Heuristic computing methods for contact plan design in the spatial-node-based Internet of Everything

摘要

To satisfy the increasing demands of high-speed transmission, high-efficiency computing, and real-time communications in the high-dynamic and heterogeneous networks, the Contact Plan Design (CPD) has attracted continuous attention in recent years, especially for the spatial-node-based Internet of Everything (IoE). In this paper, we study the NP-hardness of contact scheduling and the attenuation of atmospheric precipitation in the spatial-node-based IoE. Two heuristic computing methods for contact plan design are proposed by comprehensively considering the time-varying topology, the intermittent connectivity, and the adaptive transmission in different weather conditions, which are named Contact Plan Design-Particle Swarm Optimization (CPD-PSO) and Contact Plan Design-Greedy algorithm with the Minimum Delivery Time (CPD-GMDT) separately. For the population-based algorithm, CPD-PSO not only solves the CPD problem with a limited-resource condition, but also dynamically adjusts the search scope to ensure the continuous searching capability of the algorithm. For the CPD-GMDT that makes CP decisions based on the current state, the algorithm uses the idea of greedy algorithm to schedule Satellite-Platform Links (SPLs) and Inter Satellite Links (ISLs) respectively using the strategies of optimal matching and load balancing. The simulation results show that the proposed CPD-PSO outperforms Contact Plan Design-Genetic Algorithm (CPD-GA) in terms of fitness and delivery time, and CPD-GMDT presents better overall delay than Fair Contact Plan (FCP).