STFNets: Learning Sensing Signals from the Time-Frequency Perspective with Short-Time Fourier Neural Networks

• Published in: arXiv.org 2019-02
• Main Authors: Yao, Shuochao, Piao, Ailing, Jiang, Wenjun, Zhao, Yiran, Shao, Huajie, Liu, Shengzhong, Liu, Dongxin, Li, Jinyang, Wang, Tianshi, Hu, Shaohan, Lu, Su, Han, Jiawei, Abdelzaher, Tarek
• Format: Article
• Language: English
• Subjects: Brain; Data processing; Fourier transforms; Frequency domain analysis; Internet of Things; Machine learning; Neural networks; Physics; Resonant frequencies; Signal processing; Speech recognition; State of the art; Time-frequency analysis
• ISSN: 2331-8422
• DOI: 10.48550/arxiv.1902.07849

Recent advances in deep learning motivate the use of deep neural networks in Internet-of-Things (IoT) applications. These networks are modelled after signal processing in the human brain, thereby leading to significant advantages at perceptual tasks such as vision and speech recognition. IoT applications, however, often measure physical phenomena, where the underlying physics (such as inertia, wireless signal propagation, or the natural frequency of oscillation) are fundamentally a function of signal frequencies, offering better features in the frequency domain. This observation leads to a fundamental question: For IoT applications, can one develop a new brand of neural network structures that synthesize features inspired not only by the biology of human perception but also by the fundamental nature of physics? Hence, in this paper, instead of using conventional building blocks (e.g., convolutional and recurrent layers), we propose a new foundational neural network building block, the Short-Time Fourier Neural Network (STFNet). It integrates a widely-used time-frequency analysis method, the Short-Time Fourier Transform, into data processing to learn features directly in the frequency domain, where the physics of underlying phenomena leave better foot-prints. STFNets bring additional flexibility to time-frequency analysis by offering novel nonlinear learnable operations that are spectral-compatible. Moreover, STFNets show that transforming signals to a domain that is more connected to the underlying physics greatly simplifies the learning process. We demonstrate the effectiveness of STFNets with extensive experiments. STFNets significantly outperform the state-of-the-art deep learning models in all experiments. A STFNet, therefore, demonstrates superior capability as the fundamental building block of deep neural networks for IoT applications for various sensor inputs.