An Adaptive Access Method for Cognitive Users in a Transmission Deadline Communication Scenario

Abstract

The present invention discloses a method for cognitive user adaptive access in a transmission deadline communication scenario. In this solution, cognitive users can independently observe the channel to obtain the latest estimate of the number of active cognitive users in the current time slot. When the licensed user does not occupy the channel, the cognitive user determines the access probability that maximizes the current expected reward under the multi-packet reception mechanism according to the confidence state distribution of the number of active cognitive users obtained from the network observation. For cognitive users, the packet arrival probability conforms to the binomial distribution and has a strict transmission deadline. The system uses an acknowledgment and retransmission mechanism, and the receiver has the multi-packet reception ability (i.e., can successfully receive multiple packets simultaneously). A method for cognitive user adaptive access in a transmission deadline communication scenario also gives the access probability calculation method, which can make the network reliability close to the theoretical optimal reliability while ensuring a low computational complexity.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless network communication, and specifically relates to a method for cognitive user adaptive access in a transmission time limit communication scenario. Background Art

[0002] With the rapid development of wireless communication technology, spectrum resources are becoming increasingly scarce. In addition, research on the actual utilization rate of the current spectrum shows that the utilization rate of the entire wireless spectrum is very low. This is because the current spectrum allocation and sharing methods are mainly static sharing, that is, fixed frequency bands are allocated to specific users for use, resulting in low spectrum resource utilization. Cognitive radio is a hot technology in the field of wireless communication in recent years. Its proposal is to solve the problem of low spectrum resource utilization. It allows the original licensed band system to open the spectrum resource, so that other systems can opportunistically use the band without affecting the normal use of the channel by the licensed users, and has received the attention of governments and international standard organizations around the world. Currently, organizations and government agencies such as the International Telecommunication Union (ITU), the Institute of Electrical and Electronics Engineers, the Software Defined Radio Forum (SDR), and the Federal Communications Commission (FCC) in the United States are actively researching and formulating corresponding dynamic spectrum sharing standards to standardize the industrial specifications that spectrum sharing technologies such as cognitive radio and cognitive networks should follow. Cognitive users obtain the opportunity to share the spectrum by listening to and observing the spectrum to understand the resource status of the spectrum. A typical system is, for example, IEEE802.22 WRAN, which is an opportunistic sharing of the television broadcast band. With the enhancement of the wireless signal processing capabilities of software and hardware, the physical layers of many protocols already have the ability to receive multiple packets. However, there is still a lack of an efficient and reliable adaptive access strategy for cognitive users in strict transmission time limit and multiple packet reception scenarios. Summary of the Invention

[0003] The present invention aims to provide a method for cognitive user adaptive access in a transmission time limit communication scenario, providing an access probability strategy that makes the reliability close to the optimal for a communication system under a multiple packet reception mechanism, and having outstanding advantages in terms of computational complexity.

[0004] The technical solution for achieving the object of the present invention is: a method for cognitive user adaptive access in a transmission time limit communication scenario, characterized in that the specific steps are as follows:

[0005] S1: Determine the parameters of the partially observable Markov decision process. The relevant parameters include: the number of cognitive users N, describing the number of active cognitive users n at time slot t t , the channel occupancy state c of the licensed user t , taking n t and c t together as the system state s t , the confidence state b of the cognitive user t, the access probability p of cognitive users t . Describes that at time slot t, the system is in state s t =s, cognitive users with p t = p probability access, when the time slot arrives at t+1, the system is in s t+1 = state transition probability f of s′ t (s′,s,p). Describes the observation result o of the cognitive user at time slot t t . Describes that at time slot t, the system is in state s t =s, cognitive users with p t = p probability access, in the next time slot, the system state is s t+1 =s′, the cognitive user obtains the observation result o at time slot t t = the probability v of o t (o,s′,s,p) describes the probability g of the cognitive user generating a data packet at the start of each superframe. The internal operation step of the cognitive user jumps to S2.

