CN1702979A - Equalization technique and associated detection technique combined receiver and receiving method thereof - Google Patents
Equalization technique and associated detection technique combined receiver and receiving method thereof Download PDFInfo
- Publication number
- CN1702979A CN1702979A CN 200510076963 CN200510076963A CN1702979A CN 1702979 A CN1702979 A CN 1702979A CN 200510076963 CN200510076963 CN 200510076963 CN 200510076963 A CN200510076963 A CN 200510076963A CN 1702979 A CN1702979 A CN 1702979A
- Authority
- CN
- China
- Prior art keywords
- receiver
- data
- output
- detection
- algorithm
- 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
This invention relates to receiver combined with equilibrium and integration technique, which comprises the following parts: equilibrium receiver to receive the expansion factors activated and to process the interference signals through equilibrium technique; integration test receiver to receive the data expansion factors activated and process the multiple-address interference between signals though integration technique; selector to selectively output; controller to activate the said receiver and test receiver according to senior expansion factors and to control the output of the receivers.
Description
Technical field
The invention belongs to wireless communication technology field, particularly the receiver and the method for reseptance of 3-G (Generation Three mobile communication system) TD-SCDMA (TD SDMA) system.
Background technology
Along with the following multimedia service demand growing to high speed data transfer, wireless data service will sharply increase, this just requires 3-G (Generation Three mobile communication system) must have some characteristics of suitable transmitting data service, as high data volume, high sudden, high reliability etc.For the TD-SCDMA system of one of 3G (Third Generation) Moblie standard, can both provide high-quality low speed speech business at all types of user of different communication demand, high-quality high speed data transfer business also is provided.Common TD-SCDMA system supports variable bandspread factor, when spreading factor greater than 1 the time, system interference mainly is that the multiple access between different user disturbs, the intersymbol interference that is caused by multipath transmisstion accounts for back burner, and the joint detection algorithm that be used widely this moment in the TD-SCDMA system can obtain reasonable reception.When spreading factor is 1 high speed data transfer business, the interference that receiver is subjected to mainly is the intersymbol interference that is caused by multipath transmisstion, receive if still use associated detection technique merely, receptivity is just very undesirable and will cause serious error sign ratio.
Summary of the invention
In order to overcome the problems referred to above, the present invention proposes the method for reseptance that a kind of balancing technique and associated detection technique combine, and multiple access disturbs and the overall performance of part intersymbol interference raising receiver to eliminate.
The purpose of this invention is to provide a kind of novel method of reseptance and device, realizing that the TD-SCDMA receiver all reaches high-quality reception to the different spreading factor of height, thereby satisfy following the Internet and multimedia application rapid increase to 3G wireless high-speed data business demand.
The invention provides the receiver that a kind of balancing technique and associated detection technique combine, it is characterized in that comprising:
Equalizing receiver is activated when the spreading factor that receives data is 1, the signal of input is carried out the signal of balancing technique processing and output elimination intersymbol interference;
The joint-detection receiver when the spreading factor that receives data is activated greater than 1 the time, carries out that associated detection technique is handled and output eliminates that multiple access disturbs and the signal of part intersymbol interference to the signal of input;
Selector is selected output from described equalizing receiver or described joint-detection receiver;
Controller, the spreading factor that transmits according to high level activates described equalizing receiver or described joint-detection receiver, and controls the output that described selector is selected described equalizing receiver or described joint-detection receiver.
Described equalizing receiver is to adopt receiver linear equalization or nonlinear equalization.Described joint-detection receiver is the receiver that adopts linear combined detection and non-linear joint-detection.
The invention provides the receiver that a kind of balancing technique and associated detection technique combine, it is characterized in that comprising:
The equalizing receiver of many grids viterbi algorithm is activated when the spreading factor that receives data is 1, the signal of input is carried out the signal of the processing and the output elimination intersymbol interference of many grids viterbi algorithm;
The joint-detection receiver of least mean-square error piecemeal linearity when the spreading factor that receives data is activated greater than 1 the time, carries out the united detection processing and the output of least mean-square error piecemeal linearity and eliminates interference signals to the signal of input;
Selector is selected output from described equalizing receiver or described joint-detection receiver;
Controller, the spreading factor that transmits according to high level activates described equalizing receiver or described joint-detection receiver, and controls the output that described selector is selected described equalizing receiver or described joint-detection receiver.
The receiver that this balancing technique and associated detection technique combine also comprises:
Data extractor, the data-signal that is used for receiving is separated into data symbol and training sequence;
Data buffer is used to store described data symbol;
The training sequence buffer is used to store described training sequence; And
Demodulator is used for the signal that the described selector of demodulation is exported.
The equalizing receiver of described many grids viterbi algorithm comprises:
Channel estimator carries out the estimation of channel impulse response with local training sequence data and the training sequence data that receives;
The channel impulse response processor is worked under the control of controller, is 1 if receive the spreading factor of data, then produces the ideal communication channel response of estimating, as the input of described many grids viterbi algorithm module;
Many grids viterbi algorithm module is carried out many grids viterbi algorithm to the described data symbol from described data buffer and is handled, and realizes maximum-likelihood sequence estimation; With
Descrambler carries out scramble process to the output of described many grids viterbi algorithm module.
Described joint-detection receiver comprises:
Described channel estimator carries out the estimation of channel impulse response with local training sequence data and the training sequence data that receives;
Described channel impulse response processor is worked under the control of controller, if the spreading factor that receives data greater than 1, is then done the denoising processing to channel estimating, and produces the input of estimating noise power as least mean-square error piece linear equalization module;
Least mean-square error piece linear algorithm module, work greater than 1 o'clock at the spreading factor that receives data, spreading code according to estimated channel impulse response and user produces sytem matrix, from the data of reception and the noise power and the transmission data of estimation the data of described channel estimator output are carried out scramble process, ask relevant and output data behind the descrambling and different spreading codes.
Described many grids viterbi algorithm module comprises:
The grid initialization unit in the beginning that receives time slot data, produces the initialization information of many grids viterbi algorithm according to the channel impulse ideal response of channel impulse processor output;
Control unit is controlled the beginning of other unit of scheduling and the access of information according to initialization information;
The branch metric unit is used for the calculating and the renewal of branch metric;
The path is selected cell relatively, the contended path cumulative metric value that the selection of executing state transfer path and the hard decision of symbol and output reliability information test cell need and first hard decision symbol on the contended path;
The reliability information measuring unit is implemented as the receiver soft-decision output algorithm that improves the channel-decoding performance and adopt.
Described least mean-square error piece linear algorithm module comprises:
The training sequence interference cancellation module is used for eliminating the chip that receives training sequence in the data and disturbs to the leading interference of first data block with to the hangover of second data block, and the data of disturbing are eliminated in output;
The sytem matrix generation module, the output system matrix;
Matched filter module, be used for the sytem matrix that the described sytem matrix generation module of computing generates conjugate transpose with eliminate the long-pending of the data disturbed and export this result;
The covariance matrix generation module is according to the system matrix computation covariance matrix of described sytem matrix generation module output;
The Matrix Solving module is estimated to handle to the output of matched filter module, the output and the noise power of covariance matrix generation module, obtains sending the estimation of sequence.
Described Matrix Solving module comprises:
The Cholesky decomposing module is decomposed into lower triangular matrix to the symmetrical matrix of conjugation symmetry and its conjugation symmetrical matrix is long-pending;
The matrix inversion unit, the result that the Cholesky decomposing module is decomposed carries out matrix inversion;
The matrix multiple unit multiplies each other and exports the output matrix of described matrix inversion unit.
The method of reseptance that the present invention also provides a kind of balancing technique and associated detection technique to combine is characterized in that may further comprise the steps:
The data-signal that receives is separated into data symbol and training sequence;
In data buffer, store described data symbol and the described training sequence of storage in the training sequence buffer;
If the spreading factor of described data equals 1, activate equalizing receiver, the signal of input is carried out the signal of balancing technique processing and output elimination intersymbol interference;
If the spreading factor of described data is greater than 1, activation joint-detection receiver carries out associated detection technique to the signal of importing and handles and export the signal of eliminating multiple access interference and part intersymbol interference;
Select the output signal of joint-detection receiver or the output signal of equalizing receiver according to spreading factor.
The step that described associated detection technique is handled comprises:
Local training sequence data and the training sequence data that receives are carried out the estimation of channel impulse response;
If receiving the spreading factor of data is 1, then produce the ideal channel response of estimating, as the input of described many grids viterbi algorithm module; If the spreading factor that receives data greater than 1, is then made denoising to channel estimating and handled, and produce the input of estimating noise power as least mean-square error piece linear equalization module;
Spreading code according to estimated channel impulse response and user produces sytem matrix, from the data of reception and the noise power and the transmission data of estimation the data of described channel estimator output are carried out scramble process, ask relevant and output data behind the descrambling and different spreading codes.
Described step of carrying out the balancing technique processing comprises:
Adopt many grids viterbi algorithm to realize maximum-likelihood sequence estimation; With
Scramble process is carried out in output to described many grids viterbi algorithm.
The receiver that balancing technique of the present invention and associated detection technique combine adopts joint-detection and balancing technique can eliminate multiple access and disturbs and intersymbol interference, guarantee that effectively receiver can reach the excellent reception performance to different spreading factors, improve the overall performance of receiver.
Description of drawings
Fig. 1 is a receiver structure schematic diagram balanced and that joint-detection combines;
Fig. 2 is the block diagram of receiver according to an embodiment of the invention;
Fig. 3 is the structure of time slot figure of TD-SCDMA system;
Fig. 4 is the block diagram of an embodiment of MVA module;
Fig. 5 is the block diagram of an embodiment of MMSE_BLE module.
Embodiment
The present invention will be described in detail below with reference to accompanying drawing.
An embodiment of the receiver structure that equilibrium of setting up according to the present invention and joint-detection combine as shown in Figure 1.This receiver comprises the joint-detection receiver and the equalizing receiver of receiving inputted signal respectively, controls the controller that joint-detection receiver and equalizing receiver work and control selector are selected, and is used for the demodulator of demodulation selector output signal.Equalizing receiver activates when spreading factor is 1, eliminates the signal of intersymbol interference by equilibrium treatment output; When spreading factor joint-detection receiver greater than 1 time activates, eliminate interference signals by united detection processing output; Selector is selected output from described equalizing receiver or described joint-detection receiver; Controller activates described equalizing receiver or described joint-detection receiver according to spreading factor, and controls the output that described selector is selected described equalizing receiver or described joint-detection receiver.The signal of described selector output is exported after the demodulator demodulation.
Below explanation the present invention is applied in the embodiment of TD-SCDMA terminal receiver.The maximum-likelihood sequence estimation (MLSE) that the balancing technique of this receiver adopts many grids viterbi algorithm (MVA) to realize, associated detection technique adopts least mean-square error piece linear equalization (MMSE_BLE), and the data-modulated mode adopts QPSK.
Fig. 2 represents that balancing technique is the TD-SCDMA terminal receiver structure that the equalizer of many grids viterbi algorithm and joint-detection receiver that associated detection technique is least mean-square error piecemeal linearity combine.The middle MVA module and the MMSE_BLE module of this receiver realize MVA algorithm and MMSE_BLE algorithm function respectively.In addition, comprise that also controller and other peripheral components such as channel estimator, channel activate detector, CIR (channel impulse response) processor, descrambler, selector and demodulator.
In the TD-SCDMA system, channel estimation module carries out the estimation of channel impulse response with local Midamble data and the Midamble data that receive.Suppose that training sequence process rotation transformation before emission that the user uses is the complex-valued data B of Midamble
Midamble, the data that receive are R
Midamble, other establishes h is channel impulse response, n is a white noise.Then can obtain following formula:
R
Midamble＝Gh+n
G is the transition matrix that is made of basic Midamble sign indicating number.Because the noise robustness of the training sequence Midamble sign indicating number of selecting for use is better, The noise can be ignored, and therefore can obtain following formula:
R
Midamble＝Gh
Adopt following formula calculating channel impulse response
Channel estimation module is exported this channel impulse response
FFT is a fast fourier transform in the formula, and IFFT is an inverse fast fourier transform.
Channel activate detection module only at the spreading factor that receives data greater than 1 o'clock operate as normal, it is based on and received data before entering channel, not only through the QPSK modulation, also pass through the processing of spread spectrum and scrambling, and the The data that receives of synchronization different orthogonal intersections and identical scrambler.Thereby channel activation detection module at first carries out scramble process to receiving data, then the data behind the descrambling are asked relevant with different spreading codes, if a certain channel window is by specific Walsh (Walsh) sign indicating number spread spectrum, then correlated results is bigger, thereby can be detected, corresponding channel window is for activating the channel window, and corresponding Walsh sign indicating number then is effective spreading code.
The CIR processing module is worked under the control of control module, if spreading factor is 1, the CIR processing module produces the DIR (ideal channel response) that estimates, as the input of MVA module; If spreading factor is greater than 1, then the CIR processing module is done the denoising processing to channel estimating, and produces the input of estimating noise power as the MMSE_BLE module.
Descrambling module realizes that transmitting terminal is to sending the contrary processing of data scrambling.Because the MVA algorithm is not considered scrambling code information in the search of maximum likelihood path, therefore need make scramble process to dateout by descrambling module; And to the MMSE_BLE algorithm, its scrambling code information has been included in the sytem matrix, so the also implicit expression realization in matrix inversion of corresponding descrambling.
Demodulation module is finished the inverse process of QPSK modulation, and promptly according to 0 →+1,1 →-1 corresponding relation rotates 45 ° with the dateout of multiplexer MUX.The output of demodulation module will be handled as other processing modules that the final soft output information of receiver is sent into terminal, as deinterleaving, channel-decoding etc.
Fig. 3 is the structure of time slot figure of TD-SCDMA system.This time slot comprises data symbol, training sequence (Midamble) and GP.Training sequence is 144 chips, is respectively first and second data blocks before and after it, is respectively the data symbol of 352 chips, is the protection interval (GP) of 16 chips at last, and total length is 864 chips.
Receive the data of a time slot when receiver after, data extractor at first according to structure of time slot shown in Figure 3 with data symbol part R
DataWith training sequence part R
MidambleDeposit data buffer and Midamble buffer (training sequence buffer) respectively in, channel estimator is then according to the training sequence R that receives
MidambleCarry out the estimation of channel impulse response (CIR) with the training sequence of this locality
The spreading factor SF that controller transmits according to high level selects to adopt joint-detection to receive still balanced the reception.If spreading factor is greater than 1, receiver adopts combined detector MMSE_BLE to receive, and this moment, channel activated the detector operate as normal, promptly according to receiving data R
DataSpreading code and the scrambling code information known with this locality detect the activation window, estimate thereby obtain efficient channel
The CIR processor will
Carry out denoising and handle, obtain final channel estimation results
Input as joint detection algorithm MMSE_BLE module.If spreading factor is 1, receiver adopts equalizing receiver MVA, and channel activates detector and will not be activated the output of channel estimating
Directly as the input of CIR processor, this moment, the CIR processor was finished the estimation of ideal communication channel impulse response (DIR), its output
Input as equalizer MVA module.The output d ' of MVA module
MvaThe processing of process descrambler obtains the output d behind the descrambling
MvaAnd the MMSE_BLE module of combined detector has been included in the sytem matrix owing to its scrambling code information, and corresponding scramble process is implicit expression realization in the MMSE_BLE module also, thereby no longer needs the processing of descrambler.Controller is controlled selector equally, when spreading factor is 1, selects the descrambling output d of MVA module
Mva, when spreading factor greater than 1 the time, select the output d of MMSE_BLE module
MmseLast demodulator is to the output d of selector
SelMake demodulation process, thereby obtain the final output d of receiver
Dem
The MVA module adopts the MVA algorithm, and it is to develop from the common VA algorithm (viterbi algorithm) that adopts of MLSE receiver.Less and extend long multipath channel for the main tap coefficient of CIR (channel impulse response), the MVA algorithm can overcome the excessive shortcoming of traditional VA algorithm computation amount, and has kept the best balanced performance of MLSE receiver simultaneously.
Fig. 4 is the block diagram of an embodiment of MVA module.The MVA module comprises: control unit, and the grid initialization unit, the branch metric unit, the path is selected cell relatively, the reliability information measuring unit.Wherein, this control unit is according to the beginning of other unit of initialization information control scheduling and the access of information.The grid initialization unit is in the beginning that receives time slot data, produce MVA grid initialization information according to DIR, comprise the number of determining non-zero tap according to the DIR that imports, the position of non-zero tap, the judgement of symbol constantly, determine that each judgement step needs the state of storage and dicision depth and determine the state transitions relation according to state of each judgement step storage, and the corresponding relation of each store status and branch metric.This control unit will be controlled scheduling to branch metric unit, path comparison selected cell and reliability information measuring unit according to this initialization information.
The branch metric unit is used for the calculating and the renewal of branch metric.With the difference of traditional VA be, traditional VA is behind the grid of search certain-length, find out the survivor path of a cumulative metrics value minimum, and export all symbols on this path, be that a certain branch metric only uses once constantly in the judgement of correspondence, and MVA is behind the short grid of search respective length, only exports first symbol on the survivor path, and then from second grid search that sign-on is new, promptly there is the repeated use of branch metric in MVA.
The selection of selected cell executing state transfer path and the hard decision of symbol are compared in the path, and the path of each state is relatively selected to be made of four parts, and promptly the moment judgement adds up, and compare and selection.Wherein, moment judgement part compare according to the information of control unit branch metric every contended path correspondence, and select minimum branched measurement value.If the only corresponding branch metric of current contended path then need not the moment judging process, what the cumulative metric value in the path that directly is at war with and respective branches were measured adds up.Accumulation result relatively at last, the path of the metric minimum of selecting to add up is a survivor path, and exports first hard decision symbol on this survivor path and the metric that adds up of survivor path.In addition, also the contended path cumulative metric value that needs of output reliability information test cell and first hard decision symbol on the contended path of this unit.
The reliability information measuring unit is implemented as the receiver soft-decision output algorithm that improves the channel-decoding performance and adopt.The SOVA (soft output Viterbi algorithm) that the MVA soft decision method of QPSK modulating system is based on the binary transmissions system derives, and by adopting approximate calculation, has avoided being difficult at hardware the log () and the exp () computing of realization.
Fig. 5 is the block diagram of an embodiment of MMSE_BLE module.The MMSE_BLE module have only when spreading factor greater than 1 the time just the Be Controlled module activate, be used to realize that spreading factor was greater than 1 o'clock joint-detection function to a plurality of subscriber signals.Based on the MMSE_BLE algorithm, produce sytem matrix according to estimated channel impulse response and user's spreading code, according to data and the noise power of estimation and the transmission data that sytem matrix solve user of following formula from receiving:
The Midamble interference cancellation module is used for eliminating preceding W-1 chip receiving Midamble in the data W-1 chip of leading interference and back of first data block is disturbed (W is 16 to the maximum) to the hangover of second data block, the data e of output elimination interference.Matched filter module is used for finishing the conjugate transpose A of the sytem matrix that the sytem matrix generation module generates
HWith the long-pending A that eliminates the data of disturbing
HE also exports this result; Covariance matrix (R matrix) generation module is according to the covariance matrix of the sytem matrix A calculating A of sytem matrix generation module output, i.e. R=A
HA; The Matrix Solving module is to the output A of matched filter module
HThe output R=A of e, R matrix generation module
HA and noise power are estimated σ
2Carry out Cholesky and decompose (tall formula decomposition), matrix inversion, matrix multiple finally obtains sending estimation (the R+ σ of sequence
2I)
-1A
HE, I is a unit matrix in the formula.Wherein the Cholesky decomposing module is finished R+ σ
2The decomposition of I matrix is the symmetrical matrix R+ σ of conjugation symmetry
2I is decomposed into the long-pending LL of the conjugation symmetrical matrix of lower triangular matrix L and L
HThe matrix inversion unit carries out matrix inversion to the result that Cholesky decomposes, and carries out matrix multiple and output by the matrix multiple unit at last.
Joint-detection receiver among the present invention can adopt multiple different joint detection algorithm to realize the ZF piecemeal linear equalizer (ZF-BLE) in for example linear combined detection algorithm and ZF piecemeal decision feedback equalization (ZF_BDFE) in least mean-square error piecemeal linear equalizer (MMSE-BLE) and the non-linear joint detection algorithm and least mean-square error piecemeal decision feedback equalization (MMSE_BDFE) etc.Equally, equalizing receiver among the present invention also can adopt multiple different equalization algorithm to realize, for example in the linear equalization based on the self adaptation zero forcing algorithm of steepest descent recursive algorithm, based on least square (LMS) algorithm of mean-square error criteria and the recursive least squares (RLS) etc. of development on this basis, various decision feedback algorithms in the nonlinear equalization, in the best equalization algorithm based on the Sequence Detection algorithm of maximum-likelihood sequence estimation (MLSE) with based on the algorithm of symbol detection one by one of maximum a posteriori probability (MAP) etc., and on above basis equalization algorithm basis, develop and come other improve equalization algorithms.Relative independentability in view of joint-detection receiver and equalizing receiver among the present invention, different joint detection algorithms and equalization algorithm can combination in any be realized this receiver, for example the MLSE equalizing receiver of viterbi algorithm (VA) realization makes up the joint-detection receiver that the MMSE-BLE algorithm is realized, joint-detection receiver of the equalizing receiver combination ZF-BLE algorithm realization that decision feedback algorithms realizes or the like.
More than with specific embodiment the present invention has been described, but these embodiment are exemplary, the invention is not restricted to specific embodiment.Those of ordinary skill in the art can make amendment, change the present invention or substitute under the situation that does not break away from the spirit and scope of the present invention.
Claims (19)
1. the receiver that combines of balancing technique and associated detection technique is characterized in that comprising:
Equalizing receiver is activated when the spreading factor that receives data is 1, the signal of input is carried out the signal of balancing technique processing and output elimination intersymbol interference;
The joint-detection receiver when the spreading factor that receives data is activated greater than 1 the time, carries out that associated detection technique is handled and output eliminates that multiple access disturbs and the signal of part intersymbol interference to the signal of input;
Selector is selected output from described equalizing receiver or described joint-detection receiver;
Controller, the spreading factor that transmits according to high level activates described equalizing receiver or described joint-detection receiver, and controls the output that described selector is selected described equalizing receiver or described joint-detection receiver.
2. receiver according to claim 1 is characterized in that described equalizing receiver is to adopt receiver linear equalization or nonlinear equalization.
3. receiver according to claim 2 is characterized in that:
Described linear equalization receiver is based on the linear equalizer of one of following linear equalizer algorithm: the self adaptation zero forcing algorithm of steepest descent recursive algorithm, the least-squares algorithm and the recursive least squares of mean-square error criteria.
4. receiver according to claim 2 is characterized in that:
Described nonlinear equalization receiver is based on the nonlinear equalizer of one of following algorithm: the algorithm of symbol detection one by one of the Sequence Detection algorithm of the maximum-likelihood sequence estimation of decision feedback algorithms, best equalization algorithm and maximum a posteriori probability.
5. receiver according to claim 1 is characterized in that described joint-detection receiver is the receiver that adopts linear combined detection or non-linear joint-detection.
6. receiver according to claim 5 is characterized in that: described linear combined detection receiver is based on the linear combined detection of one of following linear combined detection algorithm: ZF piecemeal linear equalization and least mean-square error piecemeal linear equalization.
7. receiver according to claim 5 is characterized in that:
Described non-linear joint-detection receiver is based on the non-linear joint-detection of one of following non-linear joint detection algorithm: ZF piecemeal decision feedback equalization and least mean-square error piecemeal decision feedback equalization.
8. the receiver that combines of balancing technique and associated detection technique is characterized in that comprising:
The equalizing receiver of many grids viterbi algorithm is activated when the spreading factor that receives data is 1, the signal of input is carried out the signal of the processing and the output elimination intersymbol interference of many grids viterbi algorithm;
The joint-detection receiver of least mean-square error piecemeal linearity when the spreading factor that receives data is activated greater than 1 the time, carries out the united detection processing and the output of least mean-square error piecemeal linearity and eliminates interference signals to the signal of input;
Selector is selected output from described equalizing receiver or described joint-detection receiver;
Controller, the spreading factor that transmits according to high level activates described equalizing receiver or described joint-detection receiver, and controls the output that described selector is selected described equalizing receiver or described joint-detection receiver.
9. receiver according to claim 8 is characterized in that also comprising:
Data extractor, the data-signal that is used for receiving is separated into data symbol and training sequence;
Data buffer is used to store described data symbol;
The training sequence buffer is used to store described training sequence; And
Demodulator is used for the signal that the described selector of demodulation is exported.
10. receiver according to claim 9 is characterized in that:
Described equalizing receiver comprises:
Channel estimator carries out the estimation of channel impulse response with local training sequence data and the training sequence data that receives;
The channel impulse response processor is worked under the control of controller, is 1 if receive the spreading factor of data, then produces the ideal communication channel response of estimating, as the input of described many grids viterbi algorithm module;
Many grids viterbi algorithm module is carried out many grids viterbi algorithm to the described data symbol from described data buffer and is handled, and realizes maximum-likelihood sequence estimation; With
Descrambler carries out scramble process to the output of described many grids viterbi algorithm module.
11. receiver according to claim 10 is characterized in that:
Described joint-detection receiver comprises:
Described channel estimator carries out the estimation of channel impulse response with local training sequence data and the training sequence data that receives;
Described channel impulse response processor is worked under the control of controller, if the spreading factor that receives data greater than 1, is then done the denoising processing to channel estimating, and produces the input of estimating noise power as least mean-square error piece linear equalization module;
Least mean-square error piece linear algorithm module, work greater than 1 o'clock at the spreading factor that receives data, spreading code according to estimated channel impulse response and user produces sytem matrix, from the data of reception and the noise power and the transmission data of estimation the data of described channel estimator output are carried out scramble process, ask relevant and output data behind the descrambling and different spreading codes.
12. receiver according to claim 10 is characterized in that:
Described many grids viterbi algorithm module comprises:
The grid initialization unit in the beginning that receives time slot data, produces the initialization information of many grids viterbi algorithm according to the channel impulse ideal response of channel impulse processor output;
Control unit is controlled the beginning of other unit of scheduling and the access of information according to initialization information;
The branch metric unit is used for the calculating and the renewal of branch metric;
The path is selected cell relatively, the contended path cumulative metric value that the selection of executing state transfer path and the hard decision of symbol and output reliability information test cell need and first hard decision symbol on the contended path;
The reliability information measuring unit is implemented as the receiver soft-decision output algorithm that improves the channel-decoding performance and adopt.
13. receiver according to claim 11 is characterized in that:
Described least mean-square error piece linear algorithm module comprises:
The training sequence interference cancellation module is used for eliminating the chip that receives training sequence in the data and disturbs to the leading interference of first data block with to the hangover of second data block, and the data of disturbing are eliminated in output;
The sytem matrix generation module, the output system matrix;
Matched filter module, be used for the sytem matrix that the described sytem matrix generation module of computing generates conjugate transpose with eliminate the long-pending of the data disturbed and export this result;
The covariance matrix generation module is according to the system matrix computation covariance matrix of described sytem matrix generation module output;
The Matrix Solving module is estimated to handle to the output of matched filter module, the output and the noise power of covariance matrix generation module, obtains sending the estimation of sequence.
