Intelligent Agent Technology for Cellular-Assisted GPS Positioning using Bayesian and Self-Organizing Map
Yee-Wai Sim1, Lam Hong Lee2, Yen Pei Tay3, Khang Wen Goh4, Dino Isa5
1Yee-Wai Sim*, School of Computing and Creative Media, University College of Technology Sarawak, Sibu, Malaysia.
2Lam Hong Lee, School of Computing, Quest International University Perak, Ipoh, Malaysia.
3Yen Pei Tay, School of Computing, Quest International University Perak, Ipoh, Malaysia.
4Khang Wen Goh, School of Computing, Quest International University Perak, Ipoh, Malaysia.
5Dino Isa, Faculty of Engineering, University of Nottingham, Selangor, Malaysia.

Manuscript received on November 20, 2019. | Revised Manuscript received on November 26, 2019. | Manuscript published on 30 November, 2019. | PP: 3391-3403 | Volume-8 Issue-4, November 2019. | Retrieval Number: D6753118419/2019©BEIESP | DOI: 10.35940/ijrte.D6753.118419

Open Access | Ethics and Policies | Cite  | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: This paper presents the use of intelligent agent technology, cellular-assisted Global Positioning System (GPS) and data mining for positioning purpose. Due to overlapping coverage areas of cell towers, conventional cell-based positioning techniques have been reported to be inaccurate. Current cell-assisted GPS positioning setup with high accuracy is costly as it requires huge investments on hardware deployments. A new solution of using intelligent agent technology was proposed by the authors for an economical and satisfactory cell-assisted GPS positioning system. Location information in the form of cell identity (ID) and GPS coordinates pairs can be acquired via devices such as smart phones and GPS trackers. The cell ID-GPS coordinates pairs are then grouped by each individual cell ID. An intelligent agent equipped with data mining capabilities is then deployed to computer the optimal GPS coordinates of the cell ID to provide more precise location information. The proposed solution was evaluated via a prototype system. The system was built to collect raw data of cell-ID and GPS coordinate pairs from trackers and mobile phone applications. Using the reference GPS coordinate that was calculated by taking the mean of longitude and mean of latitude for all the GPS coordinates clustered in the same group, the geographical distance between each GPS coordinate and the reference GPS coordinate in the same group was computed to evaluate the performance of the proposed solution.. Experimental results showed that the proposed solution based on intelligent agent equipped with data mining capability helped in improving the prediction of location with sub-kilometer accuracy, in contract to the conventional cell-assisted GPS positioning system which have low accuracy with distance rate various in kilometers.
Keywords: Intelligent Agent, Cellular-Assisted GPS Positioning, Bayesian, Self-Organizing Map, Data Mining.
Scope of the Article: Data Mining.