Snow Layer Detection by Pattern Matching in a Multipath Radar Interference Scenario. Issue 9 (3rd May 2021)

• Record Type: Journal Article
• Authors: Kulsoom, Farzana; Dell'Acqua, Fabio; Pasian, Marco; Espín-López, Pedro Fidel
• Is Part Of: International journal of remote sensing. Volume 42:Issue 9(2021)
• Page Start: 3193
• Page End: 3218
• Publication Date: 2021-05-03
• Subjects: Remote sensing -- Periodicals; Télédétection -- Périodiques; 621.3678
• DOI: 10.1080/01431161.2020.1854890
• Languages: English
• ISSNs: 0143-1161

Abstract

ABSTRACT: Snowpack in sloped mountain terrain may present stratification in which layers of snow accumulate over a certain period. The loosely adhering layers in snowpack may exhibit a severe avalanche risk, and detecting such stratification phenomena is key to avalanche detection. The danger of manual, in-situ sampling calls for investigation of alternative approaches, including the use of in-situ radars. Considering a buried radar sensor, this manuscript analyses the impact of multipath phenomena on the results of the sensing operation. Multiple bounces in snow layers may degrade the received signal and corrupt the estimation results. Multipath phenomena are well known and analysed in communication systems, but snow layering measurement is a different problem, with little previous research done on related multipath phenomena. This paper analyses the problem with a multi-receiver radar architecture, investigating the impact of multi-path effects from layered snow on measurements carried out by a multi-static radar of Frequency-Modulated Continuous Wave (FMCW) type. A few simplifying assumptions are formulated, in order to make the problem manageable, and a possible approach for solving the related issues is proposed, using a robust algorithm based on pattern matching to detect snow layering in such a challenging environment. Results obtained from various simulations carried out with realistic environmental and geometrical parameters appear to support the effectiveness of the proposed method. Results show that with 3 metres of snow thickness, the probability of successfully detecting the correct numbers of layers exceeds 80%.