
Emergency communication network planning method based on deep reinforcement learning
CHEN Hao-ran, ZHU Wei, YU Sheng
Emergency communication network planning method based on deep reinforcement learning
Emergency communication has the characteristics of strong sudden and uncertainty, to meet the requirements of flexible network planning, on the basis of different characteristics of diverse layered network to accomplish the modeling description, apply the deep reinforcement learning algorithm to implement network topology planning task, and enhance the generation efficiency through the algorithm optimization. According to the business characteristics and strategies to allocate the emergency communications network business resources, then achieve the complete emergency communications network planning, finally the emergency communications network planning model and the method is verified with high rationality and efficiency by a sample simulation, which provides a certain reference for the emergency communications network planning, which provides a certain reference for the emergency communications network planning.
emergency communication network; communication network planning; deep reinforcement learning; DQN algorithm {{custom_keyword}};
Tab.1 Node connection rules of emergency communication network表1 应急通信网节点连接规则 |
主动接 入方 | 被动接 入方 | 数量限制 | 距离 限制 | 优先级 |
---|---|---|---|---|
Cm | Bn | LC | RC+RB | 高 |
Cm | LB-C | RB+RC | 高 | |
Bn | Ap | LB-A | RB+RA | 中 |
Bn | LB-LB-C-LB-A | RB+RB | 低 | |
Ap | Uq | LA-LA-B | RA+RU | 高 |
Bn | LA-B | RA+RB | 低 | |
Uq | Ap | LU | RU+RA | 高 |
Tab.2 Sample nodes weight value on different layers表2 不同层次节点权重值样表 |
节点类型 | 节点权重 |
---|---|
指挥所 | WC |
骨干传输 | WB |
综合接入 | WA |
终端应用 | WU |
Tab.3 Emergency communication service generating result表3 应急通信业务生成结果 |
节点类型 | 地址范围 | IP资源数量 | 频谱资源 |
---|---|---|---|
指挥所1 | 27.XX.0.0/25 27.XX.0.128/26 | 192 | 45 MHz-52 MHz |
指挥所2 | 27.XX.0.192/26 27.XX.1.0/24 27.XX.2.0/25 | 448 | 52 MHz-64 MHz |
指挥所3 | 27.XX.2.128/25 27.XX.3.0/24 | 384 | 65 MHz-75 MHz |
…… | …… | …… | …… |
[1] |
彭高召. 基于无人机的应急通信网研究[D]. 浙江大学, 2019.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
[2] |
李莉, 朱正甲, 任赟, 等. 基于绿色无线网络覆盖最优的配电通信网规划方法研究[J]. 电力系统保护与控制, 2018, 46(7): 56-62.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
[3] |
余斌, 尹项根, 吴小忠, 等. 输电线路在线监测的层次化通信网络规划模型[J]. 中国电力, 2019, 52(3): 161-168.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
[4] |
魏焕新, 胡招娣. 梯度DPSO算法在舰船电力通信网络规划中的应用[J]. 舰船科学技术, 2020, 42(8): 64-66.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
[5] |
王巍, 赵文红. 应急通信网拓扑模型及其分析[J]. 通信学报, 2012, 33(S1): 201-209.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
[6] |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
[7] |
刘全, 翟建伟, 章宗长, 等. 深度强化学习综述[J]. 计算机学报, 2018, 41(1): 1-27.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
[8] |
盘小娜, 陈哲, 李金泽, 等. 一种利用优先经验回放深度Q-Learning的频谱接入算法[J]. 电讯技术, 2020, 60(5): 489-495.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
[9] |
尤天晴, 叶欣, 初德龙. 电力系统数据通信网IP地址规划分析[J]. 吉林电力, 2016, 44(4): 9-13.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
[10] |
杨非. 无线中继网络的同频干扰问题研究[D]. 中国科学技术大学, 2014.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
[11] |
苗丽. 无线通信网频率资源动态分配方法[D]. 哈尔滨工业大学, 2013.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
[12] |
刘汝清, 贺晓星. 互调干扰与GSM网络规划[J]. 移动通信, 2012, 36(10): 43-47.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_ref.label}} |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
/
〈 |
|
〉 |