Symbiotic radio intelligent anti-interference method based on deep reinforcement learning

Abstract

*The invention provides a symbiotic radio intelligent anti-interference method based on deep reinforcement learning, a symbiotic radio network comprises a transmitter, a receiver and an interference unit, and the method comprises the following steps: in each time slot, the transmitter selects an action to interact with an environment according to a sensed state; the transmitter receives an award and senses a next state, wherein the award is used for representing a successful transmission probability; obtaining a preset DDQN model, and determining an action value corresponding to the action by searching an optimal action value function Q * (s, a) through the DDQN model; and judging whether the action value is an optimal action value or not, and outputting an optimal reward value when the action value is the optimal action value. The adversarial between the DDQN-based simulated symbiotic radio network and the jammer and the Q network are realized by using a Transformer encoder so as to effectively estimate the action value from the original spectrum data, and the method can effectively defend various interference attacks, so that the successful transmission rate of a communication system is improved to the greatest extent.

Description

technical field [0001] The present invention relates to the technical field of wireless communication, and in particular, to a symbiotic radio (SR) intelligent anti-jamming method based on a double-deep Q-network (DDQN). Background technique [0002] In recent years, with the explosive growth of data in the information age, there are new and higher requirements for precious resources such as energy and spectrum. However, traditional wireless communication networks generally have problems such as insufficient energy supply and low spectrum utilization, which slows down the development of the Internet of Things industry to a certain extent. The emerging symbiotic radio technology (SR) provides a new idea to solve this problem. SR can take advantage of Cognitive Radio (CR) and ambient backscatter communication (AmBC) and effectively solve the shortcomings of these two technologies. Similar to CR, SR consists of two spectrum sharing systems, the primary system and the secondar...

AI Technical Summary

Problems solved by technology

[0004] In view of the defects in the prior art, the purpose of the present invention is to provide an intelligent anti-jamming method based on deep reinforcement learning, which can solve the problem that the symbiotic radio is vulnerable to interference attacks, resulting in a serious decline in spectrum utilization and privacy leakage.

Images