Methods of active pre-caching in heterogeneous wireless network and device

Abstract

The invention relates to the resource allocation field of wireless networks, and particularly to methods of active pre-caching in heterogeneous wireless network and a device. Concernment on content pre-caching in wireless communication is caused for using renewable energy sources and saving traditional energy sources. The methods focus on the purpose of content pushing and caching for improving agreen energy source utilization rate and saving the traditional energy sources. State transition probability and future reward in a mobile environment are unknown, thus the methods utilize reinforcement learning to solve the problems of green-energy-source allocation and content pushing. Q learning is a model-free reinforcement-type learning technology, and can find a best action selection strategy in the MDP problem. A Boltzmann distribution method is used for updating the strategy, and the methods can obtain best action in each state according to a Q table after the Q table is stable.

Description

technical field [0001] The invention relates to the field of wireless network resource allocation, in particular to a content pre-caching method and system in a heterogeneous wireless network. Background technique [0002] Due to the rapid growth of multimedia services and the sharp increase of carbon dioxide emissions, green communication is an effective measure to solve such problems. There are many ways of green wireless access, such as energy harvesting, multicast and heterogeneous network. EH technology using solar energy, wind energy, kinetic energy and other natural energy can greatly reduce the power consumption of traditional power wireless communication, thereby reducing carbon dioxide emissions. It is considered as one of the candidate technologies for implementing green communication. Wireless multicast achieves a huge energy advantage by simultaneously providing multimedia content that is often of interest to multiple users while extending a single data stream...

AI Technical Summary

Problems solved by technology

However, each technique has its limitations

[0003] On the one hand, due to the limited battery capacity, a mismatch between energy and request arrival may occur, resulting in energy waste or shortage

On the other hand, in order to achieve wireless multicast, some user requests need to be delayed to wait for concurrent transmissions, which may seriously impair the quality of service required by earlier

Finally, due to the inflexible deployment of micro base stations, the cost of deploying support for power lines and high-speed backhaul links is relatively high

Images