WiFi Channel State Information (CSI) represents the characteristics of wireless channels in wireless networks. WiFi CSI plays a pivotal role in wireless communications, primarily due to the variability of channel characteristics across time and space. By leveraging data analysis techniques based on Deep Neural Networks, we can capture the variations in channel characteristics associated with people's movements and behaviors indoors using WiFi CSI data. This offers a novel approach to Human Behavior Recognition, providing an alternative to camera-based methods that potentially infringe on privacy. In this paper, we delve into the structural design analysis of deep neural networks for human behavior recognition using WiFi CSI data and explore the training strategies vital for delivering extended services.

Symptoms of laryngeal cancer manifest primarily through voice changes, and its diagnosis relies solely on laryngoscopy examinations, lacking objective indicators of voice alterations. Recent advances in deep learning have opened possibilities for vocal-based laryngeal cancer screening. However, the practical medical application remains constrained due to relatively low accuracy. In this paper, we propose a method that combines patient information and voice analysis with a CNN model to address this issue. Experiments demonstrate a 8% improvement in accuracy when additional information is embedded alongside voice signals, compared to using voice data alone in deep learning models. This approach holds promise for more effective laryngeal cancer screening and diagnosis.

Geospatial research interests in emergency responses in micro-scale environments have been increased after 9/11. The application requires to define optimal paths in emergency situation, to support evacuees and rescuers. The optimal path defined in this study is used to guide rescuers. So, the path is from entrances to the disaster site (room), not from rooms to entrances in the building. In this study, we propose a time-dependent optimal routing algorithm to develop real-time evacuation systems. The network data that represents navigable spaces in building is used for routing the optimal path. Associated information about environments (for example, number of evacuees or rescuers, capacity of hallways and rooms, type of rooms and so on) is assigned to nodes and edges in the network. The time-dependent optimal path is defined after concerning temporally changed environmental information and the positions of evacuees and rescuer at each time slot for avoiding places jammed with evacuees. To detect the positions of human beings in a building per time period, we use the results of evacuation simulation system to identify the movement patterns of human beings in the emergency situation. We use the simulation data of five or ten seconds time interval, to determine the optimal route for rescuers.