Interference efficiency evaluation method based on automatic modulation identification

Abstract

The invention provides an interference efficiency evaluation method based on automatic modulation identification, and relates to the field of wireless communication security. The method provided by the invention comprises the following steps: defining a cognitive interference system and a communication process, and initializing system parameters; for an interference period, identifying a modulation pattern of each frame of communication signal to obtain a distribution vector; calculating scores of different interference periods according to the evaluation indexes; and obtaining an optimal interference decision according to the scores of different interference periods. Aiming at the challenge that it is very difficult to evaluate the interference decision without direct information feedback, the multi-antenna cognitive jammer analyzes the behavior change of a communication system based on an automatic modulation recognition technology under the condition of not depending on channel information, so that the interference effect is deduced; and support is provided for beam forming optimization and interference waveform optimization of the multi-antenna jammer, so that the interference efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of wireless communication security technology, and in particular to an interference effectiveness evaluation method based on automatic modulation identification. Background Art

[0002] In cognitive electronic warfare or cognitive jamming systems, cognitive jammers usually adopt some jamming strategies to disrupt the enemy's normal communications as much as possible and maximize their own jamming utility. In addition, in order to maximize the jamming utility, cognitive jammers need to estimate the quality of jamming decisions through the effectiveness evaluation process, so as to select the optimal jamming decision in the current environment. However, in non-cooperative jamming scenarios, for cognitive jammers, due to the lack of direct information feedback from the communicating party, jamming channel information and communication state parameter information are difficult to obtain.

[0003] Most of the current work on interference effectiveness evaluation focuses on the following two aspects: First, from the perspective of the interfered party, evaluate the interference situation of the communication system and provide support for the optimization of communication system parameters. Current research includes: for the evaluation of pulse interference effect in communication electronic warfare, the non-stationary characteristics of the interference signal in the time and frequency domain are used, and the Wigner-Will distribution tool is used to analyze the difference in the time and frequency characteristics of the target signal and the interfered signal, and the average effective correlation power density index is used to evaluate the interference effect of pulse interference on the communication system; on the basis of building a radio network simulation platform, the theoretical methods such as hierarchical analysis method, rough set and cloud model are comprehensively used to evaluate the effectiveness of communication interference; the evaluation of the interference effect of single-tone interference on FM signals is studied, and the corresponding relationship between the proposed interference evaluation index and the interference-to-signal ratio is obtained through simulation analysis, so as to evaluate the single-tone interference effect of FM signals; in order to detect interference attacks, Sharaf Malebary studies the interference impact evaluation problem based on the behavior and mobility mode of the jammer in the Vehicular Ad-Hoc Network (VANET) under the IEEE802.11p protocol, and proposes a two-stage detection algorithm to accurately detect the type of jammer to provide early warning. The second is to evaluate the anti-interference performance of the communication party or the interference effectiveness of the interference party from a global perspective, usually relying on a third party that has global information, in order to verify the performance of the anti-interference scheme or the interference scheme in different scenarios. The current research includes: for the objective evaluation of the effect of communication voice interference, two evaluation methods based on multi-measurement and multi-modal fusion are proposed; a jammer training effect evaluation index system is established, and the relative weight of the evaluation index is determined by the entropy weight method. At the same time, in order to eliminate the uncertainty and ambiguity in the confrontation environment, the evaluation method based on the cloud model is used to realize the evaluation of the jammer communication confrontation training effect; the main influencing factors of satellite interference are analyzed and selected, such as interference frequency, interference power, bit error rate influencing factor, interference style, etc., and a satellite interference effect evaluation method based on fuzzy comprehensive evaluation is proposed using fuzzy mathematical theory tools; in order to improve the level of communication confrontation training, a third-party interference effect evaluation method based on conditional cloud model and TOPSIS method is proposed. The proposed method can incorporate qualitative and quantitative indicators into the overall evaluation system, and use the combined weighting method to reasonably configure the indicator weight; using the bit error rate as the evaluation indicator, a real-time evaluation tool for communication network confrontation effectiveness is initially developed.

[0004] However, in non-cooperative or confrontational scenarios, the jammer cannot directly obtain information such as the jamming channel and communication status. Existing evaluation methods from the perspective of the victim and the third party may be difficult to apply, making it difficult for the jammer to evaluate the jamming effectiveness and thus difficult to make cognitive jamming decisions. Therefore, it is necessary to study efficient and robust effectiveness evaluation methods for different communication systems from the perspective of the jammer. In addition, due to the non-cooperative nature between the communicating party and the cognitive jammer, the jamming effectiveness evaluation initiated by the jammer still faces the following challenges: incomplete information, changeable behaviors of the communicating party, dynamic and changeable environment, multi-source signal interweaving, and difficulty in verifying the evaluation results. Summary of the invention

[0005] The present application provides an interference effectiveness evaluation method based on automatic modulation identification, which can be used to solve current technical problems.

