CN106375065A - Iterative MMSE soft detection method based on spatial modulation system having frequency offset - Google Patents

Abstract

The invention belongs to the technical field of communication anti-interference, and relates to spatial modulation (Spatial Modulation, SM) technology, minimum mean square error (Minimum Mean Square Error, MMSE) technology, and related MIMO (Multiple Input Multiple Output) technology. The invention provides a new MMSE soft detection algorithm based on an SM-OFDM system. According to the technology, the coefficient of an MMSE detector is theoretically calculated, soft information and external information of an iterative detector are solved by prior information LLR, and to-be-detected data is iteratively updated, so that the BER perform ace of the system is significantly improved.

Description

The soft detection method of iteration mmse based on the spatial modulation system that there is frequency deviation

Technical field

The invention belongs to Communication Anti-Jamming Techniques field, it is related to spatial modulation (spatial modulation, sm) technology, Least mean-square error technology (minimum mean square error, mmse), and its mimo (multiple of correlation Input multiple output) technology.

Background technology

Mimo modulation technique is the high speed transmission technology under a kind of wireless environment, and it is in transmitting terminal and/or receiving terminal configuration More antenna elements, and combine advanced Space Time Coding modulation scheme, by making full use of to spatial degrees of freedom, can carry Come extra diversity, multiplexing and beam forming gain.

Recently, sm technology is proposed as a kind of new modulation technique as a kind of new mimo modulation scheme.This skill The ultimate principle of art is by activating different antennas, antenna index value being modulated and is used for transmission information bit.This transmission side The essence of case is the independence using different channels in mimo system.Because an only antenna is activated every time, and then sending out Penetrating end and only need to a radio frequency unit and this process energy hop bit, so this approach increases transfer rate, and dropping The low cost of mimo system and complexity.

However, the same with ofdm system, sm-ofdm system equally can be affected by frequency deviation cfo, and then affects system Performance.In recent years, the compensation of many cfo and detection technique is had to be used for the up-link of mimo-ofdm.However, it is substantial amounts of imitative Really facts have proved that these technology are not particularly suited for sm-ofdm system, because contrast mimo-ofdm system, the letter of sm-ofdm system Road matrix is unsatisfactory for sequency spectrum.The method of the iterative interference cancellation (pic) based on ml is gone out recently, but because in pic Distracter is contacted with sending signal presence all the time and leads to can not completely eliminate interference with the performance of lift system.Carried on the back based on above Scape, therefore a kind of signal to detect the sm-ofdm system that there is frequency deviation for the method for the proposition soft detection of mmse of iteration.

Content of the invention

The present invention is directed to frequency deviation affects this problem larger to sm-ofdm systematic function, provides a kind of being based on to there is frequency deviation Spatial modulation system the soft detection method of iteration mmse, comprise the following steps:

S1, receiving terminal utilize the method detection signal of iteration mmse, specifically comprise the following steps that

S11, solve the coefficient of iterative detection mmse wave filter using prior information, receipt signal can be expressed as:Wherein, s (k) represents frequency deviation matrix

S12, mmse detection method estimates emission signal vector and is represented by:Wherein,It is the coefficient of mmse wave filter, n_tRepresent the number of transmitting antenna, n_rRepresent the number of reception antenna Mesh；

S13, try to achieve mmse filter coefficient c_k, b_kMethod as follows:

If trying to achieve c_k, b_k, then only need to meet with Under formulaAccording to above formula can abbreviation be:

$b\left(k\right)=e\[x\left(k\right)\]-c\left(k\right)\underset{l=0}{\overset{n-1}{\σ}}h\left(l\right)s(l-k)e\[x\left(l\right)\]$

$c\left(k\right)=\{e\[x\left(k\right)x{\left(k\right)}^h\]s{\left(0\right)}^{h}h{\left(k\right)}^h-e\[x\left(k\right)\]\underset{l=0}{\overset{n-1}{\σ}}e\[x{\left(l\right)}^h\]s{(l-k)}^{h}h{\left(l\right)}^h\}$

$\begin{array}{c}\{\underset{l=0}{\overset{n-1}{\σ}}h\left(l\right)s(l-k)e\[x\left(l\right)x{\left(l\right)}^h\]s{(l-k)}^{h}h{\left(l\right)}^h+e\[n\left(k\right)n{\left(k\right)}^h\]-\\ \underset{l=0}{\overset{n-1}{\σ}}h\left(l\right)s(l-k)e\[x\left(l\right)\]\underset{l=0}{\overset{n-1}{\σ}}e\[x{\left(l\right)}^h\]s{(l-k)}^{h}h{\left(l\right)}^h{\}}^{-1}\end{array}$ ,

