Wireless systems are an integral part of aviation. Apart from their apparent use in air-to-ground communication, wireless systems play a crucial role in avionic functions including navigation and landing. An interference-free wireless environment is therefore critical for the uninterrupted operation and safety of an aircraft. Hence, there is an urgency for airport facilities to acquire the capability to continuously monitor aviation frequency bands for real-time detection of interference and anomalies. To meet this critical need, we design and build AviSense, an SDR-based real-time , versatile system for monitoring aviation bands. AviSense detects and characterizes signal activities to enable practical and effective anomaly detection. We identify and tackle the challenges posed by a diverse set of critical aviation bands and technologies. We evaluate our methodology with real-world aviation signal measurements and two custom datasets of anomalous signals. We find that our signal classification capability achieves a true positive rate of ∼99%, with few exceptions, and a false positive rate of less than 4%. We also demonstrate that AviSense can effectively distinguish between different types of anomalies. We build and evaluate a prototype implementation of AviSense that supports distributed monitoring.

We present DeepRadar, a novel deep-learning-based environmental sensing capability system for detecting radar signals and estimating their spectral occupancy. DeepRadar makes decisions in real-time and maintains continuous operability by adapting its computations based on the available computing resources. We thoroughly evaluate DeepRadar using a variety of test data at different signal-to-interference ratio (SIR) levels. Our evaluation results show that at 20 dB peak-to-average SIR, per MHz, DeepRadar detects radar signals with 99% accuracy and misses only less than 2 MHz, on average, while estimating their spectral occupancy. Our implementation of DeepRadar using a commercial-off-the-shelf software-defined radio also achieves a similarly high detection accuracy.

We make a case for using the emerging blockchain technology for spectrum sharing in Citizen's Broadband Radio Service (CBRS). We show that blockchain can dynamically allocate channels in a transparent and fair manner by achieving distributed channel de-confliction. Furthermore, blockchain helps automate communication between the Spectrum Access Systems (SASs) belonging to competing, non-trusting organizations by providing consensus, while at the same time preserving confidentiality of sensitive information. We implement our CBRS blockchain architecture using the Hyperledger Fabric platform. In order to support high channel request rates, we also develop a novel feature that adaptively pre-processes channel allocation requests at the SAS before sending those to the blockchain network. We evaluate our blockchain solution under different settings and show that it meets the FCC latency requirements. We also analyze the scalability of our system and show that the latency and throughput guarantees can be maintained in scenarios involving multiple organizations.

We present LLOCUS, a novel learning-based system that uses mobile crowdsourced RF sensing to estimate the location and power of unknown mobile transmitters in real time, while allowing unrestricted mobility of the crowdsourcing participants. We carefully identify and tackle several challenges in learning and localizing, based on RSS, in such a dynamic environment. We decouple the problem of localizing a transmitter with unknown transmit power into two problems, 1) predicting the power of a transmitter at an unknown location, and 2) localizing a transmitter with known transmit power. LLOCUS first estimates the power of the unknown transmitter and then scales the reported RSS values such that the unknown transmit power problem is transparent to the method of localization. We evaluate LLOCUS using three experiments in different indoor and outdoor environments. We find that LLOCUS reduces the localization error by 17-68% compared to several non-learning methods.

With the tremendous increase in the use of wireless devices, understanding the surrounding wireless/RF environment is becoming essential for many application areas. In this work, we develop the technical building blocks needed for a spectrum monitoring system that can incrementally learn about the signals present in a deployed environment. We achieve incremental learning (IL) by identifying and grouping the new/unknown signals and, automatically building new machine learning (ML) models for detecting them. A thorough evaluation of our approach demonstrates its adaptability and high accuracy with signal data from several over-the-air scenarios.