Frequency collision array signal separation method and device based on channel matrix

Abstract

The present invention belongs to the field of wireless communication technology and discloses a method for separating frequency collision array signals based on a channel matrix. The method comprises the following steps: two radiation sources transmit signals at the same symbol rate, and two antennas receive the frequency collision signals according to the symbol period; a convolution matrix based on the channel tap coefficients is established, and a channel matrix is ​​constructed based on the convolution matrix; a frequency collision array signal model received by the two antennas is constructed based on the channel matrix, the transmission signal sequences of different radiation sources, and an additive white Gaussian noise matrix; and the pseudo-inverse of the convolution matrix in the frequency collision array signal model received by the two antennas is calculated to separate the transmission signal sequences of the two radiation sources. Based on the channel matrix, the present invention solves the problem that a single-channel frequency collision signal cannot be demodulated independently, thus limiting communication, simplifies the calculation process for signal separation, and improves communication reliability in the event of frequency congestion.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communications and relates to a frequency collision array signal separation method and device based on a channel matrix. Background Art

[0002] During signal transmission, the same antenna may receive signals from different radiating sources. Because different radiating sources are at varying distances from the receiving antennas, the frequency-colliding signals received by different antennas exhibit differences in parameters such as signal-to-noise ratio and amplitude ratio. Furthermore, because each receiving channel exhibits time-frequency aliasing, a single channel's frequency-colliding signal cannot independently complete communication processes such as demodulation, limiting communication. Therefore, a new method for separating frequency-colliding signals is needed. Summary of the Invention

[0003] The present invention addresses the technical problem that single-channel frequency collision signals cannot be demodulated independently, thus limiting communication. It provides a frequency collision array signal separation method and device based on a channel matrix. A frequency collision array signal model received by an antenna array is constructed based on the channel matrix. By calculating the pseudo-inverse of the convolution matrix in the frequency collision array signal model received by the antenna, the transmission signal sequences of the two radiation sources can be separated, thereby simplifying the calculation process of signal separation and improving communication reliability in the event of frequency congestion.

[0004] To achieve the above object, the present invention adopts the following technical solutions:

[0005] In a first aspect, the present invention provides a frequency collision array signal separation method based on a channel matrix, comprising the following steps:

[0006] Two radiating sources transmit signals with the same symbol rate, and two antennas receive frequency-colliding signals according to the symbol period;

[0007] Establish a convolution matrix based on the channel tap coefficients, and construct a channel matrix based on the convolution matrix;

[0008] Based on the channel matrix, the transmitted signal sequences of the two radiating sources and the additive white Gaussian noise matrix, a frequency collision array signal model received by two antennas is constructed;

[0009] The pseudo-inverse of the convolution matrix in the frequency collision array signal model received by the two antennas is obtained, and the transmission signal sequences of the two radiation sources are separated.

[0010] In one technical solution, the channel matrix H is expressed as follows:

[0011]

[0012] Where i is the number of antennas, i=1,2, j is the number of radiation sources, j=1,2, H ijis the convolution matrix based on the channel tap coefficient, L is the channel constraint length, L = 2L n +1, L n are the non-causal and causal periods of the equivalent filter, N is the signal length,

[0013]

[0014] In one technical solution, the frequency collision array signal model received by the two antennas is as follows:

[0015]

[0016] Where H is the channel matrix, y1 is the observation signal sequence received by antenna 1, y1 = [y 1,1 y 1,2 y 1,3 ...y 1,N- 1y 1,N ] T , y2 is the observation signal sequence received by antenna 2, y2=[y 2,1 y 2,2 y 2,3 ...y 2,N-1 y 2,N ] T , s1 is the signal sequence sent by radiation source 1, s1=[s 1,1 s 1,2 s 1,3 ...s 1,N-1 s 1,N ] T , s2 is the signal sequence sent by radiation source 2, s2 = [s 2,1 s 2, 2s 2,3 ...s 2,N-1 s 2,N ] T , N is the signal length, v1 and v2 are additive white Gaussian noise.

[0017] In a second aspect, the present invention provides a frequency collision array signal separation device for implementing the above-mentioned channel matrix-based frequency collision array signal separation method, comprising:

[0018] A channel matrix creation module, configured to construct a channel matrix based on a convolution matrix based on channel tap coefficients;

[0019] A model creation module is used to construct a frequency collision array signal model received by two antennas based on the channel matrix, the transmission signal sequences of the two radiation sources, and the additive white Gaussian noise matrix;

[0020] The calculation module is used to obtain the pseudo-inverse of the convolution matrix in the frequency collision array signal model received by the two antennas, and separate the transmission signal sequences of the two radiation sources.

[0021] In a third aspect, the present invention provides an electronic device comprising a memory and a processor, wherein the processor and the memory communicate with each other via a bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the above-mentioned frequency collision array signal separation method.

