A method for generating signals and directly estimating signal bit error rate based on shaping filter

Abstract

The present invention belongs to the field of electronic reconnaissance technology, and relates to a method for generating signals based on shaping filters and directly estimating the signal bit error rate. The present invention takes the direct acquisition of the bit error rate of interference signals on communication signals as a starting point, first establishes a decision criterion at the signal sampling point, and proposes that the size of the communication signal bit error rate depends on the probability distribution of the amplitude of the interference signal at the sampling point during sampling decision, and then uses the characteristic parameters of the shaping filter and the delay size to calculate the probability distribution of the amplitude of the interference signal at the sampling point, and then directly calculates the communication signal bit error rate through its amplitude probability distribution. The present invention can estimate the impact of interference signals on the communication signal bit error rate in advance, saving the step of signal data processing at the receiving end, and the method is simple and effective.

Description

Technical Field

[0001] The invention belongs to the technical field of electronic countermeasures and relates to a method for generating a signal based on a shaping filter and directly estimating a signal bit error rate. Background Art

[0002] In communication electronic countermeasures, wireless communication systems are extremely susceptible to active or passive interference due to their inherent openness, and actively imposed interference can significantly affect the performance of wireless communication systems, among which bit error rate is an important indicator to measure communication performance.

[0003] At present, most commonly used wireless communication signals are band-limited by a shaping filter at the transmitting end, and are received by a matched filter at the receiving end. This method can overcome the inter-code interference generated during transmission and obtain the best receiving effect. In practical applications, the shaping filter and the matched filter usually select the root mean square raised cosine filter, but the filter is not theoretically infinite in the time domain, that is, the signal still has inter-code interference at the sampling point. Since it is difficult to achieve complete synchronization between the transmission of the interference signal and the communication signal in time, it is also necessary to consider the time shift problem of the interference signal. For the measurement of the bit error rate of the communication signal, the usual processing is to superimpose the interference signal and the communication signal at the signal receiving end, perform down-conversion, low-pass filtering, matched filtering, and finally sample and judge the number of error code elements to obtain the bit error rate. The measurement of this bit error rate requires the actual transmission and reception of the signal, and the received signal data will be processed, and there is a problem of large data volume.

[0004] In practical applications, the important parameters of the shaping filter are the roll-off coefficient and the symbol cutoff length. Different parameters determine the size of the inter-symbol crosstalk value, and different signals have different distributions at the sampling point due to different generation methods. Summary of the invention

[0005] The purpose of the present invention is to address the above-mentioned problems and to provide a method for directly estimating the signal bit error rate based on a shaping filter. The method utilizes the characteristics of the shaping filter, the differences in different digital signal generation methods, and the time difference between the transmission of the interference signal and the communication signal. By calculating the probability distribution of the amplitude of the interference signal at the sampling point, the method completes the estimation of the communication signal bit error rate.

[0006] The technical solution of the present invention is:

[0007] A method for generating a signal based on a shaping filter and directly estimating the signal bit error rate, defining the communication signal as S c (t), the interference signal is S p (t), the code rate of the communication signal is R S1 , the code rate of the interference signal is R S2 , the sampling rate is FS , oversampling is The method comprises:

[0008] S1. Establish the signal sampling decision criteria:

[0009] For the signal generated by the shaping filter, the complex baseband signal received by the receiver is expressed as:

[0010]

[0011] Among them, a i , b j , T a , T b are the code element value and symbol interval of the communication signal and interference signal respectively, t 0 is the random time delay within a symbol, n(t) is the noise, g T (t) is the RMS raised cosine filter impulse response, and its Fourier transform is:

[0012]

[0013] Among them, α is the roll-off factor of the filter, and f is the frequency;

[0014] Assuming that the receiver has achieved clock synchronization, the value of y(t) at the optimal sampling point after matched filtering is:

[0015]

[0016] Among them, N is the filter cutoff length, n(τ) is the noise. Since the inter-code interference value of the communication signal has little effect on its sampling, the above formula is simplified to:

[0017]

[0018] Set the decision rule to:

[0019]

[0020] c is the decision threshold, a is the amplitude of the communication signal at the sampling point, b is the amplitude of the interference signal at the sampling point,

[0021] S2. Establish the probability distribution of the interference signal at the sampling point:

