Optimizing spectral utilization of LPWANs

• Main Author: Jathun Gamage Isuru Amalinda
• Other Authors: Mo Li
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Computer science and engineering
• Online Access: https://hdl.handle.net/10356/172897
• Institution: Nanyang Technological University

The emergence of the Internet of Things (IoT) has been pivotal in advancing urban and industrial efficiency, largely enabled by the growth of long-range low-power wide-area networks (LPWANs). These networks have facilitated IoT integration into diverse applications, including environmental monitoring, traffic control, and smart metering. In the LPWAN domain, LoRa modulation stands out for its long-range communication capabilities, addressing challenges posed by the rapid expansion of IoT devices. Despite being relatively new, LoRa has rapidly expanded to nearly 150 countries, redefining LPWAN standards. However, there is significant potential for further optimization. This thesis focuses on enhancing LoRa networks. First we identify a fundamental limitation in current LoRa networks: their reliance on the basic ALOHA mechanism for media access control, a result of LoRa's lack of carrier sense capability. Our research reveals that the recently introduced channel activity detection feature in LoRa, initially aimed at energy-efficient preamble chirp detection, can also reliably detect payload chirps. This discovery leads to the development of an efficient carrier-sense multiple access protocol, named LMAC, tailored for LoRa networks. We present three progressive versions of LMAC, each building upon the last. These versions implement carrier-sense multiple access and optimize communication load distribution among channels defined by frequencies and spreading factors. This optimization is based on local information from end nodes and, additionally, global information from gateways. Our empirical studies, including a 50-node lab testbed and a 16-node university deployment, demonstrate that LMAC significantly outperforms the traditional ALOHA mechanism. The results show up to 2.2 times higher goodput and a 2.4 times reduction in radio energy per successfully delivered frame, indicating that replacing LoRaWAN's ALOHA with LMAC could yield considerable network performance enhancements. In addition to the initial focus on LMAC, this thesis then tackles the implementation and optimization of this protocol within the industrial requirements of LoRaWAN. We explore the complexities of aligning LMAC with global regulatory compliance and ensuring its interoperability with existing network deployments. This involves a thorough analysis of LMAC's performance under a variety of network conditions and scenarios. The research presents a significant step towards integrating LMAC into the LoRaWAN standard, highlighting its potential in enhancing the overall efficiency and effectiveness of LoRaWAN networks. Furthermore, the thesis addresses the increasing demand for spectral resources in the rapidly expanding IoT ecosystem. The surge in IoT nodes places significant strain on the limited and crowded spectrum. To address this, we propose a strategy for more efficient spectrum utilization: the concept of unchannelized or 'borderless' spectrum utilization for LoRa, allowing nodes to transmit frames at any selected central frequency, thereby enhancing spectral efficiency. This borderless approach poses challenges, such as detecting the central frequency of arbitrarily placed frames prior to demodulation. We provide solutions to these challenges, ensuring effective implementation of this novel spectrum utilization method.