SHADE算法在无线传感器网络节点分布优化的应用
|
摘要
为了提高无线传感器网络覆盖率,减少目标区域覆盖盲区,采用基于历史成功的自适应参数差分进化(SHADE)算法对节点分布进行优化,制定节点优化方案。以网络覆盖率为优化目标函数,建立联合概率模型,采用SHADE算法对目标函数进行求解。SHADE使用了控制参数设置的历史记忆来指导未来控制参数值的选择,确保了精确和快速收敛到全局最优,增强了算法鲁棒性。通过与差分进化算法和人工蜂群差分进化算法进行仿真对比,验证了SHADE算法能够快速收敛,网络覆盖率高。仿真结果表明,在500次模拟情况下,SHADE算法的平均覆盖率分别高于差分进化算法5.17%、人工蜂群差分进化算法1.88%。
In order to improve the coverage of wireless sensor network and reduce the blind area in the target area,the node distribution is optimized by using the success-history based parameter adaptation for differential evolution( SHADE) algorithm.Taking the network coverage as the optimization objective function,a joint probability model is established,and the SHADE is used to solve the objective function. SHADE algorithm uses the historical memory of control parameter settings to guide the selection of future control parameter values,which ensures accurate and fast convergence to the global optimum and enhances the robustness of the algorithm. By comparing with differential evolution algorithm and artificial bee colony differential evolution algorithm,it is verified that SHADE algorithm can converge rapidly and has the high network coverage rate. The simulation results show that under the same conditions of 500 times of simulation,the average coverage rate of SHADE algorithm is 5. 17% higher than the differential evolution algorithm,and 1. 88% higher than artificial bee colony differential evolution algorithm.
In order to improve the coverage of wireless sensor network and reduce the blind area in the target area,the node distribution is optimized by using the success-history based parameter adaptation for differential evolution( SHADE) algorithm.Taking the network coverage as the optimization objective function,a joint probability model is established,and the SHADE is used to solve the objective function. SHADE algorithm uses the historical memory of control parameter settings to guide the selection of future control parameter values,which ensures accurate and fast convergence to the global optimum and enhances the robustness of the algorithm. By comparing with differential evolution algorithm and artificial bee colony differential evolution algorithm,it is verified that SHADE algorithm can converge rapidly and has the high network coverage rate. The simulation results show that under the same conditions of 500 times of simulation,the average coverage rate of SHADE algorithm is 5. 17% higher than the differential evolution algorithm,and 1. 88% higher than artificial bee colony differential evolution algorithm.
引文
[1]YANG H,XUNBO L I,WANG Z,et al.A novel sensor deployment method based on image processing and wavelet transform to optimize the surface coverage in WSNs[J].Chinese Journal of Electronics,2016,25(3):495-502.
[2]王明亮,闵新力,薛君志.基于改进人工鱼群算法的WSN覆盖优化策略[J].微电子学与计算机,2015(6):78-81.
[3]潘丽姣,吴红英.混沌逃逸粒子群优化算法在WSN覆盖优化中的应用[J].重庆邮电大学学报(自然科学版),2014,26(2):177-181.
[4]熊伟丽,刘欣,陈敏芳,等.基于差分蜂群算法的无线传感器网络节点分布优化[J].控制工程,2014,21(6):1036-1040.
[5]袁浩.基于改进蜂群算法无线传感器感知节点部署优化[J].计算机应用研究,2010,27(7):2704-2705.
[6]石为人,袁久银,雷璐宁.无线传感器网络覆盖控制算法研究[J].自动化学报,2009,05:540-545.
[7]KIRAN M S,HAKLI H,GUNDUZ M,et al.Artificial bee colony algorithm with variable search strategy for continuous optimization[J].Information Sciences An International Journal,2015,300(1):140-157.
[8]TANABLE R,FUKUNAGA A.Success-history based parameter adaptation for differential evolution[C]//Evolutionary Computation,IEEE,2013:71-78.
[9]乔俊飞,傅嗣鹏,韩红桂.基于混合变异策略的改进差分进化算法及函数优化[J].控制工程,2013,20(5):943-947.
[10]林川.粒子群优化与差分进化算法研究及其应用[D].成都:西南交通大学,2009.
[11]刘波,王凌,金以慧.差分进化算法研究进展[J].控制与决策,2007,22(7):721-729.
[2]王明亮,闵新力,薛君志.基于改进人工鱼群算法的WSN覆盖优化策略[J].微电子学与计算机,2015(6):78-81.
[3]潘丽姣,吴红英.混沌逃逸粒子群优化算法在WSN覆盖优化中的应用[J].重庆邮电大学学报(自然科学版),2014,26(2):177-181.
[4]熊伟丽,刘欣,陈敏芳,等.基于差分蜂群算法的无线传感器网络节点分布优化[J].控制工程,2014,21(6):1036-1040.
[5]袁浩.基于改进蜂群算法无线传感器感知节点部署优化[J].计算机应用研究,2010,27(7):2704-2705.
[6]石为人,袁久银,雷璐宁.无线传感器网络覆盖控制算法研究[J].自动化学报,2009,05:540-545.
[7]KIRAN M S,HAKLI H,GUNDUZ M,et al.Artificial bee colony algorithm with variable search strategy for continuous optimization[J].Information Sciences An International Journal,2015,300(1):140-157.
[8]TANABLE R,FUKUNAGA A.Success-history based parameter adaptation for differential evolution[C]//Evolutionary Computation,IEEE,2013:71-78.
[9]乔俊飞,傅嗣鹏,韩红桂.基于混合变异策略的改进差分进化算法及函数优化[J].控制工程,2013,20(5):943-947.
[10]林川.粒子群优化与差分进化算法研究及其应用[D].成都:西南交通大学,2009.
[11]刘波,王凌,金以慧.差分进化算法研究进展[J].控制与决策,2007,22(7):721-729.