Impact of Precipitation on Millimetre Wave Fixed Links and Mobile Measurement

Abstract

Millimetre wave communication stands at the forefront of technological innovation, underpinning the development of fifth-generation and subsequent communication networks. Despite its promise, the propagation characteristic of millimetre wave radio links is significantly compromised by rain attenuation, an important factor during link budget calculations. This thesis undertakes a comprehensive investigation into the impact of precipitation on mmWave fixed links and mobile measurements, combining empirical weather data analysis with advanced channel sounding techniques to unravel the dynamics between rainfall and signal attenuation.
Central to this research is the deployment of a PWS100 laser disdrometer, a high-resolution meteorological instrument, to capture annual rainfall statistics encompassing intensity, duration, and drop size distribution (DSD). Complementing this, an advanced channel sounder operating at 25.84 GHz (K-band) and 77.52 GHz (E-band) was implemented to measure real-time signal attenuation across a 35-meter terrestrial link. These datasets enable an examination of how raindrop microphysics such as size, velocity, and the interaction with electromagnetic waves at mmWave frequencies.
This thesis evaluates established DSD models, including the Marshall-Palmer exponential model, log-normal distribution, and gamma distribution, to assess their fidelity in representing observed rainfall patterns in the Durham region, UK. While the Marshall-Palmer model offers simplicity, its limitations in capturing small-droplet dominance (<1.5 mm) under stratiform rain conditions are highlighted. On the other hand, the gamma distribution demonstrates superior adaptability to diverse rain regimes, accommodating both the skewed droplet populations of convective storms and the narrower spectra of prolonged stratiform events. This finding underscores the necessity of region-specific DSD model calibration, particularly in temperate climates characterized by mixed precipitation types. This thesis encompasses annual statistics of rain events, focusing on intensity, duration, and drop size distribution. This thesis employs various distribution models, including the Marshall and Palmer, log-normal, and gamma distribution models, to fit the collected DSD data and evaluate their suitability for representing rainfall patterns.
A key finding of this thesis is the significant role of the wet antenna effect on short communication links and the importance of selecting appropriate scattering methods for the calculation of rain attenuation. The study reveals that the ITU-530 model may not be entirely suitable for short links and that a combination of Mie and Rayleigh scattering methods provides more accurate predictions for millimetre wave frequencies.
Overall, this research contributes to a better understanding of the complex interactions between rain and millimetre wave signals, offering valuable insights for the development of robust communication systems capable of operating efficiently under various weather conditions. The results of this research has also led to two journal publications and serval conference papers.