[0006] The present application provides an interference effectiveness evaluation method based on automatic modulation identification, the method comprising:

[0007] Step 1: Define the cognitive interference system and initialize the interference system parameters;

[0008] Step 2: Define the communication behavior model and initialize the communication behavior parameters;

[0009] Step 3: Define the cognitive evaluation model and initialize the evaluation parameters;

[0010] Step 4: For the interference period m = 1, 2, ..., M, identify the modulation pattern of each frame of the communication signal and obtain the distribution vector Ω a,m and the evaluation index Γ m ;

[0011] Step 5: Get the optimal interference decision based on the scores of different interference cycles

[0012] Furthermore, a cognitive interference system is defined and interference system parameters are initialized, including:

[0013] The cognitive jamming system includes a communication pair and a cognitive jammer;

[0014] In each transmission frame, the transmitter (Alice) of the communication pair performs adaptive transmission by adjusting the signal modulation mode according to the channel quality, and the receiver (Bob) feeds back the channel quality information, i.e., signal-to-interference-noise ratio, to the sender after each transmission frame. The cognitive jammer (Cj) aims to achieve maximum jamming efficiency at the minimum power cost. The decision of the communication pair is the modulation mode X. a ; The interference decision of the cognitive jammer is the precoding vector W j and modulation mode X j;Through perception and recognition, the cognitive jammer obtains the transmission power and modulation mode of the communication party; because the feedback information of the communication receiver is difficult to intercept, the cognitive jammer does not have prior information about the communication receiver;

[0015] The communication frame k includes T transmission time slots; the first T′ time slots are used for random bit data transmission, and the remaining time slots are used for the receiver Bob to feedback the channel state information; in different transmission time slots of the same frame, the transmitter Alice does not change the modulation mode, but determines the modulation mode of the next frame transmission time slot according to the feedback information at the end of the frame; the interference frame includes two stages: interference and evaluation: in the interference stage, the cognitive jammer implements the interference attack according to the interference strategy; in the evaluation stage, the cognitive jammer evaluates the interference effectiveness according to the changes in the state parameters of the communication pair, providing the necessary basis for the subsequent optimization of the interference strategy;

[0016] The received signal of the communication receiver in the tth frame is expressed as:

[0017]

[0018] Among them, x k,t The transmission power is P a The communication party QAM signal, the real part and the imaginary part of the QAM signal are selected from the set The set is represented as:

[0019]

[0020] Where M represents the modulation order of the communication signal, d a represents the communication symbol distance; y k,t The power is P j The cognitive jammer transmits a signal, the real and imaginary parts of which are selected from the set The set is represented as:

[0021]

[0022] Where U represents the modulation order of the interference signal, d j represents the interference symbol distance; is Additive White Gaussian Noise (AWGN), ρ, ρ∈[0,1] is the pulse duration ratio; α is the path fading coefficient;

[0023] For a cognitive jammer deploying an L-element uniformly linear array (ULA), the beamforming vector is:

[0024]

[0025] in, The small-scale channel fading coefficient between the transmitter and the receiver in the communication pair is h a ; The small-scale channel fading coefficient vector between the cognitive jammer and the communication receiver is expressed as:

[0026]

[0027] Among them, h j is the complex channel fading coefficient, Δl is the array antenna distance, λ is the signal wavelength, θ j represents the incident angle of the interference signal; D a and D j They represent the propagation distances between the transmitter and the receiver and between the cognitive jammer and the receiver, respectively, as follows:

[0028] D a =||q a -q b ||,D j =||q a -q j ||. (6)

[0029] Among them, the positions of the transmitter Alice and the receiver Bob are q a =(x a ,y a ) and q b =(x b ,y b ); In addition, the position of the cognitive jammer is denoted as q j =(x j ,y j ); the interference signal and the communication signal are coherent; the communication symbol distance and the interference symbol distance are expressed as follows:

[0030]

[0031] Where E a and E j Represent the bit energy of the communication symbol and the interference symbol respectively; M and U represent the modulation order of the communication signal and the interference signal respectively; the instantaneous signal-to-jamming-plus-noise ratio (SJNR) of the communication signal is:

[0032]

[0033] in, and

[0034] Assuming that communication symbols and interference symbols appear with equal probability during transmission, and the signal sampling period is equal to the symbol period, the average signal-to-interference-noise ratio of the kth frame is expressed as:

[0035]

[0036] The coding rate of the communication signal and the interference signal is set to 1;

[0037] Considering the quasi-static channel model, the channel fading coefficient is constant in the same transmission frame, but its value in each different frame is a random variable; during the transmission of different frames, the channel fading coefficient satisfies the independent and identically distributed (iid) condition; in addition, the complex channel fading coefficient is a circularly symmetric complex Gaussian variable with a mean of 0 and a variance of 1, that is,

