Hidden transmission channel detection and access method based on optimal stopping theory

Abstract

The invention provides a hidden transmission channel detection and access method based on an optimal stopping theory. The method comprises the following steps: step 1, setting a multi-channel detection and hidden transmission model; step 2, determining a concealment index of concealed user transmission; step 3, determining that the hidden user is the maximum transmitting power meeting the hidden transmission condition; 4, setting a channel detection and transmission strategy of the hidden user; step 5, analyzing to obtain a hidden user transmission validity index; step 6, modeling a hidden user channel detection process by using an optimal stopping theory; and step 7, adopting a one-step forward exhibition channel detection criterion to carry out solution with low computational complexity, and carrying out covert channel detection and access based on an optimal stop theory. According to the invention, user channel estimation and information transmission scenes in multi-channel hidden transmission can be well described, and an effective signal detection and access method is provided.

Description

Technical Field

[0001] The present application relates to the field of wireless communication technology, and in particular to a covert transmission channel detection and access method based on optimal stopping theory. Background Art

[0002] In the field of wireless communications, the communicating parties transmit signals through electromagnetic waves. However, the open propagation characteristics of electromagnetic waves in space make wireless transmission easy to be detected, eavesdropped, and even attacked by the outside world. For example, after locating and capturing mobile phone signals, fake base stations can send false messages to them, threatening user privacy and property safety; in communication confrontation, once our electromagnetic signals are discovered by the enemy, they may be located, interfered with, and attacked by firepower. Therefore, hiding electromagnetic signals in the surrounding electromagnetic environment and making it difficult for the enemy to detect the existence of communication signals can provide security for wireless information transmission. Wireless covert communication technology uses the enemy's uncertainty about the communication system to make it impossible for the enemy to determine whether information transmission exists, thereby hiding the communication signal in the electromagnetic environment.

[0003] Existing research on public signal-assisted covert transmission mainly focuses on single-channel scenarios, and has not yet involved the impact of the channel estimation time overhead before covert user transmission in a multi-channel system on the covert transmission throughput and the joint design of covert user channel estimation and covert transmission strategy. Summary of the invention

[0004] The present application provides a covert transmission channel detection and access method based on the optimal stopping theory, which can be used to solve the defects of the existing scheme that the channel detection time is too long, reducing the information transmission time, and the channel detection time is too short, reducing the channel diversity gain.

[0005] The present application provides a covert transmission channel detection and access method based on optimal stopping theory, the method is as follows:

[0006] Step 1: Set up the multi-channel detection and covert transmission model.

[0007] The multi-channel detection and covert transmission model includes a pair of hidden users, M pairs of public communication users and M detectors, where each pair of public communication users uses a channel, and the channels are isomorphic and mutually orthogonal; the public communication user is an authorized user who is allowed to transmit information by the detected person, and the detector does not allow the hidden user to use the channel to transmit information; each detector is responsible for detecting a channel, the purpose is to find out whether the hidden user is transmitting on the channel it is responsible for; the hidden user performs channel estimation before transmitting information, and selects a channel to transmit information based on the channel estimation result; the hidden user can only estimate one channel at a time, and estimating the channel takes time. The longer the channel estimation time, the shorter the time left for information transmission; the optimal stopping theory modeling of the hidden user detection and transmission process is used to obtain the optimal number of channel detections before the hidden user transmits information to maximize its average transmission throughput.

[0008] Step 2: Determine the concealment index of the concealed user transmission.

