abstract |
The invention relates to a Lyapunov-based optimized multi-server computing offloading method, which specifically includes: system modeling: a multi-user edge computing system consists of N mobile devices and M small base stations, and all mobile devices are randomly distributed in the M small base stations. Within the common coverage of small cells, the computing and communication resources of small cells are limited, and mobile devices can communicate with edge servers through wireless communication technologies such as 5G; build a computing model to obtain the energy required for tasks to be calculated locally and transmitted to edge servers. energy consumption and energy consumption, construct an energy model, and obtain an energy queue; obtain the execution cost of a mobile device in a single time slot, and construct the minimization problem of the average execution cost of a mobile device; eliminate the energy causal constraint through the Lyapunov method, and use the alternate optimization idea and The Lagrangian dual method obtains the optimal solution for computing offloading decision and resource allocation strategy. The present invention can not only reduce the execution cost of the mobile device, but also reduce the task abandonment rate. |