[0038] Furthermore, a communication behavior model is defined, and communication behavior parameters are initialized, including: adaptive modulation technology is widely used in communication systems, such as LTE and 5G systems, to help communication links maximize throughput under bit error rate constraints in a dynamic environment; the communication pair adopts the following adaptive modulation process: according to the channel state information and adaptive criteria fed back by the receiver Bob, such as the maximum throughput criterion or the target bit error rate (BER) / block error rate (BLER) criterion, the switching threshold and the corresponding modulation scheme under the channel quality indicator (CQI) are set; let the set of modulation schemes that can be selected (the order is arranged in ascending order) be The modulation scheme switching process is as follows:

[0039]

[0040] Among them, ν f ,f∈[0,F] is the switching threshold with the following characteristics:

[0041]

[0042] When the communicating parties switch the modulation mode, the average symbol energy remains unchanged and is set to 1.

[0043] Furthermore, the cognitive evaluation model is defined and the evaluation parameters are initialized, including:

[0044] When implementing interference, the cognitive jammer cannot estimate the interference channel because it cannot obtain prior information about Bob, such as location. Therefore, in order to maximize the effectiveness of interference, the cognitive jammer can only make the optimal interference decision through interference effectiveness evaluation; in addition, the cognitive jammer will dynamically adjust the interference decision of each communication frame based on the interference strategy, but for the same communication frame, the interference decision remains unchanged;

[0045] Taking into account the degradation of communication performance and the interference power cost, the utility function of the cognitive jammer in the kth frame is defined as:

[0046]

[0047] in, C is a constant coefficient, SJNR k represents the average signal to noise ratio of the communication signal in the kth frame; for simplicity, the beamforming vector of the cognitive jammer does not consider the optimization of the modulus value, that is, it satisfies And the average interference signal energy is set to 1;

[0048] The cognitive jammer uses modulation recognition technology to obtain the modulation pattern of each frame of the communication pair; the recognition confusion matrix is ​​defined as follows:

[0049]

[0050] Among them, g m,n This means that the actual modulation pattern used in the communication is m, while the modulation pattern identified by the cognitive jammer is n; Represents the set of modulation methods that can be selected for communication signals;

[0051] For a fixed interference decision (X j ,W j ), when the number of observation frames is large enough, the actual signal-to-interference-noise ratio distribution of the communication pair is B β ~p(β), then the mean value μ of the signal-to-interference-to-noise ratio is expressed as follows:

[0052]

[0053] Introducing the switching threshold ν f ,f∈[0,F], the mean μ is rewritten as:

[0054]

[0055] in, is the probability of the modulation pattern f appearing, It is a constant when the signal-to-interference-noise ratio distribution and the switching threshold are known;

[0056] Under ideal conditions, the actual signal-to-interference-to-noise ratio distribution or variation of a known communication pair Then the cognitive jammer can easily evaluate the mean signal-to-interference-to-noise ratio of the communication pair. However, in actual scenarios, due to the non-cooperation and limited perception time, the above parameters are difficult to obtain accurately. Therefore, it is possible to consider using an approximate method to obtain the variable Estimated value of To obtain the final performance evaluation results.

[0057] Through analysis, we can get The following conditions are met:

[0058]

[0059] Among them, β represents the value of the random variable, p(·) represents the probability value, and ν f ,f∈[0,F] is the switching threshold;

[0060] Therefore, for a certain interference decision, the coefficient The range is:

[0061]

[0062] Assume that the interference decision set to be evaluated is The qth interference decision S j,q =(X j,q ,W j,q ); From formula (16), when evaluating the communication signal-to-interference-to-noise ratio mean μ under the qth interference decision, the variable ε q,f satisfy:

[0063]

[0064] Among them, the variable ε q,f represents ε under the qth interference decision f value;

[0065] Therefore, in the subsequent fitting parameters When , the constraint of formula (17) is satisfied. However, for different interference decisions q and q′, ε q,f With ε q′,f It is difficult to prove that they are equal in theory, which gives the parameter ε q,f The precise fitting of brings great difficulties.

[0066] Furthermore, for interference periods m=1, 2, ..., M, the modulation pattern of each frame of communication signal is identified, and the distribution vector Ω is obtained. a,m and the evaluation index Γ m ;include:

[0067] Based on the above analysis, the interference effectiveness estimation process of different decisions is described as follows: First, according to the interference decision set S jThe cognitive jammer implements different jamming according to the order of elements. Each time it jams T communication frames, it is a jamming cycle. The total number of jamming cycles is M.