[0009] For the i-th detector in the system, it determines whether the hidden user is transmitting on the channel it is responsible for by the average received power in a time slot. The average received power is expressed as shown in formula (1):

[0010]

[0011] in and denote the event that the hidden user does not transmit information on channel i and transmits information on channel i, respectively, P l and P c are the transmission powers of the public communication user and the hidden user respectively, is the receiver white noise power of detector i, and are the channel gains from the public communication user and the hidden user to the detector i, respectively, which obey the negative exponential distribution; the decision criterion of the i-th detector is:

[0012]

[0013] where τ i is the detection threshold of detector i, and are the judgment results of whether the detector i believes there is hidden information transmission or not;

[0014] Assuming that the detector is unknown to the instantaneous channel gain from the public communication user to itself, there will be two kinds of errors in the detection: false alarm and missed detection; false alarm means that the hidden user transmits information on channel i, but the detector i believes that the hidden user does not transmit on the corresponding channel, and missed detection means that the hidden user does not transmit information on channel i, but the detector i believes that the hidden user transmits information on the corresponding channel; the probabilities of the two kinds of erroneous judgments are expressed as follows:

[0015]

[0016] The detection performance of detector i is expressed as the detection error probability ξ i express:

[0017]

[0018] In order to minimize the probability of detection error, detector i adjusts its own detection threshold to minimize the probability of detection error. The optimal detection threshold is:

[0019]

[0020] When the detector uses the optimal threshold, the corresponding minimum probability of wrong judgment is When the minimum probability of wrong decision meets the following conditions, the information transmission is said to meet the confidentiality requirements:

[0021]

[0022] Where ∈ is the maximum exposure probability that the hidden user can tolerate, that is, the probability that the detected person is accurately detected.

[0023] Step 3: Determine the maximum transmission power of the hidden user that meets the hidden transmission conditions.

[0024] To meet the concealment requirement (6), the concealed user controls its own transmission power value not to exceed the maximum value P max , the maximum value determination process is as follows: According to the detection performance of detector i, the optimal detection threshold and the corresponding minimum detection error probability of detector i are obtained using the detection error probability formula (4) and the optimal detection threshold formula (5):

[0025]

[0026] in λ1=λ pw / P l , λ pw and λ cw They are and The inverse of the mean; denoted by θ i=λ1 / λ2. Since the transmission power of the concealed user is usually less than the transmission power of the external RF signal, only λ1<λ2, i.e. 0<θ i <1; in this case, the detector has the minimum probability of error in detection With concealed transmission power The first-order derivative expression of is:

[0027]

[0028] in Since the equation The only solution is θ1 = 1, so in the considered interval about The first-order derivative of cannot be zero, that is, the minimum error decision probability is monotonous with the concealed user transmit power; and θ i = 0.5, the first-order derivative is less than 0, so the minimum detection error probability of the detector decreases monotonically with the hidden user's transmission power. Therefore, the binary method is used to obtain P that meets the accuracy requirements. max value.

[0029] Step 4: Set the channel detection and transmission strategy for the hidden user.

[0030] Before transmitting the concealed information, the concealed receiver sends a pilot signal to estimate the channel. After the concealed sender receives the pilot signal from the concealed sender, it knows the instantaneous channel gain from itself to the receiver. The benefit of this operation is that the hidden sender does not need to send a pilot signal, so that the detector cannot obtain the instantaneous channel gain from the hidden sender to itself, which improves the concealment of information transmission; if the hidden sender transmits information on channel i, it is based on The instantaneous value of the concealed user adjusts its own transmission power so that the concealed signal power received by the concealed receiving end is a constant value; the concealed user transmission power is as follows:

[0031]

[0032] Where Q is the constant hidden signal power value received by the hidden receiver; therefore, after the hidden user detects channel i, the signal-to-interference-to-noise ratio γ when transmitting information on the corresponding channel i as follows:

[0033]

[0034] Step 5: Analyze and obtain the concealed user transmission effectiveness index.

[0035] When the hidden user starts to transmit information after detecting n channels, it selects the channel with the largest signal-to-interference-noise ratio among the detected channels for transmission. The normalized transmission throughput that can be achieved is as follows:

[0036]

[0037] Where τ is the ratio of the time cost of a hidden user probing a channel to the entire time slot, R n The unit is nit / second / Hz. From formula (11), we know that when the number of detected channels is small, the number of available channels is limited, resulting in A smaller value limits the improvement of throughput. When the number of channels detected is large, The value is large, and (1-nτ) is small because it takes more time to detect the channel, which also limits the transmission throughput. In order to improve the covert throughput, the covert user needs to make a reasonable decision on the number of detection channels before information transmission.

