CN107682119A - A kind of MIMO space -time code recognition methods based on packet extreme value model - Google Patents

Abstract

Aiming at the identification problem of SM and STBC space-time codes in the MIMO transmission system, the invention proposes a MIMO space-time code identification method based on grouping extremum model. First, the signals of any two different receiving antennas are time-delay correlated, and then the correlation spectrum modulus is calculated, and the grouping extremum model is applied to group the correlation spectrum modulus sequence appropriately and take the maximum value of each group to obtain the grouping extremum The sequence is normalized, and the maximum value of the normalized grouping extremum sequence is found as the identification feature quantity; the corresponding threshold is set, and if the identification feature quantity is greater than the threshold value, it is recognized as the STBC pattern, otherwise, Then it is identified as SM pattern. Simulation results show that the two types of space-time codes in MIMO can be effectively identified without signal prior information.

Description

A MIMO Space-Time Code Recognition Method Based on Packet Extremum Model

technical field

The invention belongs to the field of signal identification and processing, and in particular relates to a MIMO space-time code identification method based on a grouping extremum model.

Background technique

Signal recognition is a classic topic in military and civilian fields such as communication reconnaissance and cognitive radio, and is also an indispensable technical link in reconfigurable communication. Its tasks generally include estimation of the number of transmitting antennas, identification of space-time codes, and identification of modulation methods, etc. links. Under non-cooperative conditions, space-time code recognition is usually the premise and basis of modulation and subsequent decoding. The existing methods can be mainly divided into two categories: likelihood ratio recognition and feature recognition. Likelihood ratio recognition has the best performance but requires prior information of the signal and channel, and is easily affected by model mismatch and has high complexity. , and feature recognition methods mainly include cyclostationary frequency detection method, fourth-order moment peak feature method, etc. The complexity of these methods is slightly lower, but the performance is poor at low noise ratio.

The present invention is based on the grouping extremum model in the extremum theory (EVT), based on the correlation spectrum of any two receiving antennas, selects a specific feature quantity and threshold, completes the identification of SM and STBC code patterns, and calculates the algorithm The complexity is low, and it still has good performance at low signal-to-noise ratio. In particular, the method can overcome the disadvantage of algorithm failure caused by spectral line splitting in cyclostationary frequency detection.

Contents of the invention

Aiming at the deficiencies in the prior art, the present invention provides a MIMO space-time code identification method based on grouping extremum model.

To achieve the above object, the present invention adopts the following technical solutions:

1) Calculate the delay correlation spectral modulus of the observed signals between different receiving antennas;

2) Carry out appropriate grouping of relevant spectral modulus values, and take the maximum value of each grouping, construct grouping maximum value sequence;

3) Obtain a normalized grouped maximum sequence based on the extreme value theory EVT;

4) Extracting the maximum value of the normalized group maximum value sequence as the identification feature quantity;

5) Setting the threshold for MIMO space-time code identification;

6) Comparing the identification feature quantity with the threshold to identify SM and STBC codes in the MIMO signal.

In order to optimize the above technical solutions, the specific measures taken also include:

In step 1):

It is assumed that in the MIMO transmission environment, there are L transmitting antennas and P receiving antennas, L>1, P>1, and the receiving signal vector of the receiving antenna is: r(n)=H(n)s(n)+w( n), n=0,..., N-1, where H(n) is a Rayleigh fading channel matrix, w(n) is additive Gaussian white noise, and the variance is s(n) is the signal vector, N is the signal sample length;

Note that the received signal of the i-th antenna is r _i (n), the received signal of the j-th antenna is r _j (n), and the delay is τ, then the correlation spectrum modulus value is: U(k)=|DFT[ r _i (n)r _j (n-τ)]|, k=0,..., N-1, where i≠j, the signals involved in the correlation spectrum operation must come from different receiving antennas, τ>1.

In step 2):

Group the correlation spectral modulus values U(k), divide them evenly into K groups, and take the maximum value γ _l for each group, l=0,...,K-1, and divide the K group maximum values Constitute group maximum value sequence {γ _l }, l=0, .., K-1.

In step 3):

Based on the EVT method, the grouped maximum value sequence {γ _l } is normalized to obtain the normalized grouped maximum value sequence U′(l)=[γ _l -a _K ]/b _K ., l=0, ..., K-1, where the normalization coefficient σ _z is the standard deviation of the correlation spectrum sequence.

In step 4):

The maximum value of the normalized group maximum value sequence U'(l) is selected as the identification feature quantity, which is recorded as R _evt =max[U'(l)].

