Joint task offloading and resource allocation in mixed edge/cloud computing and blockchain empowered device-free sensing systems

Abstract

Recent studies have shown that wireless signals can not only be used as communication medium for data transmission, but also can be employed for device-free sensing (DFS). However, the sensed raw big data processing needs large amount of processing resources. Moreover, the sensed data are usually privacy-sensitive, where blockchain can be employed to guarantee the privacy related issues, and however, blockchain tasks are usually also computation-intensive. To provide the processing capabilities for sensed big data processing and blockchain task executing in DFS, mobile edge computing (MEC) and cloud computing are essential enabling technologies. In this paper, we consider a multiuser mobile offloading network consisting a edge node (or fog node) and a remote cloud server. The tasks could be processed locally, or offloaded to the edge nodes, or further migrated to the cloud relayed by the edge node. We formulate the offloading problem as the joint optimization of task offloading decision making of all the users, the computation resource allocation among the edge-executing applications, and radio resource assignment among all the remote-processing applications, aiming to minimize the maximum weighted cost of all the users. It is demonstrated that the problem is NP-hard. To tackle this challenge, we decoupled the problem into subproblems, where offloading decisions are obtained using semi-definite relaxation (SDR), and next, a swarm intelligence algorithm, i.e., fireworks algorithm, is adopted in radio resource allocation. Simulation results exhibit that as a result of the collaboration of the collaborate of edge and cloud, the proposed joint algorithm could achieves nearly optimal performance in the aspects of energy consumption and delay compared with other benchmark algorithms.