A winning proposal for the Innovative Research Program, 2008: Developing a Dual-Frequency FMCW Radar to Study PrecipitationInvestigator: Christopher.Williams@colorado.edu, CIRES, 303-497-3829 Objective: The objective of this Innovative Research Proposal is to develop an inexpensive dual-frequency vertically pointing precipitation profiling radar utilizing technologies developed for the mobile phone, the police radar, and video gaming industries. The two operating frequencies will have different attenuation properties through rain and will constrain the inverse modeling estimates of rain rate versus altitude. The radar hardware is expected to cost less than $12k. After showing proof-of-concept, funds will be requested from NSF and/or NASA to develop multiple (at least nine) profiling radars to investigate the spatial and temporal structure of precipitating cloud systems over a 10 km square satellite footprint. Importance: Better observations of precipitating cloud systems are needed to improve our understanding of microphysical and dynamical precipitation processes which will lead to improved model parameterizations of precipitation in high resolution cloud resolving models and global climate models. But with commercial profiling radars costing over $200k per radar, it is cost prohibitive for educational researchers to purchase commercial radars for precipitation research. With a small investment in time to understand technology developed for other industries, low-cost precipitation profiling radars can be developed for research and educational applications. Collaborators: While this Innovative Research Proposal supports only one investigator, there will be an informal collaboration during this project with CIRES and NOAA engineers in ESRL/PSD including: Paul Johnston, David Carter, and Ken Moran. Methodology: A bistatic Frequency-Modulated Continuous-Wave (FMCW) radar uses two antennas to continuously transmit and receive frequency modulated radio signals. The difference in the transmitted and received signal represents the distance and the motion of the targets within the radar’s field of view. The proposed radar will generate two FMCW signals in the S- and Xfrequency bands (near 2.8 and 10 GHz) with both signals being transmitted and received on the same set of antennas. The frequency modulation will use a linear ramp-up and ramp-down sawtooth pattern. The processing unit will need to be fast enough to sample the S-band signal during the ramp-up and sample the X-band signal during the ramp-down. A simple circuit diagram is shown in Figure 1. What Makes this Project Innovative? There are two elements of this project that are innovative. First, the transmitted signal is generated directly from voltage controlled oscillators (VCOs). The frequency modulation (FM) is produced by changing the input voltage from 0 to 4 volts. This simple and inexpensive device is the result of the mobile phone and police radar industries requiring low cost components. Another innovative element of this project is using a Sony PlayStation 3 (PS3) for the signal processing unit. The PS3 contains a Cell Broadband Engine (Cell BE) developed in collaboration with Sony, Toshiba, and IBM. The Cell BE has nine processing elements with the main processing element acting as a general-purpose dual-thread processor that distributes tasks in parallel to the eight other vector processors. The PS3 can be dual-bootable with Linux OS and many C library commands have already been vectorized for the Cell BE, including an FFT. The Cell BE is the fastest computational processor commercially available, and it costs less than $500! For more information see http://en.wikipedia.org/wiki/Cell_microprocessor. Validation of Performance: The developed FMCW radar will be deployed at the NOAA Boulder Atmospheric Observatory (BAO) next to a NOAA owned commercial profiler to evaluate the performance of this inexpensive dual-frequency FMCW precipitation radar system. Note: Two PlayStations are needed to evaluate new software while deploying the radar at BAO. |