Maneuvering communication network intelligent planning method based on deep reinforcement learning

Abstract

The invention discloses a maneuvering communication network intelligent planning method and device based on deep reinforcement learning, and the method comprises the following steps: 1, preprocessingresource elements: preprocessing the resource elements of a guarantee node, a guaranteed user, an erection region and the like of a maneuvering communication network; 2, preprocessing planning rules,namely preprocessing the planning rules of the mobile communication network; 3, generating a training sample: performing random Monte Carlo search calculation on the preprocessing result to generate the training sample; 4, performing model training: based on the recurrent neural network, training a network planning model by using the training sample; and 5, performing model generation: constructing a joint loss function, repeatedly searching and training the sample according to the indication of the joint loss function, and generating a maneuvering communication network planning model. The deep reinforcement learning-based maneuvering communication network intelligent planning method and device effectively solve the problems that the current maneuvering communication network planning depends on a large amount of manual operation, the planning time exceeds the task requirement, the adaptability to sudden tasks and strange environments is poor, the resource utilization rate is low and the like, and improves the overall efficiency of maneuvering communication network planning.

Description

technical field [0001] The invention relates to the field of information technology, in particular to a network intelligent planning method for a mobile communication network. Background technique [0002] The mobile communication network usually refers to a mobile communication network used to guarantee large-scale special missions in special fields, and usually consists of various subnetworks such as fixed optical fiber networks, microwave networks, satellite networks, aerial relay networks, and shortwave and ultrashortwave radio networks. The smallest unit of the integrated mobile network formed is a single communication support platform or device, which is regarded as a support node in the mobile communication network network. There are usually hundreds of people or more insured in the mobile communication network. The demand for erection is relatively random and the time is relatively tight. The planning time is usually within 24 hours or less. [0003] Network plannin...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the defects of the prior art, aiming at the practical problems such as complicated network planning conditions, urgent time, uncertain location and limited equipment conditions of the mobile communication network, and realizes an intelligent planning method for the mobile communication network based on deep reinforcement learning

Images