14. receiver according to claim 13 is characterized in that:
Described Matrix Solving module comprises:
The Cholesky decomposing module is decomposed into lower triangular matrix to the symmetrical matrix of conjugation symmetry and its conjugation symmetrical matrix is long-pending;
The matrix inversion unit, the result that the Cholesky decomposing module is decomposed carries out matrix inversion;
The matrix multiple unit multiplies each other and exports the output matrix of described matrix inversion unit.
15. the method for reseptance that balancing technique and associated detection technique combine is characterized in that may further comprise the steps:
The data-signal that receives is separated into data symbol and training sequence;
In data buffer, store described data symbol and the described training sequence of storage in the training sequence buffer;
If the spreading factor of described data equals 1, activate equalizing receiver, the signal of input is carried out the signal of balancing technique processing and output elimination intersymbol interference;
If the spreading factor of described data is greater than 1, activation joint-detection receiver carries out associated detection technique to the signal of importing and handles and export the signal of eliminating multiple access interference and part intersymbol interference;
Select the output signal of joint-detection receiver or the output signal of equalizing receiver according to spreading factor.
16. method of reseptance according to claim 15 is characterized in that also comprising the step of the described output signal of demodulation.
17. method of reseptance according to claim 15 is characterized in that:
Described step of carrying out the associated detection technique processing comprises:
Local training sequence data and the training sequence data that receives are carried out the estimation of channel impulse response;
If receiving the spreading factor of data is 1, then produce the ideal channel response of estimating, as the input of described many grids viterbi algorithm module; If the spreading factor that receives data greater than 1, is then made denoising to channel estimating and handled, and produce the input of estimating noise power as least mean-square error piece linear equalization module;
Spreading code according to estimated channel impulse response and user produces sytem matrix, from the data of reception and the noise power and the transmission data of estimation the data of described channel estimator output are carried out scramble process, ask relevant and output data behind the descrambling and different spreading codes.
18., it is characterized in that according to claim 15 or 17 described method of reseptances:
Described carry out step that balancing technique handles be based on the self adaptation zero forcing algorithm of steepest descent recursive algorithm, based on the least-squares algorithm of mean-square error criteria with based on the linear equalization of one of recursive least squares or based on decision feedback algorithms, based on the Sequence Detection algorithm of the maximum-likelihood sequence estimation of best equalization algorithm with realize based on the nonlinear equalization of one of algorithm of symbol detection one by one of maximum a posteriori probability.
19., it is characterized in that according to claim 15 or 17 described method of reseptances:
Described step of carrying out the balancing technique processing comprises:
Adopt many grids viterbi algorithm to realize maximum-likelihood sequence estimation; With
Scramble process is carried out in output to described many grids viterbi algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100769630A CN100362756C (en) | 2005-06-09 | 2005-06-09 | Equalization technique and associated detection technique combined receiver and receiving method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100769630A CN100362756C (en) | 2005-06-09 | 2005-06-09 | Equalization technique and associated detection technique combined receiver and receiving method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1702979A true CN1702979A (en) | 2005-11-30 |
| CN100362756C CN100362756C (en) | 2008-01-16 |
Family
ID=35632527
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005100769630A Expired - Fee Related CN100362756C (en) | 2005-06-09 | 2005-06-09 | Equalization technique and associated detection technique combined receiver and receiving method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100362756C (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100364243C (en) * | 2006-01-13 | 2008-01-23 | 北京天碁科技有限公司 | Receiver, transmitting device and communication system |
| CN101060505B (en) * | 2006-04-19 | 2011-06-01 | 鼎桥通信技术有限公司 | Joint channel estimation method and estimation device in a wireless mobile communication system |
| CN102404259A (en) * | 2010-09-13 | 2012-04-04 | 凌阳科技股份有限公司 | Mixing balance system |
| CN101677249B (en) * | 2008-09-19 | 2012-12-12 | 展讯通信（上海）有限公司 | Joint detection method of time division synchronous code division multiple access system and receiving device thereof |
| CN101753169B (en) * | 2008-11-28 | 2013-05-01 | Tcl集团股份有限公司 | Method for equalizing ordered block decision feedback in TD-SCDMA |
