Dynamic resource allocation method in millimeter wave system

Abstract

The invention provides a dynamic resource allocation method in a millimeter wave system. The method comprises the following steps: initializing system parameters related to a dynamic scene; observingthe state constructed by a channel state, a beam allocation state and a power allocation state in the dynamic environment; determining whether the action adopted in the current state meets each constraint condition in the optimization problem or not; setting different income values according to whether constraint conditions are met or not; storing the data into an experience pool; training an Eval-Net Q network; updating the Target-Net Q network; comparing a randomly generated numerical value and an exploration probability; updating the connection state of the base station beam and the user, the user power distribution state and the exploration probability; and training and learning for a period of time until convergence, and outputting a final base station beam and user connection state and a user power distribution state. According to the method, the technical problem that dynamic resource allocation and long-term user fairness cannot be realized by jointly allocating beams and powerin a dynamic channel environment in the prior art is solved.

Description

technical field [0001] The present invention relates to the technical field of mobile communication technology, in particular to a method for dynamic resource allocation in a millimeter wave system. Background technique [0002] At present, the operating frequencies of mobile communication systems are mainly concentrated in the microwave frequency band of 700MHz-2.6GHz, and the shortage of spectrum resources is becoming more and more serious. The millimeter wave communication greatly enriches the spectrum resources by operating in the 30GHz-300GHz frequency band. However, due to the disadvantages of poor penetration, easy absorption, and fast attenuation of millimeter-wave signals, millimeter-wave communication systems have problems such as large signal path loss and small base station coverage. Large-scale antenna technology can improve system capacity, spectrum efficiency, system anti-interference ability and other advantages by configuring a large number of antennas (mor...

AI Technical Summary

Problems solved by technology

In this mmWave system, solve the technical problem in the prior art that beams and power cannot be jointly allocated in a dynamic channel environment to achieve dynamic resource allocation and long-term user fairness

Images