Spaceborne microwave radiometers generally operate within protected or shared spectral bands that are designated for scientific research. However, previous experiences with research satellites (e.g. SSM/I, AMSR, WindSat, SMOS) and airborne campaigns have indicated Radio Frequency Interference (RFI) is increasingly plaguing the operation of spaceborne radiometers, and would undoubtedly undermine the mission science objectives if left unchecked. Examples of RFI sources include air-traffic control radars, satellite telecommunication links, and improperly shielded or designed consumer electronics.
The RFI detection algorithm Aquarius currently uses is a time-domain glitch detector with four tunable parameters [Misra 2008] independently adjusted at each 1 degree latitude/longitude grid. The four parameters are:
- WM: Local mean running average window
- TM: Local mean running average glitch threshold
- TD: RFI detection glitch threshold
- WD: RFI detection neighborhood window
The algorithm is applied to each Aquarius L1a short accumulation (sample). This short accumulation is referred to as the Sample Under Test (SUT).
Step 1: The WM samples surrounding the SUT are identified. If any of these samples have been flagged in a previous iteration of the algorithm, that sample is eliminated.
Step 2: The surviving samples from Step 1 are averaged and this mean is known as the “dirty mean”. Samples greater than the dirty mean by TM multiples of the NEDT are eliminated.
Step 3: If there are surviving samples from Step 2, they are averaged to arrive at the “clean mean”.
Step 4: If the SUT is greater than the “clean mean” by TD multiples of NEDT, it is flagged as RFI.
Step 5: If the SUT is flagged as RFI, WD samples surrounding the SUT are flagged as RFI.
The algorithm attempts to find the true value of the SUT iteratively had it not been corrupted with RFI. If the SUT is a certain multiple of NEDT above that value, it is flagged, along with some neighboring samples.
The on-orbit performance of the RFI algorithm is being evaluated as part of the early-orbit commissioning phase. In the near future, the Remote Sensing Group plans to develop and implement other RFI detection methods, including those based on signal statistics and polarimetry, in preparation for the Soil Moisture Active Passive (SMAP) mission.
Park, J., J. T. Johnson, N. Majurec, N. Niamsuwan, J. Piepmeier, P. Mohammed, C. Ruf, S. Misra, S. Yueh and S. Dinardo, “Airborne L-band Radio Frequency Interference Observations from the SMAPVEX08 Campaign and Associated Flights,” IEEE Trans. Geosci. Remote Sens., 49(9), 3359-3370, doi:10.1109/TGRS.2011.2107560, 2011.
Misra, S., P. N. Mohammed, B. Guner, C. S. Ruf, J. R. Piepmeier and J. T. Johnson, “Microwave Radiometer Radio Frequency Interference Detection Algorithms: A Comparative Study,” Trans. Geosci. Remote Sens., 47(11), 3742-3754, doi:10.1109/TGRS.2009.2031104, 2009.
Misra, S. and C. S. Ruf, “Detection of Radio Frequency Interference for the Aquarius Radiometer,” Trans. Geosci. Remote Sens., 46(10), 3123-3128, 2008.
De Roo, R., S. Misra and C. Ruf, “Sensitivity of the Kurtosis Statistic as a Detector of Pulsed Sinusoidal RFI,” Trans. Geosci. Remote Sens., 45(7), 1938-1946, 2007.
Ruf, C.S., S. M. Gross and S. Misra, “RFI Detection and Mitigation for Microwave Radiometry with an Agile Digital Detector,” Trans. Geosci. Remote Sens., 44(3), 694-706, 2006.