[0022] For a sampling at a certain time, the inter-symbol crosstalk of the shaping filter is defined as r = [r -N ,r -N+1 ,…,r -1 ,r 1 ,…,r N-1 ,r N ], N is the code element truncation length, ri is the interference value of the i-th code element at the current sampling point, i = 1, 2, ... N, and the sampling point is the oversampling multiple;

[0023] The code element values ​​of the interference signal are defined as M, and the probability of each value appearing is p i ,i=1,2,…M;

[0024] Establish the crosstalk amplitude matrix R, the specific process is as follows:

[0025] First, we obtain the possible permutation and combination matrix A of the interference signal code element values. There are M permutations and combinations in total. 2N Possibly, set the value from 0 to M 2N -1 is converted to M-base, and M is obtained. 2N ×(2N)-dimensional matrix;

[0026] Replace the M-ary numbers in A with the code element values ​​of the interference signal to obtain the crosstalk amplitude matrix R;

[0027] Establish the weight probability matrix G:

[0028] G=P×R

[0029] Where R is (2N) M ×(2N)-dimensional matrix, each row represents the possible permutations and combinations of inter-code crosstalk r, P = [p 1 ,p 2 ,…,p M ] is the probability weight vector;

[0030] At this time, the interference signal value at the sampling point is b 0 is the code value of the interference signal, r will be delayed by t 0 changes with the changes in

[0031] After merging the same items in B, the amplitude probability distribution F of the interference signal at the sampling point is obtained;

[0032] S3. Calculate the communication bit error rate:

[0033] It can be seen from S1 that when the communication signal is determined, the bit error rate is only related to the amplitude distribution of the interference signal at the sampling point. From the perspective of information theory and communication system coding, when the code element is designed, each code element has an equal probability of appearing. Then, the probability of an error in the received signal is defined as:

[0034]

[0035] Among them, p 1 =P(b <c-x),p 2 =P(b>-cx).

[0036] Using the interference signal amplitude probability distribution F, the communication bit error rate can be calculated as:

[0037] P e =F(b<c-x)+F(b> -cx).

[0038] The beneficial effects of the present invention are as follows: the present invention takes directly obtaining the bit error rate of the interference signal on the communication signal as the starting point, first establishes the decision criteria at the signal sampling point, proposes that the size of the communication signal bit error rate depends on the probability distribution of the amplitude of the interference signal at the sampling point during the sampling decision, and then uses the characteristic parameters of the filter and the delay size to calculate the probability distribution of the amplitude of the interference signal at the sampling point, and then directly calculates the bit error rate of the communication signal through its amplitude probability distribution. The present invention can predict the impact of the interference signal on the communication signal bit error rate in advance, saving the step of signal data processing at the receiving end. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 is a flow chart of the present invention;

[0040] Figure 2 Amplitude distribution for BPSK sampling;

[0041] Figure 3 Simulate the amplitude distribution at the BPSK sampling location;

[0042] Figure 4 This is a comparison chart of the theoretical and simulated bit error rates of BPSK with time shift;

[0043] Figure 5 Amplitude distribution for QPSK sampling processing;

[0044] Figure 6 Simulate the amplitude distribution at the QPSK sampling location;

[0045] Figure 7 This is a comparison chart of the bit error rate between the theory and simulation of QPSK with time shift;

[0046] Figure 8 Amplitude distribution for 16-QAM sampling;

[0047] Fig. 9 Simulate the amplitude distribution at the 16-QAM sampling point;

[0048] Fig.10 This is a comparison chart of the theoretical and simulated bit error rates for 16-QAM with time shift. DETAILED DESCRIPTION

[0049] The technical solution of the present invention will be further described below in conjunction with the accompanying drawings and simulations.

[0050] This example will simulate and verify the proposed method. There is a time delay between the transmission time of the interference signal and the communication signal, and the time shift length is set within one symbol time.

[0051] Take the communication signal as BPSK signal, and the interference signals as BPSK, QPSK, and 16-QAM as examples. The simulation settings are: the shaping filter is a root raised cosine filter, the code element truncation length N = 4, the corollary roll-off coefficient α = 0.35, and the oversampling is 12; the time shift between the interference signal and the communication signal is set to be within one code element. When the oversampling is 12, the time shift is set to n / 12 code element lengths, n = 0, 1, ..., 11, and the time shift size has equal probability, all 1 / 12. In each time shift, the sampling value shifts to the right (left) by 1 / 12 of the code element length, and the crosstalk value also shifts to the right (left) accordingly. When n ≥ 2, the second code element (truncation value) to the right (left) side of the code element at the sampling point has no crosstalk effect on the sampling point. In the simulation, the communication signal and the protection signal are kept completely overlapped in the frequency domain, and the number of Monte Carlo times is 240. Among them, SIR is defined as:

[0052]

[0053] Among them, P T is the energy of the communication signal, P D is the energy of the interference signal.

