Joint Security Service Assignment, Power Control and Multi-Step Computation Offloading in Ultra-Dense Multi-Task NOMA-Enabled Networks

To meet the connection requirements of massive mobile devices (MDs), balance network loads, fully utilize network resources, achieve green and secure communications, joint security service assignment, power control and multi-step computation offloading is performed in ultra-dense multi-task networks, minimizing the network-wide (total/system) energy consumption under users' latency and security cost constraints. To properly manage the severe interferences caused by the deployment of ultra-dense base stations, a frequency band division mechanism is first introduced. Then, in order to eliminate co-channel interferences and further improve bandwidth utilization, non-orthogonal multiple access (NOMA) technology is utilized. To solve the formulated problem in a nonlinear mixed-integer form, we develop a further improved whale optimization algorithm (FIWOA) by improving an existing algorithm (i.e., improved whale optimization algorithm, IWOA). In addition, we also develop another algorithm by enhancing the adaptive genetic algorithm with diversity-guided mutation (AGADGM) using FIWOA, which is named as genetic whale optimization algorithm (GWOA). As for the designed algorithms, some detailed analyses have been made for the convergence, complexity, and parallel implementation. Simulation results show that FIWOA may achieve lower network-wide energy consumption than other algorithms, and GWOA may achieve lower one than AGADGM under strict constraints of delay and security breach costs.