[0022] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the above-mentioned frequency collision array signal separation method when executed by a processor.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] The present invention constructs a frequency collision array signal model received by an antenna array based on a channel matrix. By obtaining the pseudo-inverse of the convolution matrix in the frequency collision array signal model received by the antenna, the transmission signal sequences of the two radiation sources can be separated. This solves the problem that a single-channel frequency collision signal cannot be demodulated independently, thus limiting communication. The calculation process of signal separation is simplified, and communication reliability is improved in the event of frequency congestion. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 Figure 3 is a simulation result showing the effect of the signal-to-noise ratio of the frequency collision signal received by antenna 2 on the separation and demodulation performance under different signal-to-noise ratios.

[0026] Figure 2 To be a 21 =1.0, a 22 =1.0, a 11 and a 12 Impact on signal separation performance.

[0027] Figure 3 To be a 21 =2.0, a 22 =1.0, a 11 and a 12 The impact of changes on signal separation performance.

[0028] Figure 4 To be a 21 =3.0, a 22 =1.0, a 11 and a 12 The impact of changes on signal separation performance. DETAILED DESCRIPTION

[0029] The following examples are used to illustrate the present invention but are not intended to limit the scope of protection of the present invention. Unless otherwise specified, the technical means used in the examples are conventional means well known to those skilled in the art. The test methods in the following examples are all conventional methods unless otherwise specified.

[0030] Example 1

[0031] If two radiating sources (i.e., j = 1, 2) respectively transmit MPSK or MQAM modulated signals with the same symbol rate, and the symbol period of the two signal components is T, the frequency-colliding signals received by the two antennas (i.e., i = 1, 2) are down-converted to obtain the complex baseband signal observation data at time t:

[0032]

[0033] Where, v i (t) is an additive white Gaussian noise with a mean of 0 and a power spectral density of N0; x1(t) and x2(t) are the baseband modulation waveforms of the signals sent by radiation source 1 and radiation source 2, respectively.

[0034] Assuming that the modulation mode of the two signals is the same and the two signals are statistically independent of each other, then x i The expression of (t) is:

[0035]

[0036] Where: s 1,k 、s 2,k are the signal sequences sent by radiation source 1 and radiation source 2 respectively. Their values ​​are related to the modulation mode. k is the number of signal cycles, k = 0, 1, 2, ...;

[0037] τ 1j,t , τ 2j,t The time delays of the two signal components received by antenna 1 and antenna 2 at time t (|τ i,t / T|≤1 / 2);

[0038] a 1j,t 、a 2j,t are the amplitudes of the two signal components received by antenna 1 and antenna 2 at time t;

[0039] f 1j,t 、f 2j,t are the carrier frequency deviations of the two signal components received by antenna 1 and antenna 2 at time t;

[0040] The initial phases of the carrier waves of the two signal components received by antenna 1 and antenna 2 at time t corresponding to time t0;

[0041] g 1j (t), g2j (t) is the impulse response of the equivalent channel filter of antenna 1 and antenna 2. The filter includes a shaping filter, a channel filter, and a matching filter.

[0042] Assume that both the non-causal and causal periods of the equivalent filter are L n , the filter response duration is [-L n T,L n T], then the channel constraint length is L = 2L n +1.

[0043] If the received signal is sampled according to the symbol period, the observation data of antenna 1 and antenna 2 in the kth sampling period are as follows:

[0044]

[0045] Where y i,k =y i (kT), h i,k =h i,kT , v i,k =v i (kT).

[0046] Then x i,k The expression is transformed into:

[0047]

[0048] Where, τ ij,k =τ ij,kT represents discrete time, and m takes values ​​over the duration of the filter response.

[0049] To simplify the channel model, it is assumed that the amplitude and delay remain unchanged in a short time, that is, a ij =a ij,k , τ i =τ ij,k , define a 1×L-dimensional channel tap coefficient vector:

[0050]

[0051] Where T represents transposition, the discrete forms of the frequency collision signals received by antenna 1 and antenna 2 are as follows:

[0052] y 1,k =h 11,k s 1,k +h 12,k s 2,k +v 1,k ,

[0053] y 2,k =h 21,k s1,k +h 22,k s 2,k +v 2,k ,

[0054] y 1,k and y 2,k Both are frequency collision signals, that is, the superposition of signals sent by radiation sources 1 and 2.

[0055] Since a single-channel frequency collision signal cannot be demodulated alone, it is necessary to build an array signal processing model to separate the frequency collision signals.

