|Téma:||Detekce deště z dat páteřních mikrovlnných spojů|
|Vedoucí:||prof. Dr. Ing. Jan Kybic|
|Vypsáno jako:||Diplomová práce, Bakalářská práce, Semestrální projekt|
|Popis:||Rainfall information is a crucial input for water management in cities, particularly for the design, optimization, and control of sewer systems and waste water treatment plants, flood early warning, etc. Standard rain-gauge networks are usually insufficiently dense and weather radars are affected by large uncertainties and measure rainfall several hundreds of meters above ground. Data on microwave link attenuation, monitored by mobile phone operators for network management purposes, can be used to estimate rainfall intensity, since microwave link attenuation is affected by precipitation. The task is to develop an automatic procedure for classification of rainfall estimation from data collected from microwave links. This is still a challenge, especially due to other phenomena contributing to radiowave attenuation and frequent hardware malfunction causing erratic signal fluctuations. Automated detection of malfunction (quality control) and reliable classification between dry and wet periods are crucial steps for upscaling microwave link rainfall observation technique from single links to city scale and later also to country-wide scale. The algorithm should be capable to learn from historical microwave link data and data from traditional rainfall network. It should be able to classify acquired data with a latency not exceeding one minute. You will be part of the interdisciplinary research project Tel4Rain, where we collaborate with hydrologists, urban water engineers, radio engineers and IT specialists and with relevant stakeholders from the City of Prague, T-Mobile CZ and Veolia.
|Literatura:||Duda, Hart, Stork: Pattern Classification.
Bishop: Pattern Recognition and Machine Learning
Overeem et al: Country-wide raingall maps from cellular communication networks. PNAS. 2013, 110 (8) 2741-2745, https://doi.org/10.1073/pnas.1217961110
Leijnse et al: Microwave link rainfall estimation: Effects of link length and frequency,
temporal sampling, power resolution, and wet antenna attenuation. Advances in Water Resources. 31 (2008) 1481–1493
Fencl et al: Gauge-adjusted rainfall estimates from commercial microwave links. Hydrol. Earth Syst. Sci., 21, 617–634, 2017
Bianchi et al.: A Variational Approach to Retrieve Rain Rate by Combining Information
from Rain Gauges, Radars, and Microwave Links. Journal of Hydrometeorology, 2013, doi:10.1175/JHM-D-12-094.1