of this coded identification, form the hard decision symbol of sequence, each hard decision symbol contains corresponding reliability value;
(c) send the reliability value of this hard decision symbol sebolic addressing and correspondence to the decoder of revising;
(d) utilize the decoder of this modification, produce correct coded sequence, and, then finish this method if the decoder of this modification is successfully replied;
(e) if the sum of the hard decision symbol that receives greater than the limit, then finishes this method; (f) receive several extra coded identifications; (g), upgrade this hard decision symbol sebolic addressing and corresponding reliability value according to this extra coded identification; And
(h) get back to step (c).
The method of the coded sequence that generation of the present invention is correct, wherein this step (g) use location is that the extra coded identification of m is according to the following step: the reliability value of (g1) calculating new hard decision symbol and new correspondence for this extra coded identification; (g2) the hard decision symbol of identification combination, its receiving position equals the remainder of m divided by n, and n is the symbolic number in this hard decision symbol sebolic addressing; (g3) if this new hard decision symbol is constantly equal to the hard decision symbol of this combination, then increase the corresponding reliability value of the hard decision symbol of this combination; (g4) if this new hard decision symbol is nonidentical in the hard decision symbol of this combination, and this new corresponding reliability value then reduces the corresponding reliability value of the hard decision symbol of this combination less than the corresponding reliability value of the hard decision symbol of this combination; And (g5) if this new hard decision symbol is nonidentical in the hard decision symbol of this combination, and this new corresponding reliability value is greater than the corresponding reliability value of the hard decision symbol of this combination, the hard decision symbol that then replaces this combination with this new hard decision symbol, and replace corresponding reliability value with the corresponding reliability value of new minimizing.
The method of the coded sequence that generation of the present invention is correct, wherein this decoder revised in this step (d) is carried out the following step: (i) receive a new symbol by this hard decision symbol sebolic addressing; (ii) if the pairing reliability value of this new symbol greater than the reliability critical value, assert that then this new symbol is a valid symbol, otherwise be invalid symbol; (iii) write down the sum of significant character; (iv) if the sum of the significant character of this record more than or equal to the critical value of significant character sum, then according to valid symbol, send coded sequence to wrong with wipe decoder, otherwise execution in step is (vi); (v) if this mistake with wipe decoder and successfully reply, then produce this correct coded sequence and finish this step (d); And (vi) if receive whole symbols of this hard decision symbol sebolic addressing, end step (d) then, otherwise get back to step (i).
The method of the coded sequence that generation of the present invention is correct, wherein this hard decision symbol sebolic addressing has 15 hard decision symbols at the most, and the critical value of this significant character sum be one between 3 to 15 integer, and be the function of the hard decision symbolic number that receives.
The method of the coded sequence that generation of the present invention is correct, this decoder revised in this step (d) is carried out the following step: (i) the hard decision symbol of receiving sequence comprises valid symbol and invalid symbol; (ii) if the sum of the unblind of this hard decision symbol sebolic addressing is not more than the critical value of unblind sum, then according to valid symbol, send coded sequence to wrong with wipe decoder, otherwise finish this step (d); (iii) if this mistake with wipe decoder and successfully reply, then produce this correct coded sequence and finish this step (d); (iv) by the present significant character of this hard decision symbol sebolic addressing, choose k symbol, this k symbol has k minimum reliability value in the present significant character of this hard decision symbol sebolic addressing; (assert that v) this k symbol is invalid, and determine the sum of unblind by this hard decision symbol sebolic addressing; And (vi), then get back to step (ii), otherwise finish this step (d) if the sum of the unblind of this hard decision symbol sebolic addressing is not more than the critical value of this unblind sum.
The method of the coded sequence that generation of the present invention is correct, wherein this hard decision symbol sebolic addressing has 15 hard decision symbols at the most, and the critical value of this unblind sum be one between 0 to 12 integer.
