Smart City Gnosys
Smart city article details
| Title | Weather-Aware Lorawan Channel Selection Using Bandit Algorithms |
|---|---|
| ID_Doc | 61574 |
| Authors | Kumar S. |
| Year | 2023 |
| Published | 2023 IEEE World Forum on Internet of Things: The Blue Planet: A Marriage of Sea and Space, WF-IoT 2023 |
| DOI | http://dx.doi.org/10.1109/WF-IoT58464.2023.10539541 |
| Abstract | LoRaWAN (Long Range Wide Area Networks) has a broad range of applications in outdoor IoT scenarios, including smart cities, agriculture, and disaster management. However, the performance of LoRa Wansignals is significantly affected by weather conditions, leading to decreased Signal-to-Noise Ratio (SNR) and resulting packet losses. Previous studies have investigated the influence of weather parameters on LoraWAN signal strength and path loss. Nevertheless, there is a lack of analysis on online weather parameter learning at the LoRaWAN edge and choosing the RF channel appropriately for improving the SNR. This research attempts to enable the online learning of weather conditions at the edge and facilitate the usage of suitable RF channels. Such a process pre-emptively avoids signal degradation and improves the SNR. This is achieved by employing a multi-criteria channel bandit (MCCB) algorithm at the LoraWAN edge. The MCCB adjusts the selection of RF channels based on real-time weather parameters. We evaluate the effectiveness of three MCCB variants: MCCB-Greedy, MCCB-Bayesian, and MCCB-Exponential. The effectiveness of three MCCB variants is evaluated using real-world weather datasets from vineyard farm field to optimize LoRaWAN performance. © 2023 IEEE. |
| Author Keywords | Agriculture; Internet of Things; LoRaWAN; Multi Arm Bandits; QoS |
