EE Seminar: Rain-Mapping through Compressed Sensing: Reconstruction Criteria Relating Image Sparsity, Resolution, and Random-Observations

 

Speaker: Lior Gazit,

M.Sc. student under the supervision of Prof. Hagit Messer - Yaron

 

Monday, October 10th, 2016 at 15:00

Room 011, Kitot Bldg., Faculty of Engineering

 

Rain-Mapping through Compressed Sensing: Reconstruction Criteria Relating Image Sparsity, Resolution, and Random-Observations

 

Abstract

 

An innovative method published in 2006 by Prof. Messer-Yaron introduced the use of existing cellular networks, and specifically microwave-transmission data, for 2-D mapping of precipitation (such as rain).

Current weather monitoring techniques lack the ability to capture precipitation maps with sufficient spatial resolution.

The purpose of the research presented here is to form criteria for successful reconstruction of a precipitation map. These criteria, novel in its application to microwave precipitation monitoring, calculate the probability of achieving proper estimation. When given a file of microwave measurements, one would apply the criteria and immediately know the expected probability of reconstruction for the rain map. Moreover, the suggested method also serves as a guide for system design, telling what adjustments may allow a higher probability of reconstruction. The methods applied are borrowed from the world of image-processing, linear-operators, and statistical-modeling.

The scientific contribution is three fold. First, this work provides the criteria for reconstruction. Second, it marks another contribution to the study of linear systems of equations and their application is real-world measurement problems. Specifically, it presents a unique method of image-sampling where an image is observed by random straight lines. Third, it presents a statistical study and modeling of microwave-links.

 

10 באוקטובר 2016, 15:00 
חדר 011, בניין כיתות חשמל 
אוניברסיטת תל אביב עושה כל מאמץ לכבד זכויות יוצרים. אם בבעלותך זכויות יוצרים בתכנים שנמצאים פה ו/או השימוש
שנעשה בתכנים אלה לדעתך מפר זכויות, נא לפנות בהקדם לכתובת שכאן >>