A method for dynamically adjusting backoff parameters of SPMA protocol

Abstract

The present invention belongs to the field of wireless communication technology and relates to a method for dynamically adjusting the backoff parameters of the SPMA protocol. The present invention dynamically optimizes the backoff strategy by combining a comprehensive priority score (CPS) and current network status information. First, the CPS of each data packet to be sent is calculated, and the data packets are sorted according to these scores to determine their priority in the transmission queue, and then the backoff time and other parameters are dynamically adjusted according to the network conditions to improve network performance. The beneficial effects of the present invention are: by calculating the comprehensive priority score (CPS) and adjusting the weight method, the data packets to be sent are accurately evaluated and sorted, so as to clearly guide the selection of the backoff time, and the backoff parameters are dynamically adjusted under the real-time network state to ensure the transmission efficiency of high-priority data, and further improve the overall throughput of the network and reduce latency.

Description

Technical Field

[0001] The invention belongs to the technical field of wireless communications and relates to a method for dynamically adjusting backoff parameters of an SPMA protocol. Background Art

[0002] The Statistical Priority-based Multiple Access (SPMA) protocol is an optimized multiple access technology that aims to effectively distinguish service flows of different priorities in wireless self-organizing networks to achieve more efficient data packet transmission. The protocol mainly consists of mechanisms such as channel state detection, multi-priority transmission judgment, and flow control, and implements differentiated services for data packets of different levels. By setting different thresholds and priorities, the SPMA protocol can give priority to the transmission of high-priority services when the network load is high, while suppressing the transmission of low-priority services, thereby ensuring the transmission performance of key services.

[0003] The SPMA protocol aims to improve the transmission efficiency of high-priority data packets and ensure their priority by setting different backoff times. In the protocol design, the limited transmission rate and longer backoff time of low-priority data packets are intentional to reduce interference with the transmission of high-priority data packets. However, this design may lead to overall network efficiency problems in some cases, such as low channel utilization and significantly reduced throughput of low-priority data packets. Therefore, it is necessary to consider how to reasonably utilize channel resources, reduce conflicts, improve the performance and efficiency of the overall network, and balance the needs of high- and low-priority data packets while ensuring the priority and efficiency of high-priority data packet transmission.

[0004] In response to the above problems, this invention studies a method for adjusting the backoff parameters of the SPMA protocol by combining the comprehensive priority score and the current network status. Through this system design, not only can the comprehensive priority score of each data packet to be sent be calculated and sorted accordingly to determine the priority in the transmission queue, but also the backoff parameters can be dynamically adjusted using deep learning technology to optimize network performance. It can effectively control the access volume of low-priority data packets, maintain the stability of the system load, and maximize the overall channel utilization, ensure the transmission efficiency of high-priority data packets, and maintain the fairness of data of different priorities while maintaining network performance. Summary of the invention

[0005] The present invention dynamically optimizes the backoff strategy by combining the comprehensive priority score (CPS) and the current network status information. First, the CPS of each data packet to be sent is calculated, and the data packets are sorted according to these scores to determine their priority in the transmission queue. Then, parameters such as the backoff time are dynamically adjusted according to the network status to improve network performance.

[0006] The technical solution of the present invention is:

[0007] A method for adjusting the backoff parameters of the SPMA protocol in combination with the comprehensive priority score and the current network status information, characterized in that it comprises the following steps:

[0008] S1. Collect and determine relevant information of the real-time data packets to be sent (including the preset priority of the data packets to be sent, the quality of service requirements (QoS) and the waiting time), the channel occupancy rate in the past period of time, the length changes of queues of different priorities and the time interval between the arrival of data packets;

[0009] S2, using the relevant information of the data packets to be sent collected by S1 and the weights assigned to the information, calculate the comprehensive priority score (CPS) of each real-time data packet to be sent;

[0010] S21, assign corresponding weights to the three pieces of information, namely, preset priorities, quality of service requirements (QoS) and waiting time for the data packets to be sent collected in S1, and design corresponding functions respectively. These factors are the basis for calculating the comprehensive priority score (CPS);

[0011] S22, calculate CPS and check whether weight adjustment is needed according to the current network load. If the network load is high, dynamically adjust the weight and enter S23 to optimize the backoff strategy. Otherwise, enter S3;

[0012] S23. Recalculate the CPS using the new weight.

[0013] S3, sorting all the data packets to be sent according to the CPS calculated in S2 to determine their priorities in the transmission queue;

[0014] S4, using the long short-term memory network (LSTM) model based on temporal order to dynamically adjust the backoff parameters;

[0015] S41, based on the channel occupancy rate in the past period, the length change of queues of different priorities, the time interval between arrival of data packets, the historical backoff time and the number of retries, and the CPS, the LSTM model is trained offline to predict the most appropriate backoff time;

[0016] S42. Use the trained LSTM model to evaluate the real-time network status online, intelligently select the backoff time and adjust the priority queue to ensure that the protocol maintains efficient data transmission and meets QoS requirements in the time-varying network.

[0017] The beneficial effects of the present invention are as follows: by calculating the comprehensive priority score (CPS) and adjusting the weight, each data packet to be sent is accurately evaluated and sorted, thereby clearly guiding the selection of backoff time, dynamically adjusting the backoff parameters under real-time network conditions, ensuring the transmission efficiency of high-priority data, and further improving the overall throughput of the network and reducing latency. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic diagram of the overall process of the present invention.

[0019] Figure 2 The figure is a flow chart of the calculation of the Comprehensive Priority Score (CPS) part of the present invention. DETAILED DESCRIPTION

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

[0021] As attached Figure 1 As shown, the method of the present invention specifically comprises:

[0022] S1. Collect and determine relevant information of the real-time data packets to be sent (including the preset priority of the data packets to be sent, the quality of service requirements (QoS) and the waiting time), the channel occupancy rate in the past period of time, the length changes of queues of different priorities and the time interval between the arrival of data packets;

[0023] S2, using the relevant information of the data packets to be sent collected by S1 and the weights assigned to the information, calculate the comprehensive priority score (CPS) of each real-time data packet to be sent;

[0024] As attached Figure 2 As shown in Figure 1, the preset priority of the data packets to be sent, the quality of service requirements (QoS) and the waiting time collected in S1 are assigned corresponding weights and corresponding functions are designed respectively. These factors are the basis for calculating the comprehensive priority score (CPS). First, the weight w is assigned to the preset priority of the data packets p , assign weight w to the packet quality of service (QoS) requirement t , assign weight w to the packet waiting time w Then the initial configuration of weights is performed. Since the preset priority is the decisive factor, the initial setting w p =0.5; because QoS is also important but its impact is slightly lower than priority, the initial setting w t = 0.3; because waiting time is less critical in an idle network and more critical in a congested network, the initial setting w w =0.2, and will be adjusted later according to the actual situation of the network and performance requirements. When the network load exceeds 80%, it is considered that the network has entered a high congestion state, and w is increased. wThe weight of p and w t to 0.4 and 0.2; when the network load is between 50% and 80%, the network is considered to be in a medium congestion state, and w w The weight is adjusted to 0.3, and w p and w t Adjusted to 0.4 and 0.3; when the network load is less than 50%, the network is considered to be in a low congestion state and the initial weight setting remains unchanged.

[0025] The Composite Priority Score (CPS) is calculated according to the following formula:

[0026] CPS=w p (max-P)+w t ·f(T)+w w h(W,P)

[0027] Among them, w p is the weight of the preset priority of the data packet, P is the preset priority of the data packet, which is a value increasing from 0 (highest priority), and max is the initial value when P = 0, which represents the maximum contribution of the highest priority data packet to the comprehensive priority score (CPS), which depends on the number of priorities set. t is the weight of the packet’s quality of service (QoS) requirement, and f(T) is the function that maps the packet’s QoS to a numerical value. w is the weight of the packet waiting time, and h(W,P) reflects the impact of the packet waiting time on the comprehensive priority score (CPS).

[0028] Specifically, f(T) is calculated according to the following formula:

[0029] f(T)=QoS score (type)

[0030] Among them, QoS score(type) represents the QoS numerical mapping of each type. Five types of data packets and their corresponding quality of service requirements (QoS) are set, namely command and control communications; timely intelligence, surveillance and reconnaissance; combat support; logistics support and conventional intelligence; conventional communications and management. Command and control communications refer to communications between battlefield commanders and other key military decision makers, and such communications usually have the highest priority. Timely intelligence, surveillance and reconnaissance include communications with high timeliness for collecting information from various intelligence sources, which require high priority processing to ensure timeliness. Combat support refers to communications that support critical military operations, such as precision strike requests and mission-critical data transmission, and have a higher priority. Logistics and support, conventional intelligence include communications that support tasks such as material allocation and personnel deployment. Conventional communications and management involve packets of daily management and non-critical communications, and have a lower priority. Set the specific numerical mapping as: QoS score (Command and Control Communications) = 5, QoS score (Timely Intelligence, Surveillance and Reconnaissance) = 4, QoS score (Combat Support) = 3.5, QoS score (Logistics support and general intelligence) = 2, QoS score (Regular Communication and Management) = 1. This mapping scheme ensures that packets with different QoS requirements are properly reflected in the comprehensive priority score so that the system can arrange their transmission order and processing priority in the network accordingly.

