A UAV csma access method based on adaptive adjustment strategy

Abstract

The invention discloses a UAV CSMA access method based on an adaptive adjustment strategy, which includes: before establishing a communication connection, the UAV agent initializes its own access state and access strategy; when there is a data packet to be transmitted At the same time, listen to the busy and idle status of each channel in the UAV ad hoc network; select a channel with the highest access success rate among the available channels; the UAV samples the current access strategy according to the random strategy distribution and executes the access strategy. In the human-machine self-organizing network; after the UAV executes the access strategy, it obtains transmission feedback and obtains the corresponding strategy reward, and at the same time, the access state is transferred; the UAV updates the access strategy according to the state transfer and the immediate reward obtained; After a period of iterative learning, each UAV agent learns the access strategy most suitable for the current UAV self-organizing network environment; compared with the traditional access mechanism, this method can achieve low collision rate, high throughput and It does not affect the fairness of each node's use of the channel.

Description

technical field [0001] The invention relates to the field of unmanned aerial vehicle ad hoc networks, in particular to a method for accessing unmanned aerial vehicle CSMA based on an adaptive adjustment strategy. Background technique [0002] UAVs have the advantages of diverse uses, strong flexibility, and autonomous control, and have received extensive attention and development in military and civilian fields. Compared with a single UAV system, the cooperative application of multiple UAV systems has the advantages of stronger survivability, higher scalability, and faster completion of tasks, but it also brings many technical challenges. For collaborative communication between UAVs, how to design an efficient multi-UAV communication network that meets the needs of future development has become an urgent problem to be solved. [0003] Most current research on UAV networks is based on central control scenarios, in which UAVs rely on ground control centers or satellites for c...

AI Technical Summary

Problems solved by technology

[0007] The invention provides a CSMA access method for unmanned aerial vehicles based on an adaptive adjustment strategy, the purpose of which is to solve the technical problems of high access collision rate and poor adaptability in the prior art, and the method can achieve a low collision rate , high throughput and does not affect the fairness of channel usage by drones

Reinforcement Learning (RL) in machine learning is an effective way to solve this kind of stochastic optimization problems. Its solution is based on the Markov decision process (MDP). After an action, you can get a certain reward, and then adjust the follow-up decision according to the reward

Images