Anti-interference CSMA / CA protocol execution method, device and storage medium

Abstract

The present invention provides an anti-interference CSMA / CA protocol execution method, device, and storage medium. The anti-interference CSMA / CA protocol is constructed based on the CSMA / CA protocol and decision adjustment parameters. The method comprises sensing the current environmental state and obtaining a decision of the current environmental state from a preset decision set based on epsilon-greedy. The decision is distributed to a communication node, which performs an anti-interference CSMA / CA protocol task based on the decision and senses the next environmental state. The method further comprises calculating an experience value obtained by selecting a decision in the current environmental state. The method further comprises performing Q value optimization processing through Q learning, and transferring the current environmental state to the next environmental state. The method optimizes the accumulated experience value to find the optimal decision adjustment parameter, obtains a new decision based on the decision adjustment parameter, and executes the anti-interference CSMA / CA protocol according to the new decision, thereby achieving more reasonable backoff decisions for different environments and improving the anti-interference performance of the CSMA / CA protocol.

Description

Technical Field

[0001] The present invention mainly relates to the technical field of intelligent anti-interference processing of communications, and in particular to an anti-interference CSMA / CA protocol execution method, device and storage medium. Background Art

[0002] With the rapid development of multi-user wireless access devices, multi-user application scenarios require high bandwidth efficiency and low latency services. However, the open nature of wireless channels makes them susceptible to malicious interference from jammers, which can degrade network reliability, privacy, and security. The CSMA / CA protocol, widely used in MAC access mechanisms, addresses multi-user communication issues through a binary back-off mechanism (BEB), which reduces inter-node interference to a certain extent and improves communication performance. However, the binary back-off mechanism of the CSMA / CA protocol is susceptible to exploitation by jammers. Malicious jammers can manipulate legitimate nodes to believe the channel is busy, triggering the back-off mechanism and resulting in significant packet loss or delayed transmission. In this process, malicious jammers only require a small amount of power to paralyze the network. Current protocol anti-interference methods primarily focus on interference detection, followed by improved anti-interference performance through methods such as routing, channel allocation, and power control. However, these methods do not address the issue of protocol anti-interference performance. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide an anti-interference CSMA / CA protocol execution method, device and storage medium in response to the deficiencies of the existing technology.

[0004] The present invention solves the above technical problems with the following technical solution: a method for executing an anti-interference CSMA / CA protocol, comprising the following steps:

[0005] S1: Construct an anti-interference CSMA / CA protocol based on the CSMA / CA protocol and decision adjustment parameters;

[0006] S2: Sense the current environment state and obtain a decision based on the current environment state from a preset decision set using the epsilon-greedy algorithm;

[0007] S3: Distribute the decision to the communication node, and the communication node performs the anti-interference CSMA / CA protocol task according to the decision and senses the next environmental state;

[0008] S4: Calculating the experience value obtained by selecting the decision under the current environmental state;

[0009] S5: Accumulate the experience value under the current environmental state and the experience value under all historical environmental states, optimize the accumulated experience value using a Q learning algorithm, and transfer the current environmental state to the next environmental state;

[0010] S6: Repeat steps S2-S5 until the anti-interference CSMA / CA protocol task is completed.

[0011] Another technical solution of the present invention to solve the above technical problem is as follows: an anti-interference CSMA / CA protocol execution device, comprising:

[0012] A protocol building module, used to build an anti-interference CSMA / CA protocol based on the CSMA / CA protocol and decision adjustment parameters;

[0013] The protocol execution module is used to sense the current environment state through the sensing node and obtain the decision of the current environment state from the preset decision set according to the epsilon-greedy algorithm;

[0014] Distributing the decision to the communication node, the communication node performs the task of the anti-interference CSMA / CA protocol according to the decision and senses the next environmental state;

[0015] Calculating the experience value obtained by selecting the decision under the current environmental state;

[0016] The experience value under the current environmental state is accumulated with the experience value under all historical environmental states, the accumulated experience value is optimized using the Q learning algorithm, and the current environmental state is transferred to the next environmental state. The above process is repeated until the task execution of the anti-interference CSMA / CA protocol is completed.

[0017] Another technical solution of the present invention to solve the above-mentioned technical problem is as follows: An anti-interference CSMA / CA protocol execution device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the anti-interference CSMA / CA protocol execution method described above is implemented.

[0018] Another technical solution of the present invention to solve the above technical problem is as follows: a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the anti-interference CSMA / CA protocol execution method as described above is implemented.

[0019] The beneficial effects of the present invention are as follows: an anti-interference CSMA / CA protocol is constructed based on the CSMA / CA protocol and decision adjustment parameters, the optimal decision adjustment parameters are found by optimizing the accumulated experience value, a new decision is obtained through the decision adjustment parameters, and the anti-interference CSMA / CA protocol is executed according to the new decision, thereby achieving more reasonable backoff decisions for different environments and improving the anti-interference performance of the CSMA / CA protocol. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A flowchart of a method for executing an anti-interference CSMA / CA protocol provided by an embodiment of the present invention;

[0021] Figure 2 A schematic diagram of the functional modules of the anti-interference CSMA / CA protocol execution device provided in an embodiment of the present invention;

[0022] Figure 3 A schematic diagram of the structure of a learning framework provided by an embodiment of the present invention;

[0023] Figure 4 This is a flowchart of a perception node provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0025] Example 1:

[0026] like Figure 1 As shown, a method for executing the anti-interference CSMA / CA protocol includes the following steps:

[0027] S1: Construct an anti-interference CSMA / CA protocol based on the CSMA / CA protocol and decision adjustment parameters;

[0028] S2: Sense the current environment state and obtain a decision based on the current environment state from a preset decision set using the epsilon-greedy algorithm;

[0029] S3: Distribute the decision to the communication node, and the communication node performs the anti-interference CSMA / CA protocol task according to the decision and senses the next environmental state;

[0030] S4: Calculating the experience value obtained by selecting the decision under the current environmental state;

[0031] S5: Accumulate the experience value under the current environmental state and the experience value under all historical environmental states, optimize the accumulated experience value using a Q learning algorithm, and transfer the current environmental state to the next environmental state;

[0032] S6: Repeat steps S2-S5 until the anti-interference CSMA / CA protocol task is completed.

[0033] It should be understood that perceiving the next environmental state can be understood as distributing the decision to the communication node. At this time, some states are the current state and the next moment state. The state update process needs to be updated after the Q learning algorithm is optimized.

[0034] It should be understood that the Epsilon-greed method is: before making a decision, a random number seed is randomly generated, 0≤seed≤1. It is compared with the set greed factor ε. If seed < ε, a random decision is selected, otherwise a greedy decision is selected. The decision is maxQ(s i+1 ,a i+1 ) i+1 After the execution is completed, the greed factor value is reduced by the decay exponential until it is reduced to 0.1.

[0035] In the above embodiment, an anti-interference CSMA / CA protocol is constructed based on the CSMA / CA protocol and decision adjustment parameters, the accumulated experience value is optimized to find the optimal decision adjustment parameters, a new decision is obtained through the decision adjustment parameters, and the anti-interference CSMA / CA protocol is executed according to the new decision, thereby making more reasonable backoff decisions for different environments and improving the anti-interference performance of the CSMA / CA protocol.

[0036] On the basis of the above embodiment, the anti-interference CSMA / CA protocol is constructed based on the CSMA / CA protocol and the decision adjustment parameters, specifically:

[0037] Define the decision adjustment parameter as a, and redefine the contention window CW based on the contention window CW and the decision adjustment parameter a in the CSMA / CA protocol to obtain the contention window CW', where CW∈[CW min , CW max ],CW'∈a·[CW min ,CW max ];

[0038] Define the state space S, s i ∈S:S={s1,s2,...,s n}, where s i =a i CW' represents the environmental state at time i, a i CW' means that by decision a i Adjusted competition window CW',

[0039] Define the decision space A, a i ∈A:A={a1,a2,...,an}, where a i Indicates the decision to adjust the contention window CW' at time i,

[0040] Define the decision transition probability P, P = p{s i+1 |s i ,a i}, where p represents the state of the environment s i Next selection decision a i Transfer to s i+1 The probability of

[0041] Define the return value R, R(s i ,a i ) indicates that in the environmental state s i Next selection decision a i Experience points gained;

[0042] The anti-interference CSMA / CA protocol is obtained according to the defined state space S, decision space A, decision transition probability P, reward value R and contention window CW'.

[0043] It should be understood that the CSMA / CA algorithm process is as follows: when a node needs to send data, it needs to detect the channel first. If it senses that the channel is idle, it will send it directly; if it senses that the channel is busy, it needs to wait. When the sensing node senses that the channel is busy, it will randomly generate a backoff time BackoffTime, and its value range is BackoffTime∈[0,CW-1]. Among them, CW is called the contention window, and its value range is CW∈[CW min , CW max ]. On the first fallback attempt, it will be set to CW=CW min Each time an attempt fails, the contention window will double, that is, CW = 2·CW; until it reaches the maximum value, that is, CW = CW max , then discard the packet. When the fallback attempt succeeds, set CW to CW=CW min .

[0044] The present invention uses the idea of ​​reinforcement learning to achieve the maximum experience value through learning strategies in the process of continuous interaction with the environment. The present invention sets a decision adjustment parameter a for the CSMA / CA algorithm, namely CW'∈a·[CW min ,CW max ]. The decision adjustment parameter a is learned through reinforcement learning, so that the optimal decision can be made for different environments, thereby improving the anti-interference performance of the CSMA / CA protocol.

[0045] Based on the above embodiment, the calculation of the experience value obtained by selecting the decision under the current environmental state is specifically as follows:

[0046] Obtain the number of successful transmissions per unit time, the length of the unit time and the transmission rate of the channel from the channel;

[0047] Calculate the ratio of the number of successful transmissions per unit time to the length of the unit time to obtain the throughput;

[0048] A ratio of the throughput to the transmission rate of the channel is calculated to obtain a normalized throughput, and the normalized throughput is used as an empirical value.

[0049] That is, S is the throughput, and the calculation method of throughput S is:

[0050]

[0051] The normalized throughput is calculated as:

[0052]

[0053] Based on the above embodiment, the optimization process of the accumulated experience value using the Q learning algorithm is specifically as follows:

[0054] The Q-value is updated using the Q-learning algorithm, thereby optimizing the accumulated experience value and obtaining the maximum accumulated experience value Θ. The maximum accumulated experience value Θ is used as the optimization target, and the optimization target is:

[0055]

[0056] Among them, R i is the experience value at time i, E π is the cumulative experience value taken under decision π, and A represents the decision space.

[0057] Since the decision transition probability P is unknown in the above processing, it is necessary to use the Q value update process to replace the role of the transition probability P.

[0058] Based on the above embodiment, the Q-value update using the Q-learning algorithm is specifically as follows:

[0059] The Q value is updated using the Q value update formula of the Q learning algorithm, which is:

[0060]

[0061] Where θ represents the learning rate, which is used to adjust the influence of state and experience value on Q value, γ represents the discount factor, that is, the influence of long-term return (experience value) on decision selection; r i Indicates that the environment state s i Next choice decision a i Experience points gained, si+1 Indicates that the environment state s i Selection decision a i Then transfer to the next environment state, Indicates that the environment state s i+1 The maximum Q value selected.

[0062] In the above embodiment, the purpose is to find the optimal adjustment parameters to maximize the throughput of the network, and the optimization goal is to maximize the accumulated reward value, thereby improving its network performance.

[0063] Example 2:

[0064] like Figure 2 As shown, an anti-interference CSMA / CA protocol execution device includes:

[0065] A protocol building module, used to build an anti-interference CSMA / CA protocol based on the CSMA / CA protocol and decision adjustment parameters;

[0066] The protocol execution module is used to sense the current environment state through the sensing node and obtain the decision of the current environment state from the preset decision set according to the epsilon-greedy algorithm;

[0067] Distributing the decision to the communication node, the communication node performs the task of the anti-interference CSMA / CA protocol according to the decision and senses the next environmental state;

[0068] Calculating the experience value obtained by selecting the decision under the current environmental state;

[0069] The experience value under the current environmental state is accumulated with the experience value under all historical environmental states, the accumulated experience value is optimized using the Q learning algorithm, and the current environmental state is transferred to the next environmental state. The above process is repeated until the task execution of the anti-interference CSMA / CA protocol is completed.

[0070] On the basis of the above embodiment, in the protocol construction module, an anti-interference CSMA / CA protocol is constructed based on the CSMA / CA protocol and the decision adjustment parameters, specifically:

[0071] Define the decision adjustment parameter as a, and redefine the contention window CW based on the contention window CW and the decision adjustment parameter a in the CSMA / CA protocol to obtain the contention window CW', where CW∈[CW min , CW max ],CW'∈a·[CW min ,CW max ];

[0072] Define the state space S, s i∈S:S={s1,s2,...,s n}, where s i =a i CW' represents the environmental state at time i, a i CW' means that by decision a i Adjusted competition window CW',

[0073] Define the decision space A, a i ∈A:A={a1,a2,...,a n}, where a i Indicates the decision to adjust the contention window CW' at time i,

[0074] Define the decision transition probability P, P = p{s i+1 |s i ,a i}, where p represents the state of the environment s i Next selection decision a i Transfer to s i+1 The probability of

[0075] Define the return value R, R(s i ,a i ) indicates that in the environmental state s i Next selection decision a i Experience points gained;

[0076] The anti-interference CSMA / CA protocol is obtained according to the defined state space S, decision space A, decision transition probability P, reward value R and contention window CW'.

[0077] Based on the above embodiment, the calculation of the experience value obtained by selecting the decision under the current environmental state is specifically as follows:

[0078] Obtain the number of successful transmissions per unit time, the length of the unit time and the transmission rate of the channel from the channel;

[0079] Calculate the ratio of the number of successful transmissions per unit time to the length of the unit time to obtain the throughput;

[0080] A ratio of the throughput to the transmission rate of the channel is calculated to obtain a normalized throughput, and the normalized throughput is used as an empirical value.

[0081] Specifically, if Figure 3 As shown in Figure 1, the protocol execution module can be viewed as an intelligent perception site, which includes a sensor, a learner, and a decision selector. The perception site and the communication channel serve as a learning framework:

[0082] During the adjustment of the competition window, the current state is perceived by the sensor as s i , then the decision selector selects decision a from the decision set according to the adjustment parameter a i , execute decision a i Then the environment state changes from s i The next state of the transition s i+1 , the communication link is based on the agent decision a i Feedback corresponding reward (ie experience value or return value) r i , the learner is rewarded according to r i Update and optimize the next decision selection, and repeat the corresponding process.

[0083] The goal of the sensing node is to find the optimal adjustment parameters to maximize the network throughput and thus improve its network performance. The optimization goal of the intelligent sensing node is to maximize the cumulative reward value.

[0084] On the basis of the above embodiment, the accumulated experience value is optimized and the optimization result is used as the decision adjustment parameter for the next environmental state, specifically:

[0085] The Q-value is determined by the Q-learning algorithm, and the Q-value update process is used to replace the role of the transition probability P.

[0086] The Q-value is updated using the Q-learning algorithm, thereby optimizing the accumulated experience value and obtaining the maximum accumulated experience value Θ. The maximum accumulated experience value Θ is used as the optimization target, and the optimization target is:

[0087]

[0088] Among them, Ri is the experience value at time i, E π is the cumulative experience value taken under decision π, and A represents the decision space.

[0089] Since the decision transition probability P is unknown in the above processing, the Q value update process is used to replace the role of the transition probability P.

[0090] Based on the above embodiment, the Q value is determined by using the Q learning algorithm, specifically:

[0091] The Q value is determined by the Q value update formula of the Q learning algorithm, which is:

[0092]

[0093] Where θ represents the learning rate, which is used to adjust the influence of state and experience value on Q value, γ represents the discount factor, that is, the influence of long-term return on decision selection; r i Indicates that the environment state si Next choice decision a i Experience points gained, s i+1 Represents the sensor in the environment state s i Selection decision a i Then transfer to the next environment state, Indicates that the environment state s i+1 The maximum Q value selected.

[0094] like Figure 4 As shown, the following is a flow chart of the perception node:

[0095] S1.1: Initialize network parameters and sensor node parameters;

[0096] S1.2: The sensing node perceives the current environment state;

[0097] S1.3: The perception node selects a decision based on the epsilon-greedy parameter from the preset decision set to determine the current state of the environment.

[0098] S1.4: The sensing node distributes the decision to each communication node. Each communication node executes the anti-interference CSMA / CA protocol based on the decision, and the sensing node perceives the next environmental state.

[0099] S1.5: Calculate an empirical value obtained by selecting the decision under the current environment state, that is, calculate the normalized throughput of the channel as the empirical value;

[0100] S1.6: Accumulate the experience value of the current environment state and the experience value of all historical environment states, optimize the accumulated experience value using the Q learning algorithm, and transfer the current environment state to the next environment state;

[0101] S1.7: Repeat steps S1.2-S1.6 until the anti-interference CSMA / CA protocol task is completed.

[0102] Simulation experiments compared the proposed protocol with the traditional CSMA / CA protocol under conditions of no interference, random interference, and intermittent interference. The wireless network channel transmission rate was 1 Mbits / s, each slot length was 20 μs, the packet length was 400 μs, the DIFS length was 50 μs, the SIFS length was 10 μs, the ACK length was 20 μs, the minimum contention window value was 8, and the maximum contention window value was 256. The agent reward discount factor γ was 0.9, the learning rate θ was 0.5, the greed factor ε was 1.0, and the decay exponent was 0.0001.

[0103] When the number of nodes is 10, the proposed method improves the normalized throughput by 15.89%, 10.72%, and 12.06% in the interference-free, intermittent, and random interference scenarios, respectively, compared to traditional CSMA / CA. The collision rate is reduced by 46.35%, 35.12%, and 41.94%, respectively. This effectively improves network throughput and significantly reduces the collision rate.

[0104] Example 3:

[0105] An anti-interference CSMA / CA protocol execution device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the anti-interference CSMA / CA protocol execution method described above is implemented.

[0106] Example 4:

[0107] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the anti-interference CSMA / CA protocol execution method as described above is implemented.

[0108] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0109] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for executing an anti-interference CSMA / CA protocol, characterized in that: The steps include: S1: Construct an anti-interference CSMA / CA protocol based on the CSMA / CA protocol and decision adjustment parameters; S2: Sense the current environment state and obtain a decision based on the current environment state from a preset decision set using the epsilon-greedy algorithm; S3: Distribute the decision to the communication node, and the communication node performs the anti-interference CSMA / CA protocol task according to the decision and senses the next environmental state; S4: Calculating the experience value obtained by selecting the decision under the current environmental state; S5: Accumulate the experience value under the current environmental state and the experience value under all historical environmental states, optimize the accumulated experience value using a Q learning algorithm, and transfer the current environmental state to the next environmental state; S6: Repeat steps S2-S5 until the anti-interference CSMA / CA protocol task is completed; The anti-interference CSMA / CA protocol is constructed based on the CSMA / CA protocol and the decision adjustment parameters, specifically: Define the decision adjustment parameter as a, and redefine the contention window CW based on the contention window CW and the decision adjustment parameter a in the CSMA / CA protocol to obtain the contention window CW', where CW∈[CW min ,CW max ],CW'∈a·[CW min ,CW max ]; Define the state space S, s i ∈S:S={s1,s2,...,s n }, where s i =a i CW' represents the environmental state at time i, a i CW' means that by decision a i Adjusted competition window CW', Define the decision space A, a i ∈A:A={a1,a2,...,a n }, where a i Indicates the decision to adjust the contention window CW' at time i, Define the decision transition probability P, P = p{s i+1 |s i ,a i }, where p represents the state of the environment s i Next selection decision a i Transfer to s i+1 The probability of Define the return value R, R(s i ,a i ) indicates that in the environmental state s i Next selection decision a i Experience points gained; The anti-interference CSMA / CA protocol is obtained according to the defined state space S, decision space A, decision transition probability P, reward value R and contention window CW'.

2. The anti-interference CSMA / CA protocol execution method according to claim 1, characterized in that: The calculation of the experience value obtained by selecting the decision under the current environmental state is specifically: Obtain the number of successful transmissions per unit time, the length of the unit time and the transmission rate of the channel from the channel; Calculate the ratio of the number of successful transmissions per unit time to the length of the unit time to obtain the throughput; A ratio of the throughput to the transmission rate of the channel is calculated to obtain a normalized throughput, and the normalized throughput is used as an empirical value.

3. The anti-interference CSMA / CA protocol execution method according to claim 1, characterized in that: The optimization process of the accumulated experience value using the Q learning algorithm is specifically as follows: The Q-value is updated using the Q-learning algorithm, thereby optimizing the accumulated experience value and obtaining the maximum accumulated experience value Θ. The maximum accumulated experience value Θ is used as the optimization target, and the optimization target is: Among them, R i is the experience value at time i, E π is the cumulative experience value taken under decision π, and A represents the decision space.

4. The anti-interference CSMA / CA protocol execution method according to claim 3, characterized in that: The Q-value update using the Q-learning algorithm is specifically as follows: The Q value is updated using the Q value update formula of the Q learning algorithm, which is: Among them, θ represents the learning rate, which is used to adjust the influence of state and experience value on Q value, γ represents the discount factor, r i Indicates that the environment state s i Next choice decision a i Experience points gained, s i+1 Indicates that the environment state s i Selection decision a i Then transfer to the next environment state, Indicates that the environment state s i+1 The maximum Q value selected.

5. An anti-interference CSMA / CA protocol execution device, characterized in that: include: A protocol building module, used to build an anti-interference CSMA / CA protocol based on the CSMA / CA protocol and decision adjustment parameters; The protocol execution module is used to sense the current environment state through the sensing node and obtain the decision of the current environment state from the preset decision set according to the epsilon-greedy algorithm; Distributing the decision to the communication node, the communication node performs the task of the anti-interference CSMA / CA protocol according to the decision and senses the next environmental state; Calculating the experience value obtained by selecting the decision under the current environmental state; The experience value under the current environmental state is accumulated with the experience value under all historical environmental states, the accumulated experience value is optimized using the Q learning algorithm, and the current environmental state is transferred to the next environmental state. The cycle of sensing the current environmental state, obtaining the corresponding decision, distributing the decision and executing the task, sensing the next environmental state, calculating the current experience value, accumulating the experience value and optimizing it, and state transfer is repeated until the task of the anti-interference CSMA / CA protocol is completed; Among them, the anti-interference CSMA / CA protocol is constructed based on the CSMA / CA protocol and decision adjustment parameters, specifically: Define the decision adjustment parameter as a, and redefine the contention window CW based on the contention window CW and the decision adjustment parameter a in the CSMA / CA protocol to obtain the contention window CW', where CW∈[CW min , Cw max ],CW'∈a·[CW min ,CW max ]; Define the state space S, s i ∈S:S={s1,s2,...,s n }, where s i =a i CW' represents the environmental state at time i, a i CW' means that by decision a i Adjusted competition window CW', Define the decision space A, a i ∈A:A={a1,a2,...,a n }, where a i Indicates the decision to adjust the contention window CW' at time i, Define the decision transition probability P, P = p{s i+1 |s i ,a i }, where p represents the state of the environment s i Next selection decision a i Transfer to s i+1 The probability of Define the return value R, R(s i ,a i ) indicates that in the environmental state s i Next selection decision a i Experience points gained; The anti-interference CSMA / CA protocol is obtained according to the defined state space S, decision space A, decision transition probability P, reward value R and contention window CW'.

6. The anti-interference CSMA / CA protocol device according to claim 5, characterized in that: The calculation of the experience value obtained by selecting the decision under the current environmental state is specifically: Obtain the number of successful transmissions per unit time, the length of the unit time and the transmission rate of the channel from the channel; Calculate the ratio of the number of successful transmissions per unit time to the length of the unit time to obtain the throughput; A ratio of the throughput to the transmission rate of the channel is calculated to obtain a normalized throughput, and the normalized throughput is used as an empirical value.

7. An anti-interference CSMA / CA protocol execution device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the anti-interference CSMA / CA protocol execution method according to any one of claims 1 to 4 is implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the anti-interference CSMA / CA protocol execution method according to any one of claims 1 to 4 is implemented.

Images