The another kind of method that produces correct coded sequence of the present invention comprises the following step: (a) first and second frames of received code symbol; (b), form the first and second hard decision symbol sebolic addressings by first and second frames of this coded identification; (c) each corresponding in each the hard decision symbol in this first sequence and this second sequence hard decision symbol is made comparisons,, assert that then this hard decision symbol is a valid symbol, otherwise be invalid symbol if corresponding hard decision symbol is identical; (d) if the sum of invalid hard decision symbol less than the critical value of invalid hard decision total number of symbols, then according to the effective hard decision symbol of step (c) decision, send coded sequence to wrong with wipe decoder, otherwise finish this method; And (d) if this mistake with wipe decoder and successfully reply, then produce correct coded sequence and finish this method.
The method of the coded sequence that another generation of the present invention is correct comprises the following step: the frame that (a) receives a plurality of coded identifications; (b) form a plurality of hard decision symbol sebolic addressings, each sequence corresponds to one of frame of all coded identifications that receive; (c) form new hard decision symbol sebolic addressing, the value of each hard decision symbol is adopted the majority vote decision by hard decision symbol corresponding in these a plurality of sequences in this new sequence; (d) send this new sequence to wrong with wipe decoder; And (e) if this mistake with wipe decoder and successfully reply, then produce correct coded sequence and finish this method.
Of the present invention another produce the method for correct coded sequence, also comprise the following step: (f) if this mistake with wipe decoder and can't successfully reply the efficient coding sequence, then receive the frame of new coded identification; And (g) get back to step (b).
In conjunction with detailed description and claims of following accompanying drawing, embodiment, will on address other purpose of the present invention and advantage and be specified in after.
Description of drawings
For further specifying technology contents of the present invention, be described in detail as follows below in conjunction with drawings and Examples, wherein:
Fig. 1 is a system block diagram of the synchronous device of code group group identification of the present invention and frame boundaries.
Fig. 2 illustrates the preferred embodiment of a reed solomon decoder of revising, and is called the flow chart of " critical value is wiped decoder ".
Fig. 3 illustrates the preferred embodiment of the reed solomon decoder that another was revised, and is called the flow chart of " decoder is wiped in segmentation ".
Fig. 4 illustrates symbol and reliability refresh routine.
Fig. 5 explanation is when a decoding policy that receives after surpassing a frame.
Fig. 6 illustrates a kind of decoding policy of a plurality of frames of only having a hard decision.
Fig. 7 illustrates the decoding policy of a plurality of frames that a kind of tool is voted for.
Fig. 8 illustrates the frame boundaries finder.
Fig. 9 coded sequence figure.
Embodiment
Purport of the present invention is to utilize the reed-solomon mistake of standard and wipe decoder (error-and-erasure decoder), measures in conjunction with reliability, and in UMTS WCDMA system, it is synchronous with frame boundaries to do the identification of code group group.Fig. 1 is a system block diagram of the synchronous device of code group group identification of the present invention and frame boundaries.This device comprises a correlator, wherein contains 101, one hard decisions of several correlators and 103, one frame boundaries finders 104 of 102, one coded sequence identifiers of reliability measuring unit and a code group group recognition unit 105.
As is well known, each among the SSC is all corresponding to an efficient coding in (15,3) sign indicating number.Usually, after through the 16Walsh code correlator, the error rate of hard decision symbol (hard-decision symbol) is all too high, so that the decoder of most situation 1 standard all can't successfully be replied efficient coding.Yet decoder still has lot of advantages, such as, less memory requirements and lower computational complexity (computational complexity).
