A Hardware Implemented Autocorrelation Technique for Estimating Power Spectral Density for Processing Signals from a Doppler Wind Lidar System

• Main Authors: Sameh Abdelazim, David Santoro, Mark Arend, Fred Moshary, Sam Ahmed
• Format: Article
• Language: English
• Published: MDPI AG 2018-11-01
• Series: Sensors
• Subjects: Doppler Lidar; remote sensing; wind Lidar; FPGA; autocorrelation
• Online Access: https://www.mdpi.com/1424-8220/18/12/4170
• ISSN: 1424-8220

A signal processing technique utilizing autocorrelation of backscattered signals was designed and implemented in a 1.5 µm all-fiber wind sensing Coherent Doppler Lidar (CDL) system to preprocess atmospheric signals. The signal processing algorithm’s design and implementation are presented. The system employs a 20 kHz pulse repetition frequency (PRF) transmitter and samples the return signals at 400 MHz. The logic design of the autocorrelation algorithm was developed and programmed into a field programmable gate array (FPGA) located on a data acquisition board. The design generates and accumulates real time correlograms representing average autocorrelations of the Doppler shifted echo from a series of adjustable range gates. Accumulated correlograms are streamed to a host computer for subsequent processing to yield a line of sight wind velocity. Wind velocity estimates can be obtained under nominal aerosol loading and nominal atmospheric turbulence conditions for ranges up to 3 km.