Localization of Industrial Wireless Sensor Networks: An Artificial Neural Network Approach
With the augmentation of industrial applications of wireless sensor networks, the problem of localization in such networks gains more attention. Although a considerable amount of research has been conducