Uplink NOMA resource allocation method based on deep reinforcement learning

Abstract

The invention discloses an uplink NOMA resource allocation method based on deep reinforcement learning. According to the method, under the condition that the minimum transmission rate of each user ismet, the energy efficiency of the whole system is improved by selecting the optimal sub-channel allocation strategy and power allocation strategy, and the power consumed by transmission is effectivelyreduced. The method is based on a deep Q network in deep reinforcement learning, network parameters are adjusted according to feedback of an NOMA system, and optimal sub-channel and power distribution is achieved. According to the method, the deep Q network is adapted to the continuous resource allocation task through power discretization, and the output dimension of the network is reduced by using a distributed network structure, so that the performance of the whole resource allocation network is improved. Compared with other methods, the method has the advantages that better average overallenergy efficiency can be obtained, and good performance can be obtained under the condition of different transmission power limits.

Description

technical field [0001] The invention relates to mobile communication and reinforcement learning neighborhood, in particular to an uplink NOMA wireless resource allocation method based on deep reinforcement learning. Background technique [0002] The fifth generation communication network (5G) needs to meet the rapidly increasing demand for wireless data traffic, support high-density mobile user communication, and provide various wireless network services. The recently proposed Non-Orthogonal Multiple Access technology (Non-Orthogonal Multiple Access, NOMA) is considered to be an emerging technology that can effectively improve network capacity and satisfy low latency, large-scale connections and high throughput. On the one hand, compared with the traditional Orthogonal Multiple Access (OMA), NOMA uses Superposition Coding (SC) technology at the transmitter to allocate the same subchannel to multiple Users transmit at the same time, share channel resources, and then use Seri...

AI Technical Summary

Problems solved by technology

For example, for the traditional model-based resource allocation scheme, the computational complexity of the optimization process is high, and the iterative algorithm takes a long time

Although the optimization algorithm based on deep learning reduces the computational complexity, it still takes a lot of time to construct enough sample data to train the network to achieve good performance.

Images