Dynamic spectrum anti-jamming model in fading environment and enhanced learning anti-jamming algorithm

Abstract

The invention discloses a dynamic spectrum anti-jamming model in a fading environment and an enhanced learning anti-jamming algorithm. The model comprises a transmitter, a receiver and a jammer duringthe dynamic spectrum anti-jamming process of a wireless communication system, wherein the jammer generates jamming signals to jam user communication; the transmitter and the receiver transmit data frames through a data link, and the control information is transmitted through a control link; and the receiver acquires the channel information through data communication and broadband spectrum sensing, and the enhanced learning algorithm is executed for optimizing a channel selection strategy. The algorithm is as follows: a user transmits data on a data channel and obtains the current channel transmission rate, and the return value of the current working channel is calculated; the jamming channel at the current time is obtained through spectrum sensing; a new transmission channel is decided through enhanced learning, a Q value table is updated, and due to such cycle, the state of the optimal strategy can be obtained. The dynamic spectrum anti-jamming problem in the fading environment can be solved, and the throughput of the system is improved.

Description

technical field [0001] The invention belongs to the technical field of wireless communication, in particular to a dynamic spectrum anti-jamming model in a fading environment and a reinforcement learning anti-jamming algorithm. Background technique [0002] Due to the rapid growth of demand for wireless communication services, spectrum resources are increasingly scarce. In addition, with the continuous improvement of the intelligent level of interference, the wireless communication system is facing severe challenges. In order to ensure the reliable transmission of wireless communication in the new interference environment, how to achieve efficient communication anti-interference is becoming more and more urgent. [0003] Aiming at the problem of reasonable allocation of spectrum resources, scholars in the field of intelligent anti-jamming have carried out research on dynamic spectrum anti-jamming. Dynamic spectrum access (DSA) breaks through the current static and fixed freq...

AI Technical Summary

Problems solved by technology

Some researchers combined deep learning with Q learning, and proposed to use deep learning algorithm to optimize the Q value table in Q learning (references: X.Liu, Y.Xu, L.Jia, et al, "Anti-Jamming Communications Using Spectrum Waterfall: A Deep Reinforcement Learning Approach,"in IEEE Communications Letters,vol.22,no.5,pp.998-1001,May 2018.), to achieve dynamic spectrum anti-interference, and solve the problem that the state space in Q-learning cannot be too large Insufficient, the performance is very good, but the convergence speed is relatively slow

[0005] Since the current anti-jamming research is mostly concentrated in the power domain, there are relatively few researches on dynamic spectrum intelligent anti-jamming

At the same time, most of the existing dynamic spectrum anti-interference research assumes that the channel environment is ideal, without considering the influence of multipath effects, fading, Doppler frequency shift and other factors, which makes the real-time variability of wireless channels lack of consideration

Images