Human activity recognition (HAR) is a broad research area. While there exist solutions based on sensors and vision-based technologies, these solutions suffer from considerable limitations. Thus in order to mitigate or avoid these limitations, device free solutions based on radio signals like (home) WiFi, in particular 802.11 are considered. Recently, channel state information (CSI), available in WiFi 802.11n networks have been proposed for fine-grained analysis. We are able to detect human activities like Walk, Sit, Stand, Run (in the sequel, any human activity used for classification is capitalised, i.e. is denoted by its corresponding label. For example, “standing“ is denoted as Stand, the activity “sitting“ is denoted by Sit and so on), etc. in a line-of-sight (LOS) scenario and a non-line-of-sight (N-LOS) scenario within an indoor environment. We propose two algorithms—one using a support vector machine (SVM) for classification and another one using a long short-term memory (LSTM) recurrent neural network. While the former uses sophisticated pre-processing and feature extraction techniques based on wavelet analysis, the latter processes the raw data directly (after denoising). We show that it is possible to characterize activities and/or human body presence with high accuracy and we compare both approaches with regard to accuracy and performance. Furthermore, we extend the experimental setup to detect human falls, too which is a relevant use-case in the context of ambient assisted living (AAL) and show that with the developed algorithms it is possible to detect falls with high accuracy. In addition, we also show that the algorithms can be used to count the number of people in a room based on the CSI-data, which is a first step towards detecting more complex social behavior and activities. Our paper is an extended version of the paper (Damodaran and Schäfer, Device free human activity recognition using wifi channel state information, in: 16th IEEE International Conference on Ubiquitous Intelligence and Computing (UIC 2019), 5th IEEE Smart World Congress, Leicester, vol 16, IEEE, 2019).

Using Wi-Fi IEEE 802.11 standard, radio frequency waves are mainly used for communication on various devices such as mobile phones, laptops, and smart televisions. Apart from communication applications, the recent research in wireless technology has turned Wi-Fi into other exploration possibilities such as human activity recognition (HAR). HAR is a field of study that aims to predict motion and movement made by a person or even several people. There are numerous possibilities to use the Wi-Fi-based HAR solution for human-centric applications in intelligent surveillance, such as human fall detection in the health care sector or for elderly people nursing homes, smart homes for temperature control, a light control application, and motion detection applications. This paper’s focal point is to classify human activities such as EMPTY, LYING, SIT, SIT-DOWN, STAND, STAND-UP, WALK, and FALL with deep neural networks, such as long-term short memory (LSTM) and support vector machines (SVM). Special care was taken to address practical issues such as using available commodity hardware. Therefore, the open-source tool Nexmon was used for the channel state information (CSI) extraction on inexpensive hardware (Raspberry Pi 3B+, Pi 4B, and Asus RT-AC86U routers). We conducted three different types of experiments using different algorithms, which all demonstrated a similar accuracy in prediction for HAR with an accuracy between 97% and 99.7% (Raspberry Pi) and 96.2% and 100% (Asus RT-AC86U), for the best models, which is superior to previously published results. We also provide the acquired datasets and disclose details about the experimental setups.

Human Activity Recognition (HAR) is a rather broad research area. While there exist solutions based on sensors and vision-based technologies, these solutions suffer from considerable limitations. Thus in-order to mitigate or avoid these limitations, device free solutions based on radio signals like home WiFi are considered. Recently, channel state information (CSI), available in WiFi networks have been proposed for fine-grained analysis. We are able to detect the human activities like Walk, Stand, Sit, Run, etc. in a Line of Sight scenario (LOS) and a Non Line of Sight (N-LOS) scenario within an indoor environment. We propose two algorithms - one using a support vector machine (SVM) for classification and another one using a long short-term memory (LSTM) recurrent neural network. While the former uses sophisticated pre-processing and feature extraction techniques the latter processes the raw data directly (after denoising with wavelets). We show that it is possible to characterize activities and / or human body presence with high accuracy and we compare both approaches with regards to accuracy and performance.

Fall is a prominent issue due to its severe consequences both physically and mentally. Fall detection and prevention is a critical area of research because it can help elderly people to depend less on caregivers and allow them to live and move more independently. Using electrocardiograms (ECG) signals independently for fall detection and activity classification is a novel approach used in this paper. An algorithm has been proposed which uses pre-trained convolutional neural networks AlexNet and GoogLeNet as a classifier between the fall and no fall scenarios using electrocardiogram signals. The ECGs for both falling and no falling cases were obtained as part of the study using eight volunteers. The signals are pre-processed using an elliptical filter for signal noises such as baseline wander and power-line interface. As feature extractors, frequency-time representations (scalograms) were obtained by applying a continuous wavelet transform on the filtered ECG signals. These scalograms were used as inputs to the neural network and a significant validation accuracy of 98.08% was achieved in the first model. The trained model is able to distinguish ECGs with a fall activity from an ECG with a no fall activity with an accuracy of 98.02%. For the verification of the robustness of the proposed algorithm, our experimental dataset was augmented by adding two different publicly available datasets to it. The second model can classify fall, daily activities and no activities with an accuracy of 98.44%. These models were developed by transfer learning from the domain of real images to the medical images. In comparison to traditional deep learning approaches, the transfer learning not only avoids “reinventing the wheel,” but also presents a lightweight solution to otherwise computationally heavy problems.

Many recent studies have focused on the automatic classification of electrocardiogram (ECG) signals using deep learning (DL) methods. Most rely on existing complex DL methods, such as transfer learning or providing the models with carefully designed extracted features based on domain knowledge. A common assumption is that the deeper and more complex the DL model is, the better it learns. In this study, we propose two different DL models for automatic feature extraction from ECG signals for classification tasks: A CNN-LSTM hybrid model and an attention/transformer-based model with wavelet transform for the dimensional embedding. Both of the models extract the features from time series at the initial layers of the neural networks and can obtain performance at least equal to, if not greater than, many contemporary deep neural networks. To validate our hypothesis, we used three publicly available data-sets to evaluate the proposed models. Our model achieved a benchmark accuracy of 99.92% for fall detection and 99.93% for the PTB database for myocardial infarction versus normal heartbeat classification.