科研成果详情

With the development of microprocessor technology, massive IoT devices are deployed in smart cities which promotes numerous supervisory control and data acquisition (SCADA) systems to accomplish the key issue. The emergence of computation intensive and delay sensitive applications makes it quite a challenge for IoT devices in SCADA systems with very weak computing capacity. Vehicular edge computing (VEC) is a new computing paradigm with great potential of enhancing SCADA system performance by offloading applications from the resource-constrained SCADA system to lightweight and ubiquitous VEC servers. In this paper, we propose a solution to exploit the load of IoT devices offloaded to VEC server only through vehicles by three steps. Firstly, the tasks from SCADA system are loaded via vehicles; Secondly, vehicles offload task to VEC servers; Finally, the vehicle returns the result to SCADA system. We establish this problem as integrating offloading to VEC servers with load balancing, and returning the result to SCADA system with less cost. A Joint Selection decision, Computation resource and Return result (JSCR) scheme is proposed to solve the problem. First, we propose to integrate the load balance with the tasks offload problem to efficiently complete task and further formulate the problem as a mix-integer non-linear programming problem. Then, we decouple the problem as two subproblems and develop a low complexity algorithm to jointly optimize Selection decision, Computation resource allocation and Return result. Finally, we conducted extensive simulation experiments. The results illustrate that the proposed algorithm exhibits fast convergence speed and demonstrates the superior performance. Specifically, the proposed JSCR scheme improves the system utility by 15.81% and load balancing efficiency by 27.5%.