[0068] Secondly, after the mth interference cycle ends, the cognitive jammer obtains the modulation distribution vector Ω through modulation recognition and data statistics. a,m ={p m,f},f=1,2,…,F, where p m,f is the frequency of the modulation pattern f in the interference period m; then, according to the evaluation index Γ m Find the scores for different decisions:

[0069]

[0070] Furthermore, according to the scores of different interference periods, the optimal interference decision is obtained. include:

[0071] According to the scores of different interference cycles, the optimal interference decision is:

[0072]

[0073] The beneficial effects of the present invention are:

[0074] 1. The present invention proposes a method for evaluating the interference effect of an interferer based on automatic modulation identification. In the proposed evaluation method, the cognitive jammer implements different beamforming and modulation waveform decisions to obtain the relationship between different interference decisions and the modulation pattern of the communication signal, thereby deriving the estimated value of the signal-to-interference-noise ratio at the communication receiver, and realizing the evaluation of the interference effect under incomplete information. Compared with the decision without evaluation, the proposed evaluation method provides support for cognitive interference decision-making and improves the interference effectiveness.

[0075] 2. The present invention models the non-cooperative interference effect evaluation problem as an observable parameter statistical analysis problem. By performing statistical analysis on the communication signal modulation style, it solves the non-cooperative interference effectiveness evaluation problem under incomplete information and dynamic unknown channel conditions. The communication service quality index with strong applicability is used as the evaluation index, so that the cognitive interference function can efficiently and robustly enable its own interference decision-making, which has great application potential in actual physical layer security and cognitive confrontation. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 It is a model diagram of the interference system in the present invention.

[0077] Figure 2 It is a model diagram of the working protocol of the cognitive jammer and the communication pair in the present invention.

[0078] Figure 3It is the relationship between the Spearman rank correlation coefficient between the interference effect evaluation method of the present invention and the optimal evaluation method and the observation frame length.

[0079] Figure 4 It is the variation of interference decision effectiveness enabled by the interference effect evaluation method of the present invention and other evaluation methods with the interference period.

[0080] Figure 5 The Spearman rank correlation coefficient between the interference effect evaluation method of the present invention and the optimal evaluation method varies with the modulation recognition error under different observation frame length conditions.

[0081] Figure 6 It is the variation of the Spearman rank correlation coefficient between the interference effect evaluation method of the present invention and the optimal evaluation method with the observed frame length in different strategy spaces. DETAILED DESCRIPTION

[0082] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0083] The following first introduces the embodiments of the present application in conjunction with the accompanying drawings.

[0084] The multi-antenna cognitive jammer analyzes the behavior changes of the communication system based on automatic modulation recognition technology without relying on channel information to infer the interference effect, providing support for the beamforming optimization and interference waveform optimization of the multi-antenna jammer, thereby improving the interference effectiveness.

[0085] like Figure 1 As shown in Figure 1, the cognitive jammer system consists of a communication pair and a cognitive jammer. In each transmission frame, the transmitter (Alice) of the communication pair performs adaptive transmission by adjusting the signal modulation mode according to the channel quality, and the receiver (Bob) feeds back the channel quality information, i.e., the signal-to-interference-noise ratio, to the sender after each transmission frame. The cognitive jammer (Cj) aims to achieve maximum jamming efficiency at the minimum power cost. The decision of the communication pair is the modulation mode X. a ; The interference decision of the cognitive jammer is the precoding vector W j and modulation mode X j ; Through perception and recognition, the cognitive jammer obtains the transmission power and modulation mode of the communication party; since the feedback information of the communication receiver is difficult to intercept, the cognitive jammer does not have prior information about the communication receiver.

[0086] like Figure 2As shown, the communication frame k includes T transmission time slots; the first T′ time slots are used for random bit data transmission, and the remaining time slots are used for the receiver Bob to feedback the channel state information; in different transmission time slots of the same frame, the transmitter Alice does not change the modulation mode, but determines the modulation mode of the next frame transmission time slot according to the feedback information at the end of the frame; the interference frame includes two stages: interference and evaluation: in the interference stage, the cognitive jammer implements the interference attack according to the interference strategy; in the evaluation stage, the cognitive jammer evaluates the interference effectiveness according to the changes in the state parameters of the communication pair, providing the necessary basis for the subsequent optimization of the interference strategy.

[0087] The received signal of the communication receiver in the tth frame is expressed as:

[0088]

[0089] Among them, x k,t The transmission power is P a The communication party QAM signal, the real part and the imaginary part of the QAM signal are selected from the set The set is represented as:

[0090]

[0091] Where M represents the modulation order of the communication signal, d a represents the communication symbol distance; y k,t The power is P j The cognitive jammer transmits a signal, the real and imaginary parts of which are selected from the set The set is represented as:

[0092]

[0093] Where U represents the modulation order of the interference signal, d j represents the interference symbol distance; is Additive White Gaussian Noise (AWGN), ρ, ρ∈[0,1] is the pulse duration ratio; α is the path fading coefficient;

[0094] For a cognitive jammer deploying an L-element uniformly linear array (ULA), the beamforming vector is:

[0095]

[0096] in, The small-scale channel fading coefficient between the transmitter and the receiver in the communication pair is h a ; The small-scale channel fading coefficient vector between the cognitive jammer and the communication receiver is expressed as:

