Evolutionary game security policy method based on block chain evaluation and coordination mechanism

By adopting blockchain evaluation and coordination mechanism and evolutionary game model in the EH-CR network, the various malicious attack problems faced by SU are solved, the security and detection capabilities of the network are improved, and the resistance to malicious attacks is achieved is achieved. and confidential capacity.

CN118945665APending Publication Date: 2024-11-12HUNAN AUTOMOTIVE ENG VOCATIONAL COLLEGE
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202410924093.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-10
Publication Date
2024-11-12

Smart Images

  • Figure CN118945665A_ABST
    Figure CN118945665A_ABST
Patent Text Reader

Abstract

The invention relates to an evolutionary game security policy method based on a block chain evaluation and coordination mechanism, and belongs to the field of network security. According to the method, a block chain coordination mechanism is provided to limit the influence of a friendly secondary jammer (SJ) in a complex network on a primary user (PU); proposing an evolutionary game attack and defense model under the condition of considering three different security threats, namely primary user simulation attack PUEA and interference attack which are faced by a secondary user SU in an energy harvesting cognitive radio EH-CR network in a sensing stage and eavesdropping attack and interference attack which are simultaneously faced in a transmission stage; the probability of hostile attacks is limited by researching the dynamic characteristics of players of a secondary user SU and a malicious user MU and adjusting the energy consumption cost of an SU sensing stage or the reward and punishment of an MU behavior strategy, so that the security profit of a network system is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of network security, and specifically relates to an evolutionary game security strategy method based on blockchain evaluation and coordination mechanism. Background Art

[0002] Nowadays, the large-scale deployment of wireless devices has led to an increasingly huge consumption of spectrum resources. At the same time, when secondary users SU join the authorized network to use idle spectrum, they will be subject to various complex attacks, such as interference and eavesdropping attacks by malicious users (MU). In the EH-CR (Energy Haresting Cognitive Radio, EH-CR) network, the attack intensity and attack mode of interference attacks and eavesdropping attacks are often related to energy consumption and strategy selection. These two attacks have a great impact on the detection probability and confidentiality capacity of SU in the network in the perception stage and transmission stage respectively. In order to ensure the maximization of its own benefits, SU often needs to allocate part of its power to further adopt defense strategies to offset the adverse effects of such attacks.

[0003] However, with the complexity of the network environment, the openness and dynamic characteristics of the SU perception stage are more easily exposed to the MU's field of vision, and the SU often faces more diverse attack methods at this stage. In addition to the interference attack method, PUEA (Primary User Emulation Attacks, PUEA) from MU is also very common in the perception stage. By occupying the main channel or imitating the transmission signal of the primary user PU, the MU opportunistically confuses the SU that is currently performing the perception task, making it mistakenly believe that the PU is using the channel, thereby affecting the perception detection rate and false alarm rate of the SU. Through this PUEA attack method, the MU can privately occupy the PU idle channel for its own use or interfere with the normal use of the authorized channel by the SU. Because of the existence of PUEA, the idle authorized channels available to the SU will be sharply reduced, and it is more difficult to defend against such attacks. At the same time, the MU uses interference and eavesdropping to attack together, aggravating the degradation of the detection probability of the SU in the perception stage and the confidentiality capacity in the transmission stage. In the EH-CR network with three attack methods, how the SU can effectively defend will become a very difficult problem.

[0004] In order to defend against simultaneous eavesdropping attacks, each user has its own cooperative interference node. In coverage mode, there is no need to worry about the SU transmission power affecting the operation of the PU. However, as the number of SUs and SJs (Secondary Jammers, SJs) increases, each SU's cooperative interference node will affect the PU transmission data to varying degrees depending on the distance between itself and the PU. Summary of the invention

[0005] The purpose of the present invention is to provide an evolutionary game security strategy method based on blockchain evaluation and coordination mechanism to address the various malicious attacks and complex EH-CR network problems faced by SU.

[0006] To achieve the above purpose, the technical solution of the present invention is: an evolutionary game security strategy method based on blockchain evaluation and coordination mechanism, a blockchain coordination mechanism is proposed to limit the impact of friendly secondary jammers SJ on primary users PU in complex networks, and an evolutionary game attack and defense model is proposed considering three different security threats faced by secondary users SU in the energy harvesting cognitive radio EH-CR network in the perception stage, namely, primary user simulation attack PUEA and interference attack, and eavesdropping attack and interference attack at the same time in the transmission stage; by studying the dynamic characteristics of both players, the secondary user SU and the malicious user MU, the energy consumption cost in the SU perception stage or the reward and punishment of the MU behavior strategy is adjusted to limit the probability of malicious attacks, thereby improving the security benefits of the network system.

[0007] The present invention also provides an evolutionary game security strategy system based on blockchain evaluation and coordination mechanism, including a memory, a processor, and computer program instructions stored in the memory and capable of being executed by the processor. When the processor executes the computer program instructions, the method steps described above can be implemented.

[0008] The present invention also provides a computer-readable storage medium, on which computer program instructions that can be executed by a processor are stored. When the processor executes the computer program instructions, the method steps described above can be implemented.

[0009] Compared with the prior art, the present invention has the following beneficial effects: In view of the various malicious attacks and complex EH-CR network problems faced by SU, the present invention first uses blockchain technology to build an evaluation and adjustment mechanism, and limits the interference and self-interference behavior of the SU network on the PU network by setting the reward bandwidth. Then, the attack and defense means between MU and SU in different time slots are analyzed, and the utility function under each decision is established by combining the perception detection probability and the transmission confidentiality capacity, and the EH technology is used to solve the problem of limited user energy. Then, based on evolutionary game theory, the dynamic characteristics and game convergence between MU and SU, as well as the performance of security strategies for simultaneously defending against multiple attacks are studied. Finally, the present invention analyzes the different EH-CR networks. p The simulation results of the impact of SU detection probability and different θ on MU attack probability show that the proposed method has good anti-malicious attack capability in PU interference limited network. At the same time, in terms of detection probability and confidentiality capacity, the proposed scheme is better than the comparison scheme. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 This is a system model diagram of the present invention.

[0011] Figure 2 This is a timing diagram of the evaluation and coordination mechanism based on blockchain of the present invention.

[0012] Figure 3 This is the model space distribution diagram.

[0013] Figure 4 Evolution strategy for SU.

[0014] Figure 5 Evolution strategy for MU.

[0015] Figure 6 For E p The impact of changes on the secondary user I attack strategy.

[0016] Figure 7 For E p The impact of changes on the secondary user II attack strategy.

[0017] Figure 8 is the impact of the change of attenuation coefficient θ on the PUEA attack strategy.

[0018] Fig. 9 is the detection probability and x D relationship.

[0019] Fig.10 To transmit confidentiality capacity and y A relationship. DETAILED DESCRIPTION

[0020] The technical solution of the present invention is described in detail below in conjunction with the accompanying drawings.

[0021] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0022] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0023] The invention provides an evolutionary game security strategy method based on blockchain evaluation and coordination mechanism, proposes a blockchain coordination mechanism to limit the influence of SJ (Secondary Jammer, SJ) on primary user PU (PU) in complex network, and considers three different security threats faced by secondary user SU (SU) in EH-CR (Energy Haresting Cognitive Radio, EH-CR) network, namely PUEA (Primary User Emulation Attacks, PUEA) and interference attack in the perception stage, and eavesdropping attack and interference attack in the transmission stage, and proposes an evolutionary game attack and defense model; by studying the dynamic characteristics of both players SU and MU (Malicious User, MU), adjusting the energy consumption cost in the SU perception stage or the reward and punishment of MU behavior strategy to limit the probability of malicious attack, so as to improve the security benefits of the network system.