In step 5):

Set the threshold th _evt for MIMO space-time code identification, th _evt =-1n[-ln(1-P _fa )], where P _fa is the false alarm probability.

In step 6):

When R _evt ≥ th _evt , the code type is STBC; otherwise, the code type is SM.

The beneficial effects of the present invention are: by extracting the maximum value of the correlation spectrum normalized packet extremum as a basis to complete the identification of the SM and STBC two code patterns, only need to use the received signals of any two receiving antennas, without the need for signal Prior information, it still has better performance when the signal-to-noise ratio is lower, and the method is easy to implement.

Description of drawings

Fig. 1 is a flowchart of the identification method of the present invention.

Fig. 2 is a statistical histogram of the maximum value of the correlation spectrum normalized grouping extremum sequence under different code patterns.

Fig. 3 is a performance comparison diagram between the present invention and the cyclostationary frequency detection method under the same simulation conditions.

detailed description

The present invention is described in further detail now in conjunction with accompanying drawing.

In the identification method of the present invention, the signals of two different receiving antennas are firstly selected, and the two are subjected to delay correlation, and then the correlation spectrum modulus value is calculated, and the correlation spectrum modulus value sequence is properly grouped and the maximum value is obtained to obtain Group the extremum sequence and normalize based on the EVT method, find the maximum value of the normalized grouped extremum sequence as the identification feature, and then set the corresponding threshold. If the identification feature is greater than this threshold, the identification It is STBC pattern, otherwise, it is identified as SM pattern.

Fig. 1 shows a MIMO space-time code identification method based on grouping extremum model, which specifically includes the following steps.

1. Calculation of correlation spectrum modulus

Assuming that in a MIMO transmission environment, there are L transmitting antennas and P receiving antennas, L>1, P>1, then the receiving signal vector of the receiving antenna is:

r(n)=H(n)s(n)+w(n), n=0,...,N-1

In the formula: H(n) is the Rayleigh fading channel matrix, w(n) is additive white Gaussian noise (variance is ), s(n) is the signal vector, and N is the signal sample length.

Note that the received signal of the i-th antenna is r _i (n), and the received signal of the j-th antenna is r _j (n), and its correlation spectrum modulus is:

U(k)=|DFT[r _i (n)r _j (n-τ)]|, k=0,...,N-1

In the formula, i≠j is required, that is, the signals participating in the correlation spectrum calculation must come from different receiving antennas, τ is the delay amount, and τ>1.

2. Construct grouping maximum value sequence

Group the correlation spectrum modulus value U(k), and evenly divide it into K groups (generally, the number of samples in each group is 5-15), and take the maximum value γ _l for each group, l=0, .. ., K-1, K grouping maxima constitute a grouping maxima sequence {γ _l }, l=0, . . . , K-1.

3. Normalization

Based on the EVT method, the grouped maximum value sequence {γ _l } is normalized to obtain the normalized grouped maximum value sequence:

U'(l)=[γ _l -a _K ]/b _K ., l=0, ..., K-1

Among them, the normalization coefficient where σ _z is the standard deviation of the correlation spectrum sequence, in practice, σ _z is given by It is estimated that, in the formula It is the statistical average after removing 3-5 large spectral lines in the correlation spectrum U(k).

Fourth, define the identification feature quantity

The maximum value of the normalized group maximum value sequence U'(l) is selected as the identification feature quantity, which is recorded as R _evt =max[U'(l)].

5. Threshold setting

Set the threshold th _evt of MIMO space-time code identification, which is calculated by the following formula:

th _evt ＝-ln[-ln(1-P _fa )]

In the formula, _Pfa is the false alarm probability, which is generally between 0.01-0.0001.

6. Pattern recognition

When R _evt ≥ th _evt , the code type is STBC; otherwise, the code type is SM.

Table 1 illustrates the average recognition accuracy rate of this method. Assuming that there are 3 transmitting antennas and 4 receiving antennas in the MIMO transmission environment, the first receiving antenna is delayed related to the received signal of the second one, and the amount of delay for 20 sample points. The channel is a Rayleigh fading channel matrix, and the additional noise is additive white Gaussian noise. When the signal is transmitted, it is divided into 4 time slots, and the number of symbols in each time slot is 1024 points. The modulation method of the signal is QPSK modulation. The signal-to-noise ratio setting range is -3dB to 12dB with a step size of 3dB. For each signal-to-noise ratio, 1000 simulations are performed for two different code patterns, and the false alarm probability is 0.0001.

It can be seen from Table 1 that the correct recognition probability of this method under the above simulation conditions: when the signal-to-noise ratio is greater than -3dB, the correct recognition rates of the two code patterns are above 91%.

