A Multiwall Path-Loss Prediction Model Using 433 MHz LoRa-WAN Frequency to Characterize Foliage's Influence in a...

Area of Science:

• Wireless Communication
• Agricultural Technology
• Signal Propagation

Background:

• Palm oil is a major Asian cash crop, with smart agriculture offering revenue enhancement potential.
• Low-power wide-area network (LPWAN) technologies like LoRa are crucial for enabling smart farming in these plantations.
• Characterizing wireless channels is essential for reliable sensor network deployment.

Purpose of the Study:

• To characterize LoRa 433 MHz channels for various spreading factors and bandwidths in a palm oil plantation.
• To develop and validate a path-loss prediction model using empirical data for wireless sensor networks.
• To compare the proposed model's accuracy against established foliage loss models.

Main Methods:

• Empirical measurement of received signal strength (RSS) for LoRa 433 MHz.
• Data collection included approximately 1500 line-of-sight (LoS) and 300 non-line-of-sight (NLoS) propagation measurements.
• Path-loss exponent and attenuation per trunk/canopy were calculated for model construction.

Main Results:

• Path-loss exponents for LoS propagation were determined as 2.34 (125-250 kHz) and 2.9 (500 kHz).
• NLoS attenuation per trunk and canopy were quantified for spreading factors SF7-SF12.
• The developed prediction model achieved a mean RMSE of 2.74 dB, outperforming Weissberger's and ITU-R models.

Conclusions:

• The developed empirical path-loss model accurately predicts signal propagation in palm oil plantations.
• This model is superior to existing foliage loss models for optimizing LPWAN deployments in smart agriculture.
• Improved wireless communication reliability can enhance the efficiency and revenue of the palm oil industry.