Throughput-Oriented Non-Orthogonal Random Access Scheme for Massive MTC Networks

Yichen Wang,Tao Wang,Dawei Wang,Zihuan Yang,Julian Cheng

doi:10.1109/tcomm.2019.2957767

Abstract

Machine-type communications (MTC) technology, which enables direct communications among devices, plays an important role in realizing Internet-of-Things. However, a large number of MTC devices can cause severe collisions. As a result, the network throughput is decreased and the access delay is increased. To address this issue, a throughput-oriented non-orthogonal random access (NORA) scheme is proposed for massive machine-type communications (mMTC) networks. Specifically, by employing the technique of tagged preambles (PAs), multiple MTC devices (MTCDs) choosing the same PA can be distinguished and regarded as a non-orthogonal multiple access (NOMA) group, which enables multiple MTCDs to share the same physical uplink shared channel for transmissions by multiplexing in the power domain. The Sukhatme's classic theory and the characteristic function approach are adopted to formulate an optimization problem. The aim is to maximize the throughput subject to the constraints on the power back-off factor, the number of MTCDs included in a NOMA group, and the successful transmission probability. Based on the particle swarm optimization (PSO) algorithm, the formulated optimization problem is efficiently solved. The derived solution can be used to adjust the access class barring factor such that more MTCDs can obtain the access opportunities. Moreover, a low-complexity suboptimal solution is also developed, which can achieve near-PSO performance under high data rate requirement. Simulation results show that the proposed scheme can efficiently improve the network performance and comparison is made with the existing schemes.

Highlights

W ITH the development of the Internet-of-Things, machine-type communications (MTC) has been widely accepted as an efficient approach for future wireless communication systems
Because the random access (RA) scheme plays a critically important role in massive MTC networks, it should be carefully designed and a number of works have been dedicated to this important research topic [6]–[8]
We set the parameter of the exponentially distributed channel power gain u2 = 1 and the number of resource blocks (RBs) allocated for each physical uplink shared channel (PUSCH)

Summary

Introduction

W ITH the development of the Internet-of-Things, machine-type communications (MTC) has been widely accepted as an efficient approach for future wireless communication systems. As the MTC technology enables machine-type devices exchange data with little human intervention [1], it has a wide range of applications and has attracted much research attention from both industry and academia [2]–[4]. In MTC networks, the widely used random access (RA) procedure is the four-message handshake process [5]. The involved four messages include preamble (PA), random access response (RAR), initial layer-3 message, and contention resolution (CR) message. The rapid growth in the number of MTCDs will incur severe collisions and large signaling overheads, resulting in network congestion, large access delay, and the degradation of resource utilization efficiency. Because the RA scheme plays a critically important role in massive MTC (mMTC) networks, it should be carefully designed and a number of works have been dedicated to this important research topic [6]–[8]

Objectives

Results

Conclusion