Multi-carrier cognitive noma resource allocation method based on deep learning

Abstract

The present invention proposes a multi-carrier cognitive non-orthogonal multiple access (NOMA) resource allocation method based on deep learning, including the following steps: (1) Scenario of multi-carrier cognitive NOMA system Model building; (2) Mathematical description of user scheduling and resource allocation strategies; (3) Design of deep neural network and deep learning algorithms based on joint allocation. After establishing a downlink system based on multi-carrier cognitive NOMA, the present invention innovatively proposes a spectrum sharing strategy that meets the scenario, and reasonably establishes mathematical expressions for optimization objectives and constraints, and realizes multi-carrier cognitive NOMA based on deep learning The user scheduling and resource allocation strategy satisfies the various transmission needs of users, and at the same time better realizes the low-power resource allocation of the multi-carrier cognitive NOMA downlink system.

Description

technical field [0001] The present invention proposes a multi-carrier cognitive NOMA resource allocation mechanism based on deep learning, establishes a transmission model of a multi-carrier cognitive NOMA downlink system, proposes a user pairing and resource allocation strategy that optimizes the spectrum efficiency and energy efficiency of the NOMA system, and designs a A fully connected neural network based on message passing and an optimal deep learning algorithm are proposed, which enables large-scale NOMA users to transmit data with high quality, high speed and low power consumption fairly and flexibly. Background technique [0002] In the past few decades, with the rapid development of mobile communication technology, technical standards have continued to evolve. The fourth generation mobile communication technology (4G) is based on OFDMA, and its data service transmission rate reaches hundreds of per second. Mega or even gigabit, which can meet the needs of broadband...

AI Technical Summary

Problems solved by technology

[0007] Existing studies on NOMA user pairing and resource allocation in the power domain are almost all analyzed and solved using traditional methods, and have not been combined with advanced deep learning technology for in-depth research

In addition, the power allocation scheme in the NOMA system is often based on the single-carrier NOMA system, which makes it difficult to flexibly meet the diverse communication needs of different users

For the pairing mechanism of shared users, the existing work only conducts a rough comparative analysis of different pairing performances, and does not give specific pairing criteria

For the channel allocation of NOMA users, the relevant research is limited to the traditional optimization method of single-carrier NOMA channel selection, which is difficult to guarantee the optimality of the algorithm and cannot meet the needs of fast solution in actual scenarios.

Images