[0056] If y1=[y 1,1 y 1,2 y 1,3 ...y 1,N-1 y 1,N ] T , y2=[y 2,1 y 2,2 y 2,3 ...y 2,N-1 y 2,N ] T Denotes the observation signal sequences of antenna 1 and antenna 2 respectively, s1=[s 1,1 s 1,2 s 1,3 ...s 1,N-1 s 1,N ] T ,s2=[s 2,1 s 2,2 s 2,3 ...s 2,N-1 s 2,N ] T Denote the signal sequences sent by radiation source 1 and radiation source 2 respectively, and N is the signal length (i.e., the number of symbols). The frequency collision array signal model received by antenna 1 and antenna 2 can be expressed as:

[0057]

[0058] Where H is the channel matrix, Matrix element H ij (i=1,2; j=1,2) is the convolution matrix based on the channel tap coefficients, for example:

[0059]

[0060] Channel Matrix Seeking pseudo-reversal, we can get s'1 and s'1 are the signal sequences sent by the separated radiation sources 1 and 2.

[0061] Separation effect verification:

[0062] The 2-condition number of the channel matrix H can be calculated by the singular value decomposition of the matrix, that is:

[0063]

[0064] Among them, σ max is the largest singular value of H, σ min is the smallest singular value of H.

[0065] The condition number of a matrix is ​​used to determine whether it is "well-conditioned" or "ill-conditioned." Generally speaking, the larger the condition number, the closer the matrix is ​​to a singular matrix and the more "ill-conditioned" it is. In numerical calculations, the larger the condition number, the larger the upper bound on the relative error in the matrix inversion, that is, the more "uncontrollable" the error is, and the lower the accuracy of the solution. Therefore, the condition number of the channel matrix H affects the bit error rate (BER) of the signals transmitted by separated radiation sources 1 and 2. Factors affecting the condition number of H mainly include the amplitude, delay, and signal-to-noise ratio (SNR) of the signals received by antennas 1 and 2.

[0066] Taking the QPSK modulated signal as an example, the separation method of this embodiment is simulated. The simulation conditions are set as follows: the equivalent channel filter uses a square root raised cosine pulse with a roll-off factor of 0.3, the channel constraint length L = 21, the symbol period T = 1, and the sampling multiple

[0067] (1) Impact of signal-to-noise ratio on signal separation performance

[0068] Frequency deviation f 11 =2×10 -6 、f 12 =-2×10 -6 、f 21 =-2×10 -6 、f 22 =3×10 -6 , initial phase Delay τ 11 =-0.3, τ 12 =0.1, τ 21 =0.2, τ 22 =-0.3, amplitude a 11 =a 12 = 10.0. The signal-to-noise ratio (SNR1) of the signal arriving at antenna 1 is 15 dB, and the signal-to-noise ratio (SNR2) of the signal arriving at antenna 2 is in the range [1, 20]. The signal sequence length N is 10,000.

[0069] Figure 1The simulation results of the effect of the signal-to-noise ratio of the frequency collision signal received by antenna 2 on the separation and demodulation performance under different signal-to-noise ratios are shown in Figure 2. Figure 1 -a is when a 21 =2.0, a 22 =1.0, the impact of SNR2 on signal separation performance; Figure 1 -b is when a 21 =1.0, a 22 = 2.0, the impact of SNR2 on signal separation performance. It can be seen that as SNR2 increases, the bit error rate of the separated signals from sources A and B gradually decreases. This is because as SNR2 increases, the quality of the frequency-colliding signals received by antenna 2 improves, making it more conducive to separation and demodulation of the frequency-colliding signals.

[0070] contrast Figure 1 -a and 1-b found that when a 11 =a 12 And a 21 >a 22 When s1 is A BER lower than s2 B ; when a 11 =a 12 And a 21 22 When s2 has a BER of B BER lower than s1 A Therefore, the signal separation and demodulation performance is greatly affected by the signal amplitude. Next, we analyze the impact of amplitude on signal separation performance.

[0071] (2) Impact of amplitude on signal separation performance

[0072] Frequency deviation f 11 =2×10 -6 、f 12 =-2×10 -6 、f 21 =-2×10 -6 、f 22 =3×10 -6 , initial phase Delay τ 11 =-0.3, τ 12 =0.1, τ 21 =0.2, τ 22 =-0.3, amplitude a 11 and a 12 The values ​​are taken in the range [1, 10]. The SNR1 of the signal arriving at antenna 1 is 15dB, the SNR2 of the signal arriving at antenna 2 is 10dB, and the signal sequence length N is 5000.

[0073] ​ Figure 2 To be a 21 =1.0, a 22 =1.0, a 11 and a 12 The impact on signal separation performance, including Figure 2 -a is a 11 and a 12 Impact on the signal separation performance of radiation source 1, Figure 2 -b is a 11 and a 12 Impact on the signal separation performance of radiation source 2, Figure 2 -c is a 11 and a 12 The impact on the condition number of the channel matrix H.