According to the present invention, just be sent in the correlator that contains 16 correlators 101 whenever receiving a signal, to discern the coded identification CS of itself and 16 quadratures
01, CS
02..., and CS
16Between correlation.At time m, be { r from the output correlation of 16 correlators
01 m, r
02 m..., r
16 m.Hard decision value of symbol R at time m
mBy { CS
01, CS
02..., CS
16In pick out one and have high correlation person.Therefore, reliability is measured and just can be defined as 16 correlation { r
01 m, r
02 m..., r
16 mA function, be used for measuring hard decision value of symbol R
mReliability.For example, reliability is measured L
mCan be defined as
Or
As shown in Figure 1, hard decision and reliability measuring unit 102 receive the correlation of being sent here by each correlator 101 respectively, remake hard decision to choose a symbol R
mAfterwards, comply with the formula of above-mentioned prior decision again, to the symbol R that selects
mCalculate its reliability and measure L
m
Because 64 codings each in the group is all (15,3) efficient coding in the sign indicating number, reed solomon decoder is replied after the effective code, and the present invention chooses a coded identification with high-reliability measured value, and wipes remaining coded identification.Based on the characteristic of (15,3) sign indicating number,, then there are 12 coded identifications to be wiped free of at most with low reliability measurements if whole 15 coded identifications all receive.
According to the present invention, coded sequence identifier 103 comprises a decoder of revising 111.A preferred embodiment of the decoder of revising 111 is one, and " critical value-wipe decoder " (threshold-erasedecoder), wherein according to the reliability measurements of calculating in hard decision and the reliability measuring unit 102, uses a critical value σ
rWhether this is wiped free of to decide coded identification.
Fig. 2 explanation is wiped the flow chart of the implementation method of decoder as the critical value of the decoder of revising 111.When receiving a new coded identification, just write down its hard decision value of symbol and corresponding reliability value.If reliability is less than critical value σ
r, the coded identification that receives just is identified as invalid symbol, and is wiped, and corresponding reliability value is set at-∞.Otherwise, if the reliability value is greater than σ
r, (the number of valid symbol VSN) just adds 1 just to write down its hard decision value of symbol and significant character number.
Suppose another critical value σ
vBe one between 3 to 15 integer and symbolic number (the number of received symbol, function RSN) for receiving.When VSN more than or equal to σ
vThe time, entire coded sequence is just delivered to the mistake of standard and is wiped decoder.If decode procedure failure and RSN just receive another new coded identification less than 15.If the new reliability value of character that receives is greater than σ
r, the new coded sequence that contains the coded identification of new reception is just delivered to the mistake of standard again and is wiped decoder.When the mistake of standard when wiping decoder and reply an efficient coding, whole decode procedure finishes with regard to leaving.
Fig. 3 illustrates the preferred embodiment of another decoder of revising 111, is called " decoder is wiped in segmentation " (erase K by K decoder).The coded sequence that contains 15 coded identifications is sent " decoder is wiped in segmentation " to.Record hard decision value of symbol (R
0, R
1, R
2..., R
14) and its corresponding reliability (L
0, L
1, L
2..., L
14).Attention: some coded identification is regarded as wiping (erasures) and its reliability is set as-∞.
Calculate the total e that wipes
0, and and critical value σ between 0 to 12
eMake comparisons.If the total e that wipes
0Be not more than σ
e, then coded sequence is just delivered to the mistake of standard and is wiped decoder.If decode procedure failure, at first (L
0, L
1, L
2..., L
14) according to the ascending ((L that is ranked into
(0), L
(1), L
(2)..., L
(14)), L wherein
(i)Corresponding to R
(i)At this moment, e is arranged
0Individual coded identification is wiped free of, therefore, and R
(0), R
(1)..., R
(e0-1)For wiping and L
(0)=L
(1)=...=L
(e0-1)=-∞.At L
(e0), L
(e0+1)..., L
(14)In, be L corresponding to minimum k reliability of the symbol that is not wiped free of
(e0), L
(e0+1)..., L
(e0+k-1)
Then, Dui Ying reliability is L
(e0), L
(e0+1)..., L
(e0+k-1)K coded identification R
(e0), R
(e0+1)..., R
(e0+k-1)Be wiped free of, wherein k is a positive integer and is present e
0The function of value that is to say that k is variable in each erase process.Therefore, be wiped free of number and become e
0+ k.(e is counted in more present being wiped free of
0+ k) with critical value σ
e, be not more than σ if be wiped free of number
e, then new coded sequence (contain k wipes more) is sent to decoder once more.When the mistake of standard with wipe demultiplexer (DMPX) and send an efficient coding sequence or be wiped free of number and surpass critical value σ
eThe time, whole the wiping that just be through with-compare-decode procedure.