[0097]

[0098] Among them, h j is the complex channel fading coefficient, Δl is the array antenna distance, λ is the signal wavelength, θ j represents the incident angle of the interference signal; D a and D j They represent the propagation distances between the transmitter and the receiver and between the cognitive jammer and the receiver, respectively, as follows:

[0099] D a =||q a -q b ||,D j =||q a -q j ||. (6)

[0100] Among them, the positions of the transmitter Alice and the receiver Bob are q a =(x a ,y a ) and q b =(x b ,y b ); In addition, the position of the cognitive jammer is denoted as q j =(x j ,y j ); the interference signal and the communication signal are coherent; the communication symbol distance and the interference symbol distance are expressed as follows:

[0101]

[0102] Where E a and E j Represent the bit energy of the communication symbol and the interference symbol respectively; M and U represent the modulation order of the communication signal and the interference signal respectively; the instantaneous signal-to-jamming-plus-noise ratio (SJNR) of the communication signal is:

[0103]

[0104] in, and

[0105] Assuming that communication symbols and interference symbols appear with equal probability during transmission, and the signal sampling period is equal to the symbol period, the average signal-to-interference-noise ratio of the kth frame is expressed as:

[0106]

[0107] The coding rate of the communication signal and the interference signal is set to 1;

[0108] Considering the quasi-static channel model, the channel fading coefficient is constant in the same transmission frame, but its value in each different frame is a random variable; during the transmission of different frames, the channel fading coefficient satisfies the independent and identically distributed (iid) condition; in addition, the complex channel fading coefficient is a circularly symmetric complex Gaussian variable with a mean of 0 and a variance of 1, that is,

[0109] Adaptive modulation technology is widely used in communication systems, such as LTE and 5G systems, to help communication links maximize throughput under bit error rate constraints in dynamic environments. The communication pair adopts the following adaptive modulation process: according to the channel state information and adaptive criteria fed back by the receiver Bob, such as the maximum throughput criterion or the target bit error rate (BER) / block error rate (BLER) criterion, set the switching threshold and the corresponding modulation scheme under the channel quality indicator (CQI); let the set of modulation schemes that can be selected (arranged in ascending order) be The modulation scheme switching process is as follows:

[0110] X a =X f ,if SJNR aver,k ∈(ν f-1 ,ν f ), (10)

[0111] Among them, ν f ,f∈[0,F] is the switching threshold with the following characteristics:

[0112]

[0113] When the communicating parties switch the modulation mode, the average symbol energy remains unchanged and is set to 1.

[0114] When implementing interference, the cognitive jammer cannot estimate the interference channel because it cannot obtain prior information about Bob, such as his location. Therefore, in order to maximize the effectiveness of interference, the cognitive jammer can only make the optimal interference decision through interference effectiveness evaluation; in addition, the cognitive jammer will dynamically adjust the interference decision of each communication frame based on the interference strategy, but for the same communication frame, the interference decision remains unchanged.

[0115] Taking into account the degradation of communication performance and the interference power cost, the utility function of the cognitive jammer in the kth frame is defined as:

[0116]

[0117] in, C is a constant coefficient, SJNR k represents the average signal to noise ratio of the communication signal in the kth frame; for simplicity, the beamforming vector of the cognitive jammer does not consider the optimization of the modulus value, that is, it satisfies And the average interference signal energy is set to 1;

[0118] The cognitive jammer uses modulation recognition technology to obtain the modulation pattern of each frame of the communication pair; the recognition confusion matrix is ​​defined as follows:

[0119]

[0120] Among them, g m,n This means that the actual modulation pattern used in the communication is m, while the modulation pattern identified by the cognitive jammer is n; Represents the set of modulation methods that can be selected for communication signals;

[0121] For a fixed interference decision (X j ,W j ), when the number of observation frames is large enough, the actual signal-to-interference-noise ratio distribution of the communication pair is B β ~p(β), then the mean value μ of the signal-to-interference-to-noise ratio is expressed as follows:

[0122]

[0123] Introducing the switching threshold ν f ,f∈[0,F], the mean μ is rewritten as:

[0124]

[0125] in, is the probability of the modulation pattern f appearing, It is a constant when the signal to interference noise ratio distribution and the switching threshold are known.

[0126] Under ideal conditions, the actual signal-to-interference-to-noise ratio distribution or variation of a known communication pair Then the cognitive jammer can easily evaluate the mean signal-to-interference-to-noise ratio of the communication pair. However, in actual scenarios, due to the non-cooperation and limited perception time, the above parameters are difficult to obtain accurately. Therefore, it is possible to consider using an approximate method to obtain the variable Estimated value of To obtain the final performance evaluation results.