[0038] Step 6: Use the optimal stopping theory to model the covert user channel detection process.

[0039] The optimal stopping theory is applicable to a class of problems that involve selecting the right time to take action on a variable that can be observed sequentially, and is applicable to solving the problem considered in the present invention. The optimal stopping problem has two elements: the first element is a discrete sequence of random variables, where the specific value of each random variable is unknown before observation, but the joint distribution is known; the second element is the reward that can be obtained after each observation; in the present invention, the signal-to-interference-to-noise ratio value of the hidden user on M channels is the sequence of random variables to be observed. After completing each observation, the hidden user receives the reward of the achievable transmission throughput

[0040] Before executing step 7, it should be noted that in the prior art, adopting the analytical reverse recursive method to solve the present application is a common method, but next, the present application will specifically analyze the reasons why the analytical reverse recursive method is difficult to apply to the present problem.

[0041] Since the communication system has only M channels, the hidden user can detect at most M times before taking action, so the channel detection and access decision problem considered is a finite-dimensional optimal stopping problem; the reverse recursive method is used to solve the finite-dimensional optimal stopping problem: first assume that the hidden user has completed the detection of channel M-1. At this time, the user only needs to decide whether to continue exploring channel M based on the observed signal-to-interference-noise ratio sequence; after obtaining the optimal decision result of the hidden user on channel M-1, go back a step to consider the optimal decision of the hidden user on channel M-2. After obtaining the optimal decision result on channel M-2, go back a step to channel M-3 and obtain the optimal decision result of the hidden user on this channel, until going back to the first channel, the optimal decision result of the hidden user on each channel can be given;

[0042] The reverse recursive process is defined as:

[0043]

[0044] The achievable throughput after the covert user has detected all channels; at the same time, define:

[0045]

[0046] is the maximum covert throughput that the covert user can achieve starting from probing channel n. After probing channel n and obtaining the signal to noise ratio on the corresponding channel, the covert user calculates the transmission throughput that can be obtained by stopping probing and transmitting information. and the expected transmission throughput of continuing to probe the channel and then transmitting information According to the optimal stopping criterion, if the achievable throughput of stopping detection and transmitting on the current channel is not lower than the expected throughput of continuing detection and then transmitting information, the hidden user stops detecting the channel and starts hidden transmission, otherwise it continues to detect the channel and then continues to compare; the optimal number of channel detections before the hidden user transmits information given by the reverse recursive method is expressed as:

[0047]

[0048] Although the optimal number of channel probes before the hidden user transmission obtained by the reverse recursive method is expressed as above;

[0049] The calculation of N is very complicated, which leads to the opt The exact value of is very difficult; in particular, in order to calculate The hidden user needs to first calculate the sequence All values ​​in , and because the signal to noise ratio of the undetected channel will affect the decision of the hidden user on the current channel, the impact of the subsequent elements on the current element needs to be considered when calculating each element in the sequence. The above calculation process is a typical dynamic programming problem, which will produce great computational complexity and is difficult to implement in the actual calculation process. Therefore, the traditional reverse recursive method is difficult to apply to the optimal stopping problem established in this chapter, and a low-complexity method for calculating the optimal number of channel detections is urgently needed.

[0050] Step 7, adopting the one-step forward prospect channel detection criterion to perform a low computational complexity solution, and performing covert channel detection and access based on the optimal stopping theory.

[0051] Step 71, one-step forward prospect channel detection criterion advantage analysis: To reduce computational complexity, this application adopts one-step forward prospect (1-SLA) channel detection criterion, the optimality of which can be guaranteed in the problem under consideration. The 1-SLA criterion is also known as the short-sighted criterion. Under this criterion, the hidden user only needs to compare the current achievable throughput and the expected achievable throughput of continuing to detect the next channel and then transmitting when making a decision; the optimal number of channel detections given by the 1-SLA criterion before the hidden user transmits information is:

[0052]

[0053] in is the minimum value of the mutual interference link gain of the public communication user to the hidden user in the first n channels, is the probability that the power required for concealed user transmission on a certain channel is lower than the maximum value; from formula (15), we can see that About The constant, calculated Only the first integral needs to be considered, and the calculation process is as follows:

[0054]

[0055] remember in represents the exponential integral function, then the result in formula (15) is as follows:

[0056]

[0057] in

[0058] The timing to stop channel detection given by the 1-SLA criterion is the first time that the current achievable throughput is not less than the expected throughput of continuing to detect the next channel and then transmitting. At this time, stopping channel detection and accessing the channel with the highest signal-to-interference-noise ratio can maximize the average throughput.

[0059] Step 72, covert channel detection and access based on optimal stopping theory, includes the following steps:

[0060] Initialization process: number the channels in order from 1 to M; and perform the following loop, each iteration i;:

[0061] ,Step 721, the hidden user sends a pilot signal on channel i to perform channel detection and obtains the signal-to-interference-to-noise ratio γ transmitted on the channel i ;

[0062] Step 722, if the values ​​of i and M are equal, the loop ends, otherwise it goes to step 723;

[0063] Step 723: Calculate the achievable throughput R of the hidden user after detecting channel i according to formula (11): i ;

[0064] Step 724, calculate the expected throughput of the hidden user continuing to detect the next channel and then accessing the channel according to formula (15):

[0065] Step 5: If the hidden user can currently reach a throughput of R i No less than the expected throughput of probing the next channel and then accessing The loop ends;

[0066] Output the optimal number of channel detections before the hidden user transmits and the corresponding throughput

[0067] Step 8, prove that the proposed algorithm can maximize the average transmission throughput.

[0068] According to the optimal stopping theory, the 1-SLA criterion is optimal in the finite-dimensional monotone optimal stopping problem, so it is only necessary to prove that the optimal stopping problem under consideration is monotone. n For events If the condition If , then the optimal stopping problem is monotonic. Define the function:

[0069]

[0070] in is the complement of the cumulative distribution function of the signal-to-interference-to-noise ratio of the hidden user on channel n+1; observe and The expression shows that about Strictly increasing, Strictly decreasing with respect to n; The integrand in the integral is positive and The increase of will lead to a shorter integration interval, so about is also strictly decreasing. and The difference is as follows:

[0071]

[0072] because According to monotonicity, and Therefore That is, if satisfy There must be Therefore, there is Therefore, the optimal stopping problem in the considered scenario is monotonic, and it is proved that the 1-SLA criterion is the optimal criterion for maximizing the average throughput.

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

[0074] The concealed receiving end is used to send a pilot detection channel to prevent the detector from obtaining the instantaneous gain of the detection channel, thereby improving the transmission concealment of the concealed user.

[0075] The model is complete and the physical meaning is clear. The proposed covert transmission channel detection and access algorithm based on optimal stopping theory can achieve a reasonable compromise between channel diversity gain and channel detection time overhead, and can obtain the number of channel detections that maximizes the average throughput of covert users.

[0076] The proposed algorithm provides a solution to the channel estimation and covert transmission problems under the condition of unknown instantaneous channel information. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 Model diagram of the multi-channel covert transmission model provided in the embodiment of the present application

[0078] Figure 2 A schematic diagram of the channel usage and time slot structure of hidden users of the multi-channel covert transmission model provided in an embodiment of the present application.

