Resource-efficient Deep Neural Networks for Automotive Radar Interference Mitigation

Abstract

Radar sensors are crucial for environment perception of driver assistance systems as well as autonomous vehicles. With a rising number of radar sensors and the so far unregulated automotive radar frequency band, mutual interference is inevitable and must be dealt with. Algorithms and models operating on radar data are required to run the early processing steps on specialized radar sensor hardware. This specialized hardware typically has strict resource-constraints, i.e. a low memory capacity and low computational power. Convolutional Neural Network (CNN)-based approaches for denoising and interference mitigation yield promising results for radar processing in terms of performance. Regarding resource-constraints, however, CNNs typically exceed the hardware's capacities by far. In this paper we investigate quantization techniques for CNN-based denoising and interference mitigation of radar signals. We analyze the quantization of (i) weights and (ii) activations of different CNN-based model architectures. This quantization results in reduced memory requirements for model storage and during inference. We compare models with fixed and learned bit-widths and contrast two different methodologies for training quantized CNNs, i.e. the straight-through gradient estimator and training distributions over discrete weights. We illustrate the importance of structurally small real-valued base models for quantization and show that learned bit-widths yield the smallest models. We achieve a memory reduction of around 80\% compared to the real-valued baseline. Due to practical reasons, however, we recommend the use of 8 bits for weights and activations, which results in models that require only 0.2 megabytes of memory.

arXiv comment15 pages; published in IEEE Journal of Selected Topics in Signal Processing, Special Issue on Recent Advances in Automotive Radar Signal Processing, Volume: 15, Issue: 4, June 2021. arXiv admin note: text overlap with arXiv:2011.12706

References in corpus (12)

• Adam: A Method for Stochastic Optimization
• Categorical Reparameterization with Gumbel-Softmax
• The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
• Binarized Neural Networks
• Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networks
• Variational Dropout and the Local Reparameterization Trick
• Learned Step Size Quantization
• Automotive Radar Interference Mitigation Using Adaptive Noise Canceller
• Trained Quantization Thresholds for Accurate and Efficient Fixed-Point Inference of Deep Neural Networks
• Deep Interference Mitigation and Denoising of Real-World FMCW Radar Signals
• Probabilistic Binary Neural Networks
• Resource-Efficient Neural Networks for Embedded Systems

Cited by in corpus (4)

• Radar-Camera Fusion for Object Detection and Semantic Segmentation in Autonomous Driving: A Comprehensive Review
• Interference Suppression Using Deep Learning: Current Approaches and Open Challenges
• Prior-Guided Deep Interference Mitigation for FMCW Radars
• Angle-Equivariant Convolutional Neural Networks for Interference Mitigation in Automotive Radar