/DQN_RC_DSA_IOT2019

Deep Reinforcement Learning (DRL) based Dynamic Spectrum Access (DSA) using Reservoir Computing (RC) (In IoT-J-2019)

Primary LanguagePython

Distributive Dynamic Spectrum Access Through Deep Reinforcement Learning: A Reservoir Computing-Based Approach (IoT-J-2019)

Hao-Hsuan Chang, Hao Song, Yang Yi, Jianzhong (Charlie) Zhang, Haibo He, and Lingjia Liu

IEEE Internet of Things Journal, Vol. 6, No. 2, pp. 1938-1948, April 2019.

Introduction

A combination of reservoir computing (RC) and deep Q-network (DQN) is utilized to design spectrum access strategies for secondary users (SUs) in dynamic spectrum access (DSA) networks.

System Model

Citation

If you find the code useful in your research, please cite:

@article{Chang2019DSA,
author={Chang, Hao-Hsuan and Song, Hao and Yi, Yang and Zhang, Jianzhong and He, Haibo and Liu, Lingjia},
journal= {IEEE Internet of Things Journal},
title={Distributive Dynamic Spectrum Access Through Deep Reinforcement Learning: A Reservoir Computing-Based Approach},
year={2019},
volume={6},
number={2},
pages={1938--1948},
month={April}}

Training code

>> main.py

Plot results

>> plot_figure.py