A5039388363- Indoor and Outdoor Localization Technologies
- Speech and Audio Processing
- Underwater Vehicles and Communication Systems
- Energy Efficient Wireless Sensor Networks
- Radio Wave Propagation Studies
- Microwave Imaging and Scattering Analysis
- Wireless Signal Modulation Classification
University of Electronic Science and Technology of China
2018-2020
Sichuan Normal University
2014
Wireless fingerprinting localization method learns a mapping function from fingerprint measurement to the estimated location, which is more suitable for complex indoor environments than propagation model-based methods. However, most traditional methods only consider location matching at single time or space points, but ignore fact that correlations in sequences may improve accuracy and robustness of mobile positioning. Recently, recurrent neural network (RNN) has been readily applied many...
Wireless fingerprinting localization (FL) systems identify locations by building radio fingerprint maps, aiming to provide satisfactory location solutions for the complex environment. However, map is easy change, and cost of a new one high. One research focus transfer knowledge from old maps one. Feature-based learning methods help mapping source target common hidden domain, then minimize maximum mean difference (MMD) distance between empirical distributions in latent domain. In this paper,...
Indoor localization has received extensive attention in recent years. Due to the complexity of indoor environment, fingerprint positioning methods have better performance than based on propagation models. In WiFi network, signal strength (RSS) is usually used as feature. influence noise there error between information and online data, which results deviation results. We proposed a novel reduction feature calculates difference RSS at same location combines it with original then obtains...
A lot of studies have been done in Received Signal Strength Indication (RSSI) based indoor location for wireless sensor network, and most them assume that the Path-Loss Exponent (PLE) is a fixed value already known or can be measured by experiment. However, PLE usually not value. Analysis simulation indicate has much effect on distance estimation error under commonly used log-normal shadowing model. Based existing positioning algorithm, we proposed method dynamically correcting PLE, using...