AI-based structured interference rejection has grown more popular because deep learning approaches can outperform traditional methods by jointly considering the signal of interest (SOI) and the signal mixture (SOI plus interference). This work builds on a previous AI-enabled approach utilizing autoregressive transformer-based models by adding a Finite Scalar Quantization (FSQ) tokenizer layer which aims to improve the interference rejection performance while keeping overall latency to a minimum. Additionally, we experiment with other inference optimization techniques with the goal of speeding up inference without much accuracy loss. We explore this space with an experiment where the SOI is a digitally modulated radio frequency (RF) signal and the structured interference is a digital television signal, an extremely common type of Orthogonal Frequency-Division Multiplexing (OFDM) transmission. Our results achieve low latency and increased interference rejection over traditional techniques and prior work with other AI-enabled methods. We demonstrate the benefits of the AI-enabled approaches via audio metrics such as Perceptual Evaluation of Speech Quality (PESQ). Additionally, we explore a variety of applications and detail how our interference rejection algorithm may be used in operationally-relevant scenarios.
Figures & tables
Fig. 1: A real-time resilient communications use case. The blue gear box represents the AI model doing the processing on the receive end of the communications link.
Fig. 2: Adapted from [ 1 ] . Architecture diagram for the RF Finite Scalar Quantization (FSQ) tokenizer.
Fig. 3: Adapted from [ 1 , 3 ] . Architecture diagram for the RF Transformer using the RF FSQ tokenizer.
Fig. 4: Adapted from [ 3 ] . Data preparation process for the training pipeline.
Fig. 5: Audio quality evaluation for digital radio experiment. Metrics include Perceptual Evaluation of Speech Quality (PESQ), Signal to Distortion ratio (SDR), Log-Spectral Distance (LSD), and Mel-Cepstral Distance (Mel-CD).
Fig. 6: Audio intelligibility evaluation for digital radio experiment. Metrics include Short-Time Objective Intelligibility (STOI) and extended STOI (eSTOI).
Fig. 7: Overall latencies for the RF WaveNet, RF Transformer Decoder, and RF Transformer + Tokenizer models on the NVIDIA Jetson AGX Orin (left column) and NVIDIA H100 NVL (right column) evaluated for FP32. Note that all three models have good or fair latency for real-time streaming applications on the Jetson.
Fig. 8: Output throughput for the RF WaveNet, RF Transformer Decoder, and RF Transformer + Tokenizer models on the NVIDIA Jetson AGX Orin (left column) and NVIDIA H100 NVL (right column) evaluated for FP32. Note that all three models are currently unacceptable for real-time streaming applications on the Jetson, but the Transformer + Tokenizer had very poor throughput in particular thus no points are shown in the range of the plot axes for Jetson.
Fig. 9: Optimized latency and output throughput for the RF Transformer Decoder NVIDIA Jetson AGX Orin using the Torch-TensorRT library at FP16 precision.
Fig. 10: Spectrum sharing use case between radar and a structured interference source.
AI-enhanced interference rejection in radio frequency (RF) transmissions has recently attracted interest because deep learning approaches trained on both the signal of interest (SOI) and the signal mixture (SOI plus interference) can outperform traditional approaches which only consider the SOI. The goal is to detect, demodulate, and decode signals over a range of signal-to-interference-plus-noise (SINR) levels without having a detailed, design-level knowledge of the interfering signal or the propagation conditions. Our present AI interference suppression results are based on Autoregressive Transformer Decoder models which exhibit orders of magnitude faster throughput at inference time than WaveNet models developed in earlier work. As a specific example, we investigate an analog FM "Walkie Talkie" radio signal of interest in the presence of an Orthogonal Frequency-Division Multiplexing (OFDM) interferer. This type of interferer is near-ubiquitous in the current RF landscape. Our results clearly show the benefits of transformer-based interference mitigation in tactical settings. We show that unintelligible transmissions become intelligible via metrics such as Perceptual Evaluation of Speech Quality (PESQ), while overall latency is kept to a minimum using readily available lightweight GPUs such as a Jetson AGX Orin. We believe these same techniques can also be applied to a broader set of national security scenarios, as well as having commercial applications.
We study a problem of signal separation: estimating a signal of interest (SOI) contaminated by an unknown non-Gaussian background/interference. Given the training data consisting of examples of SOI and interference, we show how to build a fully data-driven signal separator. To that end we learn a good discrete tokenizer for SOI and then train an end-to-end transformer on a cross-entropy loss. Training with a cross-entropy shows substantial improvements over the conventional mean-squared error (MSE). Our tokenizer is a modification of Google's SoundStream, which incorporates additional transformer layers and switches from VQVAE to finite-scalar quantization (FSQ). Across real and synthetic mixtures from the MIT RF Challenge dataset, our method achieves competitive performance, including a 122x reduction in bit-error rate (BER) over prior state-of-the-art techniques for separating a QPSK signal from 5G interference. The learned representation adapts to the interference type without side information and shows zero-shot generalization to unseen mixtures at inference time, underscoring its potential beyond RF. Although we instantiate our approach on radio-frequency mixtures, we expect the same architecture to apply to gravitational-wave data (e.g., LIGO strain) and other scientific sensing problems that require data-driven modeling of background and noise.
Reliable altitude estimation with frequency-modulated continuous wave (FMCW) radar altimeters is increasingly a challenge due to in-band interference from modern communication systems. In this paper, we present a temporal convolutional autoencoder (TCAE) that directly processes in-phase and quadrature (IQ) samples to suppress structured interference while preserving signal phase and frequency content for range estimation. The model is trained and initially evaluated within a full radar altimeter simulation chain, then further validated via over-the-air (OTA) experiments using a universal software radio peripheral (USRP)-based testbed. Results show that the TCAE reduces altitude estimation error by more than 85% compared to least mean squares (LMS) adaptive filtering under severe interference conditions, including low signal-to-interference-plus-noise ratio (SINR) and full temporal overlap between interfering and radar signals. Unlike conventional methods, the TCAE maintains phase fidelity and beat structure, enabling accurate range estimation even when interferers occupy more than one-quarter of the radar bandwidth. The implemented TCAE performs mitigation directly on fixed-length IQ windows using a single feed-forward pass and was integrated into the MATLAB/ONNX-based evaluation chain used for both simulation and OTA testing. These findings demonstrate that learned IQ-domain interference mitigation can enhance radar-altimeter resilience under a range of tested interference conditions.
Charles E. Thornton, Jamie Sloop, Samuel Brown +3
Virginia Tech National Security Institute, Blacksburg, VA, USA · The Boeing Company, Tukwila, WA, USA