【摘要】：針對異構(gòu)蜂窩網(wǎng)絡下微蜂窩密集部署的不規(guī)則性,提出了一種基于強化學習的自優(yōu)化控制系統(tǒng),通過微蜂窩功率控制解決微蜂窩密集部署下的網(wǎng)絡的容量與覆蓋問題。將模糊邏輯與Q學習算法相結(jié)合,綜合考慮網(wǎng)絡的平均用戶性能、邊緣用戶性能和網(wǎng)絡環(huán)境相互影響來設計模糊邏輯與Q學習算法的聯(lián)合瞬時回報獎懲值,進行網(wǎng)絡容量與覆蓋的聯(lián)合自優(yōu)化。仿真結(jié)果表明,該方法能實現(xiàn)密集化微蜂窩部署下的容量與覆蓋自優(yōu)化,有效提高系統(tǒng)平均用戶吞吐量和邊緣用戶吞吐量。
[Abstract]:Aiming at the irregularity of microcellular dense deployment in heterogeneous cellular networks, a self-optimization control system based on reinforcement learning is proposed, which solves the network capacity and coverage problems under microcellular dense deployment through microcellular power control. Combining fuzzy logic with Q learning algorithm, considering the average user performance of network, edge user performance and network environment interaction, the combined instantaneous reward and punishment value of fuzzy logic and Q learning algorithm is designed. The joint self-optimization of network capacity and coverage is carried out. Simulation results show that this method can achieve capacity and coverage self-optimization under intensive microcellular deployment, and effectively improve the average user throughput and edge user throughput of the system.
【作者單位】： 重慶郵電大學移動通信技術重點實驗室;
【基金】：國家“863”計劃資助項目(2014AA0101A701) 國家科技重大專項資助項目(2014ZX03003010-004) 國家自然科學基金資助項目(6157103)
【分類號】：TN929.5

【參考文獻】

相關期刊論文 前2條

1 HAO Peng;YAN Xiao;Yu-Ngok Ruyue;YUAN Yifei;;Ultra Dense Network: Challenges, Enabling Technologies and New Trends[J];中國通信;2016年02期

2 豐雷;李文t，

本文編號：2159354