# One Size Does Not Fit All: Multi-Scale, Cascaded RNNs for Radar Classification

arXiv: https://arxiv.org/abs/1909.03082 · HTML: https://mlsanity.com/paper/1909.03082/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Sep 2019.

## Built on by (most cited first, 5 of 5)

- [Matryoshka Representation Learning](https://mlsanity.com/paper/2205.13147/) (2022)
- [Multi-Scale Adaptive Graph Neural Network for Multivariate Time Series Forecasting](https://mlsanity.com/paper/2201.04828/) (2022)
- [MS-RNN: A Flexible Multi-Scale Framework for Spatiotemporal Predictive Learning](https://mlsanity.com/paper/2206.03010/) (2022)
- [What Physics do Data-Driven MoCap-to-Radar Models Learn?](https://mlsanity.com/paper/2605.00018/) (2026)
- [CMIS-Net: A Cascaded Multi-Scale Individual Standardization Network for Backchannel Agreement Estimation](https://mlsanity.com/paper/2510.17855/) (2025)

## Abstract (arXiv)

Edge sensing with micro-power pulse-Doppler radars is an emergent domain in monitoring and surveillance with several smart city applications. Existing solutions for the clutter versus multi-source radar classification task are limited in terms of either accuracy or efficiency, and in some cases, struggle with a trade-off between false alarms and recall of sources. We find that this problem can be resolved by learning the classifier across multiple time-scales. We propose a multi-scale, cascaded recurrent neural network architecture, MSC-RNN, comprised of an efficient multi-instance learning (MIL) Recurrent Neural Network (RNN) for clutter discrimination at a lower tier, and a more complex RNN classifier for source classification at the upper tier. By controlling the invocation of the upper RNN with the help of the lower tier conditionally, MSC-RNN achieves an overall accuracy of 0.972. Our approach holistically improves the accuracy and per-class recalls over ML models suitable for radar inferencing. Notably, we outperform cross-domain handcrafted feature engineering with time-domain deep feature learning, while also being up to ~3× more efficient than a competitive solution.