For the further symbol error probability that reduces, and the operating efficiency of improvement coded sequence identifier 103, the present invention can use the not only frame of a coded identification.According to this, on coded sequence identifier 103, can add a symbol and reliability updating block 112, as shown in Figure 1.Under will discuss when receiving and surpass 15 symbols when above, the method for upgrading hard decision value of symbol and reliability measurement and use are above the translator of a frame.
Because 15 coded identifications are periodically transmitted,, do not need to abandon this coded sequence if a coded sequence of 15 coded identifications can't be deciphered.In other words, new coded identification can be received and be used for upgrading hard decision value of symbol and the measurement of its corresponding reliability.Fig. 4 illustrates that this upgrades an embodiment of hard decision value of symbol and reliability method of measurement.
The coded sequence of supposing 15 coded identifications before having received is (R
0, R
1, R
2..., R
14), and its corresponding reliability is (L
0, L
1, L
2..., L
14).Because 15 coded identifications are periodically to transmit, so ideally,
R
i＝R
imod15?i＝15，16，17，...。
After the frame that receives a coded identification, total RSN value is 15.When receiving the 16th coded identification, hard decision measured value R ' (or R
15) and its corresponding reliability L ' (or L
15) go on record.Ideally, R ' should equal R
0But, since the appearance of noise and other reason, hard decision measured value R
0And R
15May not can equate.If these two hard decision measured value (R
0And R
15) equate, then upgrade reliability L to increase some numerical value
0The numerical value of this increase is one and original reliability L
0The reliability L that receives at present
15Relevant function.For example, new reliability can with these two reliabilitys and represent:
L
0(after?updating)＝L
0+L
15。
Yet, if these two hard decision measured value (R
0And R
15) unequal, then symbol and corresponding reliability just must be by its corresponding reliability (L
0And L
15) make comparisons, upgrade according to its result after relatively.Hard decision measured value after the renewal is set as the value of symbol that has corresponding to bigger reliability value.
And the reliability after the renewal should reduce.The reduction of reliability also is a L
0And L
15Function.For example,
L
0(after?updating)＝max(L
0，L
15)-min(L
0，L
15)
Represent with identical symbol, when receiving the 17th coded identification (R
16) time, this moment, the RSN value equaled 16, R
16Should equal R
1, hard decision value of symbol and reliability update routine can be applied to R again
1And R
16, and or the like, the rest may be inferred.
With upgrading R
iAnd R
I mod 15The hard decision measured value and the method for reliability, just can introduce the program of decoding then.Fig. 5 explanation surpasses the decoding policy of frame when (containing 15 symbols) when receiving.When 15 coded identifications
R=(R
0, R
1, R
2..., R
14) original coding sequence can't be decoded the time, σ
NIndividual new coded identification (R
15, R
16, R
17, R
18) be received.Hard decision measured value and reliability refresh routine are applied to R
0And R
15, R
1And R
16, R
2And R
17And R
3And R
18, just can obtain one group of new coded sequence.What deserves to be mentioned is that even the hard decision measured value may not be changed, its corresponding reliability may be different.
New coded sequence
R' and new reliability sequence
L' be sent to the decoder of revising.If this new coded sequence
R' can't be decoded once more, σ in addition
NIndividual new coded identification can be received, to obtain the new coded sequence of another group
R" and new reliability sequence
L", σ wherein
NCan be any positive integer and can become with each refresh routine.
