AdaRadar: Rate Adaptive Spectral Compression for Radar-based Perception

2026-03-18

AI Summary

This work addresses the challenge of transmitting high-dimensional raw radar data over low-bandwidth links, where conventional fixed compression methods fail to adapt to dynamic or adversarial environments. The authors propose an online adaptive compression framework that, for the first time, employs a zeroth-order gradient mechanism driven by detection confidence feedback to selectively prune and scale-quantize radar data cubes in the frequency domain, thereby avoiding the transmission of high-dimensional gradient tensors. By integrating discrete cosine transform, frequency coefficient pruning, and an online rate-control strategy, the method exploits the strong spectral concentration of radar features. Evaluated on the RADIal, CARRADA, and Radatron datasets, the approach achieves over 100× compression with only approximately 1 percentage point degradation in task performance.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Data Mining & Knowledge Management: Data Compression
• Computer Vision: Learning & Optimization for CV

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Radar is a critical perception modality in autonomous driving systems due to its all-weather characteristics and ability to measure range and Doppler velocity. However, the sheer volume of high-dimensional raw radar data saturates the communication link to the computing engine (e.g., an NPU), which is often a low-bandwidth interface with data rate provisioned only for a few low-resolution range-Doppler frames. A generalized codec for utilizing high-dimensional radar data is notably absent, while existing image-domain approaches are unsuitable, as they typically operate at fixed compression ratios and fail to adapt to varying or adversarial conditions. In light of this, we propose radar data compression with adaptive feedback. It dynamically adjusts the compression ratio by performing gradient descent from the proxy gradient of detection confidence with respect to the compression rate. We employ a zeroth-order gradient approximation as it enables gradient computation even with non-differentiable core operations--pruning and quantization. This also avoids transmitting the gradient tensors over the band-limited link, which, if estimated, would be as large as the original radar data. In addition, we have found that radar feature maps are heavily concentrated on a few frequency components. Thus, we apply the discrete cosine transform to the radar data cubes and selectively prune out the coefficients effectively. We preserve the dynamic range of each radar patch through scaled quantization. Combining those techniques, our proposed online adaptive compression scheme achieves over 100x feature size reduction at minimal performance drop (~1%p). We validate our results on the RADIal, CARRADA, and Radatron datasets.

Problem

• radar data compression
• rate adaptation
• autonomous driving
• bandwidth limitation
• high-dimensional data

Innovation

• adaptive compression
• zeroth-order optimization
• discrete cosine transform
• radar perception
• rate adaptation