Self-organizing network medium access control method based on reinforcement learning

Abstract

The invention discloses a self-organizing network medium access control method based on reinforcement learning, is applied to the field of wireless networks, and aims at solving the problem that in the prior art, distributed network node multi-time-slot selection in a dynamic TDMA medium access control method is generally not considered. According to the invention, a time slot interleaving schemeis adopted in the frame structure design, so that the requirement of the media access control method on the packet response delay performance of hardware equipment is looser; meanwhile, a multi-time-slot reservation mechanism is adopted, so that the network node only needs to send a control frame once and reserve information time slots of a plurality of subsequent periods in each period, the interaction process is simple and easy to implement, and the control overhead is low; and the selection probability of each section of time slot is adaptively adjusted based on a multi-time slot selectionalgorithm of reinforcement learning, and a more optimized time slot selection strategy is generated, so that the time slot competition conflict is reduced, the time slot allocation efficiency is improved, and the competition success rate, the transmission bandwidth, the transmission time delay, the packet loss rate and other performances of the medium access control method are further optimized.

Description

technical field [0001] The invention belongs to the field of wireless networks, in particular to a multi-slot selection technology based on reinforcement learning. Background technique [0002] Self-organizing (Ad hoc) network is a multi-hop communication system that can be established and maintained by itself without relying on fixed communication facilities in a wireless network environment. It has strong robustness and invulnerability. Self-organizing networks are very suitable for wireless network scenarios such as drone communication, car networking scenarios, and environmental monitoring. [0003] MAC (Medium Access Control) medium access control method is the key responsible for nodes sharing and accessing limited channel resources, and its performance directly affects the overall performance of the network. Usually, compared with the traditional stochastic contention media access control method, the media access control method based on TDMA (Time Division Multiple A...

AI Technical Summary

Problems solved by technology

The factors affecting network performance are relatively complex. The above algorithms generally do not consider the problem of multi-slot selection of distributed network nodes in the dynamic TDMA medium access control method. Applying enhanced learning technology to the scene of multi-slot selection of distributed network nodes can make the selection of time slots The algorithm comprehensively considers environmental factors, and learns adaptively from them to a more optimal multi-slot selection strategy. The multi-slot selection method based on reinforcement learning has important research value and challenges

Images