[0079] Figure 3 A graph showing the variation of the normalized covert transmission throughput with the number of system channels that can be achieved by the covert channel detection and access algorithm based on the optimal stopping theory provided in an embodiment of the present application. DETAILED DESCRIPTION

[0080] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0081] The covert channel detection and access algorithm based on the optimal stopping theory proposed in the present invention aims to provide a solution for a multi-channel covert transmission system to achieve transmission concealment under the condition of unknown instantaneous channel information while maximizing the average transmission throughput. In terms of ensuring transmission concealment, during channel detection, the concealed receiving end sends a pilot channel to estimate the channel, so that the detector cannot obtain the instantaneous gain of the detected channel, and the concealed sender controls its own transmission power through the channel estimation result; in terms of maximizing the average transmission throughput, a 1-SLA channel estimation and access algorithm is proposed using the optimal stopping theory. Each time the concealed user estimates a channel, it compares the current achievable throughput with the expected throughput of detecting the next channel and then accessing. Once the former is not less than the latter, the channel detection ends and information transmission begins.

[0082] The present invention will be further described below according to embodiments:

[0083] Figure 1 Schematic diagram of a multi-channel covert transmission system. In this model, M pairs of public communication users occupy M channels, where the channels are orthogonal to each other. The covert user can choose a channel to transmit covert information under the cover of the public communication signal. There is a detector on each channel to determine whether there is covert information being transmitted on the current channel. The covert user does not know the channel state information before transmission, so it is necessary to detect the channel to access a channel with higher gain before transmission. However, detecting the channel will incur time overhead and shorten the subsequent information transmission time. Therefore, the covert user must choose the right time to stop channel detection.

[0084] Figure 2 (a) is a schematic diagram of the channel usage of the system when the hidden user transmits, where each pair of public communication users uses one channel, and the hidden user and the public communication user share a channel to transmit information. (b) is a schematic diagram of the hidden user time slot structure. The hidden user detects n channels before transmitting information. The time overhead generated by channel detection is nτT, and the time left for information transmission is (1-nτ)T.

[0085] The embodiment of the present invention is described in detail as follows. The system simulation is based on MATLAB software, and the parameters are set to the transmission power P of the public communication user. l is 30dBm, the mean value of all Rayleigh channels is 1. To ensure the concealment of transmission, the maximum transmission power of the concealed user is P max is 23dBm, the additive white noise power σ 2 is 0dBm, and the ratio τ of the time required for a hidden user to detect a channel to the entire time slot is 0.03.

[0086] from Figure 3 It can be seen that as the number of system channels increases, the normalized throughput of the proposed algorithm increases. Compared with comparison algorithms 1 and 2, the proposed algorithm can achieve higher normalized concealed throughput, and as the number of channels increases, the advantages of the proposed algorithm become more obvious. Among them, comparison algorithm 1 is that the concealed user detects all channels before transmission and then accesses the channel with the highest signal-to-noise ratio for transmission. The implementation process of comparison algorithm 2 is as follows: the concealed user first detects half of the channels before transmission, and then records the maximum signal-to-noise ratio γ1 in this half of the channels; then the concealed user continues to detect the channel, and once the signal-to-noise ratio of the subsequent channel is higher than γ1, it accesses the channel to transmit information, otherwise the concealed user continues to detect the channel until all channels are detected.

[0087] The present invention is described above by way of illustration of the application embodiments. Those skilled in the art should understand that the present disclosure is not limited to the embodiments described above, and various changes, modifications and substitutions may be made without departing from the scope of the present invention.

[0088] The above-described embodiments of the present application do not constitute a limitation on the protection scope of the present application.

Claims

1. A covert transmission channel detection and access method based on optimal stopping theory, characterized in that: The method comprises: Step 1, setting up a multi-channel detection and covert transmission model; Step 2, determining the concealment index of concealed user transmission; Step 3, determining the maximum transmission power of the hidden user to meet the hidden transmission conditions; Step 4, setting the channel detection and transmission strategy of the hidden user; Step 5, analyzing and obtaining the concealed user transmission effectiveness index; Step 6, modeling the covert user channel detection process using the optimal stopping theory; Step 7, adopting the one-step forward prospect channel detection criterion to perform a low computational complexity solution, and performing covert channel detection and access based on the optimal stopping theory.

