移动边缘计算网络中联合无线多播的服务功能链部署算法

Service function chain embedding algorithm with wireless multicast in mobile edge computing network

为解决移动边缘计算(MEC)网络中面向用户的服务功能链部署(SFC)算法系统开销过大、业务拥塞严重等问题,针对部署MEC服务器的多基站多用户边缘网络,提出了一种面向内容的联合无线多播的SFC部署算法。综合考虑数据流、服务器功能维护功耗、服务器功能服务功耗和无线传输功耗4种系统开销,建立波束成形设计和SFC映射的联合优化模型。首先,利用拉格朗日对偶分解技术,将优化问题解耦为SFC部署和波束成形设计2个独立子问题;其次,利用基于L_p范数惩罚项的连续凸近似算法,将整数形式的SFC部署问题松弛为一个等价线性规划问题;最后,利用路径跟随技术,将非凸波束向量优化问题转化为一系列凸优化子问题。仿真结果表明,所提算法具有较好收敛性能,并在系统开销方面优于传统的最优单播SFC部署算法和随机多播SFC部署算法。

To resolve the excessive system overhead and serious traffic congestion in user-oriented service function chain(SFC)embedding in mobile edge computing(MEC)networks,a content-oriented joint wireless multicast and SFC embedding algorithm was proposed for the multi-base station and multi-user edge networks with MEC servers.By involving four kinds of system overhead,including service flow,server function sustaining power,server function service power and wireless transmission power,an optimization model was proposed to jointly design SFC embedding with multicast beamforming.Firstly,with Lagrangian dual decomposition,the problem was decoupled into two independent subproblems,namely,SFC embedding and multicast beamforming.Secondly,with the L_p norm penalty term-based successive convex approximation algorithm,the integer programming-based SFC embedding problem was relaxed to an equivalent linear programming one.Finally,the non-convex beamforming optimization problem was transformed into a series of convex ones via the path following technique.Simulation results revealed that the proposed algorithm has good convergence,and is superior to both the optimal SFC embedding with unicasting and random SFC embedding with multicasting in terms of system overhead.