Distributed multi-user dynamic spectrum access method in cognitive wireless network

Abstract

The invention discloses a distributed multi-user dynamic spectrum access method in a cognitive wireless network. The method comprises the following steps: step 1, constructing a system model, whereinthe system model is as follows: n authorized users and k cognitive users in a cell share b channels, wherein n, k and b are natural numbers that are not zero, values of n and b are equal, and k is larger than n; step 2, performing spectrum selection and access on the cognitive users by adopting a DQN algorithm, specifically, setting an initial Q function, setting each cognitive user as an executor, and performing action selection by each executor according to the DQN algorithm, namely, selecting one of the b channels for transmission, calculating an average utility value of a system, and setting a reward value by adopting an evolutionary game theory; and then training a neural network, and using the neural network as a function approximator to obtain an updated Q function; and step 3, continuing to execute step 2 until the obtained system capacity tends to be a constant value. Under the condition of high frequency spectrum requirements, the bandwidth efficiency is high.

Description

technical field [0001] The invention belongs to the technical field of wireless communication, and in particular relates to a distributed multi-user dynamic spectrum access method in a cognitive wireless network. Background technique [0002] With the rapid development of wireless communication technology, the popularization of wireless equipment and the continuous growth of new services, spectrum resources have become increasingly scarce, and the fixed available spectrum resources have been unable to meet the communication requirements of users. Insufficient spectrum resources in the world have become increasingly serious, which makes wireless communication systems constrained by spectrum resources when promoting economic and social development. At present, cognitive radio technology has become a key technology to solve the problem of low spectrum utilization. The main idea of ​​this technology is to first detect which spectrums are idle, and then intelligently select and a...

AI Technical Summary

Problems solved by technology

But these have their limitations. For example, although Nash equilibrium and evolutionary stable equilibrium can be obtained, the requirements for the environment are high, and the user has not gone through the learning process, and the convergence speed is slow.

Or the military network is studied, which is based on the premise that each auxiliary radio node in the military network can allocate spectrum resources according to its priority, and the application has certain limitations.

Some do not take into account the balance and coordination between users, the system is relatively unstable

Images