[0127] Through analysis, we can get The following conditions are met:

[0128]

[0129] Among them, β represents the value of the random variable, p(·) represents the probability value, and ν f ,f∈[0,F] is the switching threshold.

[0130] Therefore, for a certain interference decision, the coefficient ε f , The range is:

[0131]

[0132] Assume that the interference decision set to be evaluated is The qth interference decision S j,q =(X j,q ,W j,q ); From formula (16), when evaluating the communication signal-to-interference-to-noise ratio mean μ under the qth interference decision, the variable ε q,f satisfy:

[0133]

[0134] Among them, the variable ε q,f represents ε under the qth interference decision f value.

[0135] Therefore, in the subsequent fitting parameters When , the constraint of formula (17) needs to be satisfied; however, for different interference decisions q and q′, ε q,f With ε q′,f It is difficult to prove that they are equal in theory, which gives the parameter ε q,f The precise fitting of brings great difficulties.

[0136] Based on the above analysis, the interference effectiveness estimation process of different decisions is described as follows: First, according to the interference decision set The cognitive jammer implements different jamming operations, and each time it jams T communication frames, that is, one jamming cycle; the total number of jamming cycles is M.

[0137] Secondly, after the mth interference cycle ends, the cognitive jammer obtains the modulation distribution vector Ω through modulation recognition and data statistics. a,m ={p m,f},f=1,2,…,F, where p m,f is the frequency of the modulation pattern f in the interference period m; then, according to the evaluation index Γ m Find the scores for different decisions:

[0138]

[0139] According to the scores of different interference cycles, the optimal interference decision is:

[0140]

[0141] The present invention will be further described below in conjunction with specific embodiments.

[0142] The effectiveness of the present invention is verified by simulation examples. First, the scenario setting of the embodiment is briefly introduced. Consider a cognitive jammer system consisting of a communication transmitter, a communication receiver and a cognitive jammer, in which three nodes are randomly distributed in a 400m×400m plane. The modulation style set of the communication pair and the cognitive jammer is {QPSK, 16-QAM, 64-QAM, 256-QAM}, the average symbol energy is 1J, and the maximum interference power and transmission power are 2W and 400mW respectively. The number of antenna elements of the cognitive jammer is L=4, and the ratio of the antenna element spacing to the signal wavelength is Δl / λ=0.5. The constant coefficient C=0.01, R=1000. The number of symbols transmitted in each communication frame is 10000. The pulse duration ratio of the interference and communication signals is ρ=1. The noise spectrum density is -100dBm. The path loss factor α=2.

[0143] The accuracy test process of the evaluation algorithm provided in this application is as follows: First, the accurate utility values ​​of different interference decisions are obtained using a third-party evaluation method (i.e., various parameters are known), and they are arranged in ascending order according to the utility size to obtain the following ordered array:

[0144] Rank(1)=(F1,F2,…,F M ) (20)

[0145] Replace each element in the array with the utility value of the corresponding decision evaluated by the proposed evaluation algorithm, and we can get the following array

[0146]

[0147] Then, the Matlab simulation platform is run to calculate the Spearman rank correlation coefficient to compare the correlation between the arrays Rank(1) and Rank(2), thereby testing the accuracy of the proposed evaluation algorithm.

[0148] Spearman rank correlation coefficient γ s The introduction is as follows:

[0149]

[0150] Among them, d i It represents the difference in the rank values ​​of the i-th data pair, and n is the sample size. The range of the Spearman rank correlation coefficient is [-1,1]. The closer it is to 1, the more correlated the variables are.

[0151] The gain effect of the proposed evaluation algorithm on cognitive interference decision-making is tested as follows. First, define different interference decision-making methods that need to be run on the Matlab simulation platform:

[0152] ① The cognitive interference decision method enabled by the proposed evaluation scheme: Each interference decision S j,q The score is initialized to Γ q (0),

[0153] make B is a constant. Then, in the first m′ jammer cycles, the cognitive jammer makes random decisions and obtains score estimates for different decisions:

[0154]

[0155] Among them, q Represents the cycle number of the interference decision q in the first m′ cycles. After learning the pros and cons of different interference decisions to a certain extent, the cognitive jammer adopts ε-greedy decision, and the decision in the mth (m>m′) cycle can be expressed as follows

[0156]

[0157] in Represents a purely random decision result.

[0158] ②Interference decision method without evaluation enablement: Since it is impossible to obtain the evaluation results of different interference decisions, the jammer performs random interference.

[0159] The relevant comparison results are as follows Figure 3 , Figure 4 , Figure 5 and Figure 6 As shown, it can be found that the effect achieved by the interference effect evaluation method based on automatic modulation recognition proposed in the present invention is almost equivalent to the optimal evaluation. At the same time, compared with the implementation effect of interference decision-making without evaluation empowerment, it can greatly improve the cognitive interference efficiency.

