Passive remote sensing plays a pivotal role in climate studies and Earth science. NASA's Soil Moisture Active Passive (SMAP) observatory employs passive microwave radiometry to gauge soil moisture levels and identify freezing or thawing conditions in a given region. Despite operating within the protected radio spectrum band (1400-1427 MHz), the radiometer's measurements are susceptible to contamination from Radio Frequency Interference (RFI). The proliferation of active wireless technologies, including radar signals for air surveillance, 5G wireless communication, and unmanned aerial vehicles, has contributed to RFI through out-of-band emissions and unlawful in-band operations. Detecting RFI on a global scale using physical modeling presents substantial challenges, as RFI can originate from single or multiple sources and manifest as either pulsed or continuous wave signals. In this study, we propose a novel approach for RFI detection based on Deep Learning (DL), employing a Convolutional Neural Network (CNN) framework capable of identifying various types of RFI globally. This data-driven methodology enables the detection framework to learn directly from SMAP data products, making informed decisions regarding the presence of RFI contamination in specific footprints. We leverage SMAP's Level 1A data products, which contain antenna counts of various raw moments along with Stokes parameters, to generate spectrograms. Additionally, Level 1B data products containing quality flags are used to dynamically label these spectrograms. The robust DL framework developed in this study demonstrates exceptional accuracy, achieving a detection rate of 99.99% for RFI using the raw moments of horizontal polarization. More details can be found in our journal article - [Link] 

• IGARSS 2022 Student Paper Finalist [Link] 

SOFTWARE RADIO TESTBED FOR 5G AND L BAND RADIOMETER COEXISTENCE RESEARCH

Passive remote sensing through microwave radiometry plays a crucial role in Earth observation, allowing for the estimation of various geophysical parameters. The radiometer's exceptional sensitivity, characterized by a low noise floor, enables the sampling of geophysical emissions within a dedicated and protected band specifically allocated for remote sensing. This protected L-band, spanning from 1400 to 1427 MHz, is particularly appealing for scientific applications due to its reduced atmospheric attenuation. Furthermore, its attributes make it an attractive candidate for next-generation (xG) wireless communication technologies. 5G cellular systems are designed to operate within two distinct frequency ranges: FR1 (0.45 GHz–6 GHz) and FR2 (24.45 GHz–52.6 GHz). While regulations prohibit the conduct of up-link or down-link operations within the protected portion of the L-band, the potential for out-of-band (OOB) emissions still exists, posing a significant concern for passive sensors given their stringent sensitivity requirements in scientific applications. This study introduces a novel physical testbed equipped with the capability to observe emissions both within the protected L-band and in out-of-band regions. This testbed is situated within an anechoic chamber, ensuring controlled and interference-free conditions. Notably, this setup offers flexibility in terms of transmitted waveforms and enables the analysis of raw measurements (IQ samples) from radiometers. These capabilities are instrumental in designing effective onboard Radio Frequency Interference (RFI) processing techniques, facilitating the coexistence of communication and passive sensing technologies while safeguarding the integrity of scientific observations.

This study presents a comprehensive testbed designed to investigate the coexistence of 5G wireless communication and passive microwave radiometry in the L-band (1400-1427 MHz) spectrum. The testbed enables controlled experiments to quantify the impact of 5G transmissions on radiometric measurements. The testbed encompasses both the active transmission side, simulating 5G wireless signals, and the passive sensing side, represented by a radiometer. The active side involves resource block allocation, transmission chain setup, and radiometric transmission configurations.  The radiometer on the passive side receives emissions and converts them from antenna counts to brightness temperature. Calibration processes, both internal and external, ensure data accuracy.  Experiments explore various scenarios, including in-band and out-of-band transmissions, different power levels, duty cycles, sparsity levels, and waveform types. Spectrograms and calibrated brightness temperature data are collected and pre-processed for analysis. Results showcase the transmitted and received signals, with spectrum analyzer and radiometer data illustrating the impact of resource block allocation and sparsity levels. The study quantifies the effects of 5G on L-band radiometry, differentiating between in-band (illegal transmissions) and out-of-band (receiver interference) scenarios, considering power levels, duty cycles, and waveform types.

Passive microwave remote sensing is a vital tool for acquiring valuable information regarding the Earth's surface, with significant applications in agriculture, water management, forestry, and various environmental disciplines. Precision agricultural (PA) practices necessitate the availability of field-scale, high-resolution remote sensing data products. This study focuses on the design and development of a cost-effective, portable L-band microwave radiometer capable of operating from an unmanned aircraft system (UAS) platform to measure high-resolution surface brightness temperature (TB). This radiometer consists of a dual-polarized (Horizontal polarized, H-pol and Vertical polarized, V-pol) antenna and a software-defined radio (SDR)-based receiver system with a 30 MHz sampling rate. The post-processing methodology encompasses the conversion of raw in-phase and quadratic (I&Q)surface emissions to radiation TB through internal and external calibrations. Radiometric measurements were conducted over an experimental site covering both bare soil within an agricultural field and a large water body. The results yielded a high-resolution TB map that effectively delineated the boundaries between land and water, and identified land surface features. The radiometric temperature measurements of the sky and blackbody demonstrated a standard deviation of 0.95 K for H-pol and 0.57 K for V-pol in the case of the sky and 0.39 K for both H-pol and V-pol in the case of the blackbody observations. The utilization of I&Q samples acquired via the radiometer digital back-end facilitates the generation of different time-frequency (TF) analyses through short-time Fourier transform (STFT) and power spectral density (PSD). The transformation of radiometer samples into TF representations aids in the identification and mitigation of radio frequency interference (RFI) originating from the instrument itself and external sources.

