Undergraduate YPEC 2022

UG20 – An Enhanced AI-based Resource Allocation Scheme for LoRaWAN Harmonization

Internet of Things (IoT) is an inevitable factor of modern society due to the relationship between a large number of IoT applications and people’s everyday life such as water quality sensors etc. LoRa and LoRaWAN as one of the most applied IoT technologies and networking protocols have been deployed on a large scale in Hong Kong with the estimated market size in the future at 1 billion USD in 2027. As the market grows, the increasing number of LoRaWAN devices deployed in HK would degrade their quality of service (QoS). There are two main reasons. The first reason is the competition for bandwidth and channel resources between a large number of connected end devices. Another one is the redundant channel resource allocation configurations of most end devices to achieve better transmission reliability. To address these challenges and to bring further economic benefits to Hong Kong, this work proposes a machine learning scheme based on Long Short-Term Memory (LSTM) Al method to mitigate the affection of ALOHA scheme and the multi-gateway interference problem. This improvement of LoRaWAN is estimated to improve the network’s overall performance and coverage by 20%, bringing economic benefits of over 200 million USD. Also, with the IEEE P2668 Standard to be published on LoRa and IoT industry, a foreseeable growing trend for a global market is at the corner. The improvements of this project integrated with the P2668 standard will bring more economical benefits and better IoT network performance to the IoT industry in HK.