Wherein,Then c (k), b (k) can abbreviation be:

$b\left(k\right)=-c\left(k\right)\underset{l=0,l\&notequal;k}{\overset{n-1}{\σ}}h\left(l\right)s(l-k)e\[x\left(l\right)\]$

$\begin{array}{c}c\left(k\right)=s{\left(0\right)}^{h}h{\left(k\right)}^h\{\underset{l=0,l\&notequal;k}{\overset{n-1}{\σ}}h\left(l\right)s(l-k)e\[x\left(l\right)x{\left(l\right)}^h\]s{(l-k)}^{h}h{\left(l\right)}^h+h\left(k\right)s\left(0\right)s{\left(0\right)}^{h}h{\left(k\right)}^h\\ +{\mathrm{\σ}}^2{i}_{n_r}-\underset{l=0,l\&notequal;k}{\overset{n-1}{\σ}}h\left(l\right)s(l-k)e\[x\left(l\right)\]\underset{l=0,l\&notequal;k}{\overset{n-1}{\σ}}e\[x{\left(l\right)}^h\]s{(l-k)}^{h}h{\left(l\right)}^h{\}}^{-1}\end{array}$ , The sending signal that can be derived estimation from above is:

S2, the Soft Inform ation using prior information llr solution iterative detection device and external information, specifically comprise the following steps that

S21, the average based on prior information llr solution sm symbol and variance and autocorrelation coefficient: Wherein, s represents the set of sm symbol constellation point, p (x_k=a_i) represent x_k= a_iProbability, can be solved by prior information, as follows:b_n∈{0, 1 }, l_a(b_n) represent prior information；

S22, the coefficient c according to s1 calculating mmse wave filter_k, and combine y_kObtain the sending signal estimatedAs follows:

S3, external information l based on maximum a posteriori detection (map) for the calculating_e(b_n) (n ∈ (1, b)), specifically comprises the following steps that

S31, it is represented by by the output external information of maximum a posteriori criterion:

Wherein,WithRepresent the subset of sm symbol, and meetWith Represent the probability matrix of k-th sm symbol, be represented by:Wherein,WithRepresent estimate symbolAverage and variance, then u_k It is represented by:

$u_k=e\left({\hat{x}}_k\right|x_k=a_i)=e(c_k\underset{l=0}{\overset{k-1}{\σ}}{h}_l{s}_{l-k}{x}_l+c_k{n}_k+b_k|x_k=a_i)=c_k{h}_k{s}_0{a}_i;$

S32, due to every transmitting antenna transmission signal identical random attribute, therefore conventional letterEach element All meet the distribution of homoscedastic multiple Gauss, so can be byObtain varianceWherein d_kIt is represented byVariance square Battle array:

$\begin{array}{c}{d}_k=d\left({\hat{x}}_k\right|x_k=a_i)=d(c_k\underset{l=0}{\overset{k-1}{\σ}}{h}_l{s}_{l-k}{x}_l+c_k{n}_k+b_k|x_k=a_i)\\ c_k\[\underset{l=0,l\&notequal;k}{\overset{k-1}{\σ}}{h}_l{s}_{l-k}d\left(x_l\right)s_{l-k}^h{h}_l^h\]c_k^h+{\mathrm{\σ}}^2{c}_k{c}_k^h\end{array}$

Therefore:Can be reduced to:

S4, the sm symbol solving on k-th subcarrier are:

The invention has the beneficial effects as follows:

The invention provides a kind of propose a kind of soft detection algorithm of new mmse based on sm-ofdm system, this technology is passed through The coefficient of Theoretical Calculation mmse detector, solves Soft Inform ation and the external information of iterative detection device, iteration is more using prior information llr The newly data of required detection, makes the ber performance of system be significantly improved.

Brief description

Fig. 1 is traditional sm system block diagram.

Fig. 2 is the system block diagram of iteration mmse detection proposed by the present invention.

Specific embodiment

Below in conjunction with the accompanying drawings, the present invention is described in further detail.

For preferably the present invention will be described, first introduce the term used in technical solution of the present invention and spatial modulation system System transmitter architecture.

