Leveraging cloud infrastructure to the mobile edge computing helps the mobile users to get real time multimedia services in Fifth Generation (5G) network system. To ensure higher Quality-of-Experience (QoE), faster migration of mobile multimedia service instances is required to cope up with user mobility. By deploying the mobile multimedia service instances proactively in multiple edge nodes (ENs) helps the users to get higher QoE. However, excessive deployment of service replicas might increase the cost of the overall network. To establish trade-off between these two conflicting objectives, we have formulated the problem as a Multi-objective Integer Linear Programming (MILP) by integrating the users’ path prediction model. This problem is proven to be an NP-hard one for large networks, thus we develop an artificial intelligence (AI) based meta-heuristic Binary Particle Swarm Optimization (BPSO) algorithm to achieve near-optimal solution within polynomial time. The performance analysis results show the significant performance improvement in terms of QoE and user satisfaction as compared to other state-of-the-art works.

利用云基础架构进行移动边缘计算可帮助移动用户在第五代（5G）网络系统中获得实时多媒体服务。为了确保更高的体验质量（QoE），需要更快地迁移移动多媒体服务实例以应对用户移动性。通过在多个边缘节点（EN）中主动部署移动多媒体服务实例，可以帮助用户获得更高的QoE。但是，过度部署服务副本可能会增加整个网络的成本。为了在这两个相互矛盾的目标之间进行权衡，我们通过集成用户的路径预测模型将问题表述为多目标整数线性规划（MILP）。这个问题被证明是大型网络的NP难题，因此，我们开发了一种基于人工智能（AI）的元启发式二进制粒子群优化（BPSO）算法，以在多项式时间内实现接近最优的解决方案。性能分析结果表明，与其他最新技术相比，在QoE和用户满意度方面的性能显着提高。