R" and
L" be sent to Fig. 2 and Fig. 3 once more, or both combinations, the decoder of shown modification.When the decoder of revising was replied an efficient coding sequence, whole decoder program just was through with.For fear of (signal-to-noise ratio, SIN) or the infinite circulation that other factors produced (endless loop), the present invention limits total RSN value, to stop circulation because of low signal-noise ratio.When the symbolic number that always receives surpasses a certain predetermined integer value MAX-RSN, just abandon the current encoder sequence.
Fig. 6 illustrates that another surpasses the decoding policy of a coded sequence to only having hard decision.At first, receive two coded sequences
R 1With
R 2Its hard decision measured value is respectively
R 1=(R
1 0, R
1 1, R
1 2..., R
1 14) and
R 2=(R
2 0, R
2 1, R
2 2..., R
2 14).Compare R
1 jAnd R
2 j, j=0,1,2 ..., 14.If hard decision measured value (R
1 jAnd R
2 j) unequal, R
jJust be set to and wipe.Relatively
R 1With
R 2In behind 15 symbols, if always be wiped free of several e
0Less than critical value σ
e, contain e
0The individual coded sequence of wiping
R=(R
0, R
1, R
2..., R
14) just be sent to the mistake of standard and wipe demultiplexer (DMPX), wherein σ
eBe one between arbitrary integer of 1 to 13.
If coded sequence
R' can't successfully be deciphered, these two coded sequences can be abandoned or be attempted using other decoding policy.On the other hand, another coded sequence that contains 15 coded identifications
R 3Sustainable being received.Utilize above-mentioned program, by comparing coded sequence
R 3With previous coded sequence
R', can
R' new coded sequence as a result of middle record.If
R' the middle total erasure number e that is declared
0Less than σ
e, contain e
0The individual coded sequence of wiping
R' just be sent to the mistake of standard and wipe demultiplexer (DMPX).Annotate σ
eValue can be lowered.When the mistake of standard is replied an efficient coding with wiping demultiplexer (DMPX), or during the maximum number of the code sequence columns that the receives coded sequence that equals to be allowed to, whole procedure finishes at this point.
Fig. 7 illustrates the decoding policy of the hard decision that another kind is voted for.At the beginning, receive σ
sIndividual coded sequence, and its hard decision measured value
R 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 σ s=(R
σ s 0, R
σ s 1, R
σ s 2..., R
σ s 14) be recorded.To each coded identification, its hard decision measured value R
j, j=0,1,2 ..., 14, be set to set { R
1 j, R
2 j, R
3 j..., R
σ s jMajority vote (majority vote).And result's coded sequence is recorded in
R=(R
0, R
1, R
2..., R
14), and be sent to the mistake of standard and wipe demultiplexer (DMPX).If this decode procedure failure can receive a new coded sequence again, and adopts majority vote.Then, result's coded sequence is decoded again.When the mistake of standard with wipe that demultiplexer (DMPX) is replied an efficient coding or during the code sequence columns of the maximum that the code sequence columns that receives equals to be allowed to, whole decoding policy just comes to an end.
As shown in Figure 1, after coded sequence identifier 103 identifies correct coded sequence, just find out the frame boundaries of coded sequence by the frame boundaries finder.Fig. 8 explanation, after separating efficient coding sequence of coding answer, a kind of method of seeking frame boundaries.
