CN105049104A - 4QAM modulation novel matched filtering detection method in large-scale MIMO system - Google Patents

Abstract

The invention belongs to the field of signal detection, and provides a novel matched filtering detection method in a large-scale MIMO system. The method is an interference offset novel matched filtering detection algorithm based on threshold area judgment. The concrete method is listed as follows: threshold area judgment is performed on each symbol obtained through MF detection, if there are symbols in a threshold area S, the sum of the absolute values of the real parts and the imaginary parts of the symbols is calculated, and the detection symbol with the maximum sum value is selected to perform interference offset first; and if there is no symbol in the threshold area S, the sum of the absolute values of the real parts and the imaginary parts of all the symbols is calculated, the detection symbol with the maximum sum value is selected to perform interference offset first until completion of iteration, and all the symbols are detected. According to the method, preliminary threshold screening is performed on the detection symbols, the symbols within the threshold are considered first, and the symbol with the maximum sum of the absolute values of the real parts and the imaginary parts is selected to perform interference offset so that BER performance of the system can be effectively enhanced, and the method is particularly suitable for the large-scale MIMO system based on 4QAM modulation.

Description

The novel matched filtering detection method of 4QAM modulation in a kind of extensive mimo system

Technical field

The invention belongs to the field of input, as radio communication, radar, wire communication etc., in particular to the communication mode adopting extensive multiple-input and multiple-output (MIMO:Multiple-InputMultiple-Output), be specially the novel matched filtering detection method of 4QAM modulation in a kind of extensive mimo system.

Background technology

As everyone knows, by multi-aerial space multiplexing technology, the transmission rate of system can with antenna number linear growth, and the increase of antenna number, although can improve system velocity significantly, also significantly increases the complexity of receiver method.To have maximum likelihood (ML:MaximumLikelihood) detection method of optimal detection performance, its complexity increases along with the increase exponentially of number of transmit antennas, is difficult to apply in practice.

At present, MIMO detection method obtains to be studied widely, comparatively typical method comprises two classes: a class is linearity test method, mainly contain matched filtering (MF:MatchedFilter), ZF (ZF:ZeroForcing), least mean-square error (MMSE:MinimumMeanSquare) and mutation method thereof, another kind of is non-linear detection method, the stepwise derivation mainly contained based on sequence offsets (OSIC:OrderedSuccessiveInterferenceCancellation), globular decoding (SD:SphereDecoding), K-Best method and mutation method etc. thereof.Although linear method has lower complexity, hardware easily realizes, the performance extreme difference when low signal-to-noise ratio; Nonlinear method obtains certain performance improvement relative to linear method, but complexity also improves thereupon.

But in extensive mimo system, although MF detection method complexity is minimum, performance extreme difference, seriously limits the practical application in extensive mimo system of MF detection method, for this reason, the invention provides the novel matched filtering detection method of 4QAM modulation in a kind of extensive mimo system.

Summary of the invention

The object of the present invention is to provide a kind of novel matched filtering detection method for 4QAM modulation in extensive mimo system, adopt detection method of the present invention, significantly can improve the performance of traditional MF detection method, low being easy to of complexity realizes, and can obtain nearly optimal performance in the uplink.

Technical scheme of the present invention: a kind of novel matched filtering detection method for 4QAM modulation in extensive mimo system, concrete steps are as follows:

Step 1. initialization:

I=1,1≤i≤N _t, i represents i-th iteration, N _tfor number of transmit antennas;

Y ⁽¹⁾=Y, Y are Received signal strength vector;

H ⁽¹⁾=H, H are channel matrix;

Step 2. i-th iteration: vector Y to received signal ⁽ⁱ⁾carry out matched filtering detection, obtain detected symbol vector namely wherein, () ^hrepresent conjugate transpose operation, detected symbol vector in symbol be expressed as j is the location index of symbol;

Judge vector in symbol whether meet pre-determined threshold region wherein, T is threshold value, represent and get real part computing, represent and get imaginary-part operation, || represent signed magnitude arithmetic(al);

If vector in there is symbol and meet threshold areas S, then calculate real part and the imaginary part absolute value sum of all symbols in threshold areas S, choose and be worth maximum symbol and carry out Interference Cancellation; Otherwise, compute vectors in the real part of all symbols and imaginary part absolute value sum, choose and be worth maximum symbol and carry out Interference Cancellation; Concrete steps are:

k _irepresent real part and imaginary part absolute value and be worth maximum character position index, namely changing for i-th time

The aerial position index that generation is detected;

<> _mrepresent that the m getting matrix is capable, G ⁽ⁱ⁾represent channel matrix H in i-th iterative process ⁽ⁱ⁾associate matrix;

${\tilde{x}}_{k_i}=w_{k_i}{Y}^{\left(i\right)};$

q () is constellation point reflection operator, namely recovers to send signal according to maximum-likelihood criterion hard decision; as estimated value;

Perform Interference Cancellation: [] _nrepresent the n-th row getting matrix;

Meanwhile, channel rectangle is upgraded: represent the n-th row of puncture table;

Step 3. repeats step 2, and iteration is to i=N _t, finishing iteration output detections symbol, demodulation recovers original information bits.

