Continuous Tracking of User Location in WLANs Using Recurrent Neural Networks
ENC '05 Proceedings of the Sixth Mexican International Conference on Computer Science
Design of indoor positioning systems based on location fingerprinting technique
Design of indoor positioning systems based on location fingerprinting technique
Context life cycle management in smart space environments
Proceedings of the 3rd workshop on Agent-oriented software engineering challenges for ubiquitous and pervasive computing
Local positioning system based on artificial neural networks
ICANN'07 Proceedings of the 17th international conference on Artificial neural networks
Bluetooth indoor localization with multiple neural networks
ISWPC'10 Proceedings of the 5th IEEE international conference on Wireless pervasive computing
Agent-based architecture for context-aware and personalized event recommendation
Expert Systems with Applications: An International Journal
Hi-index | 0.00 |
User location is a valuable contextual information in context-aware computing systems. These systems need to know the user's current physical location in order to provide relevant services. The location of mobile users can be inferred by measuring the received signal strength. This article presents a specific module of an agent-based framework - the Location Agent Module - which was developed for context-aware applications using the existing wireless network infrastructure: WiFi, Bluetooth and ZigBee. Furthermore, we present a comparison study with a quality of service (QoS) sub-module. The experimental results indicate that the ANN accuracy was 72%, 63% and 67% using WiFi, Bluetooth and ZigBee protocols, respectively. Considering the use of QoS sub-module, we had an increase from 11% to 21% in the results with the three protocols, increasing the accuracy to 89% to WiFi, 74% to Bluetooth and 88% to ZigBee. Based on the promising results achieved, we consider our approach adequate for indoor localization.