Network Throughput Optimization for Random Access Narrowband Cognitive Radio Internet of Things (NB-CR-IoT)

Narrowband Internet of Things (NB-IoT) is a new technology being implemented into the Long-Term Evolution (LTE) standards to support machine-to-machine communications. Applications of IoT are expected to rapidly increase and to expand into many sectors. Increased number of IoT devices trying to access the medium with typically small data packets will overwhelm the network. Cognitive radios coupled with random access strategies can help alleviate excessive collisions in the network by many IoT access attempts. On the other hand, most IoT devices do not have hardware capabilities to perform sophisticated spectrum sensing methods over long durations. Therefore, it is important to reduce the spectrum sensing overhead while maximizing the NB-IoT network level throughput. In this paper, we derive a unique set of optimal sensing parameters which achieve the maximum throughput of a narrowband cognitive radio IoT (NB-CR-IoT) network. We show that accurate sensing from a single device perspective is not always the best solution for maximizing the throughput in an NB-CR-IoT network.

I. Introduction

Internet of Things (IoT) is a vision for the future network to support machine-to-machine (M2M) communications. Without the help of human intervention, one of the demands of M2M communications is to maintain stable connectivity among IoT devices with limited resources such as power, bandwidth, and system complexity. To fulfill such a demand, a new narrowband IoT (NB-IoT) technology has been standardized by 3GPP into Long-Term Evolution (LTE) releases to enable a wide range of services [1], [2]. IoT presents unique challenges to the radio network since massive number of IoT devices will desire to access the medium at any time. IoT traffic which involves inherently small data packets will be subject to excessive collisions. Cognitive radios (CRs) coupled with random access strategies can help alleviate congestion in the network [3]. A number of studies based on different IoT random access protocols have been presented in literature. In [4] and [5], in-depth surveys on existing mechanisms of random access in LTE and LTE advanced (LTE-A) standards are provided. Two opportunistic random access mechanisms based on carrier sensing multiple access/collision avoidance (CSMA/CA) are proposed for IoT in [6]. In [7], the delay distribution of slotted-ALOHA protocol, one of the candidates of 5G random access protocol for M2M communications, is derived.