Spatial modulation: such as Fig. 1, b are the bit datas needing transmission, can be considered the matrix of a l × t, wherein l= Log2 (m)+log2 (mary) is the amount of bits entrained by a sm modulation symbol, and mary is quadrature amplitude modulation (quadrature amplitude modulation, qam) exponent number.As can be seen that the ratio that a sm modulation symbol is carried by Special quantity is together decided on by qam order of modulation and number of transmission antennas.Sm modulation criterion is to be transformed into b according to sm translation table The matrix x of one m × t.In x, string represents the data of a moment transmission, any string only one of which non-zero, meaning Taste any time and only has an antenna to send data.Fig. 2, gives four transmitting antennas, and modulation symbol is the example of bpsk Son.

Specific embodiments of the present invention system diagram as shown in Figure 2.

Transmitter architecture is roughly divided into several steps as follows:

Step 1: determine the parameter of system to be selected, that is, determine transmitting antenna number ntx, reception antenna number nrx, adjust The exponent number m of system.

Step 2: and then calculate the amount of bits of a frame, this frame data is divided into two groups, one group is antenna index bit, For selecting the transmitting antenna being activated, one group is modulation bit, is used for carrying out psk modulation.Again the transmission vector obtaining is entered Row sm modulates；After symbolic vector after being modulated, precoding is carried out to it, the precoding vector in this processing procedure is by connecing Receiving end feedback obtains.

Claims (1)

1. a kind of soft detection method of iteration mmse based on the spatial modulation system that there is frequency deviation is it is characterised in that include following Step:

S1, receiving terminal utilize the method detection signal of iteration mmse, specifically comprise the following steps that

S11, solve the coefficient of iterative detection mmse wave filter using prior information, receipt signal can be expressed as:

Wherein, s (k) represents frequency deviation matrix

S12, mmse detection method estimates emission signal vector and is represented by:Wherein,It is the coefficient of mmse wave filter, n_tRepresent the number of transmitting antenna, n_rRepresent the number of reception antenna Mesh；

S13, try to achieve mmse filter coefficient c_k, b_kMethod as follows:

If trying to achieve c_k, b_k, then following formula need to only be met

According to above formula can abbreviation be:

,

Wherein,

Then c (k), b (k) can abbreviation be:

,

The sending signal that can be derived estimation from above is:

S2, the Soft Inform ation using prior information llr solution iterative detection device and external information, specifically comprise the following steps that

S21, the average based on prior information llr solution sm symbol and variance and autocorrelation coefficient:

Wherein, s represents the set of sm symbol constellation point, p (x_k=a_i) represent x_k=a_iProbability, can be solved by prior information, such as Shown in lower:

l_a(b_n) represent prior information；

S22, the coefficient c according to s1 calculating mmse wave filter_k, and combine y_kObtain the sending signal estimatedAs follows:

S3, external information l based on maximum a posteriori detection (map) for the calculating_e(b_n) (n ∈ (1, b)), specifically comprises the following steps that

S31, it is represented by by the output external information of maximum a posteriori criterion:

Wherein,WithRepresent the subset of sm symbol, and meet Represent the probability matrix of k-th sm symbol, be represented by:Wherein,WithRepresent estimate symbolAverage and variance, then u_k It is represented by:

$u_k=e\left({\hat{x}}_k\right|x_k=a_i)=e(c_k\underset{l=0}{\overset{k-1}{\σ}}{h}_l{s}_{l-k}{x}_l+c_k{n}_k+b_k|x_k=a_i)=c_k{h}_k{s}_0{a}_i;$

S32, due to every transmitting antenna transmission signal identical random attribute, therefore conventional letterEach elementAll full Foot homoscedastic multiple Gauss distribution, so can be byObtain varianceWherein d_kIt is represented byVariance matrix:

$\begin{array}{c}{d}_k=d\left({\hat{x}}_k\right|x_k=a_i)=d(c_k\underset{l=0}{\overset{k-1}{\σ}}{h}_l{s}_{l-k}{x}_l+c_k{n}_k+b_k|x_k=a_i)\\ =c_k\[\underset{l=0,l\&notequal;k}{\overset{k-1}{\σ}}{h}_l{s}_{l-k}d\left(x_l\right)s_{l-k}^h{h}_l^h\]c_k^h+{\mathrm{\σ}}^2{c}_k{c}_k^h\end{array}$

Therefore:Can be reduced to:

S4, the sm symbol solving on k-th subcarrier are: