贝叶斯预测蜂群算法在无线传感器网络优化中的应用

针对无线传感器网络(WSN,wireless sensor network)节点分布不合理,存在较多的监测盲区等不足,提出了利用贝叶斯预测人工蜂群算法(BPABC,Bayesian predictive artificial bee colony algorithm)制定节点分布方案。BPABC算法借鉴贝叶斯预测算法的思想对蜂群算法中各蜜源存在最优解的概率进行预测,并以此为依据指导跟随蜂寻优工作。采用BPABC算法对WSN中的节点分布进行优化,与人工蜂群算法、全局人工蜂群算法制定的优化方案进行比较。结果表明,BPABC在平均覆盖率、最差覆盖率等方面均优于其他两种算法,并且BPABC算法在迭代收敛速度方面也有明显的优势。为了进一步验证改进算法的实用性,采用BPABC制定不同监测区域的WSN节点分布方案。WSN的覆盖率均在97%左右,并且标准差不超过0.005%。由此可见,基于BPABC的WSN节点分布优化方案具有较高的覆盖率、良好的适应性和稳定性。

Application of Bayesian predictive bee colony algorithm in WSN optimization

The node distribution of wireless sensor network(WSN) is often unreasonable, and always has many monitoring blind spots.Aiming at this problem,Bayesian predictive artificial bee colony algorithm (BPABC) is proposed to develop a node distribution scheme. Based on the idea of Bayesian prediction algorithm, this algorithm predicts the probability of optimal solution of each nectar source in the bee colony algorithm, and guides the followed bees to seek optimal solution. A designed algorithm is used to optimize the distribution of nodes in WSN, and the effect is compared with those of artificial bee colony algorithm and global artificial bee colony algorithm. The results show that BPABC is superior to the other two algorithms in terms of average coverage and worst coverage. Besides,this algorithm also has obvious advantages in iterative convergence rate. In order to further verify the practicability of the improved algorithm, this paper uses BPABC algorithm to develop WSN node distribution scheme for different monitoring areas. Coverage for all WSNs is around 97% with a standard deviation no more than 0.005% can be seen that the WSN node distribution optimization scheme based on BPABC has high coverage,good adaptability and stability.