Luo Haiyong

RSSI based Bluetooth low energy indoor positioning

The presentation of Bluetooth Low Energy (BLE; e.g., Bluetooth 4.0) makes Bluetooth based indoor positioning have extremely broad application prospects. In this paper, we propose a received signal strength indication (RSSI) based Bluetooth positioning method. There are two phases in the procedure of our positioning: offline training and online locating. In the phase of offline training, we use piecewise fitting based on the lognormal distribution model to train the propagation model of RSSI for every BLE reference nodes, respectively, in order to reduce the influence of the positioning accuracy because of different locations of BLE reference nodes. Here we design a Gaussian filter to pre-process the receiving signals in different sampling points. In the phase of online locating, we use weighted sliding window to reduce fluctuations of the real-time signals. In addition, we propose a distance weighted filter based on triangle trilateral relations theorem, which can reduce the influence of positioning accuracy due to abnormal RSSI and improve the location accuracy effectively. Besides, in order to reduce the errors of targets coordinates caused by ordinary least squares method, we propose a collaborative localization algorithm based on Taylor series expansion. Another important feature of our method is the active learning ability of BLE reference nodes. Every reference node adjusts its pre-trained model according to the received signals from detecting nodes actively and periodically, which improve the accuracy of positioning greatly. Experiments show that the probability of locating error less than 1.5 meter is higher than 80% using our positioning method.

Design and Implementation of WiFi Network Based Mobile Location Service System

With the growing popularity of intelligent terminals, as well as the increasing demand for location-based service, a design and implementation of WiFi network based mobile location service system and a location algorithm based on GMM (Gauss Mixture Model) introduced in this paper. The algorithm consists of an offline training phase and a real time localization phase, and RSSI is modeled by GMM. This system contains three parts: a server, a middleware and application client. The client is developed on Android Operation System. The server provides the underlying location service, the middleware provides user control and packaging positioning API functionalities as the bridge between the server and the clients, and the mobile clients provide location services for users directly.

A convolutional neural networks based transportation mode identification algorithm

With the increasing sensing ability of smartphone, both recognizing and understanding a users activity using sensor data have become a popular topic of ubiquitous computing systems. Individual transportation mode identification can provide essential data for road planning and traffic management. In this paper, we present a Convolutional Neural Networks (CNN) based method to extract expressive and discriminative features automatically for transportation mode identification. The signal preprocessing in the time and frequency domain is performed before the sensor data is fed into the deep learning framework. We optimize various important hyper-parameters such as learning rate, kernel size and number of convolutional layers to adapt the characteristics of multiple sensor signals. Extensive experimental results indicate that the proposed CNN based transportation mode identification algorithm can achieve 98% accuracy to distinguish between car, bus, train and metro, which outperforms the Support Vector Machines and Adaboost based transportation identification with better robustness and generalization.

A pervasive indoor and outdoor scenario identification algorithm based on the sensing data and human activity

The location and context switching, especially the indoor and outdoor scenario switching, provide basic and original information for various mobile applications. Diverse smartphone placements and limited battery supply pose challenge for the accurate and robust indoor-outdoor identification. In this paper, a pervasive indoor and outdoor scenario identification algorithm is proposed, which utilizes both the sensing data collected by the commodity smartphones and recognized human activity information. The indoor and outdoor scenario identification is modelled as a binary classification problem. To better utilize the collected sensor data and inferred human activities, two time-dependent Adaboost classifiers are developed to perform stateless and instance-based scene detection. The stateless detection result is further used as the observation of a hidden Markov model (HMM) to obtain the final scene estimation. The adoption of the stateful HMM filter can effectively eliminate the occasional noises and improve detection accuracy. Furthermore, to meet the high-accuracy detection demand on the indoor-outdoor transition scenario, invoking GPGSV on demand is introduced to improve the detection confidence. Extensive experimental results confirm that the proposed pervasive indoor-outdoor identification algorithm outperforms the state-of-the-art JODetector with more than 97% detection accuracy under various weather condition and smartphone placements, especially in the cloudy daytime, at night and being put in pocket.