Development of a digital receiver for range imaging atmospheric radar

In this paper, we describe a new digital receiver developed for a 1.3-GHz range imaging atmospheric radar. The digital receiver comprises a general-purpose software-defined radio receiver referred to as the Universal Software Radio Peripheral 2 (USRP2) and a commercial personal computer (PC). The re...

Full description

Bibliographic Details
Published in:Journal of Atmospheric and Solar-Terrestrial Physics
Main Author: Yamamoto M.K.; Fujita T.; Abdul Aziz N.H.B.; Gan T.; Hashiguchi H.; Yu T.-Y.; Yamamoto M.
Format: Article
Language:English
Published: Elsevier Ltd 2014
Online Access:https://www.scopus.com/inward/record.uri?eid=2-s2.0-84904795266&doi=10.1016%2fj.jastp.2013.08.023&partnerID=40&md5=31b0b192170b2be372b33bec2be8ae47
Description
Summary:In this paper, we describe a new digital receiver developed for a 1.3-GHz range imaging atmospheric radar. The digital receiver comprises a general-purpose software-defined radio receiver referred to as the Universal Software Radio Peripheral 2 (USRP2) and a commercial personal computer (PC). The receiver is designed to collect received signals at an intermediate frequency (IF) of 130MHz with a sample rate of 10MSs-1. The USRP2 digitizes IF received signals, produces IQ time series, and then transfers the IQ time series to the PC through Gigabit Ethernet. The PC receives the IQ time series, performs range sampling, carries out filtering in the range direction, decodes the phase-modulated received signals, integrates the received signals in time, and finally saves the processed data to the hard disk drive (HDD). Because only sequential data transfer from the USRP2 to the PC is available, the range sampling is triggered by transmitted pulses leaked to the receiver. For range imaging, the digital receiver performs real-time signal processing for each of the time series collected at different frequencies. Further, the receiver is able to decode phase-modulated oversampled signals. Because the program code for real-time signal processing is written in a popular programming language (C++) and widely used libraries, the signal processing is easy to implement, reconfigure, and reuse. From radar experiments using a 1-μs subpulse width and 1-MHz frequency span (i.e., 2-MHz frequency bandwidth), we demonstrate that range imaging in combination with oversampling, which was implemented for the first time by the digital receiver, is able to resolve the fine-scale structure of turbulence with a vertical scale as small as 100m or finer. © 2013 Elsevier Ltd.
ISSN:13646826
DOI:10.1016/j.jastp.2013.08.023