Cooperative Caching Method Based on Value Function Approximation for Ultra-Dense Heterogeneous Network Small Station Coding

Abstract

The invention discloses an ultra-dense heterogeneous network small station encoding cooperative caching method based on value function approximation. Using the reinforcement learning method of value function approximation, the value function is expressed as a function of state and action, and the optimization goal is to maximize the average cumulative number of file requests directly served by small stations. By continuously interacting with the environment, it adapts to the dynamic changes of the environment. Excavate the potential file request transfer mode, obtain the approximate value function, and then obtain the cooperative caching decision matching the file request transfer mode; the macro base station encodes the cooperative caching decision, and communicates the coded cooperative caching result to each small station . The present invention formulates caching decisions through the transfer mode of file requests in the real network mined by reinforcement learning, without any assumptions on the prior distribution of data, and is more suitable for actual systems; and through real-time interaction with the environment, time-varying files can be tracked Popularity, make a corresponding caching strategy, the process is simple and feasible, and there is no need to solve the NP‑hard problem.

Description

technical field [0001] The invention belongs to the technical field of wireless network deployment in mobile communication, and in particular relates to an ultra-dense heterogeneous network small station coding cooperative caching method. Background technique [0002] With the popularity of smart terminals and the development of Internet services, ultra-dense heterogeneous networks will become one of the key technologies of the fifth generation mobile communication system (5G) in order to meet users' requirements for high data rates and high service quality. Deploying dense small cells within the coverage area of ​​macro base stations can effectively improve the communication quality of users at the edge of the network, thereby improving spectrum efficiency and system throughput. However, since the small cells are connected to the macro base station through the wireless backhaul link, the densely deployed small cells put a huge pressure on the wireless backhaul link. At this...

AI Technical Summary

Problems solved by technology

First of all, it is often considered that the popularity of files does not change with time, and the popularity of files in the actual network changes from time to time. This method of solving optimization problems based on constant file popularity cannot track the continuous changes of file popularity, thus The resulting caching decision cannot be well applied to the actual network; secondly, even if the constant file popularity is replaced with the instantaneous file popularity, once the file popularity is changed, the optimization problem will be rerun, which brings Huge network overhead, and the optimization problem of modeling is often NP-hard (Non-Polynomial hard) problem, which is very difficult to solve; finally, because the cache problem itself is based on the file request behavior that has occurred in the network, the cache decision is made. To prepare for the upcoming file request behavior, the method of making cache decisions based on the traditional solution optimization problem cannot mine the transfer mode of file requests in the network, so that the cache decision made is not optimal for the upcoming file requests

Images