Evolutionary game spectrum sensing method based on blockchain trust mechanism

Abstract

The application relates to an evolution game spectrum sensing method based on a blockchain trust mechanism, and belongs to the technical field of mobile communication, which comprises the following steps: FC selects nearby SUs for CSS according to the geographical position of PU; if the SU participates in the CSS for the first time, the SU needs to register authentication information to the FC, the FC verifies the SU, registers the SU as a legal participant, and allocates an identity key and an initial trust score to the SU; each SU performs spectrum sensing on the PU channel, and after completing the spectrum sensing, the SU selects to send a real or false local decision result to the FC according to the profit of the strategy; after receiving the local decision result of the SU, the FC makes a global decision on the PU channel according to a decision criterion; after comparing the local decision result of the SU with the global decision result, the FC re-calculates the trust value of the SU, provides a reward for the SU with correct sensing, and records the local decision result of the SU node and the updated trust value on a blockchain.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of mobile communication, and relates to an evolutionary game spectrum sensing method based on a blockchain trust mechanism. BACKGROUND

[0002] With the continuous innovation of mobile communication technology, wireless communication has been widely used in people's daily life. With the continuous increase of mobile services and the number of users, the contradiction between the increasing demand for spectrum resources and the lack of spectrum resources has become a bottleneck restricting the development of wireless communication. However, the current spectrum resource allocation mainly adopts a fixed authorization mode, which authorizes the spectrum to a specific user and prohibits non-authorized users from using it, but the utilization rate of the allocated spectrum is not high, and there are a large number of spectrum resources that can be utilized. Therefore, dynamic spectrum access will be essential in the new generation of mobile communication networks. Cognitive wireless networks can effectively solve the problem of spectrum scarcity by finding idle spectrum through the sensing of PU spectrum by SUs to achieve dynamic spectrum access. However, in a network system lacking reliable authentication, the behavior of SUs cannot be guaranteed, and some SUs may launch SSDF attacks to cause the fusion center to make mistakes and obtain selfish benefits, which will undoubtedly greatly reduce the detection performance of the CSS. Traditional anti-SSDF attack schemes mostly focus on enhancing the data review ability of the FC to improve the accuracy of the CSS by filtering out incorrect data, but these methods lack dynamic analysis of the evolution of malicious behavior of SUs and cannot fundamentally reduce the probability of SUs launching SSDF attacks.

[0003] Evolutionary game, as a widely used method, provides a mathematical framework for analyzing the evolution of group behavior in networks. The difference between evolutionary game and classical game theory is that evolutionary game no longer assumes rationality, but limited rationality. It aims to study the participants of the game who cannot fully grasp the entire information, and their decision-making will change according to the payoff information of each round of game. In actual scenarios, SUs aim to maximize their own interests, and some SUs may launch SSDF attacks to cause the cooperative dilemma. How to promote cooperation among SUs in the CSS has always been a hot issue in the field of cognitive radio. SUMMARY

[0004] Therefore, the purpose of the present application is to provide an evolutionary game spectrum sensing scheme based on a blockchain trust mechanism. By linking the benefits of SUs to the degree of trust they receive through the proposed blockchain trust mechanism, the probability of SUs launching SSDF attacks is reduced, and the evolutionary stable strategy of SUs tends to send real local decision results.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical scheme:

[0006] An evolutionary game spectrum sensing method based on a blockchain trust mechanism, comprising the following steps:

[0007] S1: the FC selects nearby SUs for CSS according to the geographical position of the PU, and if the SU participates in the CSS for the first time, the SU needs to register authentication information with the FC, and after verification by the FC, the SU is registered as a legal participant, and the FC allocates an identity key and an initial trust score to the SU;

[0008] S2: each SU performs spectrum sensing on the PU channel, and after completing the spectrum sensing, the SU selects to send a real or false local decision result to the FC according to the benefit of the strategy;

[0009] S3: after receiving the local decision result of the SU, the FC makes a global decision on the PU channel according to the decision criterion;

[0010] S4: the FC recalculates the trust value of the SU after comparing the local decision result of the SU with the global decision result, and provides a reward to the SU that correctly senses;

[0011] S5: record the local decision result of the SU node and the updated trust value on the blockchain.

[0012] Further, in the step S1, the identity key of the SU is asymmetrically encrypted to realize the identity authentication of the SU in the network.

[0013] Further, in the step S2, the SU performs spectrum sensing on the PU channel by using an energy detection method, and uses a Fermi function to drive the change of the SU strategy.

[0014] Further, the step S2 specifically comprises the following steps:

[0015] The detection process of the i-th SU node is described as follows:

[0016]

[0017] wherein H1 and H0 represent hypotheses that the PU signal exists and does not exist respectively, s(n) represents the PU signal received by the SU, h i represents a channel fading coefficient, and u i (n) represents an additive white Gaussian noise with a mean of 0 and a variance of Without loss of generality, s(n) and u i (n) are independent of each other; the test statistic of the energy detection of the i-th SU is represented as:

[0018]

[0019] wherein N S is the sampling number, and D(y) obeys the following Gaussian distribution according to the central limit theorem:

[0020]

[0021] σ u 2 The variance of the PU signal, the false alarm probability of the energy detection of the honest SU node under the assumption of H0 is:

[0022]

[0023] Where ε is the threshold of the energy detection, the detection probability of the energy detection of the honest SU node under the assumption of H1 is:

[0024]

[0025] If the SU chooses to send a false decision result intentionally, the false alarm probability thereof is represented as:

[0026] P f,m =1-P f (6)

[0027] The detection probability is represented as:

[0028] P d,m =1-P d (7).

[0029] Further, in the step S3, the FC adopts the majority voting criterion to make the global decision, that is, as long as one half of the SU local decision results are channel idle, the global decision is channel idle, and otherwise, the global decision is channel busy.

[0030] Further, the step S3 specifically comprises the following steps:

[0031] Suppose that the number of the SUs participating in the CSS is N, the number of the SU nodes selected to send the real local decision result in the network is h, the number of the SUs selected to send the false local decision result is m, and if then the global false alarm probability of the system utility FC is:

[0032]

[0033] The global detection probability is

[0034]

[0035] Wherein If k

[0036]

[0037] The global detection probability is:

[0038]

[0039] Further, in the step S4, the trust value calculation of the SU adopts a time decay function, that is, the influence degree of past behavior on the trust value is lower and lower, and the reward of the SU is allocated according to the proportion of the trust value.

[0040] Further, in the step S5, the sensing result and the trust value of the SU node in each round of CSS are recorded through the block chain.

[0041] The application has the beneficial effects that the application proposes an evolved game spectrum sensing scheme based on a block chain trust mechanism in a cognitive wireless network, which can reduce the probability of the SU launching a SSDF attack and improve the performance of cooperative spectrum sensing. When the SU participates in the CSS for the first time, the SU needs to register the authentication information to a fusion center (FC), and the FC verifies and registers the device as a legal participant after verification, and allocates an identity key and an initial trust score to the device. The trust change after that will give a trust value reward or deduction according to whether the local sensing result of the device is consistent with the global decision result of the FC. In the scheme, the system reward obtained by the SU node participating in the CSS is related to the degree of trust of the SU node in the network, the mechanism reduces the probability of the SU launching a SSDF attack, and the sensing result and the trust value of the SU node in each round of CSS are recorded through the block chain, so that the safety of the data is guaranteed.

[0042] Other advantages, objects, and features of the application will be set forth in part in the following specification taken in conjunction with the accompanying drawings, and in part will become apparent to those skilled in the art from a consideration of the following specification and drawings, or can be learned from the practice of the application. The advantages and objects of the application can be realized and attained by means of the instrumentalities and combinations pointed out in the following specification. BRIEF DESCRIPTION OF DRAWINGS

[0043] In order to make the objects, technical solutions and advantages of the application clearer, the preferred detailed description of the application will be combined with the drawings to describe the application, and the drawings are as follows:

[0044] Figure 1 A model diagram of the evolved game spectrum sensing scheme based on the block chain trust mechanism;

[0045] Figure 2 A flowchart of the evolved game spectrum sensing scheme based on the block chain trust mechanism. DETAILED DESCRIPTION

[0046] The present application can be implemented or applied through other different specific embodiments, and various modifications or changes can be made to the details based on different views and applications without departing from the spirit of the present application. It should be noted that the drawings provided in the following embodiments only illustrate the basic concept of the present application in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0047] The drawings are only used for illustrative description, and the representation is only a schematic diagram, not a physical diagram, and cannot be understood as a limitation of the present application; in order to better illustrate the embodiments of the present application, some components in the drawings are omitted, enlarged or reduced, and do not represent the size of the actual product; for those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings can be omitted.

[0048] The same or similar reference numerals in the drawings of the embodiments of the present application correspond to the same or similar components; in the description of the present application, it should be understood that if the terms "upper", "lower", "left", "right", "front", "back" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, therefore the terms describing the positional relationship in the drawings are only used for illustrative description, and cannot be understood as a limitation of the present application, for those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0049] The purpose of the present application is to propose an evolutionary game spectrum sensing scheme based on a blockchain trust mechanism. The method proposes a series of solutions to the problem of performance reduction of CSS caused by SU launching SSDF attack in the current CSS.

[0050] The implementation process of the present application is as follows:

[0051] S1: FC selects nearby SU for CSS according to the geographical position of PU, if the SU participates in CSS for the first time, it needs to register the authentication information to FC, and after verification by FC, it is registered as a legal participant, and the device is allocated an identity key and an initial trust score. The identity key of the SU adopts asymmetric encryption to realize the identity authentication of the SU in the network.

[0052] S2: Each SU uses the energy detection method to perform spectrum sensing on the PU channel, and after completing the spectrum sensing, the SU selects to send real or false local decision results to the FC according to the benefit situation of the strategy, and uses the Fermi function to drive the change of the SU strategy.

[0053] where the detection procedure of the ith SU node can be described as:

[0054]

[0055] where H1and H0denote the hypothesis of the presence and absence of PU signal respectively, s(n) denotes the PU signal received by the SU, h i denotes the channel fading coefficient, u i (n) denotes the additive white Gaussian noise with mean 0 and variance Without loss of generality, s(n) and u i (n) are independent. The test statistic of the energy detection of the ith SU can be expressed as:

[0056]

[0057] where N S is the number of samples. According to the central limit theorem, D(y) obeys the following Gaussian distribution:

[0058]

[0059] σ u 2 denotes the variance of the PU signal. The false alarm probability of the energy detection of the honest SU node under the hypothesis of H0is:

[0060]

[0061] where ε is the threshold of the energy detection. The detection probability of the energy detection of the honest SU node under the hypothesis of H1is:

[0062]

[0063] If the SU chooses to send a false decision result intentionally, the false alarm probability and the detection probability of the SU can be expressed as:

[0064] P f,m = 1 - P f (6)

[0065] P d,m = 1 - P d (7)

[0066] S3: After receiving the local decision results of the SUs, the FC makes a global decision by using the majority voting criterion, that is, as long as half of the local decision results of the SUs are channel idle, the global decision is channel idle, and otherwise, the global decision is channel busy.

[0067] Assuming the number of SUs participating in the CSS is N, the number of SU nodes in the network that select to send a real local decision result is h, the number of SU nodes that select to send a false local decision result is m, if then the global false alarm probability and detection probability of the system utility FC of the present application are respectively:

[0068]

[0069]

[0070] wherein if k < m, then the global false alarm probability and detection probability are respectively:

[0071]

[0072]

[0073] S4: FC recalculates the trust value of the SU after comparing the local decision result of the SU with the global decision result, the trust value of the SU is calculated using a time decay function, that is, the influence degree of past behavior on the trust value will become lower and lower, and a reward is provided to the SU whose perception is correct, the reward of the SU is allocated according to the proportion of the trust value.

[0074] S5: The local decision result of the SU node and the updated trust value are recorded on the blockchain. The perception result and the trust value of the SU node in each round of CSS are recorded through the blockchain, which not only ensures that these data have high security but also protects the privacy of the SU node to a certain extent.

[0075] Figure 1 The model is a centralized CSS system, which assumes that the system contains a static PU, N SU nodes, the number of honest SU nodes that select to send a real local decision result is h, the number of malicious SUs that select to send a false local decision result is m, and a fusion center (FC) for decision. Among them, the PU has the priority right to use the channel, the SU performs spectrum sensing on the PU channel through energy detection, then the SU selects to send a strategy according to the benefit of the strategy, that is, to send a real or false local decision result to the FC. The FC fuses all the local decision results of the SUs to obtain the final perception result, and provides a certain reward to the SU whose perception is correct (that is, the SU whose local decision result is consistent with the global decision result of the FC) according to the proportion of the trust value.

[0076] The SU performs spectrum sensing on the channel of the PU by using the energy detection method, the false alarm probability is P f , and the detection probability is Pd , the false alarm probability and detection probability of the SU that chooses to send false perception results are P f,m and P d,m FC makes a global decision on the channel condition of PU according to the majority voting criterion. If the number of SUs that choose to send false decision results is Then the global false alarm probability Q of FC is f and the detection probability Q d They are respectively Equation (8) and Equation (9), otherwise the global false alarm probability Q f and the detection probability Q d They are equation (10) and equation (11) respectively.

[0077] SUs participating in CSS receive different rewards depending on their sending strategies. SUs that choose to send truthful perception reports have a chance of receiving system rewards, while those that launch SSDF attacks, upon successful attacks, not only receive system rewards but also receive additional malicious rewards. Furthermore, the probability of receiving rewards for SUs choosing either of these sending strategies depends on the strategies chosen by the majority of other SUs. When a large number of SUs choose to honestly send their perception reports, these honest SUs have a higher probability of receiving system rewards. Conversely, malicious SUs that deliberately send false perception reports are more likely to receive system rewards and additional malicious rewards. The following details the expected returns of these two sending strategies.

[0078] Assume that the system rewards the SU with correct perception (i.e., consistent with the judgment result of the fusion center) as R, and the energy consumption of the SU for detection is C d The cost of a malicious node launching an attack is C a When the PU channel is idle, the malicious node gains G1 by occupying the channel. When the channel is busy, the malicious node gains G2 by successfully attacking and causing system interference. The possible gains of the SU node that chooses to send the true perception result and the SU node that chooses to launch the SSDF attack in each case are shown in Table 1:

[0079] Table 1

[0080]

[0081] The above table shows the benefits of selecting two sending strategies SU in all cases, where T or F represents whether the perceived judgment is correct or incorrect. According to the above table, the expected benefits of the SU node sending the true judgment result in one round of CSS are:

[0082]

[0083] in and are the expected benefits of an honest SU under idle and busy conditions of the PU channel, respectively:

[0084]

[0085]

[0086] Where P0 and P1 represent the probability of the PU channel being idle or busy, respectively. Similarly, the expected benefit of a malicious SU node in a round of CSS is:

[0087]

[0088] The same applies and are the expected benefits of the malicious SU when the PU channel is idle and busy, respectively. Specifically:

[0089]

[0090]

[0091] In a CSS system containing a total of N SU nodes, the average expected benefits of the SU that actually sends the judgment result and the SU that launches the SSDF attack are:

[0092]

[0093]

[0094] In evolutionary game theory, the payoff of a strategy has a profound impact on its evolution. Generally speaking, a SU will choose a strategy with higher payoffs. However, in reality, SUs are boundedly rational and may sometimes choose a strategy with lower payoffs. Furthermore, the number of SUs collaborating in an actual CSS system is limited, so N is a finite integer. Within such a limited population, the Fermi function is generally used to represent the probability of a strategy transition. Specifically, the probability that a SU choosing the strategy of sending the true judgment results will switch to launching an SSDF attack is:

[0095]

[0096] Where β represents the selection strength. The evolution of the SU node sending strategy in CSS can be described as a Markov process. Therefore, the transition probabilities of the number of SU nodes that choose to send true judgment results increasing from h to h+1 and decreasing from h to h-1 are:

[0097]

[0098]

[0099] The selection gradient of the sending strategy for any selection intensity β in the SU population is:

[0100]

[0101] The stability of the SU sending strategy can be judged by the selection gradient g(h). In order to maximize the performance of the CSS system, we hope that all SU nodes in the network select to send the true decision result. In the evolutionary game of the limited population, given any initial distribution, the probability that all participants eventually select a certain strategy depends on the ratio of the transition probability of the strategy. Therefore, in the CSS model in this paper, the fixed probability that all SU nodes eventually select to send the true decision result is:

[0102]

[0103] Further, when the number of SU nodes that initially select the strategy of sending the true decision result is k, the fixed probability that all nodes eventually select this strategy is:

[0104]

[0105] As a distributed data structure, blockchain has the characteristics of distributed consensus, tamper resistance and anonymity, so by recording the perception results and trust values of SU nodes in each round of CSS through blockchain, not only can these data be guaranteed to have high security, but also the privacy of SU nodes can be protected to some extent. When SU node i participates in CSS for the first time, it needs to submit a registration application to the FC, which will verify it and register it as a legal participant, and assign an identity key and an initial trust value to node i. The change of the trust value thereafter will be rewarded or deducted according to whether the local decision result of node i is consistent with the global decision result of the FC. The trust value of any SU node i at time t is calculated by the formula:

[0106]

[0107] where T0 is the initial trust value given by the system when the SU node registers on the blockchain, δ t is the trust value of a single cooperative perception participated by node i at time t, δ t > 0 when the perception result of the node is consistent with the result determined by the FC, otherwise δ t < 0, and λ is the time decay factor, the influence of past perception behavior on trust is gradually reduced. According to the trust value weighting, the system reward obtained by node i when perceiving correctly is related to its current trust value, which is specifically represented as:

[0108]

[0109] where St is the correct set of SU nodes perceived at time t, R s The total reward provided by the system to all correctly sensing nodes is distributed based on the node's trust weight. Through this mechanism, SU nodes that choose to send truthful judgment results have a higher probability of receiving trust rewards, resulting in higher returns. SU nodes that send false judgment results will not only face a trust penalty after a failed SSDF attack, but will also receive reduced rewards in subsequent CSSs. This mechanism increases the rewards of nodes that choose to send truthful judgment results, making SUs more likely to send truthful spectrum sensing results in CSSs, thereby reducing the probability of SUs launching SSDF attacks.

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. 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 solutions of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solutions, which should all be included in the scope of the claims of the present invention.

Claims

1. An evolutionary game spectrum sensing method based on blockchain trust mechanism, characterized by: The following steps are involved: S1: The FC selects a nearby SU based on the PU's geographic location to participate in the CSS. If the SU is participating in the CSS for the first time, it needs to register its authentication information with the FC. The FC verifies that the SU is a legitimate participant and assigns an identity key and initial trust value to the SU. S2: Each SU performs spectrum sensing on the PU channel. After completing spectrum sensing, the SU chooses to send a true or false local decision result to the FC based on the benefits of the strategy. The SU uses the energy detection method to perform spectrum sensing on the PU channel and uses the Fermi function to drive the change of the SU strategy. That is, the probability that the SU that chooses to send the true decision result strategy will switch to launching an SSDF attack is: in Indicates the selection intensity, represents the average expected benefit of SU that actually sends the judgment result, represents the average expected benefit of SU launching SSDF attack; S3: After receiving the local decision result from the SU, the FC makes a global decision on the PU channel according to the decision criteria; S4: The FC compares the SU's local judgment result with the global judgment result, recalculates the SU's trust value, and provides rewards to the SU with correct perception. Specifically, it includes: The number of SU nodes that send the real judgment results is selected from Increase to and from Reduce to The transition probabilities are: In the SU population, for any selection intensity The selection gradient of the sending strategy is: By selecting the gradient Determine the stability of SU sending strategy; The fixed probability that all SU nodes will eventually choose to send the true judgment result is: The number of SU nodes that choose to send the true judgment result strategy at the beginning is When , the fixed probability that all nodes choose this strategy is: When SU ​​node When participating in CSS for the first time, a registration application is submitted to FC, which will verify and register it as a legal participant and become a node. Assign identity keys and initial trust values; subsequent trust value changes depend on the node Whether the local judgment result of the SU node is consistent with the global judgment result of the FC, the trust value will be rewarded or deducted; any SU node In The trust value calculation formula at a certain moment is: in It is the initial trust value given by the system when the SU node is registered in the blockchain. for The trust value of a single collaborative perception of a node at a given moment, when the perception result of the node is consistent with the result of the FC judgment >0, otherwise <0, is the time decay factor; According to the trust value weight, the node The system reward obtained when the perception is correct is related to its current trust value, which is specifically expressed as: in for Always perceive the correct set of SU nodes, The total reward provided by the system to all nodes that perceive the correct information is distributed according to the trust value weight of the node itself; S5: Record the local decision result of the SU node and the updated trust value on the blockchain.

2. The spectrum sensing method based on the blockchain trust mechanism based on evolutionary game theory according to claim 1 is characterized by: In step S1, the identity key of the SU is encrypted using asymmetric encryption to achieve identity authentication of the SU in the network.

3. The spectrum sensing method based on the blockchain trust mechanism based on evolutionary game theory according to claim 1 is characterized by: Step S2 specifically includes the following steps: No. The detection process of a SU node is described as follows: in and represent the hypothesis of the presence and absence of PU signal, respectively. Indicates the PU signal received by SU, represents the channel fading coefficient, The mean is 0 and the variance is Additive Gaussian white noise, and are independent of each other; The test statistic of SU energy detection is expressed as: in is the number of samples, according to the central limit theorem Obey the following Gaussian distribution: represents the variance of the PU signal, Under the assumption that , the false alarm probability of honest SU node energy detection is: in is the energy detection threshold. Under the assumption that , the detection probability of honest SU node energy detection can be obtained as: If the SU chooses to intentionally send a false decision result, the false alarm probability is expressed as: The detection probability is expressed as: 。 4. The spectrum sensing method based on the blockchain trust mechanism based on evolutionary game theory according to claim 3 is characterized by: In step S3, the FC uses a majority voting criterion to make a global decision, that is, as long as half of the SUs locally decide that the channel is idle, the global decision is that the channel is idle; otherwise, the global decision is that the channel is busy.

5. The spectrum sensing method based on the blockchain trust mechanism of evolutionary game according to claim 4 is characterized by: The step S3 specifically includes the following steps: Assume that the number of SUs participating in CSS is , the number of SU nodes in the network that choose to send the true local decision results is , the number of SUs that choose to send false local decision results is ,like , then the global false alarm probability of the system utility FC is: The global detection probability is in ,like , then the global false alarm probability is: The global detection probability is: 。 6. The spectrum sensing method based on the blockchain trust mechanism based on evolutionary game theory according to claim 1 is characterized by: In step S5, the perception results and trust values ​​of the SU nodes in each round of CSS are recorded through the blockchain.

Images