The National Aeronautics and Space Administration's (NASA) Soil Moisture Active Passive (SMAP) is the latest passive remote sensing satellite operating in the protected L-band spectrum from 1.400 to 1.427 GHz. SMAP provides global-scale soil moisture images with point-wise passive scanning of the earth's thermal radiations. SMAP takes multiple samples in frequency and time from each antenna footprint to increase the likelihood of capturing RFI-free samples. SMAP's current RFI detection and mitigation algorithm excludes samples detected to be RFI-contaminated and averages the remaining samples. But this approach can be less effective for harsh RFI environments, where  RFI contamination is present in all or a large number of samples. Instead of rigidly excluding samples and averaging the remaining samples, we investigate a weighted sum of samples approach where the sample weights are determined according to the RFI's mean and covariance matrix.

There is insufficient wireless frequency spectrum to support the continued growth of active wireless technologies and devices. This has provoked extensive research on spectrum coexistence. One case that has gained limited attention in this course is using currently banned frequency bands for active wireless communications. One such option is the 27 MHz-wide narrowband portion of the L-band from 1.400 to 1.427 GHz, which is exclusively devoted to space-borne passive radiometry for remote sensing and radio astronomy. Radio regulations currently prohibit active wireless communications and radars from operating in this band to avoid radio frequency interference (RFI) on highly noise-sensitive passive radiometry equipment. The National Aeronautics and Space Administration's (NASA's) Soil Moisture Active Passive (SMAP) satellite is one of the latest space-borne remote sensing missions that evaluates global soil moisture by passive scanning of the thermal emissions of the earth in this frequency band. In this paper, we investigate the opportunistic temporal use of this 27 MHz-wide passive radiometry band for active wireless transmissions when there is no Line of Sight (LoS) between SMAP and a terrestrial wireless network. We use MATLAB simulations to determine the fraction of time that SMAP has LoS (and non-LoS) with a terrestrial wireless cell at different Earth latitudes based on SMAP's orbital characteristics. We also investigate the severity of RFI induced on SMAP in the presence of a terrestrial cluster of 5G cells with LoS.

Autoencoder-based Radio Frequency Interference Mitigation for SMAP Passive Radiometer

Ali Owfi, Fatemeh Afghah

Passive space-borne radiometers operating in the 1400-1427 MHz protected frequency band face radio frequency interference (RFI) from terrestrial sources. With the growth of wireless devices and the appearance of new technologies, the possibility of sharing this spectrum with other technologies would introduce more RFI to these radiometers. This band could be an ideal mid-band frequency for 5G and Beyond, as it offers high capacity and good coverage. Current RFI detection and mitigation techniques at SMAP (Soil Moisture Active Passive) depend on correctly detecting and discarding or filtering the contaminated data leading to the loss of valuable information, especially in severe RFI cases. In this project, we proposed an autoencoder-based RFI mitigation method to remove the dominant RFI caused by potential coexistent terrestrial users (i.e., 5G base station) from the received contaminated signal at the passive receiver side, potentially preserving valuable information and preventing the contaminated data from being discarded .

 Meta-Learning for Wireless Interference Identification

Ali Owfi, Fatemeh Afghah, Jonathan Ashdown

Deep learning-based (DL-based) models have shown to be powerful tools for wireless interference identification (WII). However, one of the key concerns toward using these models in practical systems is that they perform poorly when they are encountered with signals coming from new sources not previously observed during the training phase. In a real-world communication system, the interference identifier will frequently face new unknown signals due to the existence of many wireless transmitters. This renders the conventional DL-based models impractical as a WII tool unless they go through a new training phase. Retraining the model is not only inefficient, but it can also be not feasible in some cases (e.g., at end-user devices) as the training phase consumes time and resources and requires large amounts of data. In this project, we presented a new approach for data-driven WII systems using meta-learning to address the lack of adaptability in conventional DL-based models to new (not previously seen) signals. We demonstrated that by using meta-learning, we are able to identify signals coming from not previously observed technologies and frequencies using just a handful of new samples, a task that is not generally possible with conventional DL models. Finally, we analyzed and compared the performance of the presented meta-learning model in multiple different settings using raw I/Q samples and Fast Fourier Transform of I/Q samples. Based on our experiments, we showed that the proposed meta-learning scheme outperforms the conventional DL models for WII when there are just a few samples available for training.