Secure Collaborative Computation Offloading and Resource Allocation in Cache-Assisted Ultra-Dense MEC Networks With Multi-Slope Channels

Cache-assisted ultra-dense mobile edge computing (MEC) networks have been extensively seen as a promising solution to meeting the rapidly growing requirements of massive mobile devices (MDs). To properly tackle the complicated, severe, and average interferences caused by small base stations (SBSs) ultra-densely deployed in such networks, the orthogonal frequency division multiple access (OFDMA), non-orthogonal multiple access (NOMA) and base station (BS) clustering are jointly considered in this paper. To protect the tasks of MDs offloaded to BSs for computing, which are exposed to multiple MDs, and vulnerable to eavesdropping and malicious attacks, some security measures are further introduced. After that, we develop a computation offloading scheme to minimize the energy consumed by MDs under the constraints of delay, power, computing resources, and security costs, which jointly optimizes the task execution decision, device association, channel selection, security service assignment, power control, and computing resource allocation. To solve the finally formulated problem, we develop a high-performance algorithm by improving the existing hierarchical adaptive search algorithm. Then, the convergence, computation complexity, and parallel implementation analyses are made for the proposed algorithms. Simulation results show that such algorithms may generally achieve lower total energy consumption and delay than other algorithms under strict latency and cost constraints.