Abstract
Growing demands and key requirements for Machine-to-Machine (M2M) communications have raised concerns about the coexistence of various networks in unlicensed bands. Currently in unlicensed bands, Short Range Devices (SRD) are performing a part of the services that M2M aims to implement. SRD comply with European standards and therein two random access schemes, Duty Cycle (DC) Limit and Listen Before Talk (LBT), are specified. In fact, DC Limit and LBT can be regarded as pure ALOHA and non-persistent CSMA, respectively. Thus, it is likely that radio devices using ALOHA and CSMA coexist in a single network. However, in spite of the simple concepts of ALOHA and CSMA, there have been few research works on the accurate performance analysis for such heterogeneous networks due to high correlations among devices. To tackle this issue, we rigorously investigate the types of collisions that can occur in such networks and derive accurate formulas for related probabilities. We also provide new and accurate closed form expressions for the throughput and success probability associated with each channel access scheme. The accuracy of our results is validated by simulation. Taking advantage of the accuracy of our results, we provide a method to configure the network for optimal network throughput. Furthermore, we devise a distributed dynamic channel access mechanism in which each device adjusts its access scheme by learning the network state to achieve near-maximum network throughput and high fairness.
Highlights
Machine-to-machine (M2M) communications are getting more and more attention as a promising solution to meet increasing demands on enabling automated remote communications for various Internet of Things applications
THROUGHPUT MAXIMIZATION AND FAIR COEXISTENCE we provide how the network consisting of multiple devices that can access the channel via Duty Cycle (DC) Limit or Listen Before Talk (LBT) achieves near-optimal throughput and high fairness
In this paper, we consider a heterogeneous radio network consisting of multiple radio devices where different channel access schemes, DC Limit and LBT, can be adopted
Summary
Machine-to-machine (M2M) communications are getting more and more attention as a promising solution to meet increasing demands on enabling automated remote communications for various Internet of Things applications. It is a natural circumstance that multiple radio devices with different channel access schemes coexist in the same network limiting their DC to the same extent. We consider a heterogeneous network consisting of multiple radio devices where some of those use DC Limit and the rest use LBT as the channel access schemes. We assume that the aggregate transmission trials of devices as Poisson processes using similar assumptions that are widely used in previous works With this assumption, we model the channel behavior as a renewal process where an inter-renewal period, called an event period, consists of an idle period and a busy period. By exploiting the accuracy of our formulas, we devise a channel access mechanism that each device adaptively adjusts its access scheme via learning the number of devices in the network, which achieves near-maximum network throughput and high fairness (Section V)
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call