[0160] Figure 3 The relationship between the Spearman rank correlation coefficient between the interference effect evaluation method of the present invention and the optimal evaluation method and the observed frame length is shown. It can be seen that the algorithm can accurately sort interference decisions on different time scales and provide a basis for interference optimization (the Spearman rank correlation coefficient in the figure ranges from 0.8 to 1).

[0161] Figure 4The results show how the effectiveness of interference decisions enabled by the interference effect evaluation method of the present invention and other evaluation methods change with the interference period. It can be seen that as the number of interference periods increases, the average utility of the interference decision (i.e., the blue curve) without performance evaluation remains almost stable. However, with continuous learning, the average utility of the "ε-greedy" interference decision enabled by the proposed evaluation scheme (i.e., the red curve) continues to increase and reaches stability. This shows that the proposed evaluation scheme can effectively enable cognitive interference decisions.

[0162] Figure 5 It is the variation of the Spearman rank correlation coefficient between the interference effect evaluation method of the present invention and the optimal evaluation method under different observation frame length conditions with modulation recognition error. As can be seen from the figure, as the modulation recognition accuracy of the communication signal decreases, the accuracy of the proposed evaluation scheme will decrease slightly. For example, when the modulation recognition accuracy is 0.95, the Spearman rank correlation coefficient remains at about 0.8. In addition, the comparison between the two curves shows that the observation frame length has the least effect on the accuracy of the proposed evaluation scheme. This shows that the method proposed by the present invention has the ability to improve real-time performance without affecting accuracy.

[0163] Figure 6 is the variation of the Spearman rank correlation coefficient between the interference effect evaluation method of the present invention and the optimal evaluation method with the observed frame length in different strategy spaces, where the definition The figure shows that in different interference strategy spaces, the accuracy of the evaluation scheme proposed by the present invention is relatively high, and the effect of the observed frame length on its performance is small. However, as the interference strategy space increases, the accuracy of the proposed evaluation scheme will decrease to a certain extent, which may have a certain impact on the selection of the optimal interference decision.

[0164] The above-described embodiments of the present application do not constitute a limitation on the protection scope of the present application.

Claims

1. A method for evaluating interference effectiveness based on automatic modulation recognition, characterized in that: The method comprises: Step 1: Define the cognitive interference system and initialize the interference system parameters; Step 2: Define the communication behavior model and initialize the communication behavior parameters; Step 3: Define the cognitive evaluation model and initialize the evaluation parameters; Step 4: For the interference period m = 1, 2, ..., M, identify the modulation pattern of each frame of the communication signal and obtain the distribution vector Ω a,m and the evaluation index Γ m ; Step 5: According to the scores of different interference periods, the optimal interference decision S is obtained j,m* .

2. The method according to claim 1, characterized in that Define the cognitive interference system and initialize the interference system parameters, including: The cognitive jamming system includes a communication pair and a cognitive jammer; In each transmission frame, the transmitter of the communication pair performs adaptive transmission by adjusting the signal modulation mode according to the channel quality, and the receiver feeds back the channel quality information, i.e., signal-to-interference-noise ratio, to the sender after each transmission frame. The cognitive jammer aims to achieve maximum jamming efficiency at the minimum power cost. The decision of the communication pair is the modulation mode X. a ; The interference decision of the cognitive jammer is the precoding vector W j and modulation mode X j ;Through perception and recognition, the cognitive jammer obtains the transmission power and modulation mode of the communication party; the cognitive jammer does not have prior information about the communication receiver; The communication frame k includes T transmission time slots; the first T′ time slots are used for random bit data transmission, and the remaining time slots are used for the receiver to feedback channel state information; in different transmission time slots of the same frame, the transmitter does not change the modulation mode, but determines the modulation mode of the next frame transmission time slot according to the feedback information at the end of the frame; the interference frame includes two stages: interference and evaluation: in the interference stage, the cognitive jammer implements the interference attack according to the interference strategy; in the evaluation stage, the cognitive jammer evaluates the interference effectiveness according to the changes in the state parameters of the communication pair; The received signal of the communication receiver in the tth frame is expressed as: Among them, x k,t The transmission power is P a The communication party QAM signal, the real part and the imaginary part of the QAM signal are selected from the set The set is represented as: Where M represents the modulation order of the communication signal, d a represents the communication symbol distance; y k,t The power is P j The cognitive jammer transmits a signal, the real and imaginary parts of which are selected from the set The set is represented as: Where U represents the modulation order of the interference signal, d j represents the interference symbol distance; is Gaussian white noise, ρ, ρ∈[0,1] is the pulse duration ratio; α is the path fading coefficient; For a cognitive jammer deploying an L-element uniform linear array, the beamforming vector is: in, The small-scale channel fading coefficient between the transmitter and the receiver in the communication pair is h a ; The small-scale channel fading coefficient vector between the cognitive jammer and the communication receiver is expressed as: Among them, h j is the complex channel fading coefficient, Δl is the array antenna distance, λ is the signal wavelength, θ j represents the incident angle of the interference signal; D a and D j They represent the propagation distances between the transmitter and the receiver and between the cognitive jammer and the receiver, respectively, as follows: The positions of the transmitter and receiver are q a =(x a ,y a ) and q b =(x b ,y b ); In addition, the position of the cognitive jammer is denoted as q j =(x j ,y j ); the interference signal and the communication signal are coherent; the communication symbol distance and the interference symbol distance are expressed as follows: Where E a and E j Represent the bit energy of communication symbol and interference symbol respectively; M and U represent the modulation order of communication signal and interference signal respectively; the instantaneous signal-to-interference-noise ratio of communication signal is: in, and Assuming that communication symbols and interference symbols appear with equal probability during transmission, and the signal sampling period is equal to the symbol period, the average signal-to-interference-noise ratio of the kth frame is expressed as: The coding rate of the communication signal and the interference signal is set to 1; Considering the quasi-static channel model, the channel fading coefficient is constant in the same transmission frame, but its value in each different frame is a random variable; during the transmission of different frames, the channel fading coefficient satisfies the independent and identically distributed condition; in addition, the complex channel fading coefficient is a circularly symmetric complex Gaussian variable with a mean of 0 and a variance of 1, that is, 3. The method according to claim 2, characterized in that Define the communication behavior model and initialize the communication behavior parameters, including: the communication pair adopts the following adaptive modulation process: according to the channel state information fed back by the receiver and the adaptive criterion, set the switching threshold and the modulation scheme under the corresponding channel state indication; let the set of modulation schemes that can be selected be The modulation scheme switching process is as follows: X a =X f ,if SJNR aver,k ∈(ν f-1 ,n f ), (10) Among them, ν f ,f∈[0,F] is the switching threshold with the following characteristics: When the communicating parties switch the modulation mode, the average symbol energy remains unchanged and is set to 1.

