A Novel Device-Free Counting Method Based on Channel Status Information

5G-enabled contactless multi-user presence and activity detection for independent assisted living

Wireless sensing is the state-of-the-art technique for next generation health activity monitoring. Smart homes and healthcare centres have a demand for multi-subject health activity monitoring to cater for future requirements. 5G-sensing coupled with deep learning models has enabled smart health monitoring systems, which have the potential to classify multiple activities based on variations in channel state information (CSI) of wireless signals. Proposed is the first 5G-enabled system operating at 3.75 GHz for multi-subject, in-home health activity monitoring, to the best of the authors' knowledge. Classified are activities of daily life performed by up to 4 subjects, in 16 categories. The proposed system combines subject count and activities performed in different classes together, resulting in simultaneous identification of occupancy count and activities performed. The CSI amplitudes obtained from 51 subcarriers of the wireless signal are processed and combined to capture variations due to simultaneous multi-subject movements. A deep learning convolutional neural network is engineered and trained on the CSI data to differentiate multi-subject activities. The proposed system provides a high average accuracy of 91.25% for single subject movements and an overall high multi-class accuracy of 83% for 4 subjects and 16 classification categories. The proposed system can potentially fulfill the needs of future in-home health activity monitoring and is a viable alternative for monitoring public health and well being.

WiGId: Indoor Group Identification with CSI-Based Random Forest

Human identity recognition has a wide range of application scenarios and a large number of application requirements. In recent years, the technology of collecting human biometrics through sensors for identification has become mature, but this kind of method needs additional equipment as assistance, which cannot be well applied to some scenarios. Using Wi-Fi for identity recognition has many advantages, such as no additional equipment as assistance, not affected by temperature, humidity, weather, light, and so on, so it has become a hot topic of research. The methods of individual identity recognition have been more mature; for example, gait information can be extracted as features. However, it is difficult to identify small-scale (2–5) group personnel at one time, and the tasks of fingerprint storage and classification are complex. In order to solve this problem, this paper proposed a method of using the random forest as a fingerprint database classifier. The method is divided into two stages: the offline stage trains the random forest classifier through the collected training data set. In the online phase, the real-time data collected are input into the classifier to get the results. When extracting channel state information (CSI) features, multiple people are regarded as a whole to reduce the difficulty of feature selection. The use of random forest classifier in classification can give full play to the advantages of random forest, which can deal with a large number of multi-dimensional data and is easy to generalize. Experiments showed that WiGId has good recognition performance in both LOS (line of sight) and N LOS (None line of sight) environments.