The CubeSat Radiometer Radio Frequency Interference Technology Validation (CUBERRT) Mission

Over the past couple of decades passive microwave radiometry observations have shown an increase in man-made RF interference corrupting measured radiances and thus impacting geophysical retrievals. These man-made sources also known as Radio Frequency Interference (RFI) have impacted various geophysical retrievals such as soil-moisture, sea-surface salinity, wind-speed and wind-direction, atmospheric water-vapor, precipitation retrievals etc. The spectrum environment for passive microwave measurements is expected to become worse over the coming decades due to an increasing demand from the commercial industry.

Low-frequency (~1.4 GHz) radiometry missions such as NASA’s Aquarius and SMAP (Soil Moisture Active Passive) have applied increasingly advanced RFI detection algorithms implemented on their mission payloads. SMAP is one of the first space-borne missions to implement an advanced digital signal processing algorithm for RFI detection. Though these algorithms have been successful, implementing these algorithms on higher microwave frequencies (> 6 GHz) is not trivial. The bandwidth of the payload scales with the center frequency of observation. The number of channels required for RFI detection also increases, impacting size of downlink data volume. The spectrum protection allocated to higher frequencies is of secondary nature as opposed to primary allocation for low-frequency radiometers. These facts point towards a technology gap that needs to intelligently identify RFI, mitigate RFI, downlink mitigated/unmitigated power measurements, over wide bandwidths of above a GHz to enable future high-frequency radiometers to operate in an RFI intensive environment.

The CubeRRT (CubeSat Radiometer Radio Frequency Interference Technology Validation) mission was selected under NASA’s In-space Validation of Earth Science Technologies (InVEST) program to demonstrate on-board, real-time RFI processing from 6-40 GHz over a GHz bandwidths. CubeRRT is built in a 6U configuration. The CubeRRT payload has three critical pieces of technology, a wideband antenna unit, a radiometer front-end (RFE) unit, and a radiometer digital back-end (RDB) that performs the on-board detection and filtering of RFI. The main objective of the CubeRRT mission is to demonstrate the RFI mitigation technology on a flight-ready hardware in space, increasing the technology readiness level (TRL) from 6 to 7. CubeRRT is designed to make wideband measurements over the whole 6 to 40 GHz range, but the prime mission objective is to demonstrate RFI mitigation over ten “golden” frequency bands that are allocated to Earth observation bands. The Ohio State University leads the CubeRRT mission. The algorithm validating back end technology is built at Jet Propulsion Laboratory, California Institute of Technology and the radiometer front end is built at NASA Goddard Space Flight Center.

The CubeRRT antenna subsystem consists of three circularly polarized tapered helical antennas. The antennas are being designed, developed and tested at The Ohio State University. The series of antenna are necessary to provide sufficient gain over a wide range of frequencies from 6 to 40 GHz. The current design provides a gain of 12 dBi at 6 GHz and 21 dBi at 40 GHz.

The CubeRRT radiometer front-end (RFE) is designed to sweep from 6 to 40 GHz with a 1 GHz bandwidth being injected into the RDB. The radiometer is a single tunable superheterodyne receiver. At the front-end of the radiometer the RFE has a four-position switch to choose between the three helical antennas as well as a reference load for calibration. The RFE contains front-end wideband coaxial low noise amplifiers for pre-amplification. The RFE also contains a coupled wideband noise-source to for full internal calibration of the radiometer. The RFE achieves frequency tuning via a phased-locked oscillator (PLO) and sub-harmonic image rejection (IR) mixer. The design allows flexibility between choosing upper and lower side-bands to completely cover the 6 to 40 GHz regime. The architecture sacrificed radiometer performance to meet within the size, weight and power requirements of the 6U system.

The CubeRRT digital back-end (RDB) is designed to digitize a 1 GHz bandwidth signal and perform advanced digital signal processing algorithms on an on-board FPGA for RFI mitigation. The RDB ADC is capable of ingesting the IF signal produced by the RFE from 1-2 GHz aliased region. The FPGA proceeds to produce a 128 frequency spectra of the incoming signal using a front-end polyphase filter-bank. The output of the 128 channel spectra then undergoes gain adjustment to account for the non-uniform pass-band shape of the RFE signal. The higher order statistical moments of the data per channel are calculated (2^nd and 4^th moment) as a pre-cursor to RFI detection and mitigation. The second moment is uncalibrated power of the signal. The first RFI detection algorithm applied is a simple threshold detection algorithm across the spectra to detect frequency outliers. The second RFI detection algorithm is more advanced and uses the fourth and second moments to calculate kurtosis of the signal as a test of normality. Any signal that deviates from normality is flagged as being corrupted by RFI. The flags of the two algorithms are combined and the power in each frequency bin is summed to produce mitigated and unmitigated accumulated power. CubeRRT will downlink all 128 channels to verify performance of mitigated and unmitigated uncalibrated power outputs. Most of the thresholds, coefficients, gain-adjustment values within the RDB are updatable from the user perspective.

The CubeRRT bus is being developed by Blue-Canyon Technologies. CubeRRT is expected to operate with a duty-cycle of 30%. CubeRRT will mostly be operational over landmasses where the occurrence of RFI is expected to be significantly higher. CubeRRT is designed for 12 months of commissioning and operations.

CubeRRT is slated for a February 2018 delivery and launch in March 2018. The following talk will concentrate on results of flight-model build of the payload, as well as on-board algorithm performance.