[0024] The following is the specific implementation process of the present invention.

[0025] The present invention provides an evolutionary game security strategy method based on blockchain evaluation and coordination mechanism, which is specifically implemented as follows:

[0026] 1. System model

[0027] The model of the present invention is as follows Figure 1 As shown in Figure 2, MU sets an additional PUEA malicious attacker (MaliciousAttacker, MA), and the number of malicious jammers (Malicious Jammer, MJ) is three. Eavesdroppers (Eav), MA and MJ form a benefit group {U e},{U e} and the performance degradation of SU. SU shares the authorized channel in an overlay mode and transmits its own data through the secondary transmitter ST and the secondary receiver SR. Each SU calls a cooperative SJ to reduce the quality of the eavesdropping channel. At the same time, SU will not only face a large number of interference attacks in the perception stage, but also need to detect PUEA attacks from MU. In the transmission stage, SU needs to face both interference attacks and eavesdropping attacks.

[0028] exist Figure 1In the process, both SU and MU have a utility function that they try to maximize, and they optimize the strategy and achieve their own maximum benefit by rationally utilizing the harvested energy. The benefit functions of SU and MU can be defined by the performance indicators they achieve, such as detection rate, confidentiality capacity, and signal-to-noise-interference ratio. In the confrontation between SU ​​and MU, the goals and benefit growth of SU and MU are opposite, that is, the improvement of one party's performance indicator means the reduction of the other party's performance. In the perception stage, MU decides whether to launch a PUEA attack with probability y. If PUEA is launched, MU will lose energy. And the initial malicious interference power P in the perception phase ms If the ratio is reduced by a certain percentage, the malicious interference power P in the transmission stage m ' t It can be expressed as:

[0029]

[0030] in, is the collected energy of MU, θ is the attenuation coefficient, T s is the set perception time, T t is the transmission time. If MU does not initiate PUEA (with a probability of 1-y), P will not be reduced. ms , all the collected energy will be used by MJ to send malicious interference signals.

[0031] SU will decide whether to perform PUEA detection with probability x and consume energy E p , PUEA detection will succeed with probability λ, that is, it is found that an unknown user maliciously occupies the PU channel. At this time, the transmission power P of SU in the transmission phase t ′ can be expressed as:

[0032]

[0033] Where ε is the power allocation coefficient, E h is the harvested energy of SU, E s SU senses energy consumption.

[0034] If the SU does not detect the PUEA attack when it is not aware of it (with a probability of 1-x) and allows the MU to successfully occupy the PU, the SC will reduce the SU adjustment value. The introduction of evaluation value and adjustment value will be described in detail below. If the MU does not launch a PUEA attack, the SU will not have additional adjustment value changes.

[0035] In the network model structure arrangement of the present invention, when defending against malicious attacks from MU, especially eavesdropping attacks, the SJ carried by the SU may cause a large amount of collaborative interference to the PU transmission. For the SJ closer to the PU, the impact of this interference is greater. The present invention will consider the impact of the SJ on the PU, and use the blockchain SC technology to analyze the PU interference limitation and performance evaluation mechanism, thereby improving the standardization of the SU's use of authorized channels and network reliability.

[0036] 2. Design of user evaluation and coordination mechanism based on blockchain

[0037] 2.1 Design of User Evaluation Mechanism

[0038] In the blockchain network of the present invention, the evaluation value and coordination value of a SU node represent the role of the node in the entire blockchain network. A higher evaluation value indicates that the node is honest and trustworthy in using the authorized spectrum and has a stronger means of defending against malicious attacks; a higher coordination value indicates that the node maintains a higher level of activity in the coordination process and has less interference with the PU.

[0039] In the EH-CR network, the evaluation value of SU is related to the quality of the spectrum used. When multiple SUs apply to use authorized channels, the BS will divide the spectrum of different bandwidths according to their historical performance and the interference margin of the current time slot to incentivize or inhibit the behavior of the SU. Through the transaction ledger of blockchain technology, the contribution of each SU in the system is recorded and updated. In the face of complex network environments and a large number of malicious attacks, at the end of the perception phase of the current time slot, if the node has a higher detection rate and higher confidentiality capacity in the historical time slot, it means that the node has a strong anti-interference or anti-attack ability and contributes more to the overall security performance of the network. Then the evaluation value of the node should be higher, and nodes with higher evaluation values ​​will be assigned to better channels in subsequent links.

[0040] SU in the kth time slot i Safety rating value It can be expressed as:

[0041]

[0042] Among them, Φ1 and Φ2 are the evaluation factors of detection probability and confidentiality capacity respectively. For SU i The detection probability at the k-1th time slot is, YesSU i The confidentiality capacity in the k-1th time slot can be specifically expressed as:

[0043]

[0044] Among them, x +=max(x,0), and are SUs in the k-1th time slot respectively. i The capacity of data transmission channels and eavesdropping channels; For ST i To SR i The channel gain, For ST i Channel gain to Eav; N j is the total number of SJ, N J is the total number of MJ; For SU i The self-interference influence, Ω is the self-interference influence coefficient, P t is the transmission power, z is the attenuation factor, α0 is the path loss, d i,j For SJ j To SR i distance; For MJ J About SU i The malicious interference effect, P mt is the malicious interference power, d i,J For MJ J To SR i distance; For SJ j Collaborative interference with Eav, d e,j For SJ j Distance to Eav; is the self-interference effect on MU, d e,J For MJ J Distance to Eav; is Gaussian noise.

[0045] Different from traditional spectrum sensing, SU senses not only PU signals in the network, but also interference signals and PUEA signals emitted by MJ. i Received signal There are four possible states:

[0046]

[0047] Where n0 is Gaussian noise, For MJ J About SU i The perceived interference power, Sig PU is the PU signal, Sig PUEA It is PUEA signal. Indicates that the channel is not occupied by PU and there is no PUEA signal; Indicates that only PU signal exists; It means that only PUEA signal exists, but the actual channel is idle; Indicates that the channel is busy and a PUEA signal exists at the same time.

[0048] In this section, H1 and H0 are used to represent the presence and absence of PU, respectively. When the current channel state is H0, MU probabilistically launches PUEA attack, using A P Indicates launching a PUEA attack, NA P Indicates that no PUEA attack has been launched. During the SU perception phase, PUAE detection is performed probabilistically. At the same time, D A Indicates that SU performs PUEA signal detection and uses ND A If PUEA signal detection is not performed, the false alarm probability P f ′ can be expressed as:

[0049] P f ′=P(D A / H0) (6)

[0050] According to Bayes' theorem, P f ′ can be expanded to:

[0051] P f ′=P(D A / A P ,H0)P(A P ,H0)+P(D A / NA P ,H0)P(NA P ,H0) (7)

[0052] Let P(D A / A P ,H0)=β P When MU does not take PUEA attack, that is, P(A P ,H0)=0,P(NA P ,H0)=1,P f ′=P(D A / NA P ,H0)=P f , where P f is the initial false alarm probability, at which point the SU detection probability P d ′=P d When MU takes PUEA attack, that is, P(A P ,H0)=1,P(NA P ,H0)=0,P f ′=P(D A / A P ,H0)=β P , SU detection probability Pd ′ can be expressed as:

[0053]

[0054] Among them, T s For the present invention to perceive time, f s is the initial perception frequency, Q is the Gaussian tail function, γ s ′ is the signal-to-interference-to-noise ratio in the perception stage. In order to quantify the perception benefit of MU, P e Indicates the false detection rate of SU. Combining the missed detection rate and false alarm probability, the false detection rate P e It can be expressed as:

[0055] P e =P(H0)P f ′+P(H1)P m ′ (9)

[0056] Among them, P′ m =1-P′ d is the missed detection rate in the SU perception phase, P(H0) is the probability that PU does not exist, and P(H1) is the probability that PU exists. In order to evaluate the continuous security evaluation value of the user, the present invention calculates the final evaluation value on a certain time scale. At the same time, considering that the evaluation value in the most recent time slot can better reflect the user's activity, its proportional weight will be set larger. By calculating and weighting the historical evaluation values, the final average evaluation value of SU is It can be effectively responded by a series, which is specifically expressed as:

[0057]

[0058] Among them, N k is the number of iterations of the current time slot, r is the time sequence traversal sequence, k is the current time slot, q is the attenuation coefficient, represents the evaluation value when the time slot is r. For the specific expression, refer to formula (3). When the traversal r is closer to the current time slot k, the attenuation is (1-q) k-r The smaller it is, the greater the weight of its impact on the final evaluation value. Through this historical traversal method, a more reasonable evaluation value can be obtained.