4. The method according to claim 3, characterized in that Define the cognitive evaluation model and initialize the evaluation parameters, including: The cognitive jammer can only make the optimal jamming decision through jamming effectiveness evaluation; in addition, the cognitive jammer will dynamically adjust the jamming decision of each communication frame based on the jamming strategy, but for the same communication frame, the jamming decision remains unchanged; The utility function of the cognitive jammer at the kth frame is defined as: in, C is a constant coefficient, SJNR k represents the average signal to noise ratio of the communication signal in the kth frame; the beamforming vector of the cognitive jammer does not consider the optimization of the modulus value, that is, it satisfies And the average interference signal energy is set to 1; The cognitive jammer uses modulation recognition technology to obtain the modulation pattern of each frame of the communication pair; the recognition confusion matrix is ​​defined as follows: Among them, g m,n This means that the actual modulation pattern used in the communication is m, while the modulation pattern identified by the cognitive jammer is n; Represents the set of modulation methods that can be selected for communication signals; For a fixed interference decision (X j ,W j ), when the number of observation frames is large enough, the actual signal-to-interference-noise ratio distribution of the communication pair is B β ~p(β), then the mean value of signal to interference and noise ratio μ is expressed as follows: Introducing the switching threshold ν f ,f∈[0,F], the mean μ is rewritten as: in, is the probability of the modulation pattern f appearing, It is a constant when the signal-to-interference-noise ratio distribution and the switching threshold are known; Among them, β represents the value of the random variable, p(·) represents the probability value, and ν f ,f∈[0,F] is the switching threshold; Therefore, for a certain interference decision, the coefficient ε f , The range is: Assume that the interference decision set to be evaluated is The qth interference decision S j,q =(X j,q ,W j,q ); From formula (16), when evaluating the communication signal-to-interference-to-noise ratio mean μ under the qth interference decision, the variable ε q,f satisfy: Among them, the variable ε q,f represents ε under the qth interference decision f value; Therefore, in the subsequent fitting parameters When , the constraint of formula (17) is satisfied.

5. The method according to claim 4, characterized in that For interference period m = 1, 2, ..., M, identify the modulation pattern of each frame of communication signal and obtain the distribution vector Ω a,m and the evaluation index Γ m ; include: The interference effectiveness estimation process of different decisions is described as follows: First, according to the interference decision set The cognitive jammer implements different jamming according to the order of elements. Each time it jams T communication frames, it is a jamming cycle. The total number of jamming cycles is M. Secondly, after the mth interference cycle ends, the cognitive jammer obtains the modulation distribution vector Ω through modulation recognition and data statistics. a,m ={p m,f },f=1,2,…,F, where p m,f is the frequency of the modulation pattern f in the interference period m; then, according to the evaluation index Γ m Find the scores for different decisions:

6. The method according to claim 5, characterized in that According to the scores of different interference cycles, the optimal interference decision S is obtained. j,m* ,include: According to the scores of different interference cycles, the optimal interference decision is:

Images