With reference to figure 9,64 coded sequences all tool exempt from the efficient coding position of comma reed-solomon code.That is to say, all codings all do not have inner reappearance (internal repetition), and each coded sequence in 15 coded identifications, its first coded identification contain minimum value of symbol and minimum value of symbol is looked for twice at most in coded sequence.If minimum value of symbol is unique, this symbol is exactly the beginning of frame.If minimum value of symbol has been looked for twice, then the next symbol after the frame beginning must have the value more less than the next symbol of the minimum symbol of finding out in other period.For example, if minimum value of symbol the time hop count=0 and the time found among hop count=j, then the time symbol in hop count=1 must have one than the time the less value of symbol of symbol among hop count=j+1.For instance, the coded sequence in group 0 is (1,1,2,8,9,10,15,8,10,16,2,7,15,7,16), and wherein Zui Xiao value of symbol is 1.Minimum value of symbol respectively the time hop count=0 and the time hop count=1 in found twice.The next symbol of these two values of symbol relatively, hop count in the time of just=1 and the time hop count=2, can find the symbol of frame after beginning, hop count in the time of just=1 has less value of symbol.Lift another example, the coded sequence in group 63 is (9,12,10,15,13,14,9,14,15,11,11,13,12,16,10), and wherein Zui Xiao value of symbol is 9.Minimum value of symbol the time hop count=0 and the time hop count=6 li found twice.Two values of symbol of more next symbol, hop count in the time of just=1 and the time hop count=7, can find the symbol of frame after beginning, hop count in the time of just=1 has less value of symbol.
As mentioned above, the efficient coding sequence of being replied by decoder may be the periodicity displacement (cyclic shift) of an original coded sequence.Seek the border that two minimum in coded sequence values of symbol can determine frame.If two minimum values of symbol are unequal, the then beginning of frame boundaries (head index) is the position of the value of symbol of minimum.If two minimum values of symbol equate, just can determine the beginning position via the next symbol that compares these two values of symbol.Based on above-mentioned character, the beginning position that find out frame boundaries is not so difficult.
After the decision frame boundaries, device of the present invention utilizes code group group recognition unit 105 to go to recognition coding group.With reference to figure 9, the coded sequence that can be observed in each group can be identified uniquely by first three coded identification.Utilize this character, have only preceding 3 hurdles to be compared among Fig. 9.Can confirm the group number of encoding via comparing first three coded identification.Therefore, code group group recognition unit of the present invention reduces the demand of internal memory greatly.
What deserves to be mentioned is that Fig. 2 and Fig. 3 illustrate " critical value-wipe decoder " and " decoder is wiped in segmentation " of the decoder 111 of modification of the present invention respectively.Can do various variation for these decoders.For example, if necessary also can be with " critical value-wipe decoder " and " decoder is wiped in segmentation " both combinations.
But the above only is preferred embodiment of the present invention, when not limiting scope of the invention process with this.Every equalization of doing according to claims of the present invention changes and revises, and all should still belong in the scope that patent of the present invention contains.
Claims (9)
1. method that produces correct coded sequence comprises the following step:
(a) frame of received code symbol;
(b) by the frame of this coded identification, form the hard decision symbol of sequence, each hard decision symbol contains corresponding reliability value;
(c) send the reliability value of this hard decision symbol sebolic addressing and correspondence to the decoder of revising;
(d) utilize the decoder of this modification, produce correct coded sequence, and, then finish this method if the decoder of this modification is successfully replied;
(e) if the sum of the hard decision symbol that receives greater than the limit, then finishes this method;
(f) receive several extra coded identifications;
(g), upgrade this hard decision symbol sebolic addressing and corresponding reliability value according to this extra coded identification; And
(h) get back to step (c).
2. the method for the coded sequence that generation as claimed in claim 1 is correct, wherein this step (g) use location is that the extra coded identification of m is according to the following step:
(g1) calculate the reliability value of new hard decision symbol and new correspondence for this extra coded identification;
(g2) the hard decision symbol of identification combination, its receiving position equals the remainder of m divided by n, and n is the symbolic number in this hard decision symbol sebolic addressing;
(g3) if this new hard decision symbol is constantly equal to the hard decision symbol of this combination, then increase the corresponding reliability value of the hard decision symbol of this combination;
(g4) if this new hard decision symbol is nonidentical in the hard decision symbol of this combination, and this new corresponding reliability value then reduces the corresponding reliability value of the hard decision symbol of this combination less than the corresponding reliability value of the hard decision symbol of this combination; And
(g5) if this new hard decision symbol is nonidentical in the hard decision symbol of this combination, and this new corresponding reliability value is greater than the corresponding reliability value of the hard decision symbol of this combination, the hard decision symbol that then replaces this combination with this new hard decision symbol, and replace corresponding reliability value with the corresponding reliability value of new minimizing.