SNR(dB) -6 -3 0 3 6 9 12 Average recognition accuracy 0.655 0.811 0.9075 0.933 0.9505 0.9555 0.9625

Table 1 The performance of the identification method of the present invention under different signal-to-noise ratio conditions

Figure 2 shows the statistical histograms of the normalized packet maximum sequence maximum values of the two space-time coding schemes of SM and AL (that is, the STBC code when there are two transmitting and receiving antennas, and the transmission time slot is 2) obtained by simulation picture. The signal-to-noise ratio is set at 0dB, and 1000 simulations are performed for the two different patterns, and other simulation conditions are the same as those in Table 1. It can be seen from the figure that under different space-time code conditions, there are certain differences in the statistical histograms of the maximum value of the grouped maximum value sequence of the correlation spectrum value normalization, which provides a basis for the realization of this algorithm.

Figure 3 shows the performance comparison between this method and the cyclostationary frequency detection method under the simulation conditions set in Table 1. It can be seen from Fig. 3 that the algorithm proposed by the present invention is slightly better than the cyclostationary frequency detection method.

The above are only preferred implementations of the present invention, and the protection scope of the present invention is not limited to the above-mentioned embodiments, and all technical solutions under the idea of the present invention belong to the protection scope of the present invention. It should be pointed out that for those skilled in the art, some improvements and modifications without departing from the principle of the present invention should be regarded as the protection scope of the present invention.

Claims (7)

1. a kind of MIMO space -time code recognition methods based on packet extreme value model, specifically includes following steps：

1) the delay Correlated Spectroscopy modulus value of observation signal between different reception antennas is calculated；

2) Correlated Spectroscopy modulus value is grouped, and takes the maximum of each packet, structure is grouped very big value sequence；

3) the normalized very big value sequence of packet is obtained based on extreme value theory EVT；

4) maximum of the normalized very big value sequence of packet is extracted, as identification feature amount；

5) thresholding of setting mimo space -time code identification；

6) identification feature amount is compared with thresholding, identifies two kinds of codings of SM and STBC in MIMO signal.

A kind of 2. MIMO space -time code recognition methods based on packet extreme value model as claimed in claim 1, it is characterised in that： In step 1)：

In setting mimo transmission environment, there are L transmitting antenna, P reception antenna, L ＞ 1, P ＞ 1, the reception signal of reception antenna Vector is：R (n)=H (n) s (n)+w (n), n=0 ..., N-1, wherein, H (n) is Rayleigh fading type channel matrix, and w (n) is Additive white Gaussian noise, variance areS (n) is signal vector, and N is sample of signal length；

The reception signal for remembering i-th antenna is r_i(n), the reception signal of jth root antenna is r_j(n), amount of delay τ, then Correlated Spectroscopy Modulus value is：U (k)=| DFT [r_i(n)r_j(n- τ)] |, k=0 ..., N-1, wherein i ≠ j, the signal for participating in Correlated Spectroscopy computing must Must be from different reception antennas, τ ＞ 1.

A kind of 3. MIMO space -time code recognition methods based on packet extreme value model as claimed in claim 2, it is characterised in that： In step 2)：

Correlated Spectroscopy modulus value U (k) is grouped, it is uniformly divided into K groups, and maximum γ is taken to each packet_l, l= 0 ..., K-1, K packet maximum is formed and is grouped very big value sequence { γ_l, l=0 ..., K-1.

A kind of 4. MIMO space -time code recognition methods based on packet extreme value model as claimed in claim 3, it is characterised in that： In step 3)：

Based on EVT methods, very big value sequence { γ will be grouped_lBe normalized, obtain normalization and be grouped very big value sequence U ' (l) =[γ_l-a_K]/b_K, l=0 .., K-1, wherein, normalization coefficientσ_zFor phase Close the standard deviation of spectral sequence.

A kind of 5. MIMO space -time code recognition methods based on packet extreme value model as claimed in claim 4, it is characterised in that： In step 4)：

The maximum that selection normalization is grouped very big value sequence U ' (l) is designated as R as identification feature amount_evt=max [U ' (l)].

A kind of 6. MIMO space -time code recognition methods based on packet extreme value model as claimed in claim 5, it is characterised in that： In step 5)：

The thresholding th of setting mimo space -time code identification_evt, th_evt=-ln [- ln (1-P_fa)], in formula, P_faFor false-alarm probability.

A kind of 7. MIMO space -time code recognition methods based on packet extreme value model as claimed in claim 6, it is characterised in that： In step 6)：

Work as R_evt≥th_evtWhen, then pattern is STBC；Otherwise, pattern SM.