AdaRadar：面向雷达感知的率自适应谱压缩
AdaRadar: Rate Adaptive Spectral Compression for Radar-based Perception
提出一种基于检测置信度代理梯度的自适应压缩框架，利用零阶梯度近似解决不可微操作（剪枝、量化）的梯度回传问题，可复用于视觉任务中需要动态调整压缩率或传输带宽的场景。
原文依据 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.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论雷达因其全天候特性以及测量距离和多普勒速度的能力，是自动驾驶系统中的关键感知模态。然而，高维原始雷达数据的庞大体量会饱和到计算引擎（例如NPU）的通信链路，该链路通常是低带宽接口，其数据速率仅能支持少量低分辨率距离-多普勒帧。目前明显缺乏一种用于利用高维雷达数据的通用编解码器，而现有的图像域方法并不适用，因为它们通常以固定压缩比运行，无法适应变化或对抗性条件。鉴于此，我们提出了带有自适应反馈的雷达数据压缩方法。它通过从检测置信度对压缩率的代理梯度进行梯度下降，来动态调整压缩比。我们采用零阶梯度近似，因为它使得即使核心操作（剪枝和量化）不可微也能计算梯度。这也避免了在带宽受限链路上传输梯度张量，因为如果估计梯度，其大小将与原始雷达数据相当。此外，我们发现雷达特征图高度集中在少数频率分量上。因此，我们对雷达数据立方体应用离散余弦变换，并有效地选择性剪除系数。我们通过缩放量化来保持每个雷达图像块的动态范围。结合这些技术，我们提出的在线自适应压缩方案在最小性能下降（1%p）的情况下实现了超过100倍的特征尺寸缩减。我们在RADIal、CARRADA和Radatron数据集上验证了我们的结果。
英文原文摘要 1734 字符
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.