3. the method for the coded sequence that generation as claimed in claim 1 is correct, wherein this decoder revised in this step (d) is carried out the following step:
(i) receive a new symbol by this hard decision symbol sebolic addressing;
(ii) if the pairing reliability value of this new symbol greater than the reliability critical value, assert that then this new symbol is a valid symbol, otherwise be invalid symbol;
(iii) write down the sum of significant character;
(iv) if the sum of the significant character of this record more than or equal to the critical value of significant character sum, then according to valid symbol, send coded sequence to wrong with wipe decoder, otherwise execution in step is (vi);
(v) if this mistake with wipe decoder and successfully reply, then produce this correct coded sequence and finish this step (d); And
(vi) if receive whole symbols of this hard decision symbol sebolic addressing, end step (d) then, otherwise get back to step (i).
4. the method for the coded sequence that generation as claimed in claim 3 is correct, wherein this hard decision symbol sebolic addressing has 15 hard decision symbols at the most, and the critical value of this significant character sum be one between 3 to 15 integer, and be the function of the hard decision symbolic number that receives.
5. the method for the coded sequence that generation as claimed in claim 1 is correct, this decoder revised in this step (d) is carried out the following step:
(i) the hard decision symbol of receiving sequence comprises valid symbol and invalid symbol;
(ii) if the sum of the unblind of this hard decision symbol sebolic addressing is not more than the critical value of unblind sum, then according to valid symbol, send coded sequence to wrong with wipe decoder, otherwise finish this step (d);
(iii) if this mistake with wipe decoder and successfully reply, then produce this correct coded sequence and finish this step (d);
(iv) by the present significant character of this hard decision symbol sebolic addressing, choose k symbol, this k symbol has k minimum reliability value in the present significant character of this hard decision symbol sebolic addressing;
(assert that v) this k symbol is invalid, and determine the sum of unblind by this hard decision symbol sebolic addressing; And
(vi), then get back to step (ii), otherwise finish this step (d) if the sum of the unblind of this hard decision symbol sebolic addressing is not more than the critical value of this unblind sum.
6. the method for the coded sequence that generation as claimed in claim 5 is correct, wherein this hard decision symbol sebolic addressing has 15 hard decision symbols at the most, and the critical value of this unblind sum be one between 0 to 12 integer.
7. method that produces correct coded sequence comprises the following step:
(a) first and second frames of received code symbol;
(b), form the first and second hard decision symbol sebolic addressings by first and second frames of this coded identification;
(c) each corresponding in each the hard decision symbol in this first sequence and this second sequence hard decision symbol is made comparisons,, assert that then this hard decision symbol is a valid symbol, otherwise be invalid symbol if corresponding hard decision symbol is identical;
(d) if the sum of invalid hard decision symbol less than the critical value of invalid hard decision total number of symbols, then according to the effective hard decision symbol of step (c) decision, send coded sequence to wrong with wipe decoder, otherwise finish this method; And
(d) if this mistake with wipe decoder and successfully reply, then produce correct coded sequence and finish this method.
8. method that produces correct coded sequence comprises the following step:
(a) frame of a plurality of coded identifications of reception;
(b) form a plurality of hard decision symbol sebolic addressings, each sequence corresponds to one of frame of all coded identifications that receive;
(c) form new hard decision symbol sebolic addressing, the value of each hard decision symbol is adopted the majority vote decision by hard decision symbol corresponding in these a plurality of sequences in this new sequence;
(d) send this new sequence to wrong with wipe decoder; And
(e) if this mistake with wipe decoder and successfully reply, then produce correct coded sequence and finish this method.
