# Research on data transmission system based on expert library reinforcement learning in integrated network

## TL;DR

This paper introduces a data transmission system using expert library reinforcement learning and MEMS sensors to improve wireless network performance in challenging environments.

## Contribution

The novelty lies in using expert library reinforcement learning without reward functions for wireless data transmission enhancement.

## Key findings

- The proposed system achieves strong generalizability in various wireless network conditions.
- It demonstrates fast convergence in data transmission retrieval.
- MEMS sensors effectively monitor wireless network status for improved transmission stability.

## Abstract

With the continuous advancement of network transmission technology, more and more applications are being applied in wireless network environments, especially in places that require high coverage, such as oceans and mountainous areas. However, wireless data transmission has the disadvantages of unstable transmission and easy interruption using traditional methods. Based on this, we propose a data transmission system that uses a micro-electron-mechanical system (MEMS) sensor to obtain the wireless network status and applies expert library reinforcement learning that does not rely on reward functions to achieve retrieval enhancement of data transmission. Experimental verification shows that the proposed expert library reinforcement learning has strong generalizability and fast convergence.

Expert library reinforcement learning, wireless network, MEMS, integrated network.