[0074] Figure 3 To be a 21 =2.0, a 22 =1.0, a 11 and a 12 The effect of changes on signal separation performance, where Figure 3 -a is a 11 and a 12 Impact on the signal separation performance of radiation source 1, Figure 3 -b is a 11 and a 12 Impact on the signal separation performance of radiation source 2, Figure 3 -c is a 11 and a 12 The impact on the condition number of the channel matrix H.

[0075] Figure 4 To be a 21 =3.0, a 22 =1.0, a 11 and a 12 The effect of changes on signal separation performance, where Figure 4 -a is a 11 and a 12 Impact on the signal separation performance of radiation source 1, Figure 4 -b is a 11 and a 12 Impact on the signal separation performance of radiation source 2, Figure 4 -c is a 11 and a 12 The impact on the condition number of the channel matrix H.

[0076] By comparison, it can be found that when a 11 / a 12 =a 21 / a 22When , the bit error rate reaches a peak, and the separation performance is poor. Analysis shows that at this time, the condition number of the channel matrix H reaches a peak. Since the larger the condition number of the matrix, the closer the matrix is ​​to a singular matrix, the larger the relative error of the matrix inversion and the lower the accuracy of the solution, resulting in a larger bit error rate of the separated signals sent by radiation sources 1 and 2. When a 21 =1.0, a 22 = 1.0, that is, the amplitudes of the two signal components received by antenna 2 are equal and weaker than the signal received by antenna 1. The bit error rate of the separated radiation source 1 and 2 signals is high, and the separation demodulation performance is poor. At the same time, it can be seen that as a 21 As a result, the signal component reaching antenna 2 is enhanced, which is beneficial to the separation and demodulation of the frequency collision signal and reduces the demodulation bit error rate.

[0077] The units and method steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. A person of ordinary skill in the art may use different methods to implement the described functions for each specific application, but such implementation is not considered to be beyond the scope of the present invention.

[0078] Those skilled in the art will appreciate that all or part of the steps in the above method can be performed by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a magnetic disk, or an optical disk. Alternatively, all or part of the steps in the above embodiment can be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or software functional modules. The present invention is not limited to any specific combination of hardware and software.

[0079] The embodiments described above are only preferred embodiments of the present invention and are only used to explain the present invention, not to limit the scope of implementation of the present invention. For those skilled in the art, it is of course possible to easily make other implementation methods by replacing or changing the technical content disclosed in this specification. Therefore, all changes and improvements made on the principles of the present invention should be included in the scope of the patent application of the present invention.

Claims

1. A frequency collision array signal separation method based on a channel matrix, characterized in that: The following steps are involved: Two radiating sources transmit signals with the same symbol rate, and two antennas receive frequency-colliding signals according to the symbol period. A convolution matrix based on the channel tap coefficients is established, and a channel matrix is ​​constructed based on the convolution matrix. The channel matrix H is expressed as follows: Where i is the number of antennas, i=1,2, j is the number of radiation sources, j=1,2, H ij is the convolution matrix based on the channel tap coefficient, L is the channel constraint length, L = 2L n +1, L n are the non-causal and causal periods of the equivalent filter, N is the signal length, Based on the channel matrix, the transmission signal sequences of the two radiation sources and the additive white Gaussian noise matrix, a frequency collision array signal model received by the two antennas is constructed; the frequency collision array signal model received by the two antennas is as follows: Where H is the channel matrix, y1 is the observation signal sequence received by antenna 1, y1 = [y 1,1 y 1,2 y 1,3 ...y 1,N-1 y 1,N ] T , y2 is the observation signal sequence received by antenna 2, y2=[y 2,1 y 2,2 y 2,3 ...y 2,N-1 y 2,N ] T , s1 is the signal sequence sent by radiation source 1, s1=[s 1,1 s 1,2 s 1,3 ...s 1,N-1 s 1,N ] T , s2 is the signal sequence sent by radiation source 2, s2 = [s 2,1 s 2,2 s 2,3 ...s 2,N- 1s 2,N ] T , N is the signal length, v1 and v2 are additive white Gaussian noise; The pseudo-inverse of the convolution matrix in the frequency collision array signal model received by the two antennas is obtained, and the transmission signal sequences of the two radiation sources are separated.

2. A frequency collision array signal separation device for implementing the frequency collision array signal separation method based on a channel matrix according to claim 1, characterized in that: include: A channel matrix creation module, configured to construct a channel matrix based on a convolution matrix based on channel tap coefficients; A model creation module is used to construct a frequency collision array signal model received by two antennas based on the channel matrix, the transmission signal sequences of the two radiation sources, and the additive white Gaussian noise matrix; The calculation module is used to obtain the pseudo-inverse of the convolution matrix in the frequency collision array signal model received by the two antennas, and separate the transmission signal sequences of the two radiation sources.

3. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the processor and the memory communicate with each other via a bus; the memory stores program instructions that can be executed by the processor, and the processor can execute the method according to claim 1 by calling the program instructions.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to claim 1 is implemented.

Images