Decomposed Multiobjective Wolf Pack Algorithm for Resource Allocation and Task Scheduling in Computing Networks

In computing networks, resource allocation disorder and task scheduling imbalance can lead to problems such as long latency, high energy consumption, and high cost. To address these issues, a computing network model integrating nonorthogonal multiple access (NOMA) and wireless charging at base stations is constructed, and a decomposed multiobjective wolf pack algorithm (MOWPA) is proposed to jointly optimize resource allocation and task scheduling. The uplink of the network uses NOMA technology, which allows multiple users to share the same subchannel and greatly improves the efficiency of spectrum utilization. The introduction of wireless charging technology at the base station ensures that users can complete their computing tasks without interruption and reduces maintenance costs. In the algorithm design, the decomposition strategy is introduced into the MOWPA to screen the initial population by polynomial mutation operator and differential evolution operator to improve the diversity of the initial population. To help the algorithm escape from local optimum, the mutation operator is introduced to generate new elements, so that the population can explore a wider solution space. The experimental results show that when the number of users reaches 40, the algorithm achieves average improvements of over 22.47%, 27.82%, and 25.58% in computing delay, energy consumption, and cost, respectively. Compared with the other 10 algorithms, it significantly improves the user experience and resource utilization.