CN101540661B - Switching-based adaptive MIMO decoding method - Google Patents

Switching-based adaptive MIMO decoding method Download PDF

Info

Publication number
CN101540661B
CN101540661B CN2009100610923A CN200910061092A CN101540661B CN 101540661 B CN101540661 B CN 101540661B CN 2009100610923 A CN2009100610923 A CN 2009100610923A CN 200910061092 A CN200910061092 A CN 200910061092A CN 101540661 B CN101540661 B CN 101540661B
Authority
CN
China
Prior art keywords
noise ratio
signal
decoding
conditional number
detection method
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN2009100610923A
Other languages
Chinese (zh)
Other versions
CN101540661A (en
Inventor
刘应状
傅杰
林宏志
孙俊
王德胜
葛晓虎
朱光喜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huazhong University of Science and Technology
Original Assignee
Huazhong University of Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Huazhong University of Science and Technology filed Critical Huazhong University of Science and Technology
Priority to CN2009100610923A priority Critical patent/CN101540661B/en
Publication of CN101540661A publication Critical patent/CN101540661A/en
Application granted granted Critical
Publication of CN101540661B publication Critical patent/CN101540661B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

A switching-based adaptive MIMO decoding algorithm belongs to a wireless communication technology. The method is used for responding a channel state by conditions and performing adaptive switching among a ZF decoding detection method, a sphere decoding detection method and a ZF-OSIC decoding detection method. The method helps ensure decoding performances, effectively lower computational complexity and improve practicality of a system.

Description

A kind of based on the adaptive MIMO decoding method that switches
Technical field
The present invention relates to wireless communication technology, relate in particular to a kind of self adaptation multiple-input and multiple-output (MIMO) coding/decoding method in multiaerial system based on switching.
Background technology
In vertical demixing time space (VBLAST) detected, that adopts usually at present compeled zero (ZF) algorithm, least mean-square error (MMSE) algorithm, detects and decipher based on serial interference elimination (OSIC) algorithm, maximum likelihood (ML) criterion of the ordering of the two kinds of linear decoded modes in front.The relative Maximum Likelihood Detection of the complexity of above-mentioned first three algorithm is lower, but bit error rate performance is relatively poor; And use Maximum Likelihood Detection can obtain more performance, but its complexity is higher, is not easy to realize.Globular decoding (SD) algorithm based on Maximum Likelihood Detection is a kind of guaranteed performance, the detection algorithm of moderate complexity.Because adopting the complexity of the Maximum Likelihood Detection algorithm of exhaustive search is the relation that is exponential increase with antenna number, and the complexity of SD algorithm is polynomial relation with antenna number in very big signal to noise ratio scope.So in theory, the SD algorithm can obtain or approach the maximum-likelihood decoding performance with smaller calculation.This shows that above-mentioned detection method respectively has pluses and minuses, the user need select detection mode according to the system channel actual conditions.But the channel conditions of system is constantly to change, if in decode procedure, select a kind of mode all the time, there is the bad or problem that amount of calculation is big of performance in the decoded result that obtains.
Summary of the invention
The object of the present invention is to provide a kind of adaptive MIMO decoding method, realize that the self adaptation between ZF detection mode, SD detection mode, OSIC detection mode is switched, guarantee decoding performance, effectively reduce computational complexity, improve the practicality of system.
A kind of based on the adaptive MIMO decoding method that switches, carry out according to following steps:
(1) if channel condition is counted a and is less than or equal to conditional number lower threshold th1, select the ZF decoding detection method, otherwise, step (2) got into;
(2) if signal to noise ratio snr greater than the interval critical value SNR_th of high low signal-to-noise ratio, is selected method for detecting spherical decode, otherwise entering step (3);
(3) if conditional number a belongs to greater than condition and counts lower threshold th1, be less than or equal in the conditional number upper limit threshold th2 scope, select the ZF-OSIC decoding detection method, otherwise adopt method for detecting spherical decode.
Technique effect of the present invention is embodied in:
The conditional number of channel matrix can reflect channel conditions, and is interval in the signal to noise ratio that is lower than snr threshold, if conditional number is less; The channel conditions that be described this moment is better, for receiving terminal, can consider with decoding complexity minimum; But the decoding algorithm that performance is very poor relatively is as compeling zero (ZF) algorithm.If conditional number is bigger, explain that the channel conditions of this moment is very poor, need to adopt complexity higher relatively, but the good decoding algorithm of performance, like the globular decoding algorithm.For conditional number placed in the middle, then be chosen in the ZF-OSIC algorithm of compromising on performance and the complexity.Can lose under the less situation in guaranteed performance like this, obtain the reduction of complexity.If it is interval to be in the signal to noise ratio that is higher than snr threshold; The threshold value of conditional number is th1; Only need compel to carry out similar selection between zero (ZF) algorithm and the globular decoding algorithm, the ZF-OSIC algorithm wants little and performance is more excellent because the complexity of the globular decoding of this moment has been compared.
Description of drawings
Fig. 1 is a schematic flow sheet of the present invention;
Fig. 2 is the 4QAM modulation signal, adaptive decoding and globular decoding bit error rate comparative result sketch map;
Fig. 3 is the 4QAM modulation signal, adaptive decoding and globular decoding simulation time comparative result sketch map.
Embodiment
The present invention sets up data transmission link at transmitting terminal and receiving terminal, is primarily aimed at the VBLAST system model, and receiving terminal antenna number N is more than or equal to transmitting terminal number of transmit antennas M.
The present invention utilizes conditional number to react channel condition information.Conditional number is the maximum singular value of channel matrix and the ratio of minimum singular value.Conditional number is more little, shows that channel status is good more, and can select the lower detection mode of complexity this moment for use; Otherwise channel status is poor more, is guaranteed performance, then need adopt complexity higher detection mode.Fig. 1 is a schematic flow sheet of the present invention, at first judges current condition number of channel, if it is less than or equal to conditional number lower threshold th1, then selects minimum urgent zero (ZF) detection mode of complexity for use;
If current condition number of channel a greater than condition is counted lower threshold th1; It is interval to judge whether current signal to noise ratio snr is in the signal to noise ratio that is higher than signal to noise ratio critical value SNR_th; If it is interval to be in the signal to noise ratio that is higher than the signal to noise ratio critical value, then select the globular decoding detection mode for use;
And then judge; If the conditional number of current channel a between two threshold values, promptly greater than th1, and is less than or equal to th2; And it is interval to be in the signal to noise ratio that is lower than the signal to noise ratio critical value, then selects complexity serial interference elimination (ZF-OSIC) detection mode based on urgent zero ordering placed in the middle for use;
If current condition number of channel is greater than threshold value th2, and it is interval to be in the signal to noise ratio that is lower than the signal to noise ratio critical value, then selects the highest globular decoding detection mode of complexity for use.
The interval critical value SNR_th of high low signal-to-noise ratio confirms that method is specially: according to the dual-mode antenna number of system and the modulation system of signal; Through comparing the complexity of ZF-OSIC algorithm, globular decoding algorithm; With the corresponding signal to noise ratio of the intersection point of the complexity curve of ZF-OSIC algorithm and globular decoding algorithm as the interval critical value SNR_th of high low signal-to-noise ratio; If in the target signal to noise ratio scope, do not find intersection point; It is interval that then explanation all is in the signal to noise ratio that is lower than the signal to noise ratio critical value, and the interval critical value SNR_th of signal to noise ratio is the maximum of target signal to noise ratio scope.The target signal to noise ratio scope is expressed as [0, B], and the size of B confirms that by the user B can not be too little, generally more than or equal to 20dB.
Conditional number lower threshold th1 and conditional number upper limit threshold th2 confirm that method is: according to system requirements, at first confirm to compel zero (ZF) algorithm, based on the three kinds of shared separately expectation weight w of serial interference elimination (ZF-OSIC) algorithm, globular decoding algorithm that compel zero ordering 1, w 2, w 3Emulation obtains the cumulative distribution function curve of conditional number, and the ordinate value size is w 1The corresponding abscissa value (abscissa is represented conditional number) in place is conditional number lower threshold th1, and the ordinate value size is w 1+ w 2The corresponding abscissa value in place is conditional number upper limit threshold th2.Conditional number threshold value th1 and th2 (th1≤th2), as the cut-point of choosing of three kinds of decoded modes.
Expectation weight w 1, w 2, w 3Confirm according to following method: the expectation weight w of ZF, ZF-OSIC, three kinds of algorithms of globular decoding 1, w 2, w 3Satisfy w 1+ w 2+ w 3=1, w 1Value when make using the decoding of ZF algorithm, bit error rate is near the bit error rate desired value (being confirmed by the user) of specifying under the signal to noise ratio, w 3Value when make using the globular decoding algorithm, w 3 ≥ ∫ c Hth f c H d c H ,
Figure G2009100610923D00052
Be the conditional number probability density function, c HthImplication be: when using globular decoding, conditional number [1, c Hth) in the scope, performance reaches the globular decoding algorithm in the average behavior of specifying under the signal to noise ratio; For w 2Also be similar, satisfy ( w 1 + w 2 ) ≤ ∫ 1 ZFOSICc Hth f c H d c H , ZFOSICc HthImplication be: when using the ZF-OSIC detection algorithm, the conditional number scope [1, ZFOSICc Hth) in, performance reaches the globular decoding algorithm in the average behavior of specifying under the signal to noise ratio.
Conditional number is defined as: c H = σ Max σ Min , σ MaxAnd σ MinBe respectively the maximum singular value and the minimum singular value of channel matrix H.The probability density function of conditional number can be through obtaining the system channel model emulation.
The utilization of ZF-OSIC algorithm be based on the serial interference elimination algorithm of compeling the zero-emission preface, no matter be that the serial interference elimination algorithm of which kind of improved ordering all can apply among the present invention; What the globular decoding mode was utilized is the depth-first search mode of layer internal sort, does not have the operation that enlarges search radius, no matter be that which kind of improved globular decoding all can apply among the present invention.
Instance: the mimo system of 4 transmitting antennas, 4 reception antennas, no chnnel coding, modulation system adopts 4QAM, and channel is the systems of quasi-static flat Rayleigh fading model, and piece decline length is 20, and then at each constantly, the reception signal through the model behind the channel is:
y＝Hs+n
H is a channel matrix, and s is 4 * 1 emission symbolic vector, and n is that average is 0, and variance is σ 2White complex gaussian noise, noise variance σ wherein 2Relevant with antenna number and modulation system.
Obtain complexity curve separately through emulation contrast ZF-OSIC algorithm, globular decoding running time of algorithm; The target signal to noise ratio scope is decided to be [0; 20], in [0,20] scope; There is intersection point in the complexity curve that searches ZF-OSIC algorithm and globular decoding algorithm at the 15dB place, and then that intersection point is corresponding signal to noise ratio 15dB is decided to be critical value;
It is 10dB that systematic function requires in signal to noise ratio, and the bit error rate when zero (ZF) algorithm is compeled in the expectation use is near 10 -2, on this basis, use to compel zero (ZF) algorithm, be respectively 0.3,0.3,0.4 based on serial interference elimination (ZF-OSIC) algorithm of compeling zero ordering, the expectation weight of globular decoding algorithm.According to the probability density function selector bar number of packages lower limit and the upper limit threshold of conditional number, two threshold values satisfy: conditional number APPROXIMATE DISTRIBUTION form is carried out integral operation, and operation result is near three kinds of detection algorithm desired weights coefficients.In this instance, conditional number threshold value th1 is 6, and th2 is 8.
When the signal to noise ratio that is lower than the signal to noise ratio critical value is interval, judge current condition number of channel, if be less than or equal to threshold value th1, then select minimum urgent zero (ZF) detection mode of complexity for use;
When the signal to noise ratio that is lower than the signal to noise ratio critical value is interval,, then select complexity serial interference elimination (ZF-OSIC) detection mode based on urgent zero ordering placed in the middle for use if current condition number of channel between two threshold values, promptly greater than th1, and is less than or equal to th2;
When the signal to noise ratio that is lower than the signal to noise ratio critical value is interval, if current condition number of channel, is then selected the highest globular decoding detection mode of complexity for use greater than threshold value th2;
When the signal to noise ratio that is higher than the signal to noise ratio critical value is interval; Detection algorithm is only selected between ZF algorithm and globular decoding algorithm; Be separation still,, then select minimum urgent zero (ZF) detection mode of complexity for use if be less than or equal to threshold value th1 with conditional number threshold value th1; Otherwise, then adopt the globular decoding detection mode.
Under the situation that guarantees expected performance, this method can be used as the mechanism of a kind of complexity flexibly and trade-off of performance.Use this adaptive decoding (adaptive decode) method and only adopt globular decoding (sphere decode) mode to compare; Comparative result is with reference to figure 2,3, and Fig. 2 representes that the average error bit rate (average BER) of two kinds of methods compares; Abscissa is a signal to noise ratio (snr); Unit be the decibel (dB), ordinate is bit error rate (BER), can find out adaptive decoding mode error performance compare with globular decoding the loss a small amount of dB; Fig. 3 representes that the simulation time (simulation time) of two kinds of methods compares; Abscissa is a signal to noise ratio (snr), and unit is a decibel (dB), and ordinate is simulation time (TIME); The computational complexity of reflection algorithm, the computational complexity of multiple decoded mode is less than the mode that only adopts globular decoding.Complex chart 2 can find out that with Fig. 3 the adaptive decoding mode is under the situation of guaranteed performance, obtains the compromise of complexity and error performance.If the increase modulation system, piece decline length is big more, and complexity is improved can be more obvious; If the threshold value of conditional number is further reduced; Can promote to some extent but the computational complexity meeting is corresponding, but still be less than the mode that only adopts globular decoding so that performance loss is littler.What need explanation is, no matter be that the serial interference elimination algorithm and the globular decoding of which kind of follow-on ordering all can apply in this method, belongs to protection scope of the present invention.

