A5089289729- Indoor and Outdoor Localization Technologies
- Underwater Vehicles and Communication Systems
- Speech and Audio Processing
- Energy Efficient Wireless Sensor Networks
- Advanced Image and Video Retrieval Techniques
- Robotics and Sensor-Based Localization
This paper proposes a semi-sequential probabilistic model (SSP) that applies an additional short term memory to enhance the performance of indoor localization. The conventional methods normally treat locations in database indiscriminately. In contrast, SSP leverages information previous position determine probable location since user's speed environment is bounded and near one have higher probability than other locations. Although utilizes information, it does not require exact moving...
Convolutional neural network (CNN) is a powerful tool for many data applications. However, its high dimension nature, large size and computational complexity, the need of amount training make it challenging to be used in edge computing applications, which are becoming increasingly popular, relevant important. In this paper, we propose descriptor based approach accelerate convolutional training, reduce input size, greatly facilitates use CNN computating even cloud computing. By using image...
This paper proposes passive WiFi indoor localization. Instead of using signals received by mobile devices as fingerprints, we use routers to locate the carrier. Consequently, software installation on device is not required. To resolve data insufficiency problem, flow control such request send (RTS) and clear (CTS) are utilized. In our model, signal strength indicator (RSSI) channel state information (CSI) used fingerprints for several algorithms, including deterministic, probabilistic neural...