[0006] S2: The parameters describing a cognitive user adaptive access method in a transmission time-limited communication scenario include: the number of time slots D in a superframe, the channel multi-packet receiving capability m; the confidence state b1 maintained by the cognitive user at t=0, and the system operation steps jump to S3;

[0007] S3: The cognitive user sets the current time slot to t=t+1. At the beginning of time slot t, the cognitive user performs carrier sensing on the channel and obtains the sensing result c. t , the cognitive user decides whether to estimate b based on its latest confidence state of the active cognitive user according to different listening results t Take the immediate reward r t (b,c,p) maximum access probability p t The data packet is sent. The internal operation step of the cognitive user jumps to S4;

[0008] S4: At the end of time slot t, and if the authorized user in the current time slot does not occupy the channel, each cognitive user will monitor the response signal broadcast by the main channel. If a cognitive user successfully transmits in the current time slot, the response signal is ACK j , j∈(1,2...,m); if more than m data packets are transmitted, a transmission conflict occurs, and the corresponding response signal is NACK. The cognitive user can determine whether the current time slot itself has been successfully transmitted based on the response signal. If it is successfully transmitted, it will exit the channel competition, otherwise it will continue to send data and compete for the channel. The internal operation steps of the cognitive user jump to S5;

[0009] S5: In time slot t (t≠D), there are cognitive users with remaining data packets to be sent, so the cognitive users update the confidence state to obtain bt+1 (n') and jump to S3; in the case of t = D, jump to S6;

[0010] S6: The current superframe ends, and jump to S1;

[0011] Compared with the prior art, a significant advantage of a cognitive user adaptive access method in a transmission deadline communication scenario is that: through this scheme, a low-algorithm-complexity and high-reliability random adaptive access strategy can be provided for cognitive users in a cognitive radio with a multi-packet reception and retransmission mechanism, helping cognitive users achieve high-reliability transmission in limited time slots. Description of the Drawings

[0012] Figure 1 It is a flowchart of a cognitive user adaptive access method in a transmission deadline communication scenario.

[0013] Figure 2 It is a simulation result graph of the network reliability of a cognitive user adaptive access method in a transmission deadline communication scenario and a control scheme varying with the packet generation probability g.

[0014] Figure 3 It is a simulation result graph of the network reliability of a cognitive user adaptive access method in a transmission deadline communication scenario and a control scheme varying with the number of time slots D in a superframe.

[0015] Figure 4 It is a simulation result graph of the network reliability of a cognitive user adaptive access method in a transmission deadline communication scenario and a control scheme varying with the multi-packet reception ability m.

[0016] Figure 5 It is a simulation result graph of the network reliability of a cognitive user adaptive access method in a transmission deadline communication scenario and a control scheme varying with the channel transition probability q. Detailed Embodiments

[0017] In a cognitive user adaptive access method in a transmission deadline communication scenario, N cognitive users in the network independently compete to access the channel and perform data packet transmission when the authorized user does not access the channel, using an acknowledgment, retransmission, and multi-packet reception mechanism. At the beginning of each time slot, the cognitive user actively listens to the busy or idle state of the channel. If the channel is busy, no data is sent. Otherwise, each cognitive user selects a transmission probability that maximizes the immediate reward based on the information it has to send the data packet. Through the acknowledgment signal, the cognitive user can accurately know the number of users and user IDs that have successfully transmitted in the current time slot. Once the data packet that the cognitive user needs to send is successfully transmitted, it no longer participates in the competition. Each cognitive user synchronizes the network parameters at the beginning of the superframe and independently generates data packets according to the probability g. The specific steps are as follows:

[0018] S1: Discretize the system time of each cognitive user into multiple superframes, where each superframe contains multiple time slots \(t\in\{1,2,\cdots,D\}\). The relevant parameters include: the total number of cognitive users \(N\), the binomial distribution parameter \(g\) describing the probability of a data packet arriving at the cognitive user device queue at the start of a superframe, the channel's multi-packet reception capacity \(m\), the total number of time slots \(D\) in a superframe, the access probability \(p\) of a cognitive user to the channel, the probability \(q\) that the channel changes from busy to idle, the probability \(w\) that the channel changes from idle to busy, and at time slot \(t = 1\), initialize the belief state distribution of the number of active cognitive users.

[0019] b1 = h g = [(1 - g) N , Ng(1 - g) N-1 ,\(\cdots\), g N

[0020] S2: At the start of each time slot, a cognitive user listens to the state \(c\in\{0,1\}\) of the channel being idle or occupied due to the use of an authorized user. If \(c = 0\) and there is a cognitive user with a data packet to be sent, the cognitive user sends the data packet with the access probability \(p\) that maximizes the immediate reward based on the current belief state distribution, where the definition of the immediate reward is: t \(\in\{0,1\}\), if \(c t = 0\) and there is a cognitive user with a data packet to be sent, the cognitive user sends the data packet with the access probability \(p\) that maximizes the immediate reward based on the current belief state distribution, where the definition of the immediate reward is:

[0021]

[0022] The transmission probability strategy corresponding to maximizing the immediate reward is:

[0023]

[0024] S3: If there is an active cognitive user with a data packet to be sent in the current time slot, the active cognitive user will receive the response signal broadcast on the primary channel; according to the response signal, the observation of the channel by the cognitive user can be defined as:

[0025] o t \(\in O=\{0,1,\cdots,m,*\}\)

[0026] where \(o t = o(0\leq o\leq m)\) means that \(o\) cognitive users successfully transmit in time slot \(t\), and \(o t = *\) means that a transmission conflict occurs in time slot \(t\).

[0027] S4: The cognitive user updates the belief state, and the update formula is:

[0028]

[0029] S5: If the current time slot \(t < D\), go to time slot \(t + 1\) and jump to S2; if the current time slot \(t = D\), the system jumps to S1;​

[0030] A cognitive user adaptive access method in a transmission time limit communication scenario is implemented using MATLAB software. In addition, a second scenario is set for comparison: at time slot t = 1, the initial transmission probability p1 = 1 is set. If the data packet is successfully transmitted in the current time slot, then let p t+1 = min(2p t , 1); if no user transmits in the current time slot and the licensed user does not occupy the channel, let p t+1 = p t ; if a transmission conflict occurs, then let p t+1 = 0.5 * p t .

[0031] Example 1

[0032] Set the number of cognitive users N = 50, q = 0.8, w = 0.2, and the number of simulation times is set to 10 6 times. Four simulation scenarios are set. Scenario 1: D = 10, m = 2; Scenario 2: D = 10, m = 5; Scenario 3: D = 20, m = 2; Scenario 4: D = 20, m = 5. In these four scenarios, for each scenario, the generation probability g of the data packet is incremented from 0.1 to 0.6 at an interval of 0.1. A cognitive user adaptive access method in a transmission time limit communication scenario and Scenario 2 are respectively used for comparison to evaluate the difference in network reliability between the two scenarios.

[0033] As Figure 2 shown, the increase in the number of time slots D and the increase in the multi-packet reception ability m of the data can both improve the network reliability. However, the network reliability of the access probability strategy using a cognitive user adaptive access method in a transmission time limit communication scenario is significantly higher than that of Scenario 2.

[0034] Example 2

[0035] Set the number of cognitive users N = 50, q = 0.8, w = 0.2, and the number of simulation times is set to 10 6 times. Four simulation scenarios are set. Scenario 1: g = 0.2, m = 2; Scenario 2: g = 0.2, m = 5; Scenario 3: g = 0.4, m = 2; Scenario 4: g = 0.4, m = 5. In these four scenarios, for each scenario, the number of time slots D within the superframe is incremented to 30 at an interval of 5. A cognitive user adaptive access method in a transmission time limit communication scenario and Scenario 2 are respectively used for comparison to evaluate the difference in network reliability between the two scenarios.

