Decision fusion method and system in multi-listening scene of covert communication

Abstract

The application discloses a decision fusion method and system in a multi-listening scenario in covert communication and belongs to the technical field of wireless communication. The method comprises the following steps: configuring a limited total block length for each listening node, and dividing the total block length into a detection block length and a transmission block length; each listening node performs local signal sensing, generates a local decision result by using the detection block length, and sends the local decision result to a fusion center by using the transmission block length; the fusion center allocates an optimal fusion weight for each decision result based on the local detection performance and the transmission performance of each listening node according to the criterion of minimizing the global detection error probability; the fusion center calculates a weighted sum based on the fusion weight and the corresponding decision result, compares the weighted sum with an optimal decision threshold of the fusion center, and makes a global decision on whether the target signal exists. The application realizes the optimal performance of a whole link from sensing to fusion, and improves the reliability and accuracy of signal detection.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication technology, specifically relating to a decision fusion method and system for multiple eavesdropping scenarios in covert communication. Background Technology

[0002] The statements herein provide only background information in relation to this invention and do not necessarily constitute prior art.

[0003] With the continuous evolution of wireless communication technology, signal detection technology, as a key component in ensuring communication monitoring, spectrum management, and physical layer security, has received increasing attention. In modern signal monitoring and countermeasures applications, the processing power of a single monitoring node (or Willie node) is often insufficient to effectively combat complex interference and attacks. In contrast, the multi-node (multi-Willie) collaborative processing approach, with its stronger anti-interference capabilities, wider coverage, and higher detection accuracy, has become a fundamental technology for improving detection reliability.

[0004] Among numerous collaborative processing strategies, weighted decision fusion strategies have attracted widespread attention due to their ability to effectively integrate information from multiple nodes. However, existing weighted decision fusion methods have the following shortcomings: First, most studies use weights that are not theoretically optimal; for example, directly using the node's correct detection probability as the weight cannot guarantee the optimization of global detection performance. Second, although existing methods consider the impact of non-ideal reporting channels, they do not address how to jointly optimize detection and transmission resources under the condition of limited total resources (such as total block length) of listening nodes in actual systems. Third, existing methods fail to fully adapt to the heterogeneity of detection and communication capabilities among listening nodes, and cannot achieve truly adaptive optimal fusion.

[0005] Therefore, how to minimize the final error probability of the fusion center by jointly optimizing the detection threshold, fusion weight, and detection / transmission block length allocation of each node in a multi-node scenario with performance differences is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a decision fusion method and system for multiple listening scenarios in covert communication. It is particularly suitable for scenarios with multiple independent listening nodes (Willie) with different detection and communication capabilities. It fully considers that multiple listening nodes have different detection and communication capabilities, each node is independent and the total block length is limited. The fusion center performs optimal weighted fusion of the local decisions of each listening node to improve the reliability and accuracy of signal detection.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solution: In a first aspect, the technical solution of the present invention provides a decision fusion method for multiple eavesdropping scenarios in covert communication, including: Configure a limited total block length for each listening node, and divide the total block length into two parts: a detection block length for signal sensing and a transmission block length for reporting decision results; Each monitoring node independently performs local signal sensing, generates a local decision on the presence of the target signal using the detection block length, and sends the local decision to the fusion center using the transmission block length; The fusion center receives decision results from multiple monitoring nodes and assigns an optimal fusion weight to each decision result based on the local detection performance and transmission performance of each monitoring node, according to the criterion of minimizing the global detection error probability. The fusion center calculates a weighted sum based on the fusion weights and the corresponding decision results, and compares the weighted sum with the optimal decision threshold of the fusion center to make a global decision on whether the target signal exists.

[0008] In at least one embodiment, the local detection performance of the monitoring node includes the probability of false alarms and the probability of missed detections; The false alarm probability is expressed as:

[0009] Listening Node The probability of a missed detection in local detection is expressed as:

[0010] In the formula, For listening nodes The probability of false alarms in local detection; For listening nodes The probability of missed detection in local testing; Indicates the length of the detection block; Indicates listening node The optimal detection threshold for local energy detection; Indicates the transmitter and the first Channel coefficients between monitoring nodes; Indicates the transmitter's transmission power; This represents the power of additive white Gaussian noise; It is a gamma function; It is an incomplete gamma function.

[0011] In at least one embodiment, the listening node The optimal detection threshold for local energy detection is specifically expressed as follows:

[0012] In the formula, Indicates listening node The optimal detection threshold for local energy detection.

[0013] In at least one embodiment, the transmission performance of the monitoring node includes the transmission error probability from the monitoring node to the fusion center and the probability that the fusion center correctly receives the monitoring node's decision. The transmission error probability from the monitoring node to the fusion center is specifically expressed as follows:

[0014] The probability that the fusion center correctly receives the decision from the monitoring node is specifically expressed as follows:

[0015] In the formula, Indicates the length of the detection block; Indicates the length of the transport block; Indicates listening node The optimal detection threshold for local energy detection; Indicates the transmitter and the first Channel coefficients between monitoring nodes; Indicates the transmitter's transmission power; This represents the power of additive white Gaussian noise; Indicates the first Channel coefficients between each monitoring node and the fusion center; Indicates that the fusion center receives the first The optimal decision threshold for signals from each monitoring node; Indicates the signal power transmitted by the monitoring node; Indicates the noise power at the fusion center; It is a gamma function; It is an incomplete gamma function.

[0016] In at least one embodiment, the fusion center receives the first The optimal decision threshold for the signals from each monitoring node is specifically expressed as follows:

[0017] In the formula, Indicates that the fusion center receives the first The optimal decision threshold for each listening node signal.

[0018] In at least one embodiment, the optimal fusion weight is obtained based on the equivalent false alarm probability and equivalent missed detection probability of the listening node at the fusion center through a likelihood ratio test and logarithmic transformation, specifically expressed as follows:

[0019] In the formula, Indicates the integration center at the first The equivalent false alarm probability of each monitoring node; Indicates the integration center at the first The equivalent false negative probability of each monitoring node.

[0020] In at least one embodiment, the equivalent false alarm probability is specifically expressed as:

[0021] The equivalent false negative probability is specifically expressed as:

[0022] In the formula, and These represent the global assumptions that the target signal does not exist and that it does exist, respectively. For listening nodes The probability of false alarms in local detection; For listening nodes The probability of missed detection in local testing; This represents the decision values ​​received by the fusion center from each monitoring node; Indicates listening node The probability of transmission errors to the fusion center.

[0023] In at least one embodiment, the optimal decision threshold of the fusion center is specifically expressed as:

[0024] In the formula, and These represent the prior probabilities of the signal not existing and the signal existing, respectively. Indicates the integration center at the first The equivalent false alarm probability of each monitoring node; Indicates the integration center at the first The equivalent false negative probability of each monitoring node.

[0025] In at least one embodiment, the global decision rule for determining the existence of the target signal is as follows:

[0026] In the formula, This indicates that the target signal exists; This indicates that the target signal does not exist; Indicates the optimal fusion weights; This represents the decision values ​​received by the fusion center from each monitoring node; This represents the optimal decision threshold for the fusion center.

[0027] Secondly, the technical solution of the present invention also provides a decision fusion system for multiple eavesdropping scenarios in covert communication, including: The resource configuration module is configured to: configure a limited total block length for each listening node, and divide the total block length into two parts: a detection block length for signal sensing and a transmission block length for reporting decision results; The local signal sensing module is configured such that each monitoring node independently performs local signal sensing, generates a local decision result on the presence or absence of a target signal using the detection block length, and sends the local decision result to the fusion center using the transmission block length. The weight confirmation module is configured such that: the fusion center receives the decision results from multiple monitoring nodes, and assigns an optimal fusion weight to each decision result based on the local detection performance and transmission performance of each monitoring node, according to the criterion of minimizing the global detection error probability; The global decision module is configured as follows: the fusion center calculates a weighted sum based on the fusion weights and the corresponding decision results, and compares the weighted sum with the optimal decision threshold of the fusion center to make a global decision on whether the target signal exists.

[0028] The beneficial effects of the above-described technical solution of the present invention are as follows: 1) The decision fusion method for covert communication in multi-monitor scenarios of this invention comprehensively considers the local detection performance of each monitoring node and the transmission performance between the monitoring node and the fusion center. It assigns a theoretically optimal fusion weight to each monitoring node, thereby minimizing the global error probability of the fusion center and effectively improving the reliability and accuracy of signal detection. Simulation results show that the proposed method achieves the highest global correct probability under different signal-to-noise ratio conditions and different total block lengths, significantly outperforming existing methods such as equal-gain combining and correct probability-based weighting, effectively improving the reliability and accuracy of signal detection.

[0029] 2) This invention fully considers the reality that the detection and communication capabilities of each monitoring node differ in actual scenarios. Through the derived closed-form weighting expression, it can adaptively assign different weights based on the equivalent false alarm probability and equivalent missed detection probability of each node. This adaptive weighting mechanism achieves optimal fusion in heterogeneous node scenarios, avoiding the limitations of "one-size-fits-all" or experience-based weighting in traditional methods.

[0030] 3) This invention incorporates limited block length resources into the optimization framework. Each monitoring node has a limited total block length and performs joint optimization on the detection block length and transmission block length of each monitoring node. With the goal of minimizing the global error probability, the block length allocation is jointly optimized, achieving optimal performance across the entire link from perception to fusion. Attached Figure Description

[0031] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0032] Figure 1 This is a schematic diagram of the decision fusion method in a multi-eavesdropping scenario in covert communication disclosed in Embodiment 1 of the present invention; Figure 2 This is a graph showing the local detection correctness probability of a single monitoring node under different detection block lengths, as disclosed in Embodiment 1 of the present invention. Figure 3 This is a comparison chart of the global correct probability of the fusion center under different decision thresholds and the theoretical optimal threshold disclosed in Embodiment 1 of the present invention; Figure 4 This is a comparison chart of the global correctness probability of the decision fusion method in the multi-monitoring scenario of covert communication disclosed in Embodiment 1 of the present invention with other fusion methods. Detailed Implementation

[0033] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0034] As described in the background section, the purpose of this invention is to overcome the shortcomings of the prior art and provide a decision fusion method and system for multiple listening scenarios in covert communication. It is particularly suitable for scenarios with multiple independent listening nodes (Willies) with different detection and communication capabilities. It fully considers that multiple listening nodes have different detection and communication capabilities, each node is independent and the total block length is limited. The fusion center performs optimal weighted fusion of the local decisions of each listening node to improve the reliability and accuracy of signal detection.

[0035] Example 1 In a typical embodiment of the present invention, such as Figures 1 to 4 As shown, this embodiment discloses a decision fusion method for multiple eavesdropping scenarios in covert communication, applicable to scenarios including a transmitter, a legitimate receiver, and... In a wireless communication system with several independent listening nodes (Willie) and a fusion center, the decision fusion method for multiple listening scenarios in covert communication specifically includes the following steps: S1. Configure a limited total block length for each listening node, and divide the total block length into two parts: a detection block length for signal sensing and a transmission block length for reporting decision results; S2. Each monitoring node independently performs local signal sensing, generates a local decision result on the existence of the target signal using the detection block length, and sends the local decision result to the fusion center using the transmission block length; S3. The fusion center receives the decision results from multiple monitoring nodes and assigns a fusion weight to each decision result based on the local detection performance and transmission performance of each monitoring node, according to the criterion of minimizing the global detection error probability. S4. The fusion center calculates a weighted sum based on the fusion weights and the corresponding decision results, and compares the weighted sum with the optimal decision threshold of the fusion center to make a global decision on whether the target signal exists.

[0036] The decision fusion method for multiple listening scenarios in covert communication described above will be explained in detail below with reference to specific implementation methods.

[0037] S1. Configure a limited total block length for each listening node, and divide the total block length into two parts: a detection block length for signal sensing and a transmission block length for reporting decision results.

[0038] In this step, the communication process is divided into a sensing phase and a reporting phase, for each listening node. Configure a limited total block length and the total block length Divided into detection block lengths and transport block length Two parts. The detection block length... Used for signal sensing, transmission block length Used for reporting judgment results, and the detection block length and transport block length satisfy:

[0039] During the perception phase, the listening node use Energy detection is performed on the received signal every symbol period; during the reporting phase, the listening node... use Each symbol period will listen to the node. The local judgment result is sent to the fusion center.

[0040] S2. Each monitoring node independently performs local signal sensing, generates a local decision result on the existence of the target signal using the detection block length, and sends the local decision result to the fusion center using the transmission block length.

[0041] In this step, each monitoring node independently performs local signal sensing and generates a local decision result. The specific implementation process is as follows: S21. Determine the optimal detection threshold for each monitoring node locally.

[0042] In this step, for each monitoring node, based on the Bayesian decision criterion and according to its received signal model and channel state information, the optimal detection threshold for its local energy detection is derived and determined. This makes the listening node Minimize the probability of local detection errors.

[0043] Specifically, in the The symbol period, the first The signals received by each monitoring node are represented as follows:

[0044] In the formula, Indicates the transmitter and the first Channel coefficients between monitoring nodes It is a signal sent by the transmitter. ; It is additive white Gaussian noise. ; and These represent the presence and absence of the signal, respectively.

[0045] Each monitoring node uses an energy detector, and the detection statistics are defined as follows:

[0046] In the formula, The modulus of a signal.

[0047] Based on the Bayesian decision criterion, the listening node is derived. The optimal detection threshold for local energy detection is specifically expressed as:

[0048] In the formula, Indicates listening node The optimal detection threshold for local energy detection; Indicates the transmitter's transmission power; This represents the power of additive white Gaussian noise.

[0049] S22. Calculate the local detection performance of each listening node.

[0050] In this step, based on the optimal decision threshold for local energy detection of each monitoring node determined in step S21, the local detection performance of each monitoring node is calculated, including the false alarm probability and the missed detection probability.

[0051] Specifically, listening nodes The false alarm probability in local detection is expressed as:

[0052] Listening Node The probability of a missed detection in local detection is expressed as:

[0053] In the formula, For listening nodes The probability of false alarms in local detection; For listening nodes The probability of missed detection in local testing; It is a gamma function; It is an incomplete gamma function.

[0054] S23. Analyze the transmission performance between each monitoring node and the fusion center.

[0055] In this step, for each monitoring node, based on the channel state information between it and the fusion center, the optimal decision threshold for the fusion center to receive the signal from that monitoring node is derived and determined, and the transmission error probability is calculated.

[0056] Specifically, first calculate the first... The local decision result of each listening node, i.e., the listening node Prior probabilities of the presence and absence of a decision signal:

[0057]

[0058] In the formula, and They represent the first Each listening node determines whether a signal exists or not. If the signal exists, it sends a "1"; otherwise, it sends a "-1". Specifically, based on the... Detection statistics of each listening node To listen to nodes Optimal detection threshold for local energy detection The decision is made based on the decision threshold, if the detection statistic... Larger than the listening node Optimal detection threshold for local energy detection If the condition is met, the decision signal exists; otherwise, the decision signal does not exist.

[0059] Based on this, the fusion center received data from the first... The signal of each listening node is represented as follows:

[0060] In the formula, Indicates the first Channel coefficients between each monitoring node and the fusion center; It is a signal sent by the listening node. ; It is the noise at the fusion center. .

[0061] The detection statistic at the fusion center is defined as:

[0062] In the formula, Indicates the length of the transport block.

[0063] Then, based on the Bayesian criterion, the method for the fusion center to receive the first... The optimal decision threshold for the signals from each monitoring node is specifically expressed as:

[0064] In the formula, Indicates that the fusion center receives the first The optimal decision threshold for signals from each monitoring node; Indicates the signal power transmitted by the monitoring node; This indicates the noise power at the fusion center.

[0065] Based on the fusion center receiving the first Calculate the optimal decision threshold for each listening node's signal. The transmission error probability to the fusion center is specifically expressed as:

[0066] In the formula, Indicates listening node The probability of transmission errors to the fusion center. As an alternative implementation, the probability of transmission errors... The calculation can be performed using approximate expressions to reduce complexity, for example, by simplifying under high signal-to-noise ratio conditions.

[0067] Furthermore, the false alarm probability is determined by the local detection performance of the local monitoring node. and the probability of missed detection and the The probability of transmission errors from each monitoring node to the fusion center The correct reception of the first by the fusion center is derived. The probability of a decision made by a listening node is specifically expressed as follows:

[0068] In the formula, This indicates that the fusion center has correctly received the first... The probability of a decision made by a listening node.

[0069] As a further implementation, to maximize the correct reception of the first by the fusion center The probability of a decision by a listening node To achieve the goal, by traversing... From 1 to -1 or an optimization algorithm can be used to solve for... The largest and The optimized detection block length is obtained by taking the value. and transport block length The optimized block length allocation scheme It is applied to all monitoring nodes to achieve optimal detection performance of the system.

[0070] S3. The fusion center receives the decision results from multiple monitoring nodes and assigns a fusion weight to each decision result based on the local detection performance and transmission performance of each monitoring node, according to the criterion of minimizing the global detection error probability.

[0071] S31. The fusion center receives the decision values ​​from each monitoring node.

[0072] During the reporting phase, each listening node Its local decision result is sent to the convergence center via a wireless channel. The convergence center then receives the result and submits it to the convergence center. Optimal decision threshold for signals from each monitoring node For the received signal Make a judgment and obtain the final received judgment value. Specifically, based on the detection statistics at the fusion center. To receive the first Optimal decision threshold for signals from each monitoring node As the decision threshold, if the detection statistic at the fusion center... The fusion center receives the first Optimal decision threshold for signals from each monitoring node The fusion center determined that the monitoring node reported "signal present," at which point... Conversely, if the fusion center determines that the monitoring node reports "signal not present," then... .

[0073] Thus, the integration center obtained all The decision vector for each listening node is specifically represented as follows: .

[0074] S32. Calculate the equivalent detection performance of each listening node.

[0075] In this step, each listening node is taken into account. Local detection performance and transmission error probability Calculate the equivalent false alarm probability of the monitoring node at the fusion center. and equivalent false negative probability Specifically, it is expressed as:

[0076]

[0077] In the formula, and These represent the global assumptions that the target signal does not exist and that it does exist, respectively.

[0078] S33. Calculate the fusion weight of each monitoring node.

[0079] In this step, based on Bayesian decision theory, the objective is to minimize the global average error probability, and the calculation is performed for the th... The optimal fusion weight for each listening node.

[0080] Specifically, based on the fusion center at the first Equivalent false alarm probability of each monitoring node and equivalent false negative probability Through likelihood ratio test and logarithmic transformation, the first... Optimal weight of each listening node Specifically, it is expressed as:

[0081] In the formula, Indicates the integration center at the first The equivalent false alarm probability of each monitoring node; Indicates the integration center at the first The equivalent false negative probability of each monitoring node.

[0082] As an alternative implementation, to further understand the weighting characteristics, the impact of local detection performance and transmission performance on the weights can be analyzed. For example, when the transmission error probability... When the value approaches 0.5, the local detection performance , Changes in optimal weights When the influence approaches zero, the optimal weight is... It mainly depends on the probability of transmission errors. When local detection performance meets At that time, the optimal weight Transmission error probability The changes are not sensitive. These analyses help simplify weight calculations or parameter tuning in practical systems.

[0083] S4. The fusion center calculates a weighted sum based on the fusion weights and the corresponding decision results, and compares the weighted sum with the optimal decision threshold of the fusion center to make a global decision on whether the target signal exists.

[0084] In this step, the fusion center obtains the decision vectors of all listening nodes. Finally, a global decision is made using the optimal weighted fusion rule.

[0085] S41. Calculate the optimal decision threshold for the fusion center.

[0086] In this step, based on the fusion center at the first Equivalent false alarm probability of each monitoring node and equivalent false negative probability The optimal decision threshold for calculating the fusion center is specifically expressed as:

[0087] In the formula, and These represent the prior probabilities of the signal's absence and presence, respectively, which can be preset or estimated online based on the actual scenario.

[0088] S42. Calculate the weighted sum.

[0089] In this step, the fusion center is based on the first The optimal weight of each monitoring node and the decision values ​​of each monitoring node received by the fusion center. The weighted sum is calculated as follows:

[0090] In the formula, This indicates a weighted sum.

[0091] S43. Make a global decision.

[0092] The weighted sum is compared with the optimal decision threshold of the fusion center to make a global decision on whether the target signal exists, specifically expressed as:

[0093] In the formula, This indicates that the target signal exists; This indicates that the target signal does not exist.

[0094] As a further implementation, the decision fusion method in covert communication under multiple listening scenarios also includes a step of jointly optimizing the allocation of detection block length and transmission block length.

[0095] S5. With the goal of minimizing the global error probability of the fusion center, the allocation of the detection block length and the transmission block length is jointly optimized.

[0096] Specifically, define the global error probability. for:

[0097] In the formula, and These represent the global false alarm probability and global false negative probability of the fusion center, respectively. These can be determined by methods such as numerical integration or Monte Carlo simulation observation, based on the equivalent detection performance of each node. , And obtained from the fusion rules.

[0098] This embodiment also performs simulation verification on the decision fusion method in a multi-eavesdropping scenario of covert communication. The simulation results are as follows: Figures 2 to 4 As shown.

[0099] in, Figure 2 In this method, a single listening node has different detection block lengths. The graph below shows the probability relationship of correct local detection. Figure 2 It can be seen from this that, with As the number of elements increases, the probability of correct detection first rises and then falls; there exists an optimal [condition / condition]. This optimizes detection performance. This is because the excessive length... It will compress the transmission block length This increases the probability of transmission errors, thereby reducing the overall probability of correct transmission.

[0100] Figure 3 This method uses different decision thresholds for the fusion center (i.e., the optimal decision threshold for the fusion center). This is a comparison chart of the global correct probability and the theoretical optimal threshold under simulation conditions. The solid line represents the simulated correct probability, and the dashed line represents the theoretical optimal threshold. From... Figure 3 As can be seen, the maximum value of the simulation correct probability occurs exactly at the theoretical optimal decision threshold, which verifies the correctness of the optimal decision threshold of the optimal fusion center proposed in this method.

[0101] Figure 4 This is a comparison chart of the global correctness probability of the decision fusion method proposed in this paper for covert communication in multi-eavesdropping scenarios with other fusion methods. From Figure 4 As can be seen, the proposed method achieves the highest correct probability under all signal-to-noise ratio conditions, significantly outperforming equal-gain combining and methods based on correct probability weighting, thus verifying the superiority of the present invention in fusion performance.

[0102] Example 2 In a typical embodiment of the present invention, this embodiment discloses a decision fusion system for multiple eavesdropping scenarios in covert communication, comprising: The resource configuration module is configured to: configure a limited total block length for each listening node, and divide the total block length into two parts: a detection block length for signal sensing and a transmission block length for reporting decision results; The local signal sensing module is configured such that each monitoring node independently performs local signal sensing, generates a local decision result on the presence or absence of a target signal using the detection block length, and sends the local decision result to the fusion center using the transmission block length. The weight confirmation module is configured such that: the fusion center receives the decision results from multiple monitoring nodes, and assigns an optimal fusion weight to each decision result based on the local detection performance and transmission performance of each monitoring node, according to the criterion of minimizing the global detection error probability; The global decision module is configured as follows: the fusion center calculates a weighted sum based on the fusion weights and the corresponding decision results, and compares the weighted sum with the optimal decision threshold of the fusion center to make a global decision on whether the target signal exists.

[0103] As a further implementation, the decision fusion system in covert communication multi-monitoring scenarios also includes a parameter optimization module, configured to jointly optimize the allocation of detection block length and transmission block length with the goal of minimizing the global error probability of the fusion center.

[0104] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A decision fusion method for multiple eavesdropping scenarios in covert communication, characterized in that, include: Configure a limited total block length for each listening node, and divide the total block length into two parts: a detection block length for signal sensing and a transmission block length for reporting decision results; Each monitoring node independently performs local signal sensing, generates a local decision on the presence of the target signal using the detection block length, and sends the local decision to the fusion center using the transmission block length; The fusion center receives decision results from multiple monitoring nodes and assigns an optimal fusion weight to each decision result based on the local detection performance and transmission performance of each monitoring node, according to the criterion of minimizing the global detection error probability. The fusion center calculates a weighted sum based on the fusion weights and the corresponding decision results, and compares the weighted sum with the optimal decision threshold of the fusion center to make a global decision on whether the target signal exists.

2. The decision fusion method for multiple eavesdropping scenarios in covert communication as described in claim 1, characterized in that, The local detection performance of the monitoring node includes the probability of false alarms and the probability of missed detections; The false alarm probability is expressed as: Listening Node The probability of a missed detection in local detection is expressed as: In the formula, For listening nodes The probability of false alarms in local detection; For listening nodes The probability of missed detection in local testing; Indicates the length of the detection block; Indicates listening node The optimal detection threshold for local energy detection; Indicates the transmitter and the first Channel coefficients between monitoring nodes; Indicates the transmitter's transmission power; This represents the power of additive white Gaussian noise; It is a gamma function; It is the lower incomplete gamma function.

3. The decision fusion method for multiple eavesdropping scenarios in covert communication as described in claim 2, characterized in that, Listening Node The optimal detection threshold for local energy detection is specifically expressed as follows: In the formula, Indicates listening node The optimal detection threshold for local energy detection.

4. The decision fusion method for multiple eavesdropping scenarios in covert communication as described in claim 1, characterized in that, The transmission performance of the monitoring node includes the probability of transmission errors from the monitoring node to the fusion center and the probability that the fusion center correctly receives the monitoring node's decision. The transmission error probability from the monitoring node to the fusion center is specifically expressed as follows: The probability that the fusion center correctly receives the decision from the monitoring node is specifically expressed as follows: In the formula, Indicates the length of the detection block; Indicates the length of the transport block; Indicates listening node The optimal detection threshold for local energy detection; Indicates the transmitter and the first Channel coefficients between monitoring nodes; Indicates the transmitter's transmission power; This represents the power of additive white Gaussian noise; Indicates the first Channel coefficients between each monitoring node and the fusion center; Indicates that the fusion center receives the first The optimal decision threshold for signals from each monitoring node; Indicates the signal power transmitted by the monitoring node; Indicates the noise power at the fusion center; It is a gamma function; It is the lower incomplete gamma function.

5. The decision fusion method for multiple eavesdropping scenarios in covert communication as described in claim 4, characterized in that, The Fusion Center received the first The optimal decision threshold for the signals from each monitoring node is specifically expressed as follows: In the formula, Indicates that the fusion center receives the first The optimal decision threshold for each listening node signal.

6. The decision fusion method for multiple eavesdropping scenarios in covert communication as described in claim 1, characterized in that, The optimal fusion weight is obtained based on the equivalent false alarm probability and equivalent missed detection probability of the listening node at the fusion center through likelihood ratio test and logarithmic transformation, specifically expressed as follows: In the formula, Indicates the integration center at the first The equivalent false alarm probability of each monitoring node; Indicates the integration center at the first The equivalent false negative probability of each monitoring node.

7. The decision fusion method for multiple eavesdropping scenarios in covert communication as described in claim 6, characterized in that, The equivalent false alarm probability is specifically expressed as: The equivalent false negative probability is specifically expressed as: In the formula, and These represent the global assumptions that the target signal does not exist and that it does exist, respectively. For listening nodes The probability of false alarms in local detection; For listening nodes The probability of missed detection in local testing; This represents the decision values ​​received by the fusion center from each monitoring node; Indicates listening node The probability of transmission errors to the fusion center.

8. The decision fusion method for multiple eavesdropping scenarios in covert communication as described in claim 1, characterized in that, The optimal decision threshold of the fusion center is specifically expressed as: In the formula, and These represent the prior probabilities of the signal not existing and the signal existing, respectively. Indicates the integration center at the first The equivalent false alarm probability of each monitoring node; Indicates the integration center at the first The equivalent false negative probability of each monitoring node.

9. The decision fusion method for multiple eavesdropping scenarios in covert communication as described in claim 1, characterized in that, The global decision-making rule for determining whether a target signal exists is as follows: In the formula, This indicates that the target signal exists; This indicates that the target signal does not exist; Indicates the optimal fusion weights; This represents the decision values ​​received by the fusion center from each monitoring node; This represents the optimal decision threshold for the fusion center.

10. A decision fusion system for multiple eavesdropping scenarios in covert communication, characterized in that, include: The resource configuration module is configured to: configure a limited total block length for each listening node, and divide the total block length into two parts: a detection block length for signal sensing and a transmission block length for reporting decision results; The local signal sensing module is configured such that each monitoring node independently performs local signal sensing, generates a local decision result on the presence or absence of a target signal using the detection block length, and sends the local decision result to the fusion center using the transmission block length. The weight confirmation module is configured such that: the fusion center receives the decision results from multiple monitoring nodes, and assigns an optimal fusion weight to each decision result based on the local detection performance and transmission performance of each monitoring node, according to the criterion of minimizing the global detection error probability; The global decision module is configured as follows: the fusion center calculates a weighted sum based on the fusion weights and the corresponding decision results, and compares the weighted sum with the optimal decision threshold of the fusion center to make a global decision on whether the target signal exists.

Images