进化算法迭代优化的P2P网络信任模型研究

摘要

为了解决P2P网络系统中节点频繁退出和加入引起的系统数据信息误差,并针对数据获取对中心节点依赖度较大和系统中节点的数据信息获取不完整等问题,提出了进化算法迭代优化的P2P网络信任模型.首先将P2P网络系统中节点数据信息获取的系统信任度估计模型转化为从源节点到目标节点最优信任关系的路径寻优问题;然后利用改进的粒子群算法对信任关系路径方案进行粒子映射,并通过对粒子粒距聚集度和粒子信息熵进行计算而修正粒子权值,再对粒子局部最优解和全局最优解进行更新;最后迭代的对信任关系路径解空间中的最优解进行搜索,并对最优路径的节点进行推荐信任度加权.仿真结果表明,改进算法具有较好的收敛速度和较强的有效性,且当节点跳级数较少时,可使系统数据信息估算获得最优的系统信任度.%In the paper,an improved P2P network trust model was proposed based on evolutionary algorithm with iterative optimization,to solve the problems that the data acquisition depends on the center nodes and the data information of the nodes in the system is not full.The trust estimation model of the system from the nodes data information was converted to the route optimization searching problem from source nodes to the target nodes based on the optimal trust relationship.The trust relationship was processed with particle mapping with the improved particle swarm optimization algorithm.The distance of the particles and the clustering degree and the information entropy were computed,and the weights of the particles were modified.The local and global optimal solutions were updated.The path solution of trust relationship in the space was searched with iteration.The nodes of the optimum path were weighted with the optimal trust degree.The simulation result shows that the convergence rate of the improved algorithm is fast with good availability.When the number of node placement is less,the data information estimation can get the optimum trust degree.