2. The method according to claim 1, characterized in that Step 1: Setting up a multi-channel detection and covert transmission model, including: The multi-channel detection and covert transmission model includes a pair of hidden users, M pairs of public communication users and M detectors, where each pair of public communication users uses a channel, and the channels are isomorphic and mutually orthogonal; the public communication user is an authorized user who is allowed to transmit information by the detected person, and the detector does not allow the hidden user to use the channel to transmit information; each detector is responsible for detecting a channel, the purpose is to find out whether the hidden user is transmitting on the channel it is responsible for; the hidden user performs channel estimation before transmitting information, and selects a channel to transmit information based on the channel estimation result; the hidden user can only estimate one channel at a time; the optimal stopping theory modeling of the hidden user detection and transmission process is used to obtain the optimal number of channel detections before the hidden user transmits information to maximize its average transmission throughput.

3. The method according to claim 2, characterized in that Step 2, determine the concealment index of concealed user transmission, including: For the i-th detector in the system, it determines whether the hidden user is transmitting on the channel it is responsible for by the average received power in a time slot. The average received power is expressed as shown in formula (1): in and denote the event that the hidden user does not transmit information on channel i and transmits information on channel i, respectively, P l and P c are the transmission powers of the public communication user and the hidden user respectively, is the receiver white noise power of detector i, and are the channel gains from the public communication user and the hidden user to the detector i, respectively, which obey the negative exponential distribution; the decision criterion of the i-th detector is: where τ i is the detection threshold of detector i, and are the judgment results of whether the detector i believes there is hidden information transmission or not; Assuming that the detector is unknown to the instantaneous channel gain from the public communication user to itself, there will be two kinds of errors in the detection: false alarm and missed detection; false alarm means that the hidden user transmits information on channel i, but the detector i believes that the hidden user does not transmit on the corresponding channel, and missed detection means that the hidden user does not transmit information on channel i, but the detector i believes that the hidden user transmits information on the corresponding channel; the probabilities of the two kinds of erroneous judgments are expressed as follows: The detection performance of detector i is expressed as the detection error probability ξ i express: In order to minimize the probability of detection error, detector i adjusts its own detection threshold to minimize the probability of detection error. The optimal detection threshold is: When the detector uses the optimal threshold, the corresponding minimum probability of wrong judgment is When the minimum probability of wrong decision meets the following conditions, the information transmission is said to meet the confidentiality requirements: Where ∈ is the maximum exposure probability that the hidden user can tolerate, that is, the probability that the detected person is accurately detected.

4. The method according to claim 3, characterized in that Step 3, determining the maximum transmission power of the concealed user to meet the concealed transmission conditions, including: To meet the concealment requirement (6), the concealed user controls its own transmission power value not to exceed the maximum value P max , the maximum value determination process is as follows: According to the detection performance of detector i, the optimal detection threshold and the corresponding minimum detection error probability of detector i are obtained using the detection error probability formula (4) and the optimal detection threshold formula (5): in λ1=λ pw / P l , λ pw and λ cw They are and The inverse of the mean; denoted by θ i =λ1 / λ2. Since the transmission power of the concealed user is usually less than the transmission power of the external RF signal, only λ1<λ2, i.e. 0<θ i <1; in this case, the detector has the minimum probability of error in detection With concealed transmission power The first-order derivative expression of is: in Since the equation The only solution is θ1 = 1, so in the considered interval about The first-order derivative of cannot be zero, that is, the minimum error decision probability is monotonous with the concealed user transmit power; and θ i = 0.5, the first-order derivative is less than 0, so the minimum detection error probability of the detector decreases monotonically with the hidden user's transmission power. Therefore, the binary method is used to obtain P that meets the accuracy requirements. max value.

5. The method according to claim 4, characterized in that Step 4: Set the channel detection and transmission strategy for the hidden user. Before transmitting the concealed information, the concealed receiver sends a pilot signal to estimate the channel. After the concealed sender receives the pilot signal from the concealed sender, it knows the instantaneous channel gain from itself to the receiver. The instantaneous value of; if the hidden sender transmits information on channel i, it is based on The instantaneous value of the concealed user adjusts its own transmission power so that the concealed signal power received by the concealed receiving end is a constant value; the concealed user transmission power is as follows: Where Q is the constant hidden signal power value received by the hidden receiver; therefore, after the hidden user detects channel i, the signal-to-interference-to-noise ratio γ when transmitting information on the corresponding channel i as follows:

6. The method according to claim 5, characterized in that Step 5: Analyze and obtain the concealed user transmission effectiveness indicators, including: When the hidden user starts to transmit information after detecting n channels, it selects the channel with the largest signal-to-interference-noise ratio among the detected channels for transmission. The normalized transmission throughput that can be achieved is as follows: Where τ is the ratio of the time cost of a hidden user probing a channel to the entire time slot, R n The unit is nit / second / Hz. From formula (11), we know that when the number of detected channels is small, the number of available channels is limited, resulting in A smaller value limits the improvement of throughput. When the number of channels detected is large, The value is large, and (1-nτ) is small because it takes more time to detect the channel, which also limits the transmission throughput. In order to improve the covert throughput, the covert user needs to make a reasonable decision on the number of detection channels before information transmission.

7. The method according to claim 6, characterized in that Step 6, using the optimal stopping theory to model the covert user channel detection process, including: The optimal stopping problem has two elements: the first element is a sequence of discrete random variables, the specific value of each random variable is unknown before observation, but the joint distribution is known; the second element is the reward that can be obtained after each observation; the signal-to-interference-to-noise ratio value of the hidden user on M channels is the sequence of random variables to be observed. After completing each observation, the hidden user receives the reward of the achievable transmission throughput 8. The method according to claim 7, characterized in that Step 7, using the one-step forward prospect channel detection criterion to perform a low computational complexity solution, and performing covert channel detection and access based on the optimal stopping theory, including: Step 71, the 1-SLA criterion is also called the short-sighted criterion. Under this criterion, the hidden user only needs to compare the current achievable throughput and the expected achievable throughput of continuing to detect the next channel and then transmitting when making a decision. The optimal number of channel detections before the hidden user transmits information given by the 1-SLA criterion is: in is the minimum value of the mutual interference link gain of the public communication user to the hidden user in the first n channels, is the probability that the power required for concealed user transmission on a certain channel is lower than the maximum value; from formula (15), we can see that About The constant, calculated Only the first integral needs to be considered, and the calculation process is as follows: remember in represents the exponential integral function, then the result in formula (15) is as follows: in The timing to stop channel detection given by the 1-SLA criterion is the first time that the current achievable throughput is not less than the expected throughput of continuing to detect the next channel and then transmitting. At this time, stopping channel detection and accessing the channel with the highest signal-to-interference-noise ratio can maximize the average throughput. Step 72, covert channel detection and access based on optimal stopping theory, includes the following steps: Initialization process: number the channels in order from 1 to M; and perform the following loop, each iteration i;: Step 721: The hidden user sends a pilot signal on channel i to perform channel detection and obtains the signal-to-interference-to-noise ratio γ transmitted on the channel. i ; Step 722, if the values ​​of i and M are equal, the loop ends, otherwise it goes to step 723; Step 723: Calculate the achievable throughput R of the hidden user after detecting channel i according to formula (11): i ; Step 724, calculate the expected throughput of the hidden user continuing to detect the next channel and then accessing the channel according to formula (15): Step 5: If the hidden user can currently reach a throughput of R i No less than the expected throughput of probing the next channel and then accessing The loop ends; Output the optimal number of channel detections N before the hidden user transmits osp =max j＝1,2,...,i γ j and the corresponding throughput

Images