[0059] 2.2 Design of User Coordination Mechanism

[0060] The previous section introduced the evaluation value mechanism of users in the blockchain network. By evaluating the security of the system, the specific network contribution of users was effectively quantified. This section will design a blockchain-based user coordination mechanism based on the evaluation value.

[0061] When the SJ of the present invention transmits a cooperative interference signal to degrade the eavesdropping channel, it will cause different degrees of interference to the PU, and the closer the SJ is to the PU, the greater the interference. Therefore, a variable is needed in the EH-CR network of the present invention to quantify this PU interference, so the interference coordination value V is proposed. i To express SU i Interference impact on PU network:

[0062]

[0063] Among them, g is the PU traversal sequence, I i,g Indicates SJ i About PU g The interference effect, I th Interference threshold set for the PU network.

[0064] Combined with the changes in SU behavior decision in the previous section, that is, whether to detect PUEA attacks, PU changes I th To feedback the behavior strategy selected by SU, the interference threshold I th Specifically expressed as:

[0065]

[0066] Where I0 represents the PU initial interference threshold, I v Indicates the threshold adjustment change value. When MU selects PUEA attack and SU successfully detects the attack, it contributes to the network. PU rewards this security behavior and the threshold strategy selection If the SU does not detect the attack or fails to detect, the PU will punish and select a strategy When MU chooses not to conduct PUEA attack, SU's strategy selection will not affect its permissible interference threshold.

[0067] Safety evaluation value of the current time slot and interference coordination value V i (k) It reflects the contribution made to the EH-CR network. The present invention proposes to use an improved Sigmod activation function to integrate the influence of the two factors. The Sigmod activation function is an increasing function with a dependent variable range of (0,1). It combines the form of blockchain SC bandwidth allocation to quantify the overall contribution of users. The bandwidth reward of the kth time slot is It can be expressed as:

[0068]

[0069] Among them, ρ3 and ρ4 are weight coefficients.

[0070] 2.3. Consensus Mechanism Design

[0071] In traditional public chains, PoW (Proof of Work, PoW) node consensus algorithm is generally used (commonly used distributed consensus algorithms include Proof of Stake (PoS), Proof of Work (PoW), Practical Byzantine Fault Tolerance (PBFT) algorithm, Delegated Proof of Stake (DPoS), and this example uses Proof of Work (PoW)) to achieve the generation of new blocks. Because nodes with strong computing power have a greater probability of generating blocks under the PoW algorithm, this will result in a large amount of unnecessary waste of resources. In order to encourage users to improve the quality of security performance and balance the behavior of collaborative interference, the present invention improves the target value of the traditional PoW algorithm so that users with higher evaluation values ​​and coordination values ​​have a greater probability of updating blocks to obtain block rewards. By adjusting the difficulty of mining, secure users with higher evaluation values ​​and coordination values ​​will consume less mining.

[0072] Each user node searches for the value that meets the conditions through mining. i The mining difficulty is recorded as n is the serial number of the new block to be generated. Combined with the relevant information in the block header, the mining expression that the node needs to satisfy is as follows:

[0073]

[0074] Among them, T B The timestamp generated for the new block, Nonce is a random number, Hash p The hash value of the previous block is the hash value of the previous block. In the blockchain system of the present invention, the process of updating the block will be carried out after the evaluation and coordination are completed. Expressed as:

[0075]

[0076] Among them, n is the initial difficulty of generating blocks, ζ is the difficulty attenuation coefficient, ξ is a positive constant parameter, a is the evaluation value influence coefficient, b is the coordination value influence coefficient, and c is a constant parameter.

[0077] 2.4 Smart Contract Design

[0078] The detailed process of the evaluation and coordination mechanism of the present invention based on blockchain SC is as follows: Figure 2 As shown, it is mainly divided into the following parts:

[0079] (1) Task assignment phase: After SU registers the public and private keys, it broadcasts the public key to the chain, and then BS verifies the identity and assigns the sensing task.

[0080] (2) Historical evaluation stage: The SC requests the historical security performance of the SU based on the public key, calculates the historical detection probability and confidentiality capacity, and obtains an evaluation value.

[0081] (3) Channel sensing stage: The SU senses the channel used by the PU and uploads the sensed data.

[0082] (4) Interference coordination stage: The SU uploads its own transmission power and cooperative interference allocation coefficient, and the SC allocates appropriate channels and bandwidths based on the interference impact on the PU.

[0083] 3. Establishment and analysis of mathematical models based on evolutionary games

[0084] In the evolutionary game attack and defense model, the two game parties SU and MU influence each other through the choice of their own strategies and make adjustments based on the other party's behavior and benefits. Through continuous learning and adjustment, both parties optimize their own strategies and maximize their own interests. In each iteration, both parties in the evolutionary game modify the probability of a certain decision with a certain strategy, combining EH technology and blockchain evaluation and coordination mechanism to form their own evolutionary direction.

[0085] In the EH-CR network of the present invention, SU will face a large number of malicious attacks. Based on the limited collection of energy, the two parties in the game will selfishly maximize their own interests. The best strategy is selected according to the changes in their own utility. The strategy selected by SU is S s ={D A ,ND A}, D A Represents the detection of PUEA and consumes some energy, ND A Indicates no detection; MU's strategy selection is S e ={A P ,NA P}, A P Indicates that the PUEA attack strategy is selected and a portion of the energy consumption used for interference is reduced. NA P It means that no attack will be carried out, and the energy of malicious interference will not be consumed extra. Through the decision-making selection of SU and MU, the benefit {U s ,U e}.

[0086] Case 1: SU strategy selects D A , MU strategy chooses A P , both sides will consume a lot of energy on PUEA's attack and defense, so the benefits of both sides of the game can be expressed by the following equation:

[0087] (1) SU utility is:

[0088]

[0089] Among them, the interference effect of utility function is I s ' t =ΩεP t 'h st , I′ mr =P m ' t h mr , I s ' e =εP t 'h se , I′ me =ΩP m ' t h me , α1, α2 are utility conversion factors, M i To reward bandwidth, For case 1, the SU detection probability is related to the false alarm probability P′ f function, and C1(I th ) is the SU confidentiality capacity in case 1 with respect to the interference threshold I th The interference threshold affects the cooperation power P in the network. j The maximum value of satisfies the following relationship:

[0090]

[0091] Among them, SJ j Average channel gain to each PU For SJ j Average distance to PU.

[0092] When SU ​​and MU execute this strategy, the perceived signal may be a real PU signal or a PUEA simulated signal. If the detection success rate is λ, the SU perceived signal-to-noise ratio It can be expressed as:

[0093]

[0094] Among them, P P is the PU transmission power, h i,g It is PU g To SU i The channel gain, P A is the PUEA simulation signal power, hi,A Send PUEA signal to SU for MJ i The channel gain, P′ ms =(1-θ)P ms MJ in the perception phase of PUEA attack J The malicious interference power, h i,J For MJ J To SU i The channel gain, is the Gaussian noise in the perception stage.

[0095] (2) MU utility is:

[0096]

[0097] Among them, α3 and α4 are utility conversion factors; In case 1, the SU false detection rate is about P f ′, and in this case P f ′=β P ; P(H0) is the probability that PU does not exist, and P(H1) is the probability that PU exists; The SU eavesdropping channel capacity in case 1 is about the interference threshold I th Because the SU has effectively detected the PUEA attack and made a great contribution to network security, the PU network rewards the SU and increases its permissible interference threshold I th , i.e., selection strategy

[0098] Case 2: SU strategy selects D A , MU strategy selects NA P , SU consumes energy on PUEA defense, and MU maintains the original strategy. The benefits of both parties in the game can be expressed by the following equation:

[0099] (1) SU utility is:

[0100]

[0101] (2) MU utility is:

[0102]

[0103] Among them, MU did not perform PUEA attack, but SU performed detection, which consumed additional detection energy, and the transmission power was P t ′, but SU behavior has no effect on PU network, so PU interference threshold selection strategy At this time, SU perceives the signal-to-interference-noise ratio It can be expressed as:

[0104]

[0105] Case 3: SU strategy selects ND A , MU strategy chooses A P MU consumes extra PUEA energy, and the malicious interference power in the transmission stage is P m ' t SU maintains the original strategy, that is, the transmission power remains at the initial value P t , at this time, the benefits of both parties in the game can be expressed by the following equation:

[0106] (1) SU utility is:

[0107]

[0108] (2) MU utility is:

[0109]

[0110] Among them, MU adopts the PUEA strategy, but SU does not detect the attack behavior, resulting in a decrease in network security performance, so PU chooses the interference threshold strategy At this time, the perceived signal-to-dryness ratio It can be expressed as:

[0111]

[0112] When SU ​​is not detecting, the initial power P is maintained during the transmission phase. t It can be expressed as:

[0113]

[0114] Case 4: SU strategy selects ND A , MU strategy selects NA P , MU and SU do not consume energy in PUEA attack and defense, both sides maintain the initial state, and the gains of both sides can be expressed by the following equation:

[0115] (1) SU utility is:

[0116]

[0117] (2) MU utility is:

[0118]

[0119] Among them, MU did not perform PUEA, SU did not detect the attack behavior, and had no impact on the PU network, so the interference threshold selection strategy was set At this time, the SU perception signal-to-noise ratio in case 4 is the same as equation (22), and the false alarm rate is P f Same as case 2, so Because SU consumes more energy to detect PUEA in case 2 than in case 4, the transmission power in the transmission phase is different, so the confidentiality capacity and the wiretap channel throughput are also different, that is,

[0120] In the design of the user evaluation and coordination mechanism based on blockchain, each time slot updates the bandwidth reward M allocated by the base station during the transmission phase according to the security performance of the previous time slot, and at the same time affects the profit function of each SU in the evolutionary game. The utility analysis of the decision-making of both parties is carried out as follows:

[0121] (1) SU decision analysis

[0122] From Case 1 and Case 2, we can get the average benefit U of SU for PUEA detection. s (D A )for:

[0123]

[0124] From cases 3 and 4, we can get the average benefit U of SU without PUEA detection. s (ND A )for:

[0125]

[0126] Combining equations (29) and (30), the average utility of SU behavior decision is:

[0127]

[0128] (2) MU decision analysis

[0129] From cases 1 and 3, we can get the average benefit U of MU in PUEA. e (A P )for:

[0130]

[0131] From cases 2 and 4, we can get the average benefit U of MU without PUEA. e (NA P )for:

[0132]

[0133] Combining equations (32) and (33), the average utility of MU behavior decision is:

[0134]

[0135] 4. Evolutionary Game Stability Analysis

[0136] 4.1. Replication of dynamic equations and analysis of stability conditions

[0137] This section will conduct a dynamic analysis of the evolutionary game between SU ​​and MU, give the replication dynamic equations of the two evolutionary parties, and measure the stable state of the two parties in the game according to the standard form of the evolutionary stable strategy ESS. D Behavioral strategy for SU, S A is the MU behavior strategy. If the local equilibrium strategy is adopted The utility of SU after evolution is the largest, then the strategy is ESS, that is:

[0138]

[0139] To simplify equations, use variable names instead of function names, for example: replace use replace Etc. Substituting equations (29) and (31) into the basic form of the replication dynamic equation in the evolutionary game, we can obtain the expression of the SU replication dynamic equation:

[0140]

[0141] The basic form of the replication dynamic equation in evolutionary games is as follows;

[0142]

[0143] Among them, s i is the set of evolutionary group game strategies, x i is the probability of a player choosing a strategy at time t, F(s i ,x) represents the selection strategy s i The profit obtained when , F(x,x) represents the average profit of the whole group;

[0144] U s (D A )-U s (ND A ) is expressed as:

[0145] Among them, U s (D A )-U s (ND A ) can be expressed as:

[0146]

[0147] The iterative process of the dynamic evolution of SU is described by the following equation:

[0148]

[0149] in, Adopt strategy D for SU in current time slot k A The probability of μ s Adjust the step size for SU iterations.

[0150] Substituting equations (32) and (34) into the basic form of the replication dynamic equation in the evolutionary game, we obtain the expression of the MU replication dynamic equation:

[0151]

[0152] Among them, U e (A P )-U e (NA P ) can be expressed as:

[0153]

[0154] The dynamic evolution of MU is described by the following equation:

[0155] P e (A P ,k+1)=P e (A P ,k)+μ e F e (y) (41)

[0156] Where k is the current iteration time slot, P e (A P ,k) Adopt strategy A for the current time slot P The probability of μ e The replication dynamic equation reflects the strategy adjustment trend of SU and MU when they evolve and learn from each other. When the equation is equal to 0, it means that the game parties have reached equilibrium and obtained a stable equilibrium point, that is, let dx / dt = 0 and dy / dt = 0, and four conventional equilibrium points can be solved: (1,0), (0,1), (0,0), (1,1), and a mixed equilibrium point: (x * ,y * ), the mixed equilibrium point is specifically expressed as:

[0157]

[0158] set up and For SU and MU to be in stable state, the above five stable points are where the utility function reaches convergence, that is, However, the local stable point is not necessarily the ESS. As the evolution proceeds, the utility of the later local stable point may exceed the utility of the previous stable point. The Jacobian matrix is ​​used below to verify the final ESS. The Jacobian matrix J of the present invention is m It is expressed as:

[0159]

[0160] Among them, the elements of the Jacobian matrix are specifically expressed as:

[0161]

[0162] According to the characteristics of linear systems, if the Jacobian matrix trace tr(J m ) is negative, the system evolves in a stable direction; if the trace is positive, the system evolves in an unstable direction; the determinant of the Jacobian matrix is ​​denoted by det(J m ), if the determinant is positive, it means that the disturbance of the system will not grow; if the determinant is negative, it means that the disturbance of the system will grow. In order to quickly verify whether the local stable point is the global stable point, the present invention calculates the trace tr(J m ) is less than 0 and the determinant det(J m ) is greater than 0 to identify ESS.

[0163] (1) The stable point is the mixed equilibrium point (x * ,y * ), we can get that all elements of the Jacobian matrix are equal to 0, then tr(J m )=det(J m )=0, so (x * ,y * ) is not an ESS.

[0164] (2) When the stable point is (1,0), the Jacobian matrix is ​​specifically expressed as:

[0165]

[0166] At this time, the Jacobian matrix When the other two elements are less than 0, and When (1,0) is ESS, then SU ​​detects energy consumption E p And the MU attack energy consumption attenuation factor θ satisfies the following conditions:

[0167]

[0168] Among them, the polynomial is specifically expressed as:

[0169] (3) When the stable point is (0,1), the Jacobian matrix is ​​specifically expressed as:

[0170]

[0171] when and When (0,1) is ESS, then E p and θ satisfy the following conditions:

[0172]

[0173] Among them, the polynomial can be specifically expressed with relevant parameters as follows:

[0174]

[0175] (4) When the stable point is (0,0), the Jacobian matrix is ​​specifically expressed as:

[0176]

[0177] when and When (0,0) is ESS, then E p and θ satisfy the following conditions:

[0178]

[0179] in, is the maximum allowable detection energy consumption.

[0180] (5) When the stable point is (1,1), the Jacobian matrix is ​​specifically expressed as:

[0181]

[0182] when and When (1,1) is ESS, then E p and θ satisfy the following conditions:

[0183]

[0184] Based on the above analysis, the convergence between the two parties in the game will tend to one of the four stable points. p The different conditions satisfied by and θ result in different final ESS. p When and θ satisfy equation (46), the evolution will converge to the stable point (1,0), at which point all SUs will choose to detect PUEA, and MUs will not attack. pWhen and θ satisfy equation (53), the evolution will converge to the stable point (1,1), at which the probability of all SUs detecting PUEA and MUs attacking is 1. The convergence of the stable points (1,0) and (1,1) effectively stimulates all SUs to perform PUEA detection, improving the overall security performance of the EH-CR network on the basis of maximizing SU utility. p The simulations are carried out on SU under different θ and MU under different θ, and the evolution trend of their strategy selection is obtained.

[0185] 4.2 Algorithm Process

[0186] The security algorithm of the present invention is summarized in detail in Algorithm 1:

[0187]

[0188] 5. Simulation results and analysis

[0189] 5.1 Simulation parameter settings

[0190] This section evaluates the safety performance of the proposed method through MATLAB simulation. The spatial environment simulated by the present invention is based on a plane coordinate system of an 80m×80m scene, in which PUs, SUs, and MUs are randomly distributed, specifically 1 BS, 1 MA, 3 MJs, 3 PUs, 15 SUs (including transmitters and receivers), and 15 SJs. The spatial distribution diagram is shown in Figure 3 As shown in the figure, the BS is in the center, the PU is close to the periphery of the distribution circle, and other transmitters, receivers, jammers and other devices are randomly distributed.

[0191] The initial false alarm probability P of the user of the present invention f Set to 0.1, other relevant simulation parameters are shown in Table 1:

[0192] Table 1 Simulation parameters

[0193]

[0194] 5.2 Simulation Analysis

[0195] Figure 4 and Figure 5 The strategy evolution process of SU and MU is shown. Among them, the initial PUEA detection probability x0 of SU is 0.2, and the initial PUEA attack probability y0 of MU is 0.2. By randomly selecting 4 SUs to calculate their replication dynamic equations, the changes in SU detection probability and MU attack strategy are obtained. The ST and SR distances, SJ and SR distances, and ST and MA distances of different users are quite different, resulting in large differences in their strategy evolution and convergence to different stable points. Figure 4In , two different SUs choose to detect PUEA around 1500 iterations, while the other two SUs choose not to detect. Figure 5 In the example, after 1200 iterations, MU chooses to attack or not attack according to the quality of different SU eavesdropping channels.

[0196] Figure 6 and Figure 7 Describes the energy consumption E for two different SU detection p In the network of the present invention, SU needs to reasonably utilize the collected energy and defend against three attacks at the same time to maximize its own utility. p When the power consumption is less than 0.1W, the cost of SU detecting PUEA is small, and the strategy tends to evolve in the detection direction. p As , the cost of SU's choice to detect attacks increases, and the benefits decrease. Finally, SU chooses not to detect. At this time, the strategy probability evolution approaches 0. The simulation also verifies to a certain extent that in the above evolutionary game stability analysis, according to E p Under different conditions, SU finally reaches the stable state x D =0 or x D =1.

[0197] Figure 8 The influence of the change of the interference power reduction coefficient on the evolution strategy when MU adopts the PUEA attack strategy in the perception phase is described. When the attenuation coefficient θ in equation (1) is small, the cost of MU adopting the attack strategy is small, and it has more energy to carry out the interference attack in the perception phase. At this time, MU is more inclined to adopt the PUEA attack strategy. As θ gradually increases, the consumption of the attack strategy increases, and MU tends not to carry out the PUEA attack. When the iteration reaches 1800 times, the strategy selection of MU will be completely divided into two directions. Figure 8 This also verifies the evolutionary game stability analysis above to a certain extent. According to the different conditions satisfied by θ, MU eventually reaches the stable state y A =0 or y A =1.

[0198] Fig. 9 It describes the perception detection probability as the SU detection PUEA probability (x D The evolution process of the increase of PUEA probability. The initial attack probability of PUEA of MU is set to 0.6, and the algorithm of reference [1] is set as the comparison algorithm. In the entire network model of reference [1], SU is also subject to three malicious attacks: PUEA, interference attack and eavesdropping attack. In the perception stage, SU only faces PUEA but does not consider interference attack. Fig. 9 In the case of PUEA detection probability x DThe improvement of SU perception signal drying ratio increases, and the attack probability of MU is y A The detection probability of SU sensing spectrum increases rapidly, and the proposed algorithm eventually stabilizes and approaches 1; while in the comparison algorithm, malicious interference attacks are added to the original detection formula, and the SU signal-to-noise ratio decreases significantly, resulting in a significantly lower detection probability value of the comparison algorithm than that of the proposed algorithm. D =1), the detection probability of the algorithm of the present invention increases from about 0.65 to 1, while that of the comparison algorithm increases from 0.53 to 0.87. The simulation results verify that when PUEA and interference attack are considered at the same time, the algorithm of the present invention can improve the detection probability by 15%.

[0199] Fig.10 It shows that the transmission confidentiality capacity increases with the PUEA attack probability (y A The evolution process of increasing PUEA Probability. The initial probability of PUEA attack on SU is set to 0.5. In the transmission phase of the comparative algorithm [1], both interference attack and eavesdropping attack are considered, but the impact on PU is ignored, and the power of SJ will be unrestricted. A As the MU increases, it tends to consume extra energy in the perception stage to carry out PUEA attacks, which leads to a reduction in the malicious interference power in the transmission stage, which improves the quality of the transmission channel and eventually increases the confidentiality capacity. A = 1, the maximum value is 4.56. In the comparison algorithm, because there is no blockchain security evaluation and interference coordination mechanism of the present invention, that is, no additional bonus bandwidth can be brought, the part of SJ power exceeding PU interference threshold will be lost, causing SU to increase detection energy consumption, thereby reducing the total power resources in the transmission stage, and ultimately leading to a reduction in confidentiality capacity. When y A =1, the algorithm of the present invention improves the confidentiality capacity by 72% compared with the comparison algorithm.

[0200] The present invention also provides an evolutionary game security strategy system based on blockchain evaluation and coordination mechanism, including a memory, a processor, and computer program instructions stored in the memory and capable of being executed by the processor. When the processor executes the computer program instructions, the method steps described above can be implemented.

[0201] The present invention also provides a computer-readable storage medium, on which computer program instructions that can be executed by a processor are stored. When the processor executes the computer program instructions, the method steps described above can be implemented.

[0202] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0203] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0204] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0205] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0206] References:

[0207] [1] Banerjee A, Maity S P. Jamming in eavesdropping on throughputmaximization in green cognitive radio networks[J]. IEEE Transactions on MobileComputing, 2021, 22(1):299-310..

[0208] The above is only a preferred embodiment of the present invention, and does not limit the present invention in other forms. Any technician familiar with the profession may use the above disclosed technical content to change or modify it into an equivalent embodiment with equivalent changes. However, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention still belongs to the protection scope of the technical solution of the present invention.

Claims

1. An evolutionary game security strategy method based on blockchain evaluation and coordination mechanism, characterized in that: A blockchain coordination mechanism is proposed to limit the impact of friendly secondary jammers SJ on primary users PU in complex networks, and an evolutionary game attack and defense model is proposed considering three different security threats faced by secondary users SU in energy harvesting cognitive radio EH-CR networks: primary user simulation attack PUEA and interference attack in the perception stage, and eavesdropping attack and interference attack in the transmission stage. By studying the dynamic characteristics of both secondary users SU and malicious users MU, the energy consumption cost in the SU perception stage or the rewards and penalties of MU behavior strategies are adjusted to limit the probability of malicious attacks, thereby improving the security benefits of the network system.

2. According to claim 1, an evolutionary game security strategy method based on blockchain evaluation and coordination mechanism is characterized in that: The method includes: S1. Build system model; S2. Design a user evaluation and coordination mechanism based on blockchain; S3. Establish a mathematical model based on evolutionary game; S4. Analyze the stability of the evolutionary game, derive the final detection probability of SU and the final attack probability of MU, and improve the security benefits of the network system.

3. According to claim 2, an evolutionary game security strategy method based on blockchain evaluation and coordination mechanism is characterized in that: Step S1 is specifically implemented as follows: MU sets up a PUEA malicious attacker MA, three malicious jammers MJ, eavesdroppers Eav, MA and MJ form a collective interest {U e },{U e } and the performance degradation of SU; SU shares the authorized channel in coverage mode and transmits its own data through the secondary transmitter ST and the secondary receiver SR. Each SU calls an SJ to reduce the quality of the eavesdropping channel; at the same time, SU will not only face interference attacks in the perception stage, but also needs to detect PUEA attacks from MU, while in the transmission stage, SU needs to face both interference attacks and eavesdropping attacks; SU and MU both have a utility function that they try to maximize, and they achieve their own maximum benefit by using the harvested energy optimization strategy. The benefit functions of SU and MU are defined by the performance indicators they achieve, namely, detection rate, confidentiality capacity, and signal-to-noise-interference ratio. In the confrontation between SU ​​and MU, the goals and benefit growth of SU and MU are opposite, that is, the improvement of one party's performance indicator means the reduction of the other party's performance. In the perception stage, MU decides whether to launch a PUEA attack with probability y. If PUEA is launched, MU will lose energy. And the initial malicious interference power P in the perception phase ms Reduce the predetermined ratio, then the malicious interference power P in the transmission stage m ' t It is expressed as: in, is the collected energy of MU, θ is the attenuation coefficient, T s is the set perception time, T t is the transmission time; if MU does not initiate PUEA, the probability is 1-y, then P will not be reduced any further ms , all the collected energy will be used by MJ to send malicious interference signals; SU will decide whether to perform PUEA detection with probability x and consume energy E p , PUEA detection will succeed with probability λ, that is, it is found that an unknown user maliciously occupies the PU channel. At this time, the transmission power P of SU in the transmission phase t ′ is expressed as: Where ε is the power allocation coefficient, E h is the harvested energy of SU, E s SU senses energy consumption; If SU does not detect when it does not know whether there is a PUEA attack, with a probability of 1-x, and allows MU to successfully occupy PU, the smart contract SC will reduce the SU adjustment value. If MU does not launch a PUEA attack, SU will not have additional adjustment value changes; In the structural arrangement of the evolutionary game attack and defense model, when defending against malicious attacks by MU, especially eavesdropping attacks, the SJ carried by the SU may cause a large amount of collaborative interference to the PU transmission. For the SJ closer to the PU, the impact of this interference is greater.

4. According to claim 3, an evolutionary game security strategy method based on blockchain evaluation and coordination mechanism is characterized in that: Step S2 is specifically implemented as follows: S21. Design a user evaluation mechanism In the EH-CR network, the evaluation value of SU is related to the quality of the spectrum used; when multiple SUs apply to use authorized channels, the base station BS will divide the spectrum of different bandwidths according to their historical performance and the interference margin of the current time slot to stimulate or inhibit the behavior of the corresponding SU; through the transaction ledger of blockchain technology, the contribution of each SU in the system is recorded and updated; in the face of complex network environments and a large number of malicious attacks, at the end of the perception phase of the current time slot, if the corresponding node has a higher detection rate and higher confidentiality capacity in the historical time slot, it means that the corresponding node has a strong anti-interference or anti-attack ability and contributes more to the overall security performance of the network, then the evaluation value of the corresponding node will be higher, and the node with a higher evaluation value will be assigned to a better channel; SU in the kth time slot i Safety rating value It is expressed as: Among them, Φ1 and Φ2 are the evaluation factors of detection probability and confidentiality capacity respectively. For SU i The detection probability at the k-1th time slot is, YesSU i The confidentiality capacity in the k-1th time slot is specifically expressed as: Among them, x + =max(x,0), and are SUs in the k-1th time slot respectively. i The capacity of data transmission channels and eavesdropping channels; For ST i To SR i The channel gain, For ST i Channel gain to Eav; N j is the total number of SJ, N J is the total number of MJ; For SU i The self-interference influence, Ω is the self-interference influence coefficient, P t is the transmission power, z is the attenuation factor, α0 is the path loss, d i,j For SJ j To SR i distance; For MJ J About SU i The malicious interference effect, P mt is the malicious interference power, d i,J For MJ J To SR i distance; For SJ j Collaborative interference with Eav, d e,j For SJ j Distance to Eav; is the self-interference effect on MU, d e,J For MJ J Distance to Eav; is Gaussian noise; SU senses that there are not only PU signals in the network, but also interference signals and PUEA signals emitted by MJ. According to the type of signal, SU i Received signal There are four possible states: Where n0 is Gaussian noise, For MJ J About SU i The perceived interference power, Sig PU is the PU signal, Sig PUEA is the PUEA signal, Indicates that the channel is not occupied by PU and there is no PUEA signal; Indicates that only PU signal exists; It means that only PUEA signal exists, but the actual channel is idle; Indicates that the channel is busy and there is a PUEA signal at the same time; H1 and H0 represent the presence and absence of PU respectively. When the current channel state is H0, MU probabilistically launches PUEA attack. P Indicates launching a PUEA attack, NA P Indicates that no PUEA attack is launched; during the SU perception stage, PUAE detection is performed probabilistically, and at the same time, D A Indicates that SU performs PUEA signal detection and uses ND A If PUEA signal detection is not performed, the false alarm probability P f ′ is expressed as: P′ f =P(D A / H0) (6) According to Bayes' theorem, P f ' is expanded to: P′ f =P(D A / AND P ,H0)P(A P ,H0)+P(D A / ON P ,H0)P(NA P ,H0) (7) Let P(D A / A P ,H0)=β P , when MU does not take PUEA attack, that is, P(A P ,H0)=0,P(NA P ,H0)=1,P′ f =P(D A / NA P ,H0)=P f , where P f is the initial false alarm probability, at which time the SU detection probability P′ d =P d ; When MU takes PUEA attack, that is, P(A P ,H0)=1,P(NA P ,H0)=0,P′ f =P(D A / A P ,H0)=β P , SU detection probability P′ d It is expressed as: Among them, T s is the perceived time, f s is the initial perception frequency, Q is the Gaussian tail function, γ s ′ is the signal-to-interference-to-noise ratio in the perception stage; to quantify the perception benefit of MU, P e It represents the false detection rate of SU, combined with the missed detection rate and false alarm probability, the false detection rate P e It is expressed as: P e =P(H0)P′ f +P(H1)P′ m (9) Among them, P m ′=1-P d ′ is the missed detection rate in the SU perception stage, P(H0) is the probability that PU does not exist, and P(H1) is the probability that PU exists; to evaluate the continuous security evaluation value of the user, the final evaluation value is calculated on a predetermined time scale; at the same time, considering that the evaluation value in the most recent time slot can better reflect the user's activity, its proportional weight will be set larger; through the calculation and weight allocation of historical evaluation values, the final average evaluation value of SU Use a sequence to effectively respond, specifically expressed as: Among them, N k is the number of iterations of the current time slot, r is the time sequence traversal sequence, k is the current time slot, q is the attenuation coefficient, represents the evaluation value when the time slot is r. For the specific expression, refer to formula (3). When the traversal r is closer to the current time slot k, the attenuation is (1-q) k-r The smaller it is, the greater the weight of its impact on the final evaluation value; S22. Design user coordination mechanism When SJ transmits cooperative interference signals to degrade the eavesdropping channel, it will cause different degrees of interference to PU, and the closer the SJ is to the PU, the greater the interference. Therefore, a variable is needed in the EH-CR network to quantify this PU interference. Therefore, the interference coordination value V is proposed. i To express SU i Interference impact on PU network: Among them, g is the PU traversal sequence, I i,g Indicates SJ i About PU g The interference effect, I th Interference threshold set for PU network; Combined with the change of SU behavior decision in step S21, that is, whether to detect PUEA attack, PU changes I th To feedback the behavior strategy selected by SU, the interference threshold I th Specifically expressed as: Where I0 represents the PU initial interference threshold, I v Indicates the threshold adjustment change value; when MU selects PUEA attack and SU successfully detects the attack, it contributes to the network, and PU rewards this security behavior. The threshold strategy selection If the SU does not detect the attack or fails to detect, the PU will punish and select a strategy When MU chooses not to conduct PUEA attack, SU's strategy selection will not affect its permissible interference threshold. Safety evaluation value of the current time slot and interference coordination value V i (k) It reflects the contribution made to the EH-CR network, and uses the improved Sigmod activation function to integrate the influence of the two factors; the Sigmod activation function is an increasing function with a dependent variable range of (0,1), combined with the form of blockchain SC bandwidth allocation to quantify the overall contribution of users, and the bandwidth reward of the kth time slot It is expressed as: Among them, ρ3 and ρ4 are weight coefficients.

5. According to claim 4, an evolutionary game security strategy method based on blockchain evaluation and coordination mechanism is characterized in that: Step S2 also includes the following steps: S23. Design consensus mechanism In order to encourage users to improve the quality of security performance and balance the behavior of collaborative interference, the target value of the PoW algorithm is improved so that users with higher evaluation and coordination values ​​have a greater probability of updating blocks and obtaining block rewards. Through the adjustment of mining difficulty, secure users with higher evaluation and coordination values ​​will spend less on mining. Each user node searches for the value that meets the conditions through mining. i The mining difficulty is recorded as n is the serial number of the new block to be generated. Combined with the relevant information in the block header, the mining expression that the corresponding node needs to satisfy is as follows: Among them, T B The timestamp generated for the new block, Nonce is a random number, Hash p It is the hash value pointing to the previous block. In the blockchain system, the process of updating the block will be carried out after the evaluation and coordination are completed. The mining difficulty of each user with different security performance Expressed as: Among them, n is the initial difficulty of generating blocks, ζ is the difficulty attenuation coefficient, ξ is a positive constant parameter, a is the evaluation value influence coefficient, b is the coordination value influence coefficient, and c is a constant parameter.

6. According to claim 4, an evolutionary game security strategy method based on blockchain evaluation and coordination mechanism is characterized in that: Step S3 is specifically implemented as follows: In the evolutionary game attack and defense model, the two game parties SU and MU influence each other through the choice of their own strategies and adjust according to the other party's behavior and benefits; through continuous learning and adjustment, both parties optimize their own strategies and maximize their own interests; in each iteration, both parties in the evolutionary game modify the probability of a decision with a predetermined strategy, combining energy harvesting EH technology and blockchain evaluation and coordination mechanism to form their own evolutionary direction; In the EH-CR network, SU will face a large number of malicious attacks. Based on the limited energy collection, SU selfishly maximizes its own interests. Both parties choose the best strategy according to their own utility changes. SU's strategy choice is S s ={D A ,ND A }, D A Represents the detection of PUEA and consumes some energy, ND A Indicates no detection; MU's strategy selection is S e ={A P ,NA P }, A P Indicates that the PUEA attack strategy is selected and a portion of the energy consumption used for interference is reduced. NA P It means no attack will be carried out, and the energy of malicious interference will not be consumed extra; Through the decision-making choices of SU and MU, define the benefit {U s ,U e }; Case 1: SU strategy selects D A , MU strategy chooses A P , both sides will consume a lot of energy on PUEA's attack and defense, and the benefits of both sides of the game can be expressed by the following equation: (1) SU utility is: Among them, the interference effect of utility function is I s ' t =ΩεP t 'h st , I′ mr =P m ' t h mr , I s ' e =εP t 'h se , I′ me =ΩP m ' t h me , α1, α2 are utility conversion factors, M i To reward bandwidth, For case 1, the SU detection probability is related to the false alarm probability P′ f function, and C1(I th ) is the SU confidentiality capacity in case 1 with respect to the interference threshold I th The interference threshold affects the cooperation power P in the network. j The maximum value of satisfies the following relationship: Among them, SJ j Average channel gain to each PU For SJ j Average distance to PU; When SU ​​and MU execute this strategy, the perceived signal may be a real PU signal or a PUEA simulated signal. If the detection success rate is λ, the SU perceived signal-to-noise ratio It is expressed as: Among them, P P is the PU transmission power, h i,g It is PU g To SU i The channel gain, P A is the PUEA simulation signal power, h i,A Transmit PUEA signal to SU for MJ i The channel gain, P′ ms =(1-θ)P ms MJ in the perception phase of PUEA attack J The malicious interference power, h i,J For MJ J To SU i The channel gain, is the Gaussian noise in the perception stage; (2) MU utility is: Among them, α3 and α4 are utility conversion factors; In case 1, the SU false detection rate is about P f ′, and in this case P f ′=β P ; P(H0) is the probability that PU does not exist, and P(H1) is the probability that PU exists; The SU eavesdropping channel capacity in case 1 is about the interference threshold I th Because the SU effectively detects the PUEA attack and makes a greater contribution to network security, the PU network rewards the SU and increases its permissible interference threshold I th , i.e., selection strategy Case 2: SU strategy selects D A , MU strategy selects NA P , SU consumes energy on PUEA defense, MU maintains the original strategy, then the benefits of both parties in the game are expressed by the following equation: (1) SU utility is: (2) MU utility is: Among them, MU did not perform PUEA attack, but SU performed detection, which consumed additional detection energy, and the transmission power was P t ′, but SU behavior has no effect on PU network, so PU interference threshold selection strategy At this time, SU perceives the signal-to-interference-noise ratio It is expressed as: Case 3: SU strategy selects ND A , MU strategy chooses A P MU consumes extra PUEA energy, and the malicious interference power in the transmission stage is P m ' t SU maintains the original strategy, that is, the transmission power remains at the initial value P t , at this time, the benefits of both parties in the game are expressed by the following equation: (1) SU utility is: (2) MU utility is: Among them, MU adopts the PUEA strategy, but SU does not detect the attack behavior, resulting in a decrease in network security performance, so PU chooses the interference threshold strategy At this time, the perceived signal-to-dryness ratio It can be expressed as: When SU ​​is not detecting, the initial power P is maintained during the transmission phase. t It is expressed as: Case 4: SU strategy selects ND A , MU strategy selects NA P , MU and SU do not consume energy in PUEA attack and defense, both sides maintain the initial state, and the gains of both sides of the game are expressed by the following equation: (1) SU utility is: (2) MU utility is: Among them, MU did not perform PUEA, SU did not detect the attack behavior, and had no impact on the PU network, so the interference threshold selection strategy was set At this time, the SU perception signal-to-noise ratio in case 4 is the same as equation (22), and the false alarm rate is P f Same as case 2, so Because SU consumes more energy to detect PUEA in case 2 than in case 4, the transmission power in the transmission phase is different, so the confidentiality capacity and the wiretap channel throughput are also different, that is, 7. The evolutionary game security strategy method based on blockchain evaluation and coordination mechanism according to claim 6 is characterized in that: In step S3, after establishing the mathematical model based on evolutionary game, it is necessary to conduct utility analysis of SU decision and MU decision, as follows: (1) SU decision analysis From Case 1 and Case 2, we can get the average benefit U of SU for PUEA detection. s (D A )for: From cases 3 and 4, we can get the average benefit U of SU without PUEA detection. s (ND A )for: Combining equations (29) and (30), the average utility of SU behavior decision is: (2) MU decision analysis From cases 1 and 3, we can get the average benefit U of MU in PUEA. e (A P )for: From Case 2 and Case 4, we can conclude that the average benefit of MU without PUEA is: Combining equations (32) and (33), the average utility of MU behavior decision is:

8. According to claim 7, an evolutionary game security strategy method based on blockchain evaluation and coordination mechanism is characterized in that: Step S4 is specifically implemented as follows: Set S D Behavioral strategy for SU, S A is the MU behavior strategy. If the local equilibrium strategy is adopted The utility of SU after evolution is the largest, then the strategy is ESS, that is: To simplify the equation expression, the function name is replaced by the variable name, and equations (29) and (31) are substituted into the basic form of the replication dynamic equation in the following evolutionary game to obtain the SU replication dynamic equation expression (36); Among them, s i is the set of evolutionary group game strategies, x i is the probability of a player choosing a strategy at time t, F(s i ,x) represents the selection strategy s i , F(x,x) represents the average benefit of the entire group; U s (D A )-U s (ND A ) is expressed as: The iterative process of the dynamic evolution of SU is described by the following equation: Among them, P si (D A ,k) is the strategy D adopted by SU in the current time slot k A The probability of s Adjust the step size for SU iteration; Substituting equations (32) and (34) into the basic form of the replication dynamic equation in the evolutionary game, we obtain the expression of the MU replication dynamic equation: Among them, U e (A P )-U e (NA P ) is expressed as: The dynamic evolution of MU is described by the following equation: P e (A P ,k+1)=P e (A P ,k)+μ e F e (y) (41) Where k is the current iteration time slot, P e (A P ,k) Adopt strategy A for the current time slot P The probability of μ e It is the iterative adjustment step size of MU; the replication dynamic equation reflects the strategy adjustment trend of SU and MU when they evolve and learn from each other. When the equation is equal to 0, it means that the game parties have reached equilibrium and obtained a stable equilibrium point, that is, let dx / dt=0 and dy / dt=0, and solve for 4 conventional equilibrium points: (1,0), (0,1), (0,0), (1,1), and a mixed equilibrium point: (x * ,y * ), the mixed equilibrium point is specifically expressed as: set up and For SU and MU to be in stable state, the above five stable points are where the utility function reaches convergence, that is, The local stable point at the time; however, the local stable point is not necessarily the ESS. As the evolution proceeds, the utility of the later local stable point may exceed the utility of the previous stable point. The Jacobian matrix is ​​used to verify the final ESS. The Jacobian matrix J m It is expressed as: Among them, the elements of the Jacobian matrix are specifically expressed as: According to the characteristics of linear systems, if the Jacobian matrix trace tr(J m ) is negative, the system evolves in a stable direction; if the trace is positive, the system evolves in an unstable direction; the determinant of the Jacobian matrix is ​​denoted by det(J m ), if the determinant is positive, it means that the disturbance of the system will not grow; if the determinant is negative, it means that the disturbance of the system will grow; in order to quickly verify whether the local stable point is the global stable point, calculate the trace tr(J m ) is less than 0 and the determinant det(J m ) is greater than 0 to identify ESS; (1) The stable point is the mixed equilibrium point (x * ,y * ), we can get that all elements of the Jacobian matrix are equal to 0, then tr(J m )=det(J m )=0, so (x * ,y * ) is not an ESS; (2) When the stable point is (1,0), the Jacobian matrix is ​​specifically expressed as: At this time, the Jacobian matrix When the other two elements are less than 0, and When (1,0) is ESS, then SU ​​detects energy consumption E p And the MU attack energy consumption attenuation factor θ satisfies the following conditions: The polynomial is specifically expressed as: (3) When the stable point is (0,1), the Jacobian matrix is ​​specifically expressed as: when and When (0,1) is ESS, then E p and θ satisfy the following conditions: Among them, the polynomial is specifically expressed with relevant parameters as follows: (4) When the stable point is (0,0), the Jacobian matrix is ​​specifically expressed as: when and When (0,0) is ESS, then E p and θ satisfy the following conditions: in, is the maximum allowable detection energy consumption; (5) When the stable point is (1,1), the Jacobian matrix is ​​specifically expressed as: when and When (1,1) is ESS, then E p and θ satisfy the following conditions: Based on the above analysis, the convergence between the two parties in the game will tend to one of the four stable points. p The different conditions satisfied by and θ result in different final ESS. p When and θ satisfy equation (46), the evolution will converge to the stable point (1,0), at which point all SUs will choose to detect PUEA, and MUs will not attack. p When and θ satisfy equation (53), the evolution will converge to the stable point (1,1), at which time the probability of all SUs detecting PUEA and MUs attacking is 1; the convergence of the stable points (1,0) and (1,1) effectively stimulates all SUs to perform PUEA detection, thereby improving the overall security performance of the EH-CR network on the basis of maximizing the SU utility.

9. An evolutionary game security strategy system based on blockchain evaluation and coordination mechanism, characterized in that: The method comprises a memory, a processor and computer program instructions stored in the memory and capable of being executed by the processor. When the processor executes the computer program instructions, the method steps as claimed in any one of claims 1 to 8 can be implemented.

10. A computer-readable storage medium having stored thereon computer program instructions that can be executed by a processor, and when the processor executes the computer program instructions, the method steps according to any one of claims 1 to 8 can be implemented.

Citation Information

Cited By

  • Multi-agent network security defense method and system based on consensus game

    CN120263559A

  • A Multi-Agent Network Security Defense Method and System Based on Consensus Game

    CN120263559B