Collaborative Task Offloading Based on Scalable DAG in Cell-Free HetMEC Networks

With the development of the Internet of Things (IoT), there are numerous real-time applications emerging for B5G. Driven by the low-latency requirement, mobile edge computing has become an important technology for enhancing user experience and reducing network costs. However, the computing power of the single-processor CPU is limited, it is difficult to meet the low-latency requirements of computation-intensive applications. Besides, the performance of the network is limited by cell interference. In this paper, we design a user-centric Cell-Free Heterogeneous Mobile Edge Computing (CF-HetMEC) network architecture, which comprehensively utilizes the computing and transmission resources of multi-heterogeneous processors of Central Processing Units. A category-based Directed Acyclic Graphs (DAGs) subtask offloading algorithm in CF-HetMEC network is proposed to reduce the system delay. Simulation results show that the proposed scheme can effectively reduce the processing delay and the energy consumption of CF-HetMEC network.