LoRa signal propagation path loss measurement and prediction in riparian flood area scenarios

This thesis researches LoRa (from long range), a low power, wide area networking protocol, signal point-to-point propagation path loss measurement and prediction in a riparian flood area. LoRa is an emerging internet of things technology derived from the chirp spread spectrum. The benefit of this technology is high anti-interference performance. And a riparian zone is a riverbank full of vegetation. The topic serves for the emergence water information network project, aiming to establish a real-time flood warning system.

The author adopts the LoRa signal with a software-defined radio to obtain a novel sounder system that could work in the harsh riverbank flood area. The sniffed LoRa signal’s waveform can be examined by a spectrum analyser. And the information packet can be received by the paired LoRa module. This novel sounder system establishes a relationship between the received signal strength index and the real received signal power.

Using the designed novel sounder system, propagation path loss measurements were organised in the two typical flood happening terrains. The measurements were organised in the UK plain as well as Mexico valley. Received signal strength is recorded, and the path loss model is calculated according to the measurements. We find that vegetation significantly affects the measurements' low antenna height signal path loss. Vegetation can interfere with signal propagation, especially when raining or the wind is blowing heavily. The terrain difference can lead to extra path loss since the reflection material changes. Besides, a unique measurement related to riverbank vegetation excess loss was organised. A novel modified exponential decay vegetation excess loss model has been obtained. A novel optimised empirical path loss model has been calculated according to the results.

To predict the signal propagation path loss when the flood happens, A novel ray tracing simulation towards environment changes is designed for propagation prediction. From the simulation, an explanation of loss is given according to the ray paths on the journey.

Furthermore, a prediction excess loss range is given, considering the various stages of the flood.