数联网标识解析系统中的标识数据布局策略

Identifier Data Layout Strategy in Identifier Resolution System of Internet of Data

数联网是为解决目前互联网数据资源共享交互过程中传输效率低、协调成本高、安全管控难等缺陷而提出的一种新型信息基础设施。标识解析系统是实现数据流通的关键,但随着标识注册量和解析量的迅速增长,解析节点正面临着大量高并发解析请求,服务质量将变得难以保证。针对该问题,考虑数联网标识解析系统中节点架构的特点,以降低解析时延为目标,提出一种基于遗传算法的自适应离散粒子群优化算法(DPSO-GA)来对标识进行合理布局。该方法综合考虑节点之间的带宽、标识数量和节点容量等因素对解析时延的影响,引入遗传算法的交叉操作和变异操作,对粒子群优化算法的惯性权重因子采取自适应策略,对学习因子采取线性增减的策略。实验结果表明,相较于传统粒子群优化算法及遗传算法,该算法优化效果分别提升了48.9%和19.9%,增加了种群进化的多样性及搜索范围,减少了算法的时间复杂度以及陷入局部最优解的可能性,且能较稳定地降低标识解析时延。

The Internet of Data(IoD)is a new information infrastructure proposed to address low transmission efficiency,high coordination costs,and challenging security controls in current internet data resource sharing and interactions.The key to enabling data circulation lies in the identification resolution system.However,as identification registration and resolution rapidly increase,resolution nodes encounter numerous high-concurrency resolution requests,making it difficult to ensure service quality.To tackle this issue,we propose a novel approach,the adaptive Discrete Particle Swarm Optimization algorithm based on Genetic Algorithm(DPSO-GA),considering the node architecture's characteristics in IoD's identification resolution system.This method aims to identify layouts effectively to reduce resolution latency.It comprehensively considers factors such as bandwidth,identifier count,and node capacity's impacts on resolution latency between nodes.Additionally,it introduces crossover and mutation operations from a Genetic Algorithm(GA),employs an adaptive strategy for the inertial weight factor of the Particle Swarm Optimization(PSO)algorithm,and utilizes a linear increasing/decreasing strategy for the learning factor.Experimental results demonstrate that compared to traditional PSO and GA,DPSO-GA improves optimization by 48.9%and 19.9%,respectively.This enhances population evolution diversity,expands search range,and reduces algorithm time complexity and the likelihood of falling into a local optimum,thereby consistently reducing identification resolution delays.