Multi-domain intelligent communication model and communication method based on channel-bandwidth joint decision

Abstract

The invention discloses a multi-domain intelligent communication model and a communication method based on channel-bandwidth joint decision. A communication scene comprises a primary user, a secondaryuser and one or more jammers, and multi-domain anti-interference communication is carried out by utilizing deep reinforcement learning in a dynamic interference and channel fading environment. Firstly, a time slot, bandwidth and channel selection problem in anti-interference communication is modeled into a Markov process, and a deep reinforcement learning network structure is redesigned and optimized by utilizing a long-short-term memory network. And then the two-dimensional time-frequency spatial spectrum matrix containing interference source information and channel fading information sensedby the secondary user receiving end are input into a designed algorithm structure to obtain a multi-domain joint anti-interference decision in the current environment. The method is complete in model, clear in physical significance and reasonable and effective in design algorithm, and can well describe the anti-interference scene of the multi-domain intelligent anti-interference communication algorithm based on channel bandwidth joint decision.

Description

Technical field [0001] The invention belongs to the field of wireless communication technology, in particular to a multi-domain intelligent communication model and an intelligent anti-interference method based on channel-bandwidth joint decision-making. Background technique [0002] With the rapid development and deep integration of emerging technologies such as communication confrontation, cognitive radio and artificial intelligence, the demand for wireless services is growing rapidly, and various types of wireless devices are deployed intensively. The electromagnetic environment is becoming more and more complex, the level of interference intelligence continues to increase, and spectrum resources are becoming more and more tight. This also makes anti-interference decision-making problems need to face the new interference environment with "interference dynamics" and "complex composition. Therefore, in order to effectively deal with the anti-jamming problem under the new interfer...

AI Technical Summary

Problems solved by technology

However, if it is just a simple reinforcement learning anti-jamming algorithm, the complexity of the anti-jamming algorithm will increase explosively with the increase of the system scale.

eventually renders its algorithm invalid

If the ordinary deep reinforcement learning model (deep Q neural network anti-interference model) is used, it is difficult to make more accurate state prediction and anti-interference decision-making under the condition of fast switching speed of external interference types, and based on a single anti-interference domain To a certain extent, different external interference environments will have a greater impact on the proposed anti-jamming algorithm, resulting in low performance of the anti-jamming algorithm with a single interference method, resulting in a weak universality of the algorithm. Poor adaptability to the environment

Images