Method for estimating near-earth satellite safety communication rate using sdfs in cooperation with glrt

Abstract

The application discloses a method for estimating the safety communication rate of a near-earth satellite by using SDFS in cooperation with GLRT, and comprises the following steps: constructing a communication system model between the near-earth satellite and a ground terminal; analyzing signal and channel characteristics to obtain the distance between nodes, channel gain and received signal distribution; in the case that the transmission power of each monitoring party is unknown, generalized likelihood ratio test is used to estimate the transmission power, and the generalized likelihood ratio test is independently executed based on the observation data of each monitoring party; a superior manager performs joint decision based on a soft decision fusion scheme, in combination with the generalized likelihood ratio test result of each monitoring party and a joint threshold, to estimate the detection error probability of the monitoring party; a joint optimization strategy of transmission power and the number of continuous transmission time slots is constructed, and the total system throughput is maximized under the condition of meeting the cooperative concealment constraint; and the application provides an innovative solution for reliable communication of the near-earth satellite in a complex monitoring environment.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and more specifically to a method for estimating near-Earth satellite secure communication rates using SDFS in conjunction with GLRT. Background Technology

[0002] Currently, satellites are classified into three categories based on their orbits: GEO (Geostationary Earth Orbit Satellite), MEO (Medium Earth Orbit Satellite), and LEO (Low Earth Orbit Satellite). GEO satellites are located 36,000 kilometers above the equator, providing continuous regional coverage and suitable for fixed services such as broadcasting and television. However, they suffer from significant signal latency (approximately 0.27 seconds) and limited satellite slots. MEO satellites orbit at altitudes of 5,000-20,000 kilometers, making them suitable for navigation and positioning (such as GPS and BeiDou), but they lack coverage flexibility. LEO satellites, orbiting at altitudes of 160-2,500 kilometers, offer advantages such as low latency and low signal loss, and can achieve global coverage through large-scale networking. They have become a research hotspot in recent years (e.g., the Starlink project) and have enormous potential in areas such as broadband access and the Internet of Things, serving as the core carrier for near-Earth satellite communication.

[0003] For scenarios involving joint detection by multiple monitoring parties, existing research typically employs the Soft Decision Fusion Scheme (SDFS). In this scheme, each monitoring party uploads its received power or local decision statistics to a fusion center, which then makes a joint judgment based on a set overall decision threshold to determine whether a transmission has occurred, and in some cases, simultaneously estimates the transmission power. However, most of these methods rely on the premise that the monitoring parties know the transmission power or its statistical distribution. In real-world environments, monitoring parties often cannot obtain prior information about the transmitter (such as transmission power). In such cases, if the traditional likelihood ratio test (LRT) method is still used, the lack of key parameters will lead to decreased detection performance or even failure. Summary of the Invention

[0004] Purpose of the invention: The purpose of this invention is to provide a method for estimating near-Earth satellite secure communication rate using SDFS in conjunction with GLRT, thereby solving the problem of estimating the detection error probability of satellite secure communication in multi-monitoring collaborative detection scenarios.

[0005] Technical Solution: The present invention provides a method for estimating the secure communication rate of near-Earth satellites using SDFS in conjunction with GLRT, comprising the following steps: constructing a communication system model between near-Earth satellites and ground terminals, including a transmitting ground terminal, a receiving near-Earth satellite, and multiple ground monitoring parties; analyzing signal and channel characteristics to obtain the distance between each node, channel gain, and received signal distribution; estimating the transmission power using a generalized likelihood ratio test when the transmission power of each monitoring party is unknown, and independently performing the generalized likelihood ratio test based on the observation data of each monitoring party; the upper-level manager, based on a soft decision fusion scheme, combining the generalized likelihood ratio test results of each monitoring party with a joint threshold, performing a joint decision to estimate the detection error probability of the monitoring party; and constructing a joint optimization strategy for transmission power and the number of continuous transmission time slots to maximize the total system throughput while satisfying the cooperative concealment constraint.

[0006] Furthermore, in the communication system model, both the transmitting ground terminal and the receiving near-Earth satellite are equipped with multi-antenna uniform planar arrays, while the ground monitoring device is a single-antenna device, and all three communicate in half-duplex mode.