| CN104065451A (en) * | 2013-03-20 | 2014-09-24 | 中兴通讯(美国)公司 | Soft Maximum Likelihood Sequence Estimation In Digital Communication |
| CN107018097A (en) * | 2017-02-15 | 2017-08-04 | 浙江科技学院 | A kind of sequence detecting method of wireless light communication based on generalized likelihood test principle |
| CN107979547A (en) * | 2016-10-24 | 2018-05-01 | 深圳超级数据链技术有限公司 | Channel estimation methods and device |
| CN110832586A (en) * | 2017-09-25 | 2020-02-21 | 美光科技公司 | Memory decision feedback equalizer test |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1393062A (en) * | 2000-09-29 | 2003-01-22 | 松下电器产业株式会社 | Demodulator and demodulating method |
| DE10149065A1 (en) * | 2001-10-05 | 2003-04-24 | Infineon Technologies Ag | Joint-detection equalization method for received signal in mobile radio system, using iterative technique to solve inhomogeneous linear equation system |
| CN1283110C (en) * | 2002-07-10 | 2006-11-01 | 西门子（中国）有限公司 | Disturbance power estimation apparatus and method and disturbance power detecting system |
-
2005
- 2005-06-09 CN CNB2005100769630A patent/CN100362756C/en not_active Expired - Fee Related
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100364243C (en) * | 2006-01-13 | 2008-01-23 | 北京天碁科技有限公司 | Receiver, transmitting device and communication system |
| CN101060505B (en) * | 2006-04-19 | 2011-06-01 | 鼎桥通信技术有限公司 | Joint channel estimation method and estimation device in a wireless mobile communication system |
| CN101677249B (en) * | 2008-09-19 | 2012-12-12 | 展讯通信（上海）有限公司 | Joint detection method of time division synchronous code division multiple access system and receiving device thereof |
| CN101753169B (en) * | 2008-11-28 | 2013-05-01 | Tcl集团股份有限公司 | Method for equalizing ordered block decision feedback in TD-SCDMA |
| CN102404259A (en) * | 2010-09-13 | 2012-04-04 | 凌阳科技股份有限公司 | Mixing balance system |
| CN104065451A (en) * | 2013-03-20 | 2014-09-24 | 中兴通讯(美国)公司 | Soft Maximum Likelihood Sequence Estimation In Digital Communication |
| CN107979547A (en) * | 2016-10-24 | 2018-05-01 | 深圳超级数据链技术有限公司 | Channel estimation methods and device |
| CN107018097A (en) * | 2017-02-15 | 2017-08-04 | 浙江科技学院 | A kind of sequence detecting method of wireless light communication based on generalized likelihood test principle |
| CN107018097B (en) * | 2017-02-15 | 2020-02-28 | 浙江科技学院 | Sequence detection method based on generalized likelihood ratio detection principle for wireless optical communication |
| CN110832586A (en) * | 2017-09-25 | 2020-02-21 | 美光科技公司 | Memory decision feedback equalizer test |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100362756C (en) | 2008-01-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1702979A (en) | Equalization technique and associated detection technique combined receiver and receiving method thereof | |
| CN1224228C (en) | Multicarrier transmission system using multiplication by leakage matrix with complex reduction | |
| CN1193563C (en) | Multicarrier receiver with channel estimator | |
| JP4975757B2 (en) | Linear turbo equalization using despread values | |
| CN102124706B (en) | Receiver and method for processing radio signals using soft pilot symbols | |
| CN1113482C (en) | Method and apparatus for intersymbol interference reduction using array processing of diversity signals | |
| US8126043B2 (en) | Method and apparatus for block-based signal demodulation | |
| CN1248472C (en) | Apparatus and method for intersymbol interference compensation in spread spectrum communications | |
| US7593492B1 (en) | Combinational hybrid turbo-MUD | |
| US20090037597A1 (en) | Joint symbol, amplitude, and rate estimator | |
| CN1703032A (en) | Adaptive channel estimation method and system using decision feedback | |
| JP4695183B2 (en) | Flexible demodulation method for estimated complex symbols | |
| JP2006262039A (en) | Propagation path estimation method and propagation path estimation apparatus | |
| EP2243263A1 (en) | A method and apparatus for symbol detection via reduced complexity sequence estimation processing | |
| CN1777054A (en) | Array antenna channel estimating aftertreatment method | |
| JP2006067070A (en) | Mimo system receiving method and receiver | |
| CN1688144A (en) | Receiver and method for ID SCDMA system | |
| CN1146170C (en) | Improved method for joint test spectrum-spread multi-user communicaltion system | |
| CN101079863A (en) | Frequency domain balancer design method in orthogonal frequency division multiplexing system | |
| CN1595825A (en) | A combined detection method for decreasing complexity in mobile communication system | |
| CN1437345A (en) | Space-time iterative multiuser detecting algorithm based on soft sensitive bit and space grouping | |
| Nordenvaad et al. | Iterative reception for acoustic underwater MIMO communications | |
| CN1239033C (en) | Discrete Fourier transform based space-time combined inspecting device and method for radio transmission | |
| CN101039305A (en) | Balancing technique-based receiver and receiving method | |
| Yuan et al. | Low-complexity iterative detection in multi-user MIMO ISI channels |
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: 20080116 Termination date: 20180609 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |