Energy collection assistance-based covert communication system and method for maximizing covert rate

Abstract

The invention discloses a covert communication system based on energy collection assistance and a method for maximizing the covert rate, and belongs to the technical field of wireless communication and secret communication. The covert communication system based on the energy harvesting assistance comprises a source site Alice, a destination Bob, an eavesdropper Willie, a plurality of relay terminals Relay and a plurality of energy harvesting jammers EHJs. The characteristics of the energy collection jammer are utilized, signals emitted by a source are collected and converted into electric energy, and then interference signals are released to interfere communication judgment of an eavesdropper. And the energy collection jammer is added, so that the system is safer, and meanwhile, a large amount of energy is saved. Meanwhile, two model strategies for selecting the energy harvesting jammer and the relay are provided, so that the whole system applies proper strategies in different use scenes, and the communication efficiency and performance are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communication and confidential communication, and in particular relates to a covert communication system assisted by energy collection and a method for maximizing the covert rate. Background Art

[0002] In today's society, wireless systems play an indispensable role in daily life and industrial applications, especially with the emergence of the Internet of Things (IoT). These wireless systems transmit a wealth of private information, such as personal health data, geographic location information, financial transaction records, etc. Although current privacy protection technologies are mature, there are still significant security risks in the transmission process. Traditional encryption methods mainly rely on computational complexity to ensure the security of information transmission, which leads to a large consumption of device energy. However, this method is not fully applicable to IoT devices with limited energy and cannot meet their requirements for information transmission. As an effective supplement, covert communication can use the random characteristics of the channel to hide the wireless communication process, thereby ensuring the security of information transmission. In order to support various applications that require sensitive information transmission, it is essential to study and explore covert communication.

[0003] A multi-relay system is a system that uses multiple relays to enhance signal coverage, transmission performance, and reliability. This system is often used to extend the coverage of wireless communication networks, especially in large areas or complex environments, such as cities, mountains, or oceans. In a multi-relay system, a relay station is responsible for receiving signals from a source station or other relay stations and forwarding them to a target station or other relay stations. During the transmission process, the information is processed to ensure signal quality and transmission reliability. In addition, the efficiency of transmission can be improved by selecting a better relay path. Summary of the invention

[0004] In order to further enhance the transmission efficiency and covert rate of the multi-relay system, the present invention provides a covert communication system assisted by energy harvesting and a method for maximizing the covert rate. When the energy of the IoT device is limited, the energy harvesting jammer is used to assist the multi-relay system in achieving covert communication. This innovative method provides an important solution for the covert communication security in the multi-relay system.

[0005] A covert communication system based on energy harvesting assistance includes: a source site Alice, a destination Bob, an eavesdropper Willie, multiple relays Relay and multiple energy harvesting jammers EHJ;

[0006] The source site Alice sends public information or hidden information to the selected relay end Relay. The relay end Relay forwards the information to the destination Bob after receiving it. The eavesdropper Willie tries to detect whether the source site Alice has sent hidden information to the relay end Relay. The energy harvesting jammer EHJ absorbs the energy transmitted from the source site Alice and transmits interference signals to interfere with the eavesdropper Willie, affecting the eavesdropper Willie's detection of the hidden information sent by the source site Alice. It should be noted that the source site Alice first completes the task of sending public information and then sends hidden information. The power of sending public information and hidden information will be affected by the maximum power P max If the maximum power is exceeded, the eavesdropper Willie will detect the source site Alice's behavior of sending covert information.

[0007] The energy harvesting jammer EHJ (Energy harvesting Jammer, EH Jammer) transforms a passive jammer into a jammer with energy harvesting function by using wireless empowerment technology. In the present invention, the energy harvesting jammer EHJ collects the energy used by the source site Alice to send public information during the energy absorption phase and then releases it for interference during the energy release phase.

[0008] The system includes 6 channels: the channel from the source site Alice to the relay end Relay, the channel from the source site Alice to the energy harvesting jammer EHJ, the channel from the source site Alice to the eavesdropper Willie, the channel from the relay end Relay to the destination Bob, the channel from the energy harvesting jammer EHJ to the relay end Relay and the channel from the energy harvesting jammer EHJ to the eavesdropper Willie.

[0009] In the energy harvesting-assisted covert communication system, the subscripts a, b, w, r and j represent the source site Alice, the destination Bob, the eavesdropper Willie, the relay terminal Relay and the energy harvesting jammer EHJ, respectively. The channel fading vector coefficients between the source site Alice and the relay terminal Relay, the source site Alice and the energy harvesting jammer EHJ, the source site Alice and the eavesdropper Willie, the relay terminal Relay and the destination Bob, the energy harvesting jammer EHJ and the relay terminal Relay, and the energy harvesting jammer EHJ and the eavesdropper Willie are h respectively. ar 、h aj 、h aw 、h rb 、h jr and h jw ,|h x | 2is the channel gain, where x∈{ar,aj,aw,rb,jr,jw}.

[0010] As a preferred embodiment of the present invention, the channel coefficients of the system obey quasi-static Rayleigh fading, each channel remains constant in the same time slot T and the same channels in different time slots are independent of each other, the time is evenly divided into time slots of size T, and the energy collection technology with time switching is used to divide the time slot of size T into two sub-time slots, the first sub-time slot is the energy absorption stage, and the second sub-time slot is the energy release stage;

[0011] Energy absorption stage: In this stage, the source site Alice sends public information, and the relay terminal Relay receives the information and forwards it to the destination Bob. All energy harvesting jammers EHJ absorb the energy from the information transmitted by the source site Alice. The energy E absorbed by the energy harvesting jammer is

[0012]

[0013] in is the duration of the energy absorption phase, is the time switching factor, μ is the energy conversion efficiency, P a is the transmission power of the source site Alice in transmitting public information during the energy absorption phase;

[0014] Energy release phase: In this phase, the source site Alice sends hidden information on the basis of sending public information. The selected relay terminal Relay receives the information and forwards it to the destination Bob. At this time, when the public information and hidden information pass through the selected energy harvesting jammer EHJ, the energy harvesting jammer EHJ releases the energy E absorbed in the energy absorption phase as an interference signal to interfere with the eavesdropper Willie. The transmission power of the interference signal is P j :

[0015]

[0016] A method for maximizing concealment rate based on energy harvesting assistance includes the following steps:

[0017] (1) Establish a covert communication system based on energy harvesting assistance;

[0018] (2) Establish random selection model and intelligent selection model;

[0019] (3) Under the random selection model, when the source station Alice does not send a transmission of concealed information or the source station Alice sends a transmission of concealed information, the signal-to-noise ratio of the selected relay end is calculated based on the signal received by the selected relay end;

[0020] (4) Under the intelligent selection model, when the source site Alice does not send a concealed information transmission or the source site Alice sends a concealed information transmission, the signal-to-noise ratio of the selected relay end Relay is calculated based on the signal received by the selected relay end Relay;

[0021] (5) Based on the binary hypothesis test analysis of eavesdropper Willie under the random selection model and the intelligent selection model, the false alarm probability and missed detection probability of eavesdropper Willie in the two modes are derived respectively, and then the total detection error rate of eavesdropper Willie in the two modes is obtained respectively, and then the optimal detection threshold and the minimum detection error probability of eavesdropper Willie in the two modes are obtained respectively;

[0022] (6) Based on the concealment rate under the random selection model and the intelligent selection model, an optimization problem is proposed and solved.

[0023] As a preferred embodiment of the present invention, the random selection model is that in the covert communication system based on energy harvesting assistance, in the energy absorption stage, no covert communication occurs, and all energy harvesting jammers EHJ only absorb energy emitted from the source site Alice; in the energy release stage, one of the relay ends Relay and one of the energy harvesting jammers EHJ are randomly selected for covert communication.

[0024] As a preferred embodiment of the present invention, the intelligent selection model is that in the covert communication system based on energy collection assistance, in the energy absorption stage, no covert communication occurs, and all energy collection jammers EHJ only absorb the energy emitted from the source site Alice; in the energy release stage, the energy collection jammer with the maximum min{|h ar | 2 ,|h rb | 2} value of the relay end Relay and has the maximum (|h jw | 2 / |h jr | 2 ) value of energy harvesting jammer EHJ for covert communication.

[0025] As a preferred embodiment of the present invention, under the random selection model, when the source site Alice does not send the concealed information transmission, the calculation of the signal-to-noise ratio of the selected relay end Relay includes the following steps:

[0026] S31-1: Source site Alice uses P ar The power transmission public information, at this time the signal y received by the selected relay end Relay r for

[0027]

[0028] where x p [i] represents the public signal transmitted by the source site Alice in the i-th channel, x j [i] is the interference signal of the source site Alice in the i-th channel, Represents the noise received at the relay end, which has an expectation of 0 and a variance of The complex Gaussian distribution of is the noise power of the relay; P ar ≤P max , where P max is the maximum power transmitted by the source station Alice;

[0029] S31-2: Signal-to-noise ratio (SNR) of the selected relay r for

[0030]

[0031] When the source station Alice sends a hidden message transmission, the signal-to-noise ratio calculation of the selected relay includes the following steps:

[0032] S32-1: Source site Alice sends public information with power P ac Send hidden information, and the signal y received by the selected relay is r for

[0033]

[0034] Among them, x c [i] is the hidden information transmitted by the source site Alice in the i-th channel, x j [i] is the interference information of the source site Alice in the i-th channel, n r [i] is the noise of the selected relay end in the i-th channel, P ar +P ac ≤P max ;

[0035] S32-2: The selected relay node Relay gives priority to the public message x upon receiving the message from Alice. p Decode, then the hidden information x c is the noise, and the signal-to-noise ratio (SNR) of the selected relay r for

[0036]

[0037] As a preferred embodiment of the present invention, under the intelligent selection model, when the source site Alice does not send the transmission of concealed information, the signal-to-noise ratio calculation of the selected relay end Relay includes the following steps:

[0038] S41-1: When the source site Alice transmits only public information, the signal y received by the selected relay r for

[0039]

[0040] in is the power of the source site Alice sending only public information without sending hidden information under the intelligent selection model, η=2 R -1, R is the channel rate threshold for transmitting information from the source site Alice to the selected relay end Relay;

[0041] S41-2: Signal-to-noise ratio (SNR) of the selected relay r for

[0042]

[0043] When the source station Alice sends a hidden message transmission, the signal-to-noise ratio calculation of the selected relay includes the following steps:

[0044] S42-1: The source site Alice sends hidden information in addition to the public information. Then the signal y received by the selected relay is r for

[0045]

[0046] in is the power of sending public information when the source site Alice sends hidden information under the intelligent selection model,

[0047]

[0048] S42-2: Signal-to-noise ratio (SNR) of the selected relay r for

[0049]

[0050] As a preferred embodiment of the present invention, the calculation of the optimal detection threshold and the minimum detection error probability of the eavesdropper Willie under the random selection model specifically includes:

[0051] S51-1: Under the null hypothesis and alternative hypothesis of eavesdropper Willie, the received signal y of eavesdropper Willie w for:

[0052]

[0053]

[0054] Among them, H0 is the null hypothesis, indicating that the source site Alice does not transmit hidden information; H1 is the alternative hypothesis, indicating that the source site Alice transmits hidden information; W [i] represents the additive Gaussian white noise interference signal to which the eavesdropper Willie is subjected, n W The mean is 0 and the variance is The normal distribution of is the noise power of the eavesdropper Willie;

[0055] S51-2: Using the strong law of large numbers, we can get the received power T of the eavesdropper Willie. w for:

[0056]

[0057] S51-3: Using formulas (13) and (15), we can get the false alarm probability of the eavesdropper Willie: and the probability of missed detection

[0058]

[0059] τ is the signal detection threshold of the eavesdropper Willie, P j is the jammer power, x is |h aw | 2 , y is |h jw | 2 ;

[0060]

[0061] x is |h aw | 2 , y is |h jw | 2 ;

[0062] S51-4: False alarm probability by eavesdropper Willie and the probability of missed detection Get the total detection error probability φ of the eavesdropper Willie:

[0063]

[0064] S51-5: Obtain the optimal detection threshold τ of the eavesdropper Willie according to the total detection error probability φ of the eavesdropper Willie T ,

[0065]

[0066] The optimal detection threshold τ of the eavesdropper Willie T Substituting into formula (18), we get the optimal detection error probability φ of the eavesdropper Willie T .

[0067] As a preferred embodiment of the present invention, the calculation of the optimal detection threshold and the minimum detection error probability of the eavesdropper Willie under the intelligent selection model specifically includes:

[0068] S52-1: Under the null hypothesis and alternative hypothesis of the eavesdropper Willie, since the transmission power of the source site Alice in transmitting public information is affected by the channel gain, the received signal y of the eavesdropper Willie is w for:

[0069]

[0070] S52-2: Using the strong law of large numbers, we can get the received power T of the eavesdropper Willie. w for:

[0071]

[0072] S52-3: Using formulas (19) and (20), we can obtain the false alarm probability of the eavesdropper Willie: and the probability of missed detection

[0073]

[0074] in, x is |h aw | 2 , y is |h jw | 2 ;

[0075]

[0076]

[0077] β=(1+η)P ac ,x is |h aw | 2 , y is |h jw | 2 ;

[0078] S52-4: False alarm probability by eavesdropper Willie and the probability of missed detection Get the total detection error probability of the eavesdropper Willie:

[0079]

[0080] S52-5: Obtain the optimal detection threshold τ of Willie according to the total detection error probability φ of the eavesdropper Willie T ,

[0081]

[0082] The optimal detection threshold τ of the eavesdropper Willie T Substituting into formula (23), we get Willie’s optimal detection error probability φ T .

[0083] As a preferred embodiment of the present invention, the concealment rates κ under the random selection model and the intelligent selection model are calculated through steps (3)-(5) as follows:

[0084] Under the random selection model:

[0085]

[0086] Under the smart selection model:

[0087]

[0088] As a preferred embodiment of the present invention, under the random selection model, the optimization problem is:

[0089]

[0090] stφ T >1-ε (32b)

[0091] 0 <P ac ≤P max / (1+η) (32c)

[0092] Under the intelligent selection model, the optimization problem is:

[0093]

[0094] stφ T >1-ε (33b)

[0095]

[0096] As a preferred embodiment of the present invention, the optimization problem is solved by a two-dimensional array search algorithm to find the maximum concealment rate.

[0097] Compared with the prior art, the present invention has the following beneficial effects:

[0098] (1) The present invention utilizes the characteristics of the energy harvesting jammer, collects the signals emitted by the source and converts them into electrical energy, and then releases interference signals to interfere with the eavesdropper's judgment of the communication. Moreover, the addition of the energy harvesting jammer makes the system safer while saving a lot of energy. At the same time, the present invention proposes two model strategies for selecting energy harvesting jammers and relays, so that the entire system can apply appropriate strategies in different usage scenarios to improve the efficiency and performance of communication.

[0099] (2) The present invention uses covert communication signal hiding and camouflage technology to make it difficult for a third party to detect the existence of the communication behavior itself. This transmission method for protecting sensitive information and confidential data is particularly suitable for scenarios such as military and security fields that have strict requirements for the transmission of sensitive information. BRIEF DESCRIPTION OF THE DRAWINGS

[0100] Figure 1 This is a schematic diagram of the covert communication system based on energy harvesting assistance described in the present invention.

[0101] Figure 2 This is a comparison chart of the error detection probability of the eavesdropper Willie as the switching factor changes over time.

[0102] Figure 3 This is a comparison chart of the detection error probability of the eavesdropper Willie as the detection threshold τ changes.

[0103] Figure 4 The probability of error detection by the eavesdropper Willie varies with the hidden information transmission power P ac Change comparison chart.

[0104] Figure 5 The optimal detection error probability of the eavesdropper Willie varies with the interference power P j Change comparison chart.

[0105] Figure 6 is the concealment rate with the concealed information transmission power P ac Change comparison chart.

[0106] Figure 7 is the concealment rate with the interference power P j Change comparison chart.

[0107] Figure 8 This is a comparison chart of the optimal concealment rate as the concealment requirements change. DETAILED DESCRIPTION

[0108] In order to better illustrate the purpose, technical solutions and advantages of the present invention, the present invention will be further described below in conjunction with specific embodiments.

[0109] Example 1

[0110] A covert communication system based on energy harvesting assistance includes: a source site Alice, a destination Bob, an eavesdropper Willie, multiple relays Relay and multiple energy harvesting jammers EHJ;

[0111] The source site Alice sends public information or hidden information to the selected relay end Relay. The relay end Relay forwards the information to the destination Bob after receiving it. The eavesdropper Willie tries to detect whether the source site Alice has sent hidden information to the relay end Relay. The energy harvesting jammer EHJ absorbs the energy transmitted from the source site Alice and transmits interference signals to interfere with the eavesdropper Willie, affecting the eavesdropper Willie's detection of the hidden information sent by the source site Alice. It should be noted that the source site Alice first completes the task of sending public information and then sends hidden information. The power of sending public information and hidden information will be affected by the maximum power P max If the maximum power is exceeded, the eavesdropper Willie will detect the source site Alice's behavior of sending covert information.

[0112] The channel model used in the system is an independent quasi-static Rayleigh fading model. Each channel remains unchanged in the same time slot T and the same channels in different time slots are independent of each other. The channel coefficients are modeled as complex Gaussian random variables with zero mean and unit variance. The system includes 6 channels: the channel from the source site Alice to the relay end Relay, the channel from the source site Alice to the energy harvesting jammer EHJ, the channel from the source site Alice to the eavesdropper Willie, the channel from the relay end Relay to the destination Bob, the channel from the energy harvesting jammer EHJ to the relay end Relay, and the channel from the energy harvesting jammer EHJ to the eavesdropper Willie.

[0113] In the energy harvesting-assisted covert communication system, the subscripts a, b, w, r and j represent the source site Alice, the destination Bob, the eavesdropper Willie, the relay terminal Relay and the energy harvesting jammer EHJ, respectively. The channel fading vector coefficients between the source site Alice and the relay terminal Relay, the source site Alice and the energy harvesting jammer EHJ, the source site Alice and the eavesdropper Willie, the relay terminal Relay and the destination Bob, the energy harvesting jammer EHJ and the relay terminal Relay, and the energy harvesting jammer EHJ and the eavesdropper Willie are h respectively. ar 、h aj 、h aw 、h rb 、h jr and h jw ,|h x | 2 is the channel gain, where x∈{ar,aj,aw,rb,jr,jw}, and the signal loss on the path is included in the channel gain. It is assumed that the source site Alice knows |h ar | 2 and |h aj | 2 , the relay knows |h ar | 2 and |h rb | 2 , Energy Harvesting Jammer EHJ knows |h jr | 2 and |h jw | 2 , the eavesdropper Willie knows |h aw | 2 and |h jw | 2 Assume that the noise is with variance AWGN, and assuming that the bandwidth of the system is WMHZ, for the sake of loss of generality, the present invention assumes that W=1.

[0114] Assume that time is evenly divided into time slots of size T, and adopt an energy collection strategy model with time switching to divide the time slot of size T into two sub-time slots. The first sub-time slot is called the energy absorption stage, and the second sub-time slot is called the energy release stage. It is assumed that the relevant parameters of the energy harvesting jammer EHJ (such as energy conversion efficiency and energy collection time in each time slot) are known.

[0115] Example 2

[0116] A method for maximizing concealment rate based on energy harvesting assistance includes the following steps:

[0117] (1) Establish the covert communication system based on energy harvesting assistance as described in Example 1.

[0118] (2) In a covert communication system, choosing the right relay and energy harvesting jammer is a key decision that directly affects the performance and concealment of the communication. This paper proposes two different strategy models, namely the random selection model and the intelligent selection model, to optimize the system performance. At the same time, the propagation strategy model of the source site Alice under each strategy model is given.

[0119] The random selection model is a simple and intuitive method that randomly selects one of the relays and an energy harvesting jammer to complete the communication task. The advantage of this model is that it simplifies the decision-making process of the system and reduces the computational complexity. However, due to the complete randomness, the selection of some relays and energy harvesting jammers may not be ideal, thus affecting the communication quality and concealment. For example, when randomly selecting a relay, the channel quality from the source site Alice to the relay is not considered. If the received signal strength on the relay is less than the required threshold, a transmission interruption may occur.

[0120] The intelligent selection model optimizes the quality of the communication link by intelligently selecting relays and jammers. Specifically, the model first selects a relay that has the largest min{|h ar | 2 ,|h rb | 2 This step is to ensure that the selected relay has relatively good performance in terms of channel gain.

[0121] Next, based on the selected relay end, an energy harvesting jammer EHJ is selected, requiring (|h jw | 2 / |h jr | 2 ) is the largest, aiming to maximize the influence of the jammer on the eavesdropper Willie's judgment on whether the source site Alice has covert communication behavior, and minimize the influence on the relay receiving information, thereby improving the concealment of the system. In this way, the intelligent selection model not only ensures the quality of communication, but also pays more attention to improving the concealment of the system, making the communication more secure and reliable.

[0122] (3) Under the random selection model, when the source site Alice does not send a transmission of concealed information or the source site Alice sends a transmission of concealed information, the signal-to-noise ratio of the selected relay end is calculated based on the signal received by the selected relay end, specifically including:

[0123] S3-1: Transmission in which the source site Alice does not send hidden information under the random selection model: Consider the source site Alice with P ar The power transmission public information, where P ar ≤P max , where P max is the maximum power that Alice can transmit. At this time, the signal received at the selected relay (hereinafter referred to as RS) is

[0124]

[0125] where x p [i] represents the public signal transmitted by the source site Alice in the i-th channel, x j [i] is the interference signal of the source site Alice in the i-th channel, Represents the noise received at the relay end, which has an expectation of 0 and a variance of The complex Gaussian distribution of is the noise power of the relay; P ar ≤P max , where P max is the maximum power transmitted by the source station Alice.

[0126] Therefore, the signal-to-noise ratio at RS is expressed as

[0127]

[0128] in is the noise power at RS.

[0129] S3-2: When the source site Alice sends hidden information under the random selection model: Alice sends the public information with a power of P ac Send hidden information, then P ar +P ac ≤P max At this time, the signal y received by the selected relay (hereinafter referred to as RS) r for

[0130]

[0131] Among them, x c [i] is the hidden information transmitted by the source site Alice in the i-th channel, x j [i] is the interference information of the source site Alice in the i-th channel, n r [i] is the noise of the selected relay end in the i-th channel, P ar +P ac ≤P max ;

[0132] Since the system will give priority to the transmission of public messages, RS will give priority to public messages x when receiving information from the source site Alice. p Decoding will then convert the hidden information x c is regarded as noise, then the signal-to-noise ratio at RS is expressed as

[0133]

[0134] (4) Under the intelligent selection model, when the source site Alice does not send a concealed information transmission or the source site Alice sends a concealed information transmission, the signal-to-noise ratio of the selected relay end Relay is calculated based on the signal received by the selected relay end Relay, specifically including:

[0135] S4-1: In the intelligent selection model, communication is performed in a manner where public information transmission is uninterrupted, so the public information transmission power needs to be set according to the real-time channel gain.

[0136] In the intelligent selection model, the source site Alice does not send hidden information transmission: When the source site Alice only transmits public information, the signal received at the RS is

[0137]

[0138] in It is the power of sending public information when the source site Alice does not send hidden information under the intelligent selection model.

[0139] Signal-to-noise ratio (SNR) at RS r Expressed as

[0140]

[0141] S4-2: In the intelligent selection model, a non-interruption method is used to ensure the continuity and stability of the communication link. In this covert communication system, when the required signal threshold at RS is less than or equal to the channel rate from the source site Alice to RS, no interruption will occur, that is, R≤C ar , R is the channel rate threshold for transmitting information from the source site Alice to the selected relay, C ar is the channel rate from the source site Alice to RS, where C ar =log2(1+SNR r ), C ar Substitute the expression into R≤C ar In, get Where η = 2 R -1, because So set The value of

[0142]

[0143] S4-3, transmission of source site Alice without sending hidden information under the intelligent selection model: source site Alice sends hidden information in addition to sending public information, then the signal received at RS is

[0144]

[0145] in is the power of sending public information when the source site Alice sends hidden information under the intelligent selection model. Since the priority of public information transmission task is higher, the transmission of hidden information is first regarded as interference, so the signal-to-noise ratio at RS is

[0146]

[0147] S4-4, according to R≤C ar , we can get because get So set

[0148] Formula (12) adopts the non-interruption method according to the channel gain |h ar | 2 The size of Set to three different values, in channel gain |h ar | 2 In larger cases, When the channel gain |h ar | 2 In general, that is When , the system only transmits public information. When the channel gain is small, that is, No information is sent, so that the environment with high signal-to-noise ratio can be fully utilized to improve concealment. High channel gain means greater signal strength, making it easier for hidden information to be masked in noise, reducing the probability of being detected, thereby enhancing the security of communication. It can also save unnecessary energy and has strong adaptability. Since the range of channel gain when transmitting hidden information is After substitution, we can get P under the intelligent strategy. ac The value range is

[0149] (4) Based on the binary hypothesis test analysis of eavesdropper Willie under the random selection model and the intelligent selection model, the false alarm probability and missed detection probability of eavesdropper Willie in the two modes are derived respectively, and then the total detection error rate of eavesdropper Willie in the two modes is obtained respectively, and then the optimal detection threshold and the minimum detection error probability of eavesdropper Willie in the two modes are obtained respectively, including:

[0150] S4-1: Determine whether the source site Alice transmits hidden information through the hypothesis test of the eavesdropper Willie.

[0151] Under the null hypothesis of eavesdropper Willie (source site Alice does not transmit covert information) and the alternative hypothesis (source site Alice transmits covert information), the received signal at eavesdropper Willie is given by

[0152]

[0153] in represents the noise at the eavesdropper Willie.

[0154] Considering the radiometer of the eavesdropper Willie, the received power T of the eavesdropper Willie is w for

[0155]

[0156] Here n is the number of times the channel is used, τ is the signal detection threshold at the eavesdropper Willie, D0 and D1 represent decisions that are favorable to H0 and H1, respectively. Assume n→∞ and use the strong law of large numbers, T w for

[0157]

[0158] in is the noise power at the eavesdropper Willie.

[0159] S4-2: False Alarm Probability and the probability of missed detection Get the total detection error probability of the eavesdropper Willie, and They are:

[0160]

[0161] and

[0162]

[0163] From the above two formulas (16) and (17), it can be concluded that P j Need to meet P j <P ar , and because Therefore, under the random selection model, P j The value range is min means taking the smaller item in the brackets, μ is the energy conversion efficiency, is the time switching factor, T is the time of the unit time slot, P a is the power of Alice sending public information during the energy absorption phase, and the total detection error probability of the eavesdropper Willie is

[0164]

[0165] S4-3: Derivation of the optimal detection threshold based on the detection error probability of the eavesdropper Willie under the random selection model. It can be seen from formula (17) that the detection threshold of the eavesdropper Willie is When , the system's detection error probability is 1, which means that no matter whether the source site Alice sends covert information at this time, the eavesdropper Willie cannot correctly judge Alice's transmission behavior. The purpose of covert communication is φ>1-ε, where ε is the concealment requirement. When ε takes any value between 0 and 1, the total detection error probability φ must always be greater than 1-ε. Therefore, When , we need to prepare for the worst situation, that is, when the eavesdropper Willie obtains the optimal threshold, the total probability of detection error is used as a reference, and φ is derived with respect to τ to obtain

[0166]

[0167] Then we get the optimal detection threshold τ of the eavesdropper Willie T :

[0168]

[0169] The obtained τ T As the extreme point of φ, τ T Bring in It can be obtained that when τ>τ T hour, When τ<τ T hour, Therefore τ T is the minimum point, which is the optimal detection threshold of the eavesdropper Willie. T Substituting this into φ, we can get Willie's optimal detection error probability φ T .

[0170] S4-4: Under the intelligent selection model, since the transmission power of the source site Alice in transmitting public information is affected by the channel gain, the received signal at the eavesdropper Willie is expressed as

[0171]

[0172] According to formula (19), the received power T at the eavesdropper Willie is obtained by using the strong law of large numbers: w for

[0173]

[0174] S4-5: First calculate the false alarm probability and missed detection probability of the eavesdropper Willie, and get

[0175]

[0176]

[0177] Here and

[0178]

[0179] Here β = (1 + η)P ac , from the above two formulas (21) and (22), we know that α-P j >0, solving the inequality yields that when |h ar | 2 >η|h jr | 2 When P j The value range is When |h ar | 2 <η|h jr | 2 When P j The value range is Then we can get the total detection error rate of the eavesdropper Willie under the intelligent selection model:

[0180]

[0181] S4-6: Based on the detection error probability of the eavesdropper Willie in the intelligent selection model, the optimal detection threshold is derived, and φ is derived with respect to τ and equal to 0 to obtain

[0182]

[0183] Then we get the optimal detection threshold τ of the eavesdropper Willie T:

[0184]

[0185] Determine the obtained τ T Is it a minimum value? Take τ<τ T Bring in The value obtained is less than 0, then τ>τ T Bring in is greater than 0, proving that the obtained τ T The minimum point is the optimal detection threshold of Willie. T Substitute into φ to obtain the minimum detection error probability of the eavesdropper Willie is φ T .

[0186] (5) Based on the concealment rate under the random selection model and the intelligent selection model, an optimization problem is proposed and solved, including:

[0187] S5-1: Communication will not be interrupted only when the channel rate of information transmission from the source site Alice to the relay is greater than the rate threshold R required by the relay. The channel rate is given by Shannon's formula, so the probability that Alice's information transmission is not interrupted when H1 is true is for

[0188]

[0189] From the above formula (28), we know that P ar -ηP ac >0, and because P ar +P ac ≤P max , we get P under the random selection model ac The value range is P ac ∈(0,P max / (1+η)].

[0190] S5-2: The expression for the concealment rate is given by

[0191]

[0192] Among them, κ represents the concealment rate, It indicates the channel rate when the source site Alice transmits hidden information to the relay end Relay. It represents the probability that the source site Alice transmits information without interruption when H1 is true.

[0193] S5-3: Calculate the concealment rate under the random selection model. According to (28), the concealment rate is:

[0194]

[0195] S5-4: Calculate the concealment rate under the intelligent selection model. Since the source site Alice transmits public information without interruption, the concealment rate is the channel rate at which the source site Alice transmits concealed information:

[0196]

[0197] (6) Based on the concealment rate under the random selection model and the intelligent selection model, an optimization problem is proposed and solved, including:

[0198] S6-1: Under the random selection model, the optimization problem of concealment rate is proposed. The purpose of maximizing the concealment rate is to maximize the rate at which the source site Alice transmits concealed information to the relay end Relay while ensuring that the detection error probability of the eavesdropper Willie meets the security threshold required by the system. It can be expressed by the following optimization problem formula:

[0199]

[0200] stφ T >1-ε (32b)

[0201] 0 <P ac ≤P max / (1+η) (32c)

[0202]

[0203] On the basis of satisfying (32b), a simple two-dimensional numerical search for the optimal P ac and P j The value of can then be used to obtain the optimal concealment rate.

[0204] S6-2: Under the intelligent selection model, the concealment rate optimization problem is proposed. By giving the eavesdropper Willie's optimal detection error probability and P ac and P j The maximum concealment rate is obtained through a two-dimensional numerical search. The expression of the optimization problem is given by the following formula:

[0205]

[0206] stφ T >1-ε (33b)

[0207]

[0208] The simulation parameters in Example 2 are shown in Table 1.

[0209] Table 1. System simulation parameters

[0210] Parameter name Numeric Noise power at the eavesdropper Willie and the relay 0dB <![CDATA[The source site Alice transmits the maximum power P max > 16W The preset threshold of the signal rate on the channel: R 1Mbits / s / Hz

[0211] In order to more intuitively show the impact of the energy absorbed by the energy harvesting jammer EHJ on the entire system, it is assumed that all the energy absorbed by the energy harvesting jammer EHJ is used to release the interference signal, and the detection threshold τ of the eavesdropper Willie is set to 10W, and P is set to ac is 3W, and the power P of the public information sent by the source site Alice during the energy absorption phase is a Set to {4, 6, 8} W. From Figure 2 It can be seen that under the random selection model, the curve shows an increasing trend, which is due to the switching factor over time. As the time increases, the time for the energy harvesting jammer EHJ to absorb energy in a time slot T increases, and the time for releasing energy decreases. According to formula (2), this will result in an interference power P j The interference power P j The increase of will undoubtedly make it more difficult for the eavesdropper Willie to determine the occurrence of covert communication. When the energy absorption phase is constant, the power P of the public information sent by the source site Alice is a The higher the transmission power P, the higher the probability of detection error. a The more energy the energy harvesting jammer EHJ can absorb, the more it can be converted into jamming power P j In the intelligent selection model, the curve shows a trend of decreasing first and then increasing. This is because the time switching factor is small. When the detection error probability φ of the eavesdropper Willie is determined by the missed detection probability The dominant factor is the probability of missed detection, which can be seen from formula (22): is as the interference power P j With the increase of The increase in false alarm probability gradually becomes dominant, and the false alarm probability in formula (21) is as the interference power P j The curve gradually rises as the value of increases. When P a The increase of causes the detection error probability φ to decrease first and then increase.

[0212] Figure 3 The relationship between the detection error probability and the detection threshold of the eavesdropper Willie under different interference powers is explored in the two selection strategy models. Figure 3It can be seen that the probability of detection error in both selection strategy models increases first and then decreases as the detection threshold increases. Since the probability of detection error is determined by the false alarm probability P FA and the probability of missed detection P MD Composition, of which P FA is a monotonically decreasing function of the detection threshold, P MD is a monotonically increasing function of the detection threshold. When the detection threshold is small, τ has an effect on P FA The impact of P MD As τ increases further, the effect of τ on P FA The impact of P MD The influence of τ on P is relatively reduced, so the probability of detection error decreases rapidly from the beginning to a moderate level; after the value of τ exceeds the optimal detection threshold, τ has a relatively small influence on P MD The impact exceeds P FA , the curve shows an increasing trend. Under the two selection strategy models, P j ={1, 2, 3}W, the trend of the curve is determined by Figure 3 It can be seen that no matter which strategy model is used, when τ is small, the probability of detection error increases with P j This can be explained by the fact that the greater the interference signal strength released to the eavesdropper Willie, the more difficult it is for the eavesdropper Willie to detect the transmission of the hidden information. When τ is large, P j The increase of τ will make P MD The impact of P FA The influence of is relatively reduced, so the curve tends to be gentle. By taking the derivative of the detection error probability, we know that the optimal detection threshold increases with P j The lower point of the curve always increases with the increase of P. j By observing the curves of the two strategy models, we can see that the detection error probability under the random selection model is always greater than that of the intelligent selection model. This is because in the intelligent selection model, the source site Alice transmits public information in an uninterrupted manner. However, in the random selection model, in order to ensure that public information is transmitted with a low interruption probability, the transmission power of the public information needs to be increased, which leads to a greater detection error probability at the eavesdropper Willie. Therefore, the detection error probability under the random selection model is higher than that of the intelligent selection model.

[0213] Figure 4 The results of two selection models under different P j The relationship between the detection error probability and the concealed transmission power at the eavesdropper Willie under the condition of P. It is obvious that no matter which strategy model is selected, as P acWith the increase of the detection error probability, the probability of detection error always shows a decreasing trend. Under the condition that other conditions remain unchanged, the hidden information transmission power P ac The larger the value, the greater the possibility that the eavesdropper Willie can detect the hidden information, and therefore the smaller the probability of detection error. ac , the probability of detection error always increases with the interference power P j This is because when the concealed transmission power remains unchanged, the interference power P emitted by the energy harvesting jammer EHJ j The larger the value, the greater the impact on the eavesdropper Willie to detect hidden information. In the random selection model, as P ac The value of the detection error probability becomes larger and larger, and the value of the detection error probability tends to be stable. This is because in order to prevent interruption, it is necessary to increase P ac The transmission power of public information is increased on the basis of , which results in the detection error probability of the eavesdropper Willie remaining almost unchanged.

[0214] Figure 5 The optimal detection thresholds under the two selection strategy models are respectively introduced into the expression of detection error probability, aiming to explore the channel gain |h from the source site Alice to the relay end Relay ar | 2 The relationship between the optimal detection error probability and interference power at the eavesdropper Willie under different situations. Regardless of the strategy model, the optimal detection error probability at the eavesdropper Willie always increases with the interference power P. j According to the optimal detection threshold formula obtained by derivation, P j The increase of will lead to the increase of the optimal detection threshold, and the increase of the optimal detection threshold will lead to the increase of the probability of detection error, so the curve always shows an upward trend. Under the random selection strategy, when the interference power P j When the value of is fixed, |h ar | 2 The increase of will lead to a decrease in the optimal detection error probability. The reason for this is that when the channel gain from the source site Alice to the relay end Relay is good, there is no need to use a higher public information transmission power to ensure that the system will not interrupt the transmission of public information. Therefore, |h ar | 2 The increase in P ar Reduced, while the concealed information transmission power P ac unchanged, so the eavesdropper Willie will be more likely to detect the hidden information transmission behavior. Under the intelligent selection model, when the interference power P j When |h is small, the optimal detection threshold increases with ar | 2 decreases with the increase of jWhen |h is larger, the optimal detection threshold increases with ar | 2 The reason for this is that in the intelligent model selection, the public information transmission power P is determined in an uninterrupted manner. ar , according to formula (12), P ar The value of P j The value of increases, but it will increase with the increase of |h ar | 2 decreases with the increase of P j When P is smaller, ar Subject to |h ar | 2 The impact is greater, ar | 2 Increase leads to P ar decreases, which leads to a decrease in the optimal detection error probability; however, as P j The increase of P ar Affected by P j The influence of is gradually increasing, so the required public information transmission power P ar will become larger and larger, resulting in the optimal detection error probability increasing with |h ar | 2 By comparing the curves of the two strategy models, it is found that the optimal detection error probability under the random selection model is usually better than that under the intelligent selection model, but in some special cases, such as when P j When the value of is infinitely close to the maximum value in its value range, it will lead to a very large optimal detection error probability.

[0215] Figure 6 The channel gain h from the energy harvesting jammer EHJ to the relay is analyzed under two selection strategy models. jr | 2 = {0.3, 0.5, 0.7}, the concealment rate κ and the concealment information transmission power P ac Such as Figure 6 As shown in the random selection model, it can be observed that the curve first rises and then falls. The reason for the rise is that as P ac As P increases, the signal-to-noise ratio of the system improves, which means that the hidden information is easier to extract from the noise, thereby reducing the error rate in transmission and causing the concealment rate to increase. ac After rising to a certain value, due to the maximum transmission power P supported by the source site Alice max The public information transmission power P arwill become relatively small, which will greatly increase the probability of interruption of the source site Alice transmitting public information, and this increase in probability will have a negative impact on the concealment rate, so the concealment rate will decrease. Under the intelligent selection model, the concealment rate shows an increasing trend instead of decreasing like the random selection model because the public information is transmitted in an uninterrupted manner, so there will be no interruption, so the concealment rate continues to rise. Under the two strategy models, when P ac When the value of is fixed, |h jr | 2 The larger the value, the slower the concealment rate. jr | 2 The increase of |h means that the strength of the interference signal transmitted by the energy harvesting jammer EHJ received at the relay end is increasing, so the interference noise received by the relay end is stronger, which makes it more difficult to decode the hidden information. Therefore, when selecting an energy harvesting jammer, a smaller |h jr | 2 It can better improve the effect of covert communication. In general, the covert rate under the two strategy models is compared. The covert rate under the intelligent selection model is always higher than that under the random selection model.

[0216] Figure 7 Explore the channel gain from the source site Alice to the relay end Relay under two strategy models |h ar | 2 = {0.3, 0.5, 0.7} when the concealment rate κ and interference power P j relationship. Figure 7 It is obvious that κ is monotonically decreasing under both strategies. This is because the increase in interference power will undoubtedly reduce the signal-to-noise ratio at the relay end, making it more difficult for the relay end to extract hidden information from the numerous noises. j The last 3 different |h ar | 2 The effect of the value on κ can be observed in the random selection model, with three different |h ar | 2 It does not have a huge impact on the overall κ, but in the smart choice model, |h ar | 2 The higher the value, the larger κ is. This is because as |h ar | 2 As the power of the source station Alice increases, she only needs to transmit a smaller public information transmission power P ar To ensure that the system is not interrupted, this will lead to an increase in the signal-to-noise ratio of the hidden information at the relay end, thereby increasing κ. Therefore, when choosing an intelligent selection model, it is preferred to choose |h ar | 2Higher relay ends to improve the covert communication environment.

[0217] Figure 8 The solution of optimizing the hidden information transmission power and interference power at different |h jr | 2 The relationship between the optimal concealment rate and the concealment requirement under the random selection model. Under the random selection model, the curve shows a trend of first rising and then falling. The rise is due to the increase in the concealment requirement ε, which will lead to the optimal detection error probability φ T This will lead to the optimal interference power Reduction of and optimal hidden information transmission power The increase of the optimal interference power The optimal hidden information transmission power is has only a positive effect on the concealment rate κ, so The reduction and The increase of will increase the signal-to-noise ratio of the concealed information, leading to the increase of the optimal concealment rate, thus showing an upward trend; when the concealment requirement ε rises to a certain level, The continued increase of will make the public information transmission power become relatively small, and then the interruption probability will increase greatly. The negative impact of the interruption probability is better than the increase of and the decline The positive impact brought about by the curve begins to decline; when the concealment requirement ε rises to a certain value, P ar becomes extremely small, which will undoubtedly cause an interruption. When the interruption occurs, the concealment rate of the system is 0. Under the intelligent strategy, the curve shows an upward trend. This is because the optimal concealment information transmission power The increasing and optimal interference power The decrease of leads to an increase in the signal-to-noise ratio at the relay end, which in turn makes the optimal concealment rate κ T When the concealment requirement ε remains unchanged and |h jr | 2 When becomes larger, the optimal concealment rate κ under the two strategy models T All of them are showing a downward trend. This is due to the fact that |h jr | 2 The larger the value, the greater the interference of the energy harvesting jammer EHJ on the relay end Relay, so |h jr | 2 For the optimal concealment rate κ T There is a negative impact. jr | 2= 0.7 and ε is small, the optimal concealment rate of the random selection model is greater than that of the intelligent selection model. However, in fact, under the random selection model, due to the influence of the interruption probability, the source site Alice usually needs to use a larger public information transmission power P ar Transmitting public information is not suitable for systems with limited maximum transmission power.

[0218] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the essence and scope of the technical solution of the present invention.

Claims

1. A covert communication system based on energy harvesting assistance, characterized in that: include: A source site Alice, a destination Bob, an eavesdropper Willie, multiple relays Relay and multiple energy harvesting jammers EHJ; The system includes 6 channels: the channel from the source site Alice to the relay end Relay, the channel from the source site Alice to the energy harvesting jammer EHJ, the channel from the source site Alice to the eavesdropper Willie, the channel from the relay end Relay to the destination Bob, the channel from the energy harvesting jammer EHJ to the relay end Relay and the channel from the energy harvesting jammer EHJ to the eavesdropper Willie.

2. The covert communication system based on energy harvesting assistance as claimed in claim 1, characterized in that: The channel coefficient of the energy-harvesting-assisted covert communication system obeys quasi-static Rayleigh fading, each channel remains constant in the same time slot T and the same channels in different time slots are independent of each other, the time is evenly divided into time slots of size T, and the energy harvesting technology with time switching is used to divide the time slot of size T into two sub-time slots, the first sub-time slot is the energy absorption stage, and the second sub-time slot is the energy release stage; Energy absorption stage: In this stage, the source site Alice sends public information, and the relay terminal Relay receives the information and forwards it to the destination Bob. All energy harvesting jammers EHJ absorb the energy from the information transmitted by the source site Alice. The energy E absorbed by the energy harvesting jammer is in is the duration of the energy absorption phase, is the time switching factor, μ is the energy conversion efficiency, P a is the transmission power of the source site Alice in transmitting public information during the energy absorption phase, |h aj | 2 is the channel gain between the source site Alice and the energy harvesting jammer EHJ; Energy release phase: In this phase, the source site Alice sends hidden information on the basis of sending public information. The selected relay terminal Relay receives the information and forwards it to the destination Bob. At this time, when the public information and hidden information pass through the selected energy harvesting jammer EHJ, the energy harvesting jammer EHJ releases the energy E absorbed in the energy absorption phase as an interference signal to interfere with the eavesdropper Willie. The transmission power of the interference signal is P j :

3. A method for maximizing concealment rate based on energy harvesting assistance, characterized in that: The steps include: (1) Establishing a covert communication system based on energy harvesting assistance as described in claim 1 or 2; (2) Establish random selection model and intelligent selection model; (3) Under the random selection model, when the source station Alice does not send a transmission of concealed information or the source station Alice sends a transmission of concealed information, the signal-to-noise ratio of the selected relay end is calculated based on the signal received by the selected relay end; (4) Under the intelligent selection model, when the source site Alice does not send a concealed information transmission or the source site Alice sends a concealed information transmission, the signal-to-noise ratio of the selected relay end Relay is calculated based on the signal received by the selected relay end Relay; (5) Based on the binary hypothesis test analysis of eavesdropper Willie under the random selection model and the intelligent selection model, the false alarm probability and missed detection probability of eavesdropper Willie in the two modes are derived respectively, and then the total detection error rate of eavesdropper Willie in the two modes is obtained respectively, and then the optimal detection threshold and the minimum detection error probability of eavesdropper Willie in the two modes are obtained respectively; (6) Based on the concealment rate under the random selection model and the intelligent selection model, an optimization problem is proposed and solved.

4. The method for maximizing concealment rate based on energy collection assistance as claimed in claim 3, characterized in that: The random selection model is to randomly select one of the relay terminals Relay and one of the energy harvesting jammers EHJ for covert communication in the energy release phase in the covert communication system based on energy harvesting assistance; The intelligent selection model is to select the energy with the maximum min{|h ar | 2 ,|h rb | 2 } value of the relay end Relay and has the maximum (|h jw | 2 / |h jr | 2 ) value of energy harvesting jammer EHJ for covert communication; where |h ar | 2 is the channel gain between the source site Alice and the relay end Relay, |h rb | 2 is the channel gain between the relay and the destination Bob, |h jw | 2 is the channel gain between the energy harvesting jammer EHJ and the eavesdropper Willie, |h jr | 2 is the channel gain between the energy harvesting jammer EHJ and the relay end Relay.

5. The method for maximizing concealment rate based on energy collection assistance as claimed in claim 4, characterized in that: Under the random selection model, when the source site Alice does not send the hidden information transmission, the calculation of the signal-to-noise ratio of the selected relay end Relay includes the following steps: S31-1: Source site Alice uses P ar The power transmission public information, at this time the signal y received by the selected relay end Relay r for where x p [i] represents the public signal transmitted by the source site Alice in the i-th channel, x j [i] is the interference signal of the source site Alice in the i-th channel, Represents the noise received by the selected relay, which has an expectation of 0 and a variance of The complex Gaussian distribution of is the noise power of the selected relay; ar ≤P max , where P max is the maximum power transmitted by the source station Alice; h ar h is the channel fading vector coefficient between the source site Alice and the relay end Relay; jr is the channel fading vector coefficient between the energy harvesting jammer EHJ and the relay terminal Relay; S31-2: Signal-to-noise ratio (SNR) of the selected relay r for When the source station Alice sends a hidden message transmission, the signal-to-noise ratio calculation of the selected relay includes the following steps: S32-1: Source site Alice sends public information with power P ac Send hidden information, and the signal y received by the selected relay is r for Among them, x c [i] is the hidden information transmitted by the source site Alice in the i-th channel, x j [i] is the interference information of the source site Alice in the i-th channel, n r [i] is the noise of the selected relay in the i-th channel, P ar +P ac ≤P max ; S32-2: The selected relay terminal Relay gives priority to the public message x when receiving the information from the source station Alice. p Decode, then the hidden information x c is the noise, and the signal-to-noise ratio (SNR) of the selected relay r for 6. The method for maximizing concealment rate based on energy collection assistance as claimed in claim 4, characterized in that: Under the intelligent selection model, when the source site Alice does not send the concealed information transmission, the signal-to-noise ratio calculation of the selected relay end Relay includes the following steps: S41-1: When the source site Alice transmits only public information, the signal y received by the selected relay r for in is the power of the source site Alice sending only public information without sending hidden information under the intelligent selection model, η=2 R -1, R is the channel rate threshold for transmitting information from the source site Alice to the selected relay end Relay; S41-2: Signal-to-noise ratio (SNR) of the selected relay r for When the source station Alice sends a hidden message transmission, the signal-to-noise ratio calculation of the selected relay includes the following steps: S42-1: The source site Alice sends hidden information in addition to the public information. Then the signal y received by the selected relay is r for in The power of sending public information based on the hidden information sent by the source site Alice under the intelligent selection model, S42-2: Signal-to-noise ratio (SNR) of the selected relay r for 7. The method for maximizing concealment rate based on energy collection assistance as claimed in claim 4, characterized in that: Under the random selection model, the calculation of the best detection threshold and the minimum detection error probability of the eavesdropper Willie specifically includes: S51-1: Under the null hypothesis and alternative hypothesis of eavesdropper Willie, the received signal y of eavesdropper Willie w for: Among them, H0 is the null hypothesis, indicating that the source site Alice does not transmit hidden information; H1 is the alternative hypothesis, indicating that the source site Alice transmits hidden information; W [i] represents the additive Gaussian white noise interference signal to which the eavesdropper Willie is subjected, n W The mean is 0 and the variance is The normal distribution of is the noise power of the eavesdropper Willie, h aw is the channel fading vector coefficient between the source site Alice and the eavesdropper Willie, h jw is the channel fading vector coefficient between the energy harvesting jammer EHJ and the eavesdropper Willie; S51-2: Using the strong law of large numbers, we can get the received power T of the eavesdropper Willie. w for: S51-3: Using formulas (13) and (15), we can get the false alarm probability of the eavesdropper Willie: and the probability of missed detection τ is the signal detection threshold of the eavesdropper Willie, and x is |h aw | 2 , that is, the channel gain between the source site Alice and the eavesdropper Willie; y is |h jw | 2 , which is the channel gain between the energy harvesting jammer EHJ and the eavesdropper Willie; x is |h aw | 2 , y is |h jw | 2 ; S51-4: False alarm probability by eavesdropper Willie and the probability of missed detection Get the total detection error probability φ of the eavesdropper Willie: S51-5: Obtain the optimal detection threshold of the eavesdropper Willie according to the total detection error probability φ of the eavesdropper Willie The optimal detection threshold for the eavesdropper Willie Substituting into formula (18), we get the optimal detection error probability of the eavesdropper Willie:

8. The method for maximizing concealment rate based on energy collection assistance as claimed in claim 4, characterized in that: Under the intelligent selection model, the calculation of the optimal detection threshold and the minimum detection error probability of the eavesdropper Willie specifically includes: S52-1: Under the null hypothesis and alternative hypothesis of the eavesdropper Willie, since the transmission power of the source site Alice in transmitting public information is affected by the channel gain, the received signal y of the eavesdropper Willie is w for: S52-2: Using the strong law of large numbers, we can get the received power T of the eavesdropper Willie. w for: S52-3: Using formulas (19) and (20), we can obtain the false alarm probability of the eavesdropper Willie: and the probability of missed detection in, x is |h aw | 2 , y is |h jw | 2 ; β=(1+η)P ac ,x is |h aw | 2 , y is |h jw | 2 ; S52-4: False alarm probability by eavesdropper Willie and the probability of missed detection Get the total detection error probability of the eavesdropper Willie: S52-5: Obtain the optimal detection threshold of Willie based on the total detection error probability φ of the eavesdropper Willie The optimal detection threshold for the eavesdropper Willie Substituting into formula (23), we get Willie’s optimal detection error probability 9. The method for maximizing concealment rate based on energy collection assistance as claimed in claim 3, characterized in that: The concealment rates κ under the random selection model and the intelligent selection model are: Under the random selection model: It indicates the channel rate when the source site Alice transmits hidden information to the relay end Relay. represents the probability that the source site Alice transmits information without interruption when H1 is true; Under the smart selection model:

10. The method for maximizing concealment rate based on energy collection assistance as claimed in claim 9, characterized in that: Under the random selection model, the optimization problem is: 0 <P ac ≤P max / (1+n) (32c) Under the intelligent selection model, the optimization problem is: stφ T >1-e (33b) The optimization problem is solved by a two-dimensional array search algorithm to find the maximum concealment rate.

Images