Adaptive Rate NOMA for Cellular IoT Networks

TL;DR

This paper proposes an adaptive rate NOMA scheme for cellular IoT networks, demonstrating improved capacity and support for larger packet sizes compared to traditional OMA, especially under strict delay constraints.

Contribution

It introduces a novel adaptive rate NOMA configuration for uplink IoT communications and analyzes its performance using stochastic geometry, highlighting its advantages over OMA.

Findings

NOMA provides better ergodic capacity for mobile users under strict IoT delay constraints.

NOMA supports larger packet sizes for IoT devices than OMA at the same delay.

The analysis shows NOMA's potential to enhance spectral efficiency in IoT cellular networks.

Abstract

Internet-of-Things (IoT) technology is envisioned to enable a variety of real-time applications by interconnecting billions of sensors/devices deployed to observe some random physical processes. These IoT devices rely on low-power wide-area wireless connectivity for transmitting, mostly fixed- but small-size, status updates of their associated random processes. The cellular networks are seen as a natural candidate for providing reliable wireless connectivity to IoT devices. However, the conventional orthogonal multiple access (OMA) to these massive number of devices is expected to degrade the spectral efficiency. As a promising alternative to OMA, the cellular base stations (BSs) can employ non-orthogonal multiple access (NOMA) for the uplink transmissions of mobile users and IoT devices. In particular, the uplink NOMA can be configured such that the mobile user can adapt transmission rate based on its channel condition while the IoT device transmits at a fixed rate. For this setting, we analyze the ergodic capacity of mobile users and the mean local delay of IoT devices using stochastic geometry. Our analysis demonstrates that the above NOMA configuration can provide better ergodic capacity for mobile users compare to OMA when IoT devices' delay constraint is strict. Furthermore, we also show that NOMA can support a larger packet size for IoT devices than OMA under the same delay constraint.