Joint Downlink and Uplink Edge Computing Offloading in Ultra-Dense HetNets

The deployment of ultra-dense heterogeneous networks (HetNets) are envisioned as the essentials to embrace of intelligent applications for the next-generation wireless networks. For ultra-dense HetNets, the small scale heterogeneous edge servers in macrocell and smallcell, decreasing downlink and uplink communication delay should be attracted more attention for latency-critical intelligent applications. Therefore, we aim at edge computation offloading of joint downlink (DL) and uplink (UL) together with communication and computing resource allocation. Then, we formulate the computation offloading optimization problem into minimizing the overall delay of task while saving energy consumption of user device. However,the joint downlink and uplink computing offloading problem with resource allocation is a mixed binary integer programming which is difficult to deal with. We convert the programming problem into resource allocation sub-problem and computing offloading sub-problem, and propose an efficient joint downlink and uplink offloading algorithm for ultra-dense HetNets. Numerical results validate the efficiency of the proposed algorithm in terms of the delay and energy saving of system.