Claims (1)

1. one kind based on the adaptive MIMO decoding method that switches, and carries out according to following steps:
(1) if channel condition is counted a and is less than or equal to conditional number lower threshold th1, select the ZF decoding detection method, otherwise, step (2) got into;
(2) if signal to noise ratio snr greater than the interval critical value SNR_th of high low signal-to-noise ratio, is selected method for detecting spherical decode, otherwise entering step (3);
(3) if conditional number a belongs to greater than condition and counts lower threshold th1, be less than or equal in the conditional number upper limit threshold th2 scope, select the ZF-OSIC decoding detection method, otherwise adopt method for detecting spherical decode;
Said conditional number lower threshold th1 and upper limit threshold th2 confirm according to following mode: at first confirm ZF decoding detection method, ZF-OSIC decoding detection method, method for detecting spherical decode expectation weight w separately 1, w 2, w 3, according to the cumulative distribution function curve of conditional number, be w again with curve ordinate value size 1The conditional number of place's correspondence is w as conditional number lower threshold th1 with curve ordinate value size 1+ w 2The conditional number of place's correspondence is as upper limit threshold th2;
The expectation weight w of said ZF decoding detection method, ZF-OSIC decoding detection method, method for detecting spherical decode 1, w 2, w 3Satisfy w 1+ w 2+ w 3=1, w 1Satisfy when using the decoding of ZF decoding detection method, bit error rate is near the bit error rate desired value of specifying under the signal to noise ratio S, w 3Satisfy when using method for detecting spherical decode,
Figure FDA0000076881360000011
Be the conditional number probability density function, c HthImplication when being to use method for detecting spherical decode, the conditional number scope [1, c Hth) interior performance reaches method for detecting spherical decode in the average behavior of specifying under the signal to noise ratio S,
Figure FDA0000076881360000013
σ MaxAnd σ MinBe respectively the maximum singular value and the minimum singular value of channel matrix H; w 2Satisfy
Figure FDA0000076881360000021
ZFOSICc HthImplication when being to use the ZF-OSIC decoding detection method, the conditional number scope [1, ZFOSICc Hth) interior performance reaches method for detecting spherical decode in the average behavior of specifying under the signal to noise ratio S;
Definite method of the interval critical value SNR_th of said high low signal-to-noise ratio is specially: according to the dual-mode antenna number of system and the modulation system of signal, emulation obtains the complexity curve of ZF-OSIC decoding detection method, method for detecting spherical decode algorithm; If in target signal to noise ratio scope [0, B], there is intersection point in the complexity curve of ZF-OSIC decoding detection method and method for detecting spherical decode, and then that intersection point is corresponding signal to noise ratio is as the interval critical value SNR_th of high low signal-to-noise ratio; If in target signal to noise ratio scope [0, B], do not find intersection point, then the interval critical value SNR_th of signal to noise ratio is the maximum B of target signal to noise ratio scope, and B is more than or equal to 20dB.
CN2009100610923A 2009-03-09 2009-03-09 Switching-based adaptive MIMO decoding method Expired - Fee Related CN101540661B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2009100610923A CN101540661B (en) 2009-03-09 2009-03-09 Switching-based adaptive MIMO decoding method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2009100610923A CN101540661B (en) 2009-03-09 2009-03-09 Switching-based adaptive MIMO decoding method

Publications (2)

Publication Number Publication Date
CN101540661A CN101540661A (en) 2009-09-23
CN101540661B true CN101540661B (en) 2012-03-28

Family

ID=41123666

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2009100610923A Expired - Fee Related CN101540661B (en) 2009-03-09 2009-03-09 Switching-based adaptive MIMO decoding method

Country Status (1)

Country Link
CN (1) CN101540661B (en)

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102231641B (en) * 2011-07-21 2013-08-14 西安电子科技大学 MIMO (Multiple Input Multiple Output) step-by-step parallel detection method
CN103746728B (en) * 2013-10-08 2016-07-27 北京科技大学 The MIMO of a kind of mixed self-adapting receives detection method
CN104734818B (en) * 2013-12-23 2018-08-03 联芯科技有限公司 The adaptive method for switching and device of MIMO receiving algorithms
CN103973345B (en) * 2014-04-30 2017-04-19 华中科技大学 Base station antenna dispatching method based on user distance
CN105101241B (en) * 2014-05-12 2019-03-05 电信科学技术研究院 A kind of signal detecting method and device
CN104393953A (en) * 2014-11-12 2015-03-04 东华大学 A sphere decoding algorithm with low complexity and low error rate in an LTE system
CN106034321B (en) * 2015-03-20 2019-06-28 电信科学技术研究院 A kind of multi-user test method and equipment
CN107276650B (en) * 2017-07-03 2018-04-24 华中科技大学 A kind of extensive MIMO mixing precoding efficiency optimization method of multi-user
CN109936400B (en) 2019-05-20 2019-08-09 乐鑫信息科技（上海）股份有限公司 Adaptive MIMO detection method and system

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101094021A (en) * 2006-06-20 2007-12-26 中兴通讯股份有限公司 Method and device of self-adapting communication with multiple antennae

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101094021A (en) * 2006-06-20 2007-12-26 中兴通讯股份有限公司 Method and device of self-adapting communication with multiple antennae

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
Haris Vikalo, Babak Hassibi.On the Sphere-Decoding Algorithm II.Generalizations, Second-Order Statistics, and Applications to Communications.《IEEE Trans.on Signal Processing》.2005,第53卷(第8期),2819-2834. *
HarisVikalo Babak Hassibi.On the Sphere-Decoding Algorithm II.Generalizations
J Jalden，B Ottersten.On the complexity of sphere decoding in digital communications.《IEEE Trans．on Signal Processing》.2005,第53卷(第4期),1474-1484. *

Also Published As

Publication number Publication date
CN101540661A (en) 2009-09-23

Similar Documents

Publication Publication Date Title
CN101540661B (en) Switching-based adaptive MIMO decoding method
CN108712747A (en) A kind of link method for fast establishing based on spectrum prediction
CN100571098C (en) The maximum likelihood detecting method of low complex degree and device in the communication system
Prakash et al. Effects of channel prediction for transmit antenna selection with maximal-ratio combining in Rayleigh fading
CN101997639B (en) Iterative receiver method of low density parity check-multi-input/output communication system
CN101388705B (en) Initial radium computing method and apparatus for globular decoding
CN104734818B (en) The adaptive method for switching and device of MIMO receiving algorithms
CN201947449U (en) Mobile terminal
CN103326825B (en) A kind of quasi-orthogonal space time block code low-complexity decoding method
Park et al. Statistical multimode transmit antenna selection for limited feedback MIMO systems
CN103166739B (en) Method for feedback of channel quality indication and subscriber equipment
CN110896323B (en) New method and system for combining relay and antenna selection
WO2006059678A1 (en) Receiving method and receiving end apparatus
KR20080011823A (en) Method and apparatus for adaptive antenna controlling in mobile telecommunication system
CN101645755B (en) Device and method for realizing link adaptation of radio access network system
CN101098160B (en) Dependability criterion based hybrid receiver and detecting method
CN103051425B (en) Link adaptive transmission method based on channel parameter and fuzzy algorithm
CN101582743A (en) MIMO detection method and system used for iterative receivers
CN101365201A (en) MIMO receiving signal detection method and receiving apparatus
Kabir et al. Closed form capacity analysis of MIMO wireless channels
KR101289938B1 (en) Apparatus and method for receiving a signal in a mobile communication system using multiple input multiple output scheme
Xiao et al. On the capacity of Weibull fading MIMO channels under different adaptive transmission techniques
Chau et al. Spatial diversity with a new sequential maximal ratio combining over wireless fading channels
Zouaidi et al. ARQ protocols for MIMO systems
KR20160111822A (en) Method and apparatus for determining transmission mode in a multiple input multiple output 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
C17 Cessation of patent right
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20120328

Termination date: 20140309