[0031] Specifically, h(W,P) is calculated according to the following formula:

[0032] h(W,P)=e β(1-P) log(1+W)

[0033] Where W represents the waiting time of the data packet, P represents the preset priority of the data packet, which is a value increasing from 0 (highest priority), and e is the base of the natural logarithm. β(1-P) The term exponentially strengthens the waiting time impact of packets with a priority of 0 (the highest priority). β is a positive adjustment coefficient used to adjust the sensitivity of the priority to the waiting time. By properly selecting the β value, the function can show different growth rates at different priorities. When the priority P is small (higher priority), e β(1-P) The value of is large, so the waiting time W will have a greater impact on the score h(W,P). Conversely, when P is large (lower priority), the exponential term is relatively small, so the waiting time has less impact on the score relative to the lower priority packets.

[0034] Calculate the CPS of each real-time data packet to be sent and check whether the weight needs to be adjusted according to the current network load. If the network load is high, dynamically adjust the weight to optimize the backoff strategy, and then recalculate the CPS using the new weight.

[0035] S3, sorting all the data packets to be sent according to the CPS calculated in S2 to determine their priorities in the transmission queue;

[0036] S4, using the long short-term memory network (LSTM) model based on temporal order to dynamically adjust the backoff parameters;

[0037] Collect the channel occupancy rate, the length change of queues of different priorities, the packet arrival time interval, the historical backoff time and the number of retries, and the CPS calculated according to S2 in the past period of time, and perform appropriate preprocessing on the data, including normalization and time series construction. Use the collected data set to train the LSTM model offline. The model will learn the best action (the selected backoff time or number of retries) to take under specific network conditions to maximize long-term returns (increase the packet success rate and reduce the total delay) and clarify how to choose the appropriate backoff time and number of backoffs.

[0038] The trained LSTM model is used to evaluate the real-time network status online, intelligently select the backoff time and adjust the priority queue to ensure that the protocol maintains efficient data transmission and meets QoS requirements in the time-varying network.

Claims

1. A method for dynamically adjusting the backoff parameters of the SPMA protocol, characterized in that: The following steps are involved: S1. Collect and determine relevant information of the real-time data packets to be sent, the channel occupancy rate in the past period of time, the length changes of queues of different priorities and the time interval of data packet arrival, wherein the relevant information includes the preset priority of the data packets to be sent, the quality of service requirements and the waiting time; S2, using the relevant information of the data packets to be sent collected by S1 and the weights assigned to the information, calculate the comprehensive priority score of each real-time data packet to be sent, the specific method is as follows: S21, assign corresponding weights to the three pieces of information, namely, the preset priority of the data packet to be sent, the service quality requirement and the waiting time collected in S1, and design corresponding functions respectively, specifically, assigning weight w to the preset priority of the data packet p , assign weight w to the packet quality of service requirement t , assign weight w to the packet waiting time w ; Perform initial configuration of weights and set w p =0.5, w t =0.3, w w =0.2, and the subsequent adjustment method based on the actual situation of the network and performance requirements is to increase w w The weight of p and w t to 0.4 and 0.2; when the network load is between 50% and 80%, w w The weight is adjusted to 0.3, and w p and w t Adjusted to 0.4 and 0.3; When the network load is less than 50%, the initial weight setting remains unchanged; the corresponding functions include: CPS=w p ·(max-P)+w t ·f(T)+w w ·h(W,P) f(T)=QoS score (type) h(W,P)=e β(1-P) ·log(1+W) Where CPS is the comprehensive priority score, P is the preset priority of the data packet, which is a value increasing from the highest priority 0, max is the initial value when P = 0, representing the maximum contribution of the highest priority data packet to the comprehensive priority score, which depends on the number of priorities set; f(T) is a function that maps the QoS of data packet type T to a numerical value, where QoS refers to the quality of service requirement, and h(W,P) reflects the impact of the data packet waiting time on the comprehensive priority score; QoS score (type) represents the QoS value mapping of each type, W represents the packet waiting time, e is the base of the natural logarithm, and e is used β(1-P) The term exponentially strengthens the waiting time impact of packets with a priority of 0. β is a positive adjustment coefficient used to adjust the sensitivity of the priority to the waiting time impact. S22, calculate the CPS of each real-time data packet to be sent according to the designed function, and check whether the weight needs to be adjusted according to the current network load. Dynamically determine whether the weight needs to be adjusted according to the network load. If so, enter S23, otherwise enter S3; S23, recalculate the CPS using the modulated new weight; S3, sorting all the data packets to be sent according to the CPS calculated in S2 to determine their priorities in the transmission queue; S4. Use the long short-term memory network model based on temporal sequence to dynamically adjust the backoff parameters, specifically: S41, based on the channel occupancy rate in the past period, the length change of queues of different priorities, the time interval between arrival of data packets, the historical backoff time and the number of retries, and the CPS, the LSTM model is trained offline to predict the most appropriate backoff time; S42. Use the trained LSTM model to evaluate the real-time network status online, intelligently select the backoff time and adjust the priority queue to ensure that the protocol maintains efficient data transmission and meets QoS requirements in the time-varying network.

Images