[0054] The simulation results are shown in the figure:

[0055] Figure 1 , Figure 2 It is divided into the amplitude distribution of BPSK signal sampling processing theory and the amplitude distribution at the sampling point obtained by simulation; Figure 4 , Figure 5 It is divided into the amplitude distribution of QPSK signal sampling processing theory and the amplitude distribution at the sampling point obtained by simulation; Figure 7 , Figure 8 It is divided into the amplitude distribution of the 16-QAM signal sampling processing theory and the amplitude distribution at the sampling point obtained by simulation.

[0056] Figure 3 , Figure 6 and Fig. 9 The figure shows a comparison of the bit error rate calculated from the theoretical amplitude distribution at the sampling point when the interference signals are BPSK, QPSK and 16-QAM signals, and the bit error rate obtained by simulation statistics after superposition of the actual signals. The simulation results show that for the digital signal generated by the root raised cosine filter, the amplitude distribution of the signal at the sampling point can be obtained according to the filter parameters and the code element characteristics of the interference signal, and then the bit error of the communication signal can be directly obtained.

Claims

1. Based on the method of generating signals by shaping filters and directly estimating the signal bit error rate, the communication signal is defined as S c (t), the interference signal is S p (t), the code rate of the communication signal is R S1 , the code rate of the interference signal is R S2 , the sampling rate is F S , oversampling is It is characterized in that The method comprises: S1. Establish the signal sampling decision criteria: For the signal generated by the shaping filter, the complex baseband signal received by the receiver is expressed as: Among them, a i , b j , T a , T b are the symbol value and symbol interval of the communication signal and interference signal respectively, t0 is the random time delay within a symbol, n(t) is the noise, g T (t) is the RMS raised cosine filter impulse response, and its Fourier transform is: Among them, α is the roll-off factor of the filter, and f is the frequency; Assuming that the receiver has achieved clock synchronization, the value of y(t) at the optimal sampling point after matched filtering is: Among them, N is the filter cutoff length, n(τ) is the noise. Since the inter-code interference value of the communication signal has little effect on its sampling, the above formula is simplified to: Set the decision rule to: c is the decision threshold, a is the amplitude of the communication signal at the sampling point, b is the amplitude of the interference signal at the sampling point, S2. Establish the probability distribution of the interference signal at the sampling point: For a sampling at a certain time, the inter-symbol crosstalk of the shaping filter is defined as r = [r -N ,r -N+1 ,…,r -1 ,r1,…,r N-1 ,r N ], N is the code element truncation length, r i is the interference value of the i-th code element at the current sampling point, i = 1, 2, ... N, and the sampling point is the oversampling multiple; The code element values ​​of the interference signal are defined as M, and the probability of each value appearing is p i ,i=1,2,…M; Establish the crosstalk amplitude matrix R, the specific process is as follows: First, we obtain the possible permutation and combination matrix A of the interference signal code element values. There are M permutations and combinations in total. 2N Possibly, set the value from 0 to M 2N -1 is converted to M-base, and M is obtained. 2N ×(2N)-dimensional matrix; Replace the M-ary numbers in A with the code element values ​​of the interference signal to obtain the crosstalk amplitude matrix R; Establish the weight probability matrix G: G=P×R Where R is (2N) M ×(2N)-dimensional matrix, each row represents the possible permutations and combinations of inter-symbol crosstalk r, P = [p1, p2, ..., p M ] is the probability weight vector; At this time, the interference signal value at the sampling point is b0 is the code value of the interference signal, and r will change with the delay t0; After merging the same items in B, the amplitude probability distribution F of the interference signal at the sampling point is obtained; S3. Calculate the communication bit error rate: The probability of receiving a signal error is defined as: Where p1=P(b<c-x),p2=P(b> -cx); Using the interference signal amplitude probability distribution F, the communication bit error rate is obtained as follows: P e =F(b<c-x)+F(b>-c-x)。

Images