Optimizing the Transaction Times for NOMA-Based Mobile Edge Computing

Abstract

Non orthogonal multiple access (NOMA) and mobile edge computing (MEC) are evolving as key enablers for future fifth generation (5G) networks as this combination can provide high spectral efficiency, massive connectivity, improved quality-of-service (QoS), and lower latency. In this research, we aim to minimize the transaction time difference of NOMA paired users, where the users have disparate amount of data to send to the MEC servers, which have variable computing capabilities. The transaction time includes both the transmission time of data and the computational time of the servers. The equalization of transaction time for paired users reduces the wastage of both frequency and computational resources. The transaction time difference is minimized by optimizing the transmission power of the users and the computational resources of severs using a successive convex approximation method. The results show that the proposed scheme reduces the transaction time difference of the two paired users from hundreds of seconds to just a few seconds; thereby conserving the resources. .The percentage improvement in effective throughput of the system is shown to be 19% on average.