Beneficial effect of the present invention is: the invention provides the novel matched filtering detection method in a kind of extensive mimo system, this detection method is by carrying out thresholding screening to symbol, preferentially Interference Cancellation is carried out to the symbol being positioned at thresholding, significantly can improve the performance of traditional MF detection method, low being easy to of complexity realizes, and can obtain nearly optimal performance in the uplink.

Accompanying drawing explanation

Fig. 1 is mimo system schematic diagram.

Fig. 2 is the novel matched filtering detection method flow chart that the present invention proposes.

Embodiment

Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described, so that those skilled in the art understands the present invention better.Requiring particular attention is that, in the following description, when perhaps the detailed description of known function and design can desalinate main contents of the present invention, these are described in says here and is left in the basket.

For the present invention will be described better, first introduce the system used by technical solution of the present invention.

Mimo system: as Fig. 1, we consider that a number of transmit antennas is N _t, reception antenna number is N _rextensive mimo system model, generally have N _r>=N _t.At transmitting terminal, equiprobability and the binary bit stream string of statistical iteration are also converted into N _tindividual parallel sub-bit stream, then (2 are modulated to every sub-bit stream ^m-PSK or 2 ^m-QAM modulation), the signal synchronous transmission after modulation, the Rayleigh mimo channel through slow fading arrives receiving terminal.At receiving terminal, the signal that each root reception antenna receives is all N _ttransmit antennas signal in time with the superposition of the multiple signals on frequency band, the base band Model in Time Domain of its equivalence can be expressed as:

Y＝HX+N

Wherein be respectively Received signal strength vector, the vector that transmits noise signal vector, for channel matrix, it is 0 that N obeys average, and variance is σ ²multiple Gaussian Profile.

Specific embodiment of the invention scheme flow chart as shown in Figure 2:

Step 1: initialization, i=1, Y ⁽¹⁾=Y, H ⁽¹⁾=H (wherein (1) represents the 1st iteration);

Step 2: vector Y to received signal ⁽ⁱ⁾carry out MF detection judge testing result in whether there is symbol and be positioned at threshold areas in, wherein i, j represent iterations and character position index respectively;

If exist, then filter out the symbol being positioned at threshold areas, choose the element that real part and imaginary part absolute value sum in these symbols are maximum carry out judgement to obtain perform Interference Cancellation, namely upgrade channel matrix simultaneously, delete channel matrix H ⁽ⁱ⁾kth _irow, namely

If do not exist, then choose in the real part of all symbols and the maximum element of imaginary part absolute value sum carry out judgement to obtain perform Interference Cancellation, namely upgrade channel matrix simultaneously, delete channel matrix H ⁽ⁱ⁾kth _irow, namely

Step 3: if iteration does not complete, then return step 2; If iteration completes, then output detections symbol, demodulation recovers original information bits.

Claims (1)

1. a novel matched filtering detection method for 4QAM modulation in extensive mimo system, concrete steps are as follows:

Step 1. initialization:

I=1,1≤i≤N _t, i represents i-th iteration, N _tfor number of transmit antennas;

Y ⁽¹⁾=Y, Y are Received signal strength vector;

H ⁽¹⁾=H, H are channel matrix;

Step 2. i-th iteration: vector Y to received signal ⁽ⁱ⁾carry out matched filtering detection, obtain detected symbol vector namely wherein, () ^hrepresent conjugate transpose operation, detected symbol vector in symbol be expressed as j is character position index;

Judge vector in symbol whether meet pre-determined threshold region wherein, T is threshold value, represent and get real part computing, represent and get imaginary-part operation, || represent signed magnitude arithmetic(al);

If vector in there is symbol and meet threshold areas S, then calculate real part and the imaginary part absolute value sum of all symbols in threshold areas S, choose and be worth maximum symbol and carry out Interference Cancellation; Otherwise, compute vectors in the real part of all symbols and imaginary part absolute value sum, choose and be worth maximum symbol and carry out Interference Cancellation; Concrete steps are:

k _irepresent real part and imaginary part absolute value and be worth maximum character position index, i.e. the aerial position index of i-th iterative detection;

< > _mrepresent that the m getting matrix is capable, G ⁽ⁱ⁾represent channel matrix H in i-th iterative process ⁽ⁱ⁾associate matrix;

${\tilde{x}}_{k_i}=w_{k_i}{Y}^{\left(i\right)};$

q () is constellation point reflection operator, as estimated value;

Perform Interference Cancellation: [] _nrepresent the n-th row getting matrix;

Meanwhile, channel rectangle is upgraded: represent the n-th row of puncture table;

Step 3. repeats step 2, and iteration is to i=N _t, finishing iteration output detections symbol, demodulation recovers original information bits.