Dynamic spectrum sharing method between 4G and 5G networks based on deep reinforcement learning

Abstract

The invention provides a dynamic spectrum sharing method between 4G and 5G networks based on deep reinforcement learning. Considering a cellular network covered by a single base station, the base station is regarded as an agent, and its schedulable minimum resource unit is defined as a resource block, which includes two dimensions of time and frequency. The sharing strategy of the base station will be formulated for the resource block cycle, that is, in each decide which resource blocks to share for 5G service provisioning within a decision cycle. The invention aims at 4G network service quality assurance and spectrum sharing efficiency, enables base stations to continuously improve strategies according to dynamic environmental information through deep reinforcement learning, and realizes efficient and reasonable utilization of 4G idle spectrum. The present invention does not depend on a specific model, and can formulate sharing policies more accurately. At the same time, the base station in the present invention can use the trained neural network to directly formulate a sharing strategy based on the observations of the current environment, eliminating the need for complex calculations based on traditional optimization algorithms and avoiding the intolerable delay it brings.

Description

technical field [0001] The invention belongs to the technical field of mobile communication, and in particular relates to an intelligent dynamic spectrum sharing method based on deep reinforcement learning. Background technique [0002] In recent years, with the surge in the number of wireless devices connected to the network, the data traffic in mobile wireless communication networks has increased exponentially, forcing mobile communication networks to evolve to the fifth generation (5G). In order to support massive data transmission and emerging application scenarios, 5G networks need more spectrum resources as support. Spectrum shortage has become a key problem that needs to be solved urgently in the development of 5G mobile communication networks. However, a large number of studies have shown that under the current fixed spectrum allocation strategy, the utilization rate of a large number of licensed frequency bands is very low, even less than 20% in some areas and time ...

AI Technical Summary

Problems solved by technology

However, a large number of studies have shown that under the current fixed spectrum allocation strategy, the utilization rate of a large number of licensed frequency bands is very low, even less than 20% in some areas and time periods, which forms a strong contradiction with the urgent demand for spectrum for 5G

However, due to the dynamic characteristics of the network environment in space and time dimensions, relevant information such as traffic demand in the network is uncertain, which makes optimization modeling and solving very difficult and difficult to achieve

[0005] 2. Dynamic spectrum sharing needs to adjust the sharing strategy in real time according to the state of the network. Since there are usually a large number of smart devices connected to the actual network, there are many types of data and large information dimensions. Even if the optimal strategy can be obtained through optimization means, the problem is difficult to solve. High complexity will bring intolerable delay

[0006] 3. Dynamic sharing of 4G spectrum resources will bring a certain loss of channel capacity, that is, reduce the network rate of 4G users and affect the experience of 4G users

Images