Academy of Scientific and Innovative Research (AcSIR), Ghaziabad, UP 201002, India; Cyber Physical System, CSIR-Central Electronics Engineering Research Institute (CEERI), Pilani 333031, India; DeepBlink LLC, 30 N Gould St Ste R, Sheridan, WY 82801, United States.
Department of CSIS, Birla Institute of Technology and Science Pilani, Pilani Campus, Rajasthan 333031, India.
Neural Netw. 2022 Feb;146:11-21. doi: 10.1016/j.neunet.2021.11.011. Epub 2021 Nov 16.
Human activity recognition (HAR) is an important task in many applications such as smart homes, sports analysis, healthcare services, etc. Popular modalities for human activity recognition involving computer vision and inertial sensors are in the literature for solving HAR, however, they face serious limitations with respect to different illumination, background, clutter, obtrusiveness, and other factors. In recent years, WiFi channel state information (CSI) based activity recognition is gaining momentum due to its many advantages including easy deployability, and cost-effectiveness. This work proposes CSITime, a modified InceptionTime network architecture, a generic architecture for CSI-based human activity recognition. We perceive CSI activity recognition as a multi-variate time series problem. The methodology of CSITime is threefold. First, we pre-process CSI signals followed by data augmentation using two label-mixing strategies - mixup and cutmix to enhance the neural network's learning. Second, in the basic block of CSITime, features from multiple convolutional kernels are concatenated and passed through a self-attention layer followed by a fully connected layer with Mish activation. CSITime network consists of six such blocks followed by a global average pooling layer and a final fully connected layer for the final classification. Third, in the training of the neural network, instead of adopting general training procedures such as early stopping, we use one-cycle policy and cosine annealing to monitor the learning rate. The proposed model has been tested on publicly available benchmark datasets, i.e., ARIL, StanWiFi, and SignFi datasets. The proposed CSITime has achieved accuracy of 98.20%, 98%, and 95.42% on ARIL, StanWiFi, and SignFi datasets, respectively, for WiFi-based activity recognition. This is an improvement on state-of-the-art accuracies by 3.3%, 0.67%, and 0.82% on ARIL, StanWiFi, and SignFi datasets, respectively. In lab-5 users' scenario of the SignFi dataset, which has the training and testing data from different distributions, our model achieved accuracy was 2.17% higher than state-of-the-art, which shows the comparative robustness of our model.
人体活动识别 (HAR) 是许多应用中的一项重要任务，例如智能家居、运动分析、医疗保健服务等。涉及计算机视觉和惯性传感器的人体活动识别的流行模式在文献中用于解决 HAR 问题，但是它们在不同的光照、背景、杂物、干扰以及其他因素方面存在严重的局限性。近年来，基于 WiFi 信道状态信息 (CSI) 的活动识别由于其易于部署和具有成本效益等诸多优势而受到关注。本工作提出了 CSITime，这是一种改进的 InceptionTime 网络架构，是一种通用的基于 CSI 的人体活动识别架构。我们将 CSI 活动识别视为一个多变量时间序列问题。CSITime 的方法学有三个方面。首先，我们对 CSI 信号进行预处理，然后使用两种标签混合策略——mixup 和 cutmix 进行数据增强，以增强神经网络的学习能力。其次，在 CSITime 的基本块中，来自多个卷积核的特征被串联，并通过一个自注意力层和一个带有 Mish 激活的全连接层进行传递。CSITime 网络由六个这样的块组成，然后是一个全局平均池化层和一个用于最终分类的全连接层。第三，在神经网络的训练中，我们没有采用早期停止等常规训练程序，而是使用一周期策略和余弦退火来监控学习率。所提出的模型已经在公开可用的基准数据集上进行了测试，即 ARIL、StanWiFi 和 SignFi 数据集。在所提出的模型中，基于 WiFi 的活动识别的 ARIL、StanWiFi 和 SignFi 数据集的准确率分别为 98.20%、98%和 95.42%。与 ARIL、StanWiFi 和 SignFi 数据集的现有最先进技术相比，分别提高了 3.3%、0.67%和 0.82%。在 SignFi 数据集的实验室 5 用户场景中，该场景的训练和测试数据来自不同的分布，我们的模型的准确率比现有最先进技术高出 2.17%，这表明了我们的模型的比较鲁棒性。
