周恺卿, 杨森宇, 康棣文,等.基于改进布谷鸟搜索算法的无线传感器网络覆盖优化[J].河南理工大学学报（自然科学版）,2025,44(4):48-58.

ZHOU K Q, YANG S Y, KANG D W, et al. Coverage optimization in wireless sensor networks  based on improved cuckoo search algorithm [J]. Journal of Henan Polytechnic University (Natural Science) , 2025, 44(4): 48-58.

基于改进布谷鸟搜索算法的无线传感器网络覆盖优化

周恺卿, 杨森宇, 康棣文, 欧云

吉首大学 通信与电子工程学院， 湖南 吉首 416000

摘要: 目的 为了提高无线传感器网络（wireless sensor network， WSN）覆盖率，提出一种针对WSN覆盖优化的改进布谷鸟搜索（improved cuckoo search with multi-strategies， ICS-MS）算法。   方法 ICS-MS算法基于对标准布谷鸟搜索算法的寻优过程进行分析，通过建立马尔可夫模型，分析迭代过程，明确改进方向，降低自转移概率，减少平均理论迭代次数，进行一系列改进策略选择。首先，引入分阶段逐维更新策略，以减轻高维空间中的维度耦合效应，降低解空间的自转移概率；其次，在执行Lévy飞行操作后，依据个体适应度进行精英保留，并通过反向搜索操作扩展搜索域；最后，采用基于多策略的随机偏好游走算法，结合全局最优解的信息，引导种群向最优解方向进化。实验基于节点同构、节点感知范围与通信范围相同、节点实时感知假设，对WSN覆盖优化问题进行建模，并提出布谷鸟个体的构建方法。同时，以最大化WSN覆盖率为目标，实验采用点监测离散化方法，在节点数量20，30场景下，对ICS-MS算法与标准布谷鸟搜索（cuckoo search，CS）及其6种变体（MACS，DA-DOCS，WCSDE，ICS-ABC-OBL，CSDE，ICS）进行比较。   结果 结果表明，ICS-MS算法在节点数20时，相较于对比算法，覆盖率平均提升17.12%~17.35%；在节点数30时，覆盖率平均提升10.09%~18.05%。  结论 ICS-MS算法在WSN覆盖优化领域展现出更均匀的节点分布、更高的覆盖率和更快的收敛速率。 

关键词:无线传感器网络;马尔可夫链;布谷鸟搜索算法;逐维更新;反向搜索

doi: 10.16186/j.cnki.1673-9787.2024070037

基金项目:国家自然科学基金资助项目（62066016）；湖南省自然科学基金资助项目（2024JJ7395）；湖南省教育厅科学研究优秀青年项目（22B0549）

收稿日期:2024/07/09

修回日期:2024/09/11

出版日期:2025/06/19

Coverage optimization in wireless sensor networks  based on improved cuckoo search algorithm

Zhou Kaiqing, Yang Senyu, Kang Diwen, Ou Yun

School of Communication and Electronic Engineering， Jishou University， Jishou 416000， Hunan， China

Abstract: Objectives To enhance the coverage rate of Wireless Sensor Networks （WSN）.   Methods An Improved Cuckoo Search with Multi-Strategies （ICS-MS） algorithm is proposed for the coverage optimization problem in WSN. The ICS-MS algorithm analyzes the optimization process of the standard Cuckoo Search algorithm by establishing a Markov model. Through iterative process analysis， it identifies improvement directions， reduces self-transition probability， decreases the theoretical average iteration count， and implements a series of optimized strategy selections. Initially， a phased dimension-by-dimension update strategy is introduced to mitigate the dimension coupling effect in high-dimensional spaces and to reduce the self-transition probability of the solution space. Subsequently， elite individuals are retained based on their fitness after performing Lévy flight operations， and the search domain is expanded through opposition-based search operations. Finally， a multi-strategy based stochastic preference walk algorithm is employed， incorporating information from the global optimal solution to guide the population evolution towards the optimal solution. Experiments modeled WSN coverage optimization under three assumptions： node homogeneity， identical sensing/communication ranges， and real-time node sensing capability， while establishing cuckoo individual construction methods. Targeting maximal WSN coverage rate， the discrete point monitoring method was employed to compare the proposed ICS-MS against standard CS and six variants （MACS， DA-DOCS， WCSDE， ICS-ABC-OBL， CSDE， ICS） under 20-node and 30-node scenarios.   Results The experimental results show that the ICS-MS algorithm， in 20 nodes scenario， achieves an average increase in coverage rate of 17.12%~17.35% compared to the comparative algorithms； in 30 nodes scenario， the average increase is 10.09%-18.05%.   Conclusions The ICS-MS algorithm demonstrates more uniform node distribution， higher coverage rates， and faster convergence rates in the field of WSN coverage optimization. 

Key words: wireless sensor network; Markov chain; cuckoo search algorithm; dimension-by-dimension update; opposition-based search