[0036] As Figure 3As shown in the figure, with the increase of the data packet generation probability g, the network reliability decreases; with the increase of the multi-packet reception ability m of data, the network reliability increases, but the network reliability of the adaptive access probability strategy of the cognitive user adaptive access method in a transmission time limit communication scenario is significantly higher than that of Scheme II.

[0037] Embodiment 3

[0038] Set the number of cognitive users N = 50, q = 0.8, w = 0.2, and set the number of simulation times to 10 6 times. Set four simulation cases. Case 1: g = 0.2, D = 10; Case 2: g = 0.2, D = 20; Case 3: g = 0.4, D = 10; Case 4: g = 0.4, D = 20. In these four cases, for each case, increase the channel multi-packet reception ability m by an interval of 1 to 6 respectively. Compare the cognitive user adaptive access method in a transmission time limit communication scenario with Scheme II respectively, and evaluate the network reliability difference between the two schemes.

[0039] As Figure 4 shown in the figure, with the increase of the number of time slots D, the network reliability increases; with the increase of the data packet generation probability g, the network reliability decreases, but the network reliability of the access probability strategy of the cognitive user adaptive access method in a transmission time limit communication scenario is significantly higher than that of Scheme II.

[0040] Embodiment 4

[0041] Set the number of cognitive users N = 50, w = 0.2, and set the number of simulation times to 10 6 times. Set four simulation cases. Case 1: g = 0.2, m = 2; Case 2: g = 0.2, m = 5; Case 3: g = 0.4, m = 2; Case 4: g = 0.4, m = 5. In these four cases, for each case, increase q by an interval of 0.1 to 0.6 respectively. Compare the cognitive user adaptive access method in a transmission time limit communication scenario with Scheme II respectively, and evaluate the network reliability difference between the two schemes. As Figure 5 shown in the figure, with the increase of the channel multi-packet reception ability m, the network reliability increases; with the increase of the data packet generation probability g, the network reliability decreases, but the network reliability of the access probability strategy of the cognitive user adaptive access method in a transmission time limit communication scenario is significantly higher than that of Scheme II.

Claims

1. An adaptive access method for cognitive users in a transmission time-limited communication scenario, which models the system using a partially observable Markov decision process, is characterized in that The steps are as follows: S1: Determine the parameters of the partially observable Markov decision process, the relevant parameters include: the number of cognitive users N, describing the number of active cognitive users in time slot t as n t , the authorized user occupies the channel state c t , n t With c t Together as system state s t , cognitive user confidence state b t , the access probability p of cognitive users t ; describes that at time slot t, the system is in state s t =s, cognitive users with p t = p probability access, when the time slot arrives at t+1, the system is in s t+1 = state transition probability f of s′ t (s′,s,p); describes the observation result o of the cognitive user at time slot t t ; describes that at time slot t, the system is in state s t =s, cognitive users with p t = p probability access, in the next time slot, the system state is s t+1 =s′, the cognitive user obtains the observation result o at time slot t t = the probability v of o t (o,s′,s,p), describes the probability g of the cognitive user generating a data packet at the start of each superframe, and the internal operation step of the cognitive user jumps to S2; S2: The relevant parameters for describing a cognitive user adaptive access method in a transmission deadline communication scenario include: the number of time slots D within a superframe, and the channel multi-packet reception capability m; the confidence state b1 maintained by the cognitive user at t = 0, and the system operation steps jump to S3; S3: The cognitive user sets the current time slot as t = t + 1. At the start of time slot t, the cognitive user performs carrier sensing on the channel and obtains the sensing result c t , and based on different sensing results, the cognitive user decides whether to adopt the access probability p that maximizes the immediate reward r t according to its latest confidence state estimate b of the active cognitive users t (b, c, p) to send the data packet, and the internal operation step of the cognitive user jumps to S4; t ​ S4: At the end of time slot t, and when the currently time-slot authorized user does not occupy the channel, each cognitive user listens to the response signal broadcast on the primary channel. If a cognitive user successfully transmits in the current time slot, the response signal is ACK j , j ∈ (1, 2..., m); if more than m data packets are transmitted, a transmission conflict occurs, and the corresponding response signal is NACK; based on the response signal, the cognitive user can determine whether it has successfully sent in the current time slot. If it has successfully sent, it exits the channel competition. Otherwise, it continues with data transmission and channel competition, and the internal operation of the cognitive user jumps to S5; S5: For cognitive users with remaining data packets to be sent at time slot t (t ≠ D), the cognitive users update the confidence state to obtain b t+1 (n') and jump to S3; in the case of t = D, jump to S6; S6: When the current superframe ends, jump to S1.

2. According to claim 1, the strategic method for cognitive user adaptive random access in the cognitive radio network described in step S1 is characterized in that: S1-1: An adaptive access method for cognitive users in a transmission time-limited communication scenario discretizes the system time of each cognitive user into multiple superframes, and each superframe contains multiple time slots \(t\in T = \{1, 2, \ldots, D\}\); at time slot \(t\), the state is defined as \(s\) t =(n t , c t ), the confidence state \(b\) of the cognitive user t =[b t (0), b t (1), \(\ldots\), b t (N)], the number of active cognitive users is \(n\) t \(\in Y=\{0, 1, \ldots, N\}\), the channel occupancy state \(c\) of the licensed user t \(\in C = \{0, 1\}\), \(c\) t = 0 indicates that the channel is not occupied by the licensed user; \(c\) t = 1 indicates that the channel is occupied by the licensed user; the state space \(S = Y\times C\); use the parameter \(q\) to describe the probability that the channel jumps from \(c\) t = 0 to \(c\) t = 1, use the parameter \(w\) to describe the probability that the channel jumps from \(c\) t = 1 to \(c\) t = 0; define the access equation to satisfy \(a\) t :[0, 1] N+1 \(\to[0, 1]\), the access probability \(p\) t \(\in a\) t ; S1-2: State transition probability f t (s′, s, p) = Pr(s t+1 = s′|s t = s, p t = p), and is defined in the case of multi-packet reception capability of m as: S1-3: Define the observation \(o\) of the cognitive user on the channel t \(\in\mathcal{O}=\{0,1,\ldots,m,*\}\), where t \(o = o(0\leq o\leq m)\) means that \(o\) cognitive users successfully transmit in time slot \(t\), and t \(o = *\) means that a transmission collision occurs in time slot \(t\); the observation function is defined as: \(v\) t \((o,s',s,p)=\Pr(o\) t \(= o|s\) t+1 \(= s',s\) t \(= s,p\) t \(= p)\), and when the multi-packet reception ability is \(m\), we have:

3. According to claim 1, the strategy method for cognitive user adaptive random access in the cognitive radio network described in step S2 is characterized in that, At the moment of time slot t = 1, initialize the confidence state distribution of the number of active cognitive users: b1 = h g = [(1 - g) N , Ng(1 - g) N-1 ,..., g N .

4. According to claim 1, the policy method for cognitive user adaptive random access in the cognitive radio network described in step S3 is characterized in that, The cognitive user senses the current state c of the channel t , if c t = 1 and there is a cognitive user with a data packet to be sent, then the cognitive user, based on the current confidence state distribution, adopts an access probability p that maximizes the immediate reward to send the data packet, where the immediate reward is defined as: The transmission probability corresponding to maximizing the immediate reward is:

5. The strategy method for cognitive user adaptive random access in the cognitive radio network according to claim 1, step S4, is characterized in that The update formula for the confidence state of the cognitive user is:

6. According to claim 1, the policy method for cognitive user adaptive random access in the cognitive radio network described in step S5 is characterized in that If the current time slot t < D, then enter time slot t + 1 and jump to S3; if the current time slot t = D, then jump to S1.

Images