9. as the method for claim 8 coded sequence that described generation is correct, also comprise the following step:
(f) if this mistake with wipe decoder and can't successfully reply the efficient coding sequence, then receive the frame of new coded identification; And
(g) get back to 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 true CN1825773A (en) | 2006-08-30 |
| CN100367682C 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) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011032387A1 (en) * | 2009-09-15 | 2011-03-24 | 中兴通讯股份有限公司 | Method and device for decoding reed-solomon (rs) code |
| CN101262431B (en) * | 2007-02-19 | 2011-09-07 | 索尼株式会社 | Communication device and communication method |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW518839B (en) * | 2000-08-30 | 2003-01-21 | Ind Tech Res Inst | Method and device of code group identification and frame edge synchronization |
| KR100342483B1 (en) * | 2000-09-09 | 2002-06-28 | 윤종용 | Apparatus and method for searching base station in umts |
| JP3729329B2 (en) * | 2000-09-19 | 2005-12-21 | 株式会社エヌ・ティ・ティ・ドコモ | Cell search method for mobile station in mobile communication system and mobile station in mobile communication system |
-
2003
- 2003-04-22 CN CNB2006100094292A patent/CN100367682C/en not_active Expired - Fee Related
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101262431B (en) * | 2007-02-19 | 2011-09-07 | 索尼株式会社 | Communication device and communication method |
| WO2011032387A1 (en) * | 2009-09-15 | 2011-03-24 | 中兴通讯股份有限公司 | Method and device for decoding reed-solomon (rs) code |
| 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 |
| US8677222B2 (en) | 2009-09-15 | 2014-03-18 | Zte Corporation | Method and device for decoding Reed-Solomon (RS) code |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100367682C (en) | 2008-02-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1275407C (en) | Method and apparatus for code group identification and frame synchronization | |
| CN100341377C (en) | Cell searching method in asyncronous wideband code division multiple access system | |
| CN100568848C (en) | The apparatus and method of receiving packet data control channel in the mobile communication system | |
| CN1278380A (en) | Cell searching in a CDMA communications system | |
| CN1359563A (en) | Method and apparatus for fast WCDMA acquisition | |
| DK0705510T3 (en) | Receives a direct-sequence spread-spectrum signal using the rake principle | |
| CN1156105C (en) | Device and method for performing frame sync using sync channel in mobile communication system | |
| CN1291388A (en) | Speed detection in direct sequence CDMA system | |
| WO2006099973A1 (en) | Determining a detection signal in a spread spectrum communications system | |
| US7929509B2 (en) | Reduced cell acquisition time | |
| CN1309269C (en) | Method and arrangement for starting extended spectrum radio telephone receiver | |
| US7269206B2 (en) | Flexible correlation for cell searching in a CDMA system | |
| CN1808924A (en) | Apparatus and method for acquiring code group in asynchronous w-cdma system | |
| CN1244241C (en) | Method of optimizing small zone search in mobile telecon net | |
| CN1825773A (en) | Method for code group identification and frame synchronization | |
| Sriram et al. | Cyclically permutable codes for rapid acquisition in DS-CDMA systems with asynchronous base stations | |
| US7224718B2 (en) | Slot synchronization for a CDMA system | |
| CN101174852B (en) | WCDMA subdistrict searching time slot synchronization method and system | |
| CN1260933C (en) | cellular telephony searcher | |
| Sriram et al. | Fast acquisition method for DS-CDMA systems employing asynchronous base stations | |
| CN1780174A (en) | The first and second steps series iterative method for area searching in time divided duplexing system | |
| CN1303839C (en) | Method and device for reducing sum of calculation for searching in zones and increasing accuracy of calculation | |
| US20060188009A1 (en) | Frame synchronization using soft decisions in a universal mobile telephone system receiver | |
| CN100345386C (en) | Method for eliminating interence of fixed threshold value | |
| CN1331866A (en) | Method and appts. for pseudonoise spreading in CDMA communication system |
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 |