[0007] Furthermore, the channel characteristics analysis is as follows: the channel between the ground terminal and the near-Earth satellite is a direct channel, using the Ricean channel model; the channel between the ground terminal and various ground monitoring sites is a Rayleigh fading channel, using a multipath scattering model; the Doppler effect caused by the high-speed motion of the near-Earth satellite is considered, and Doppler compensation is performed on the received signal.

[0008] Furthermore, the generalized likelihood ratio test includes: current time slot detection mode and historical time slot detection mode; in the current time slot detection mode, the generalized likelihood ratio test is performed based on the current time slot observation data; in the historical time slot detection mode, the generalized likelihood ratio test is performed using all historical time slot observation data, and the transmit power is estimated iteratively through the expectation-maximization algorithm.

[0009] Furthermore, in the historical time slot detection mode, the monitoring party models the probability density function of the observation vector of each time slot as a weighted sum of the null and alternative hypotheses, and estimates the transmission power by maximizing the likelihood function.

[0010] Furthermore, in the soft decision fusion scheme, the superior manager compares the sum of the received power of each monitoring party with a joint threshold to determine whether transmission has occurred. The joint threshold is less than the sum of the thresholds of each monitoring party.

[0011] Furthermore, the joint optimization strategy is as follows: Under the condition of fixed transmission power, the maximum allowable number of continuous transmission slots is solved by the bisection method, so that the average detection error probability is not lower than the preset concealment threshold; under the condition of fixed decoding error probability, the transmission power and the number of continuous transmission slots are jointly optimized with the goal of maximizing the total concealment throughput of the system.

[0012] The present invention discloses a system for estimating near-Earth satellite secure communication rates using SDFS in conjunction with GLRT, comprising:

[0013] Transmitting module: Located in the ground terminal, used to transmit signals at optimized transmission power in selected time slots;

[0014] Receiver module: Located on near-Earth satellites to receive and decode signals;

[0015] Multiple monitoring modules: These are installed at various ground monitoring sites to collect and receive signals and perform generalized likelihood ratio tests;

[0016] Fusion Decision Module: Set up by the superior manager to make joint decisions based on the soft decision fusion scheme and output the probability of detection errors;

[0017] Optimization control module: used to dynamically adjust the transmit power and the number of continuous transmission time slots based on concealment constraints and throughput targets.

[0018] An electronic device according to the present invention includes a memory and a processor. The memory stores a computer program, and the processor executes the program to implement the steps of the method.

[0019] The present invention discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method.

[0020] Beneficial Effects: Compared with existing technologies, this invention has the following significant advantages: This invention considers a scenario where multiple monitoring parties have unknown transmit power, and employs a joint decision mechanism combining collaborative soft-decision fusion detection and generalized likelihood ratio testing. It provides a method for estimating the detection error probability of monitoring parties in this scenario. Simultaneously, by jointly optimizing transmit power and the number of consecutive time slots, it improves the secure communication rate. Furthermore, this invention provides a strategy to maximize the total system throughput while satisfying collaborative concealment constraints, balancing the contradiction between concealment and transmission efficiency, and providing an innovative solution for reliable communication of near-Earth satellites in complex monitoring environments. Attached Figure Description

[0021] Figure 1 This is a flowchart of the present invention;

[0022] Figure 2 This is a modeling diagram of the satellite-to-ground terminal communication link and ground base station of the present invention;

[0023] Figure 3 This is a coordinate model diagram of the ground terminal, satellite, and ground base station of the present invention;

[0024] Figure 4This is a comparison chart showing the performance of the MATLAB joint likelihood ratio test and the joint generalized likelihood ratio test on the probability of error. Detailed Implementation

[0025] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0026] This invention provides a method for estimating near-Earth satellite secure communication rates using SDFS in conjunction with GLRT, comprising the following steps:

[0027] Taking into account both the independent GLRT detection performed by each monitoring node based on its own observation data and the transmission power estimated by the higher-level manager in conjunction with the GLRT detection from SDFS, the detection error probability in this case is obtained. The model diagram is as follows: Figure 1 As shown, a communication link model is considered between the transmitting ground terminal (Alice) and the receiving satellite (Bob) under the condition of being monitored by multiple monitoring ground base stations (Willies). The model comprehensively considers parameters such as transmission power and the number of consecutive time slots to improve the secure communication rate; and maximizes the total system throughput while satisfying the cooperative concealment constraint.

[0028] (1) Signal and channel characteristic analysis and received signal analysis, including the following steps:

[0029] (11) Distances and characteristics between nodes: such as Figure 2 and Figure 3 As shown, the coordinates of the transmitter can be obtained. The coordinates of the k-th monitoring party (k-th Willie) and the coordinates of the receiver ,in These represent the positions of the three entities along the x, y, and z axes of the coordinate system, respectively, with the superscript 't' indicating the time slot number. Taking the transmitter as the reference point, i.e., the transmitter's location as the origin, we can obtain the following distances between the transmitter and the k-th monitoring entity, Willie: Distance between the transmitter and receiver: 。

[0030] In this model, both the ground terminal transmitter and the satellite receiver are considered to be multi-antenna uniform planar arrays (UPAs), with the following antenna numbers: , ( (These represent the number of antennas along the x-axis and y-axis, respectively). It is assumed that the antenna spacing is half a wavelength in both the x-axis and y-axis directions, and... and All three meet this condition. The ground base station Willies has a single antenna, and all three operate in half-duplex communication mode.

[0031] (12) Channel gain: Ricean channel matrix for each time slot:

[0032] ;

[0033] in , Rice factor, and These represent the direct projection matrix and the scattering matrix, respectively.

[0034] Based on the characteristics of satellites and ground terminals, the communication between the transmitter and receiver can be considered a direct LOS channel, i.e. The value is infinite, consisting only of the direct component of the Ricean channel; however, for communication between the ground, the communication between the transmitter and the monitor is considered to be a Rayleigh multipath fading channel, i.e. Approaching 0, the channel contains only scattering components. and These represent the receiver and the k-th monitoring party at the 1st... p The horizontal and vertical launch angles (AoD) along each path when the launcher is the reference frame. These represent the horizontal and vertical starting angles of the transmitter, respectively, with the receiver as the reference frame. Based on the properties of a uniform planar array, the phase shift matrices for both the transmitter and receiver can be obtained:

[0035] Launching party:

[0036] Horizontal array phase shift matrix: ;

[0037] Vertical array phase shift matrix: ;

[0038] Recipient:

[0039] Horizontal array phase shift matrix: ;

[0040] Vertical array phase shift matrix: ;

[0041] in, This indicates the matrix transpose.

[0042] The response matrices of the transmitter and receiver are as follows:

[0043] Launching party: ;

[0044] Recipient: ;

[0045] in, This represents the Kronecker convolution.

[0046] The receiver's channel gain can then be expressed as:

[0047] ;

[0048] in, For the free-space path loss of the satellite channel, This is the free-space path loss term in Fries's transmission formula, which represents the ideal loss of a signal with wavelength λ and a propagation distance of 1 meter. This represents the gain of the satellite's receiving antenna.

[0049] Considering the Doppler effect caused by the high-speed movement of near-Earth satellites, the channel gain after the Doppler effect is:

[0050] ;

[0051] in This indicates the Doppler shift of the satellite.

[0052] The channel gain of the k-th monitoring side, Willie, is:

[0053] ;

[0054] in, Let $\frac{k}{k}$ be the free space path loss of the listening channel. Let be the number of scattering paths for the k-th listening party. This is the antenna response matrix between the transmitter and the kth listening party (since the distance is long, the starting angle of each path can be considered the same). Small-scale fading coefficients for each path.

[0055] (13) Received signal analysis

[0056] The probability distribution of the transmitter's transmission power is consistent across all time slots. In the t-th time slot, if the transmitter transmits, it will map the information to an independent, identically distributed (i.i.d.) zero-mean complex Gaussian random symbol sequence with variance . This is true for every antenna, that is... Where N represents the codeword length of a time slot, Research has shown that, under the constraint of channel concealment, this Gaussian signal can effectively maximize the mutual information between the transmitted and received signals. To ensure successful decoding, the codebook used by the transmitter is shared only with the receiver and not transmitted to the monitoring station s. The noise at the k-th monitoring station and receiver in each time slot consists of a set of independent and identically distributed sequences of zero-mean complex Gaussian random variables, as follows:

[0057] ;

[0058] ;

[0059] in, .

[0060] The signal vectors received by the kth monitoring party and the receiver in the tth time slot are respectively:

[0061] ;

[0062] ;

[0063] Their specific forms are as follows: the received signal of the kth monitoring party: ;

[0064] The receiver's received signal: ;

[0065] Considering the Doppler effect caused by the high-speed movement of near-Earth satellites, the received signal after Doppler compensation is as follows: ;

[0066] in, and These represent the time delay and frequency after Doppler compensation, respectively.

[0067] Therefore, the received symbols of the k-th monitoring party follow a Gaussian distribution, which is in and The probability density functions for the following conditions are given by the following equations:

[0068] (1);

[0069] (2);

[0070] in Let be the power transmitted by the transmitter through the k-th monitoring channel. This indicates that the transmission by the sender did not occur; alternative hypothesis. This indicates that the transmission has occurred at the sending end.

[0071] Assuming there are K monitoring parties in total, the total received signal for the superior manager when multiple monitoring parties make a joint decision is: ; ; (2);

[0072] It also follows a Gaussian distribution, and its probability density functions under H1 and H0 are as follows:

[0073] (3);

[0074] (4);

[0075] in, This represents the total power passing through all monitoring channels. To monitor the variance of the noise.

[0076] (2) The application of the optimal detector and combined GLRT for monitoring method s includes the following steps:

[0077] (21) Optimal detection LRT of monitoring method s when the transmit power is known

[0078] In each time slot, the monitoring process of each monitoring party and its superior manager is a binary hypothesis test, i.e., the null hypothesis. and alternative hypotheses Therefore, Willie's detection errors can be divided into two categories: false alarms and false misses. False alarms refer to errors in Willie's detection... Under certain conditions, a false positive indicates the presence of a transmitted signal, while a false negative indicates that the signal is not detected. The erroneous judgment signal does not exist under the given conditions. This represents the probability of a false alarm. Let H1 be the probability of a missed detection. Assuming that the prior probabilities of H1 and H0 are equal, the detection error probability of monitoring party s can be expressed as: Wiley's goal is to construct an optimal detector to minimize... The sender wants to ensure secure communication—Willy always uses the optimal detector, even in the worst-case scenario. Based on this, the stealth constraint can be expressed as... (where the given parameters are) This means that the launcher has The probability of successfully evading detection is [not specified]. Regarding the prior knowledge of the monitoring method s, it is generally assumed that it fully grasps the noise variance. Transmission power And the statistical model used by the transmitter of this code. In this context, Willie already knows... , , and According to the Neyman-Pearson criterion, Wiley's optimal detector—that is, the scheme that minimizes the probability of detection errors—is the likelihood ratio test (LRT). The likelihood ratio test for the k-th monitoring side can be expressed as:

[0079] (5);

[0080] set up This represents the detection threshold used by the k-th eavesdropper in the likelihood ratio test, where the average received power is... , , It is a chi-square distribution with 2N degrees of freedom. Substituting the probability density function formulas (1) and (2) of the k-th eavesdropper into formula (5), we can obtain the threshold for each eavesdropper as follows: (6);

[0081] The soft decision-making fusion solution for upper-level managers to make joint judgments among multiple monitoring parties involves integrating the received power at each monitoring party. Joint threshold:

[0082] (7);

[0083] The comparison determines whether the launching party should launch or not. It is the error value obtained based on comprehensive factor analysis, that is This makes the joint threshold slightly smaller than the total threshold, resulting in a more reasonable and accurate test. Based on the properties of the chi-square distribution, its detection error probability can be estimated as:

[0084] (8);

[0085] (22) Optimal GLRT for Monitoring Party s when Transmit Power is Unknown: When the transmit power of monitoring party s is unknown, LRT will no longer be applicable. Therefore, monitoring party s can use GLRT to make decisions and estimate the transmit power. The core idea is that when there are unknown parameters in the probability density function (PDF) in the hypothesis test, GLRT replaces the unknown parameters with maximum likelihood estimation (MLE) to construct an achievable detector. Its optimality has been proven in the case of finite sample detection. Through the generalized likelihood ratio statistic:

[0086] (9);

[0087] in, To perform the function of generalized likelihood ratio, This allows for the estimation of unknown parameters and the calculation of thresholds. The specific process is as follows:

[0088] (221) Optimal Detection GLRT for Joint Decision by Monitoring Parties s under Current Time Slot: Based on the observation data of the current time slot, under the condition of complete prior knowledge, the optimal detector in formula (9) distinguishes the null hypothesis by comparing the likelihood function of the current time slot observations. and alternative hypotheses Similarly, based on the observation data of the kth monitoring party in the current time slot, according to formula (9), the generalized likelihood ratio, which does not include prior knowledge of the transmit power, can be expressed as:

[0089] (10);

[0090] in, It is the maximum likelihood estimate of the transmit power by the monitoring party under the current time slot.

[0091] Based on this likelihood function, the threshold of the k-th monitoring point can be calculated. The joint threshold for the soft decision-making integration scheme of superiors is:

[0092] (11);

[0093] in, Based on data from the current time slot's superior manager, the generalized likelihood ratio, excluding prior knowledge of power allocation (PA), can be expressed as:

[0094] (12);

[0095] By differentiating the likelihood ratio function and setting the derivative to zero, an estimate of the transmit power can be obtained.

[0096] (13);

[0097] Substituting the obtained joint threshold formula (11) and transmit power estimation formula (12) into the error detection probability formula (9), we can obtain the error probability estimate of the joint detection by multiple monitoring parties in the current time slot:

[0098] (14).

[0099] (222) Optimal detection GLRT based on joint decision by monitoring parties s under historical time slots: First, it is necessary to obtain the optimal detection GLRT based on historical observation data under the null hypothesis. and alternative hypotheses The probability density function is given below. When detecting the T-th time slot, the k-th monitoring party will use all historical observation vectors within that time slot, which also include the current observation value. This method makes full use of all historical time-slot observation vectors. Information. It's important to note that current observations involve a binary hypothesis test, while historical observations are a set of vectors spanning multiple time slots. —Because the transmitter only transmits in certain time slots. The probability density function of all historical observation vectors and the current observation vector. There are discrepancies. Since the monitoring party s cannot determine whether the transmitter is transmitting in each time slot, the observation vector for a single time slot of the k-th monitoring party... The probability density function of (1≤t≤T) can be modeled as follows: and Weighted sum of probability density functions under two assumptions:

[0100] (15);

[0101] in and These are weights, representing the null hypothesis, respectively. and alternative hypotheses The prior probability is given. Since the observation vectors of all time slots are independent and identically distributed, therefore... The probability density function is the product of the probability density functions of a single time slot:

[0102] (16);

[0103] According to formula (9), the generalized likelihood ratio detector of the k-th monitoring party in the T-th time slot can be expressed as:

[0104] (17);

[0105] The logarithmic derivative of the likelihood ratio function, L, relates to... The first derivative is:

[0106] (18);

[0107] in Indicating in the alternative hypothesis When it was established The probability of transmission in the t-th time slot is the probability that the transmitter will transmit. The formula for calculating this probability is:

[0108] (19);

[0109] when The maximum likelihood estimate of the transmit power of the k-th monitoring party in the T-th time slot can be derived as follows:

[0110] (20);

[0111] This expression is derived from historical observations. Sum of probabilities Jointly determined, in formula (18) Also with and There is a correlation. It can be seen that the variables in formulas (19) and (20) are coupled, which makes it difficult for the monitoring party to derive the estimate. The closed-form expression is given. For the challenging problem of solving highly coupled variables in equations, the Expectation-Maximization (EM) algorithm provides a feasible solution. This algorithm approximates the optimal solution through an iterative method, with each iteration consisting of two steps: the first is the E-step, which calculates the expected value of the likelihood function based on the estimated parameters from the previous iteration; the second is the M-step, which maximizes the likelihood function by updating the parameters. Therefore, the estimator can be derived. Substituting formula (20) into the threshold solution formula (6), we can obtain the threshold of the k-th monitoring party in the T-th time slot:

[0112] (twenty one);

[0113] The joint threshold for the integration of soft decision-making schemes by higher-level managers is:

[0114] (twenty two);

[0115] Similarly, based on the data from the current time slot's superior manager, the generalized likelihood ratio can be expressed as:

[0116] (twenty three);

[0117] The maximum likelihood estimate of its transmit power in the T-th time slot can be derived as follows:

[0118] (twenty four);

[0119] Substituting the obtained joint threshold formula (23) and transmit power estimation formula (24) into the error detection probability formula (9), we can obtain the error probability estimate of the joint detection by multiple monitoring parties under historical time slots:

[0120] (25);

[0121] (223) Covert performance of the transmitter's perspective in historical time slots: In the detection model based on current observations, the detection error probability in formula (14) This is known to the sender and can therefore be used to assess its stealth performance. However, in the historical observation detection model, the detection error probability in formula (25) It is a function of the historical observation vector of the superior manager and its estimate of the transmission power, both of which are unknown to the transmitter. Therefore, It cannot be used to evaluate the launcher's stealth performance. Further analysis of stealth performance is needed from the launcher's perspective. The EM algorithm is an iterative method and may iterate to a near-optimal solution rather than the optimal one. The worst-case scenario needs to be considered: the monitoring party can derive formula (24) using the EM algorithm. And the optimal detector is adopted. The expected detection error probability can be used as a metric to evaluate the concealment performance under unknown variables. According to formulas (25) and (24), It's about parameters. and The function of . Therefore, the expected detection error probability of the T-th time slot can be derived as:

[0122] (26);

[0123] According to the central principles ,in The power received by the monitoring party The maximum likelihood estimation satisfies ,in,

[0124] (27);

[0125] The expected probability of detection error is:

[0126] (28);

[0127] (3) Covert transmission strategy, including the following steps:

[0128] (31) Maximum allowable number of time slots: The expected detection error probability of monitoring s within T time periods is defined as the average of this value:

[0129] (29);

[0130] At a fixed transmission power Under the given conditions, maximize the number of transmission time slots T that can maintain the same power, ensuring that the average detection error probability of the monitoring party within T time slots is not lower than the concealment threshold. This optimization problem can be expressed as:

[0131] (30);

[0132] This is an integer nonlinear programming problem. The bisection method can be used to solve this problem. of Take the integer part This is the maximum allowable number of time slots.

[0133] (32) Maximizing hidden throughput: Fixed decoding error probability At that time, the concealment rate of each time slot for:

[0134] (31);

[0135] in, The power transmitted by the transmitter through the k-th monitoring channel. If it is an inverse Q-function, then the concealment rate over T time periods is defined as the average of this value:

[0136] (32);

[0137] The problem of maximizing total hidden throughput can be expressed as:

[0138] (33);

[0139] This optimization problem is a mixed-integer nonlinear programming (MNLP) problem, which can be transformed into a relaxation problem by ignoring integer constraints and utilizing implicit functional relationships. Derivation of T and The correlation; the optimal solution is Take the minimum value (at which point T is at its maximum), and solve for T using the bisection method, taking the integer part. .

[0140] To verify the feasibility of the present invention and optimize its performance, Figure 4 The simulation results from MATLAB are presented. Figure 4 As can be seen from the curves, the detection error probability changes with the number of channel uses N under different scenarios: at the same transmit power (0dB, 10dB, -10dB), the detection error probability (dashed line) of GLRT when the monitoring party (the monitoring party) does not know the transmit power is higher than that of LRT when the transmit power is known (solid line). This indicates that when the monitoring party lacks prior information about the transmit power, its detection accuracy decreases, which indirectly verifies the rationality of GLRT in the unknown parameter scenario in this invention—providing a higher level of concealment for the communication party. As the number of channel uses N increases, the detection error probability of both verification methods decreases. However, the error probability of GLRT remains at a higher level, and the rate of decrease is relatively gentle, indicating that even if the monitoring party accumulates more observation data, the scenario with unknown transmit power can still maintain a certain concealment advantage. The higher the transmit power (e.g., 10dB), the lower the detection error probability (the easier the detection) for both methods. However, the difference between GLRT and LRT is still significant, further demonstrating the effectiveness of this invention in the scenario of "unknown transmit power of multiple monitoring parties".

[0141] This invention constructs a collaborative detection framework using a multi-monitoring party joint generalized likelihood ratio test (GLRT) and combines it with a soft decision fusion scheme (SDFS). By integrating the GLRT detection results performed by each monitoring node based on observation data and the joint threshold of the superior manager, the detection error probability in this scenario is obtained. At the same time, a joint optimization strategy for transmit power and the number of consecutive time slots is designed to maximize the total system throughput while satisfying the collaborative concealment constraint (detection error probability not lower than the threshold), thus balancing the contradiction between concealment and transmission efficiency. This provides a reliable solution for near-Earth satellite secure communication in multi-monitoring party collaborative detection scenarios.

[0142] Parameter explanation table attached:

[0143] ;

[0144] ;

[0145] ;

[0146] ;

[0147] 。

Claims

1. A method for estimating near-Earth satellite secure communication rate using SDFS in conjunction with GLRT, characterized in that, Includes the following steps: The communication system model between near-Earth satellites and ground terminals includes a transmitting ground terminal, a receiving near-Earth satellite, and multiple ground monitoring parties. Analyze signal and channel characteristics to obtain the distance between nodes, channel gain, and received signal distribution; When the transmission power is unknown to each monitoring party, the transmission power is estimated by using the generalized likelihood ratio test, and the generalized likelihood ratio test is performed independently based on the observation data of each monitoring party. Based on the soft decision fusion scheme, the superior manager combines the generalized likelihood ratio test results of each monitoring party with the joint threshold to make a joint decision in order to estimate the detection error probability of the monitoring party. A joint optimization strategy for transmit power and the number of continuous transmission time slots is constructed to maximize the total system throughput while satisfying the cooperative concealment constraint. The generalized likelihood ratio (GLR) test includes two modes: current time slot detection mode and historical time slot detection mode. In the current time slot detection mode, the GLR test is performed based on the current time slot observation data. In the historical time slot detection mode, the GLR test is performed using all historical time slot observation data, and the transmit power is iteratively estimated using the expectation-maximization algorithm. In the soft decision fusion scheme, the higher-level manager compares the sum of the received power of each monitoring party with a joint threshold to determine whether transmission should occur; the joint threshold is less than the sum of the thresholds of each monitoring party.

2. The method for estimating near-Earth satellite secure communication rate using SDFS in conjunction with GLRT according to claim 1, characterized in that, In the communication system model, the transmitting end ground terminal and the receiving end near-Earth satellite are both equipped with multi-antenna uniform planar arrays, the ground monitoring side is a single-antenna device, and all three communicate in half-duplex mode.

3. The method for estimating near-Earth satellite secure communication rate using SDFS in conjunction with GLRT according to claim 1, characterized in that, The channel characteristics analysis is as follows: the channel between the ground terminal and the near-Earth satellite is a direct channel, and the Ricean channel model is used; the channel between the ground terminal and the ground monitoring sites is a Rayleigh fading channel, and the multipath scattering model is used; the Doppler effect caused by the high-speed motion of the near-Earth satellite is considered, and Doppler compensation is performed on the received signal.

4. The method for estimating near-Earth satellite secure communication rate using SDFS in conjunction with GLRT according to claim 1, characterized in that, In the historical time slot detection mode, the monitoring party models the probability density function of each time slot observation vector as a weighted sum of the null and alternative hypotheses, and estimates the transmission power by maximizing the likelihood function.

5. A method for estimating near-Earth satellite secure communication rate using SDFS in conjunction with GLRT according to claim 1, characterized in that, The joint optimization strategy is as follows: Under the condition of fixed transmission power, the maximum allowable number of continuous transmission time slots is solved by the bisection method so that the average detection error probability is not lower than the preset concealment threshold; under the condition of fixed decoding error probability, the transmission power and the number of continuous transmission time slots are jointly optimized with the goal of maximizing the total concealment throughput of the system.

6. A system for estimating near-Earth satellite secure communication rates using SDFS in conjunction with GLRT, characterized in that, The method described by any one of claims 1-5 includes: Transmitting module: Located in the ground terminal, used to transmit signals at optimized transmission power in selected time slots; Receiver module: Located on near-Earth satellites to receive and decode signals; Multiple monitoring modules: These are installed at various ground-based listening stations to collect and receive signals and perform generalized likelihood ratio tests. Fusion Decision Module: Set up by the superior manager to make joint decisions based on the soft decision fusion scheme and output the probability of detection errors; Optimization control module: used to dynamically adjust the transmit power and the number of continuous transmission time slots based on concealment constraints and throughput targets.

7. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the program to implement the steps of the method according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The device contains a computer program that, when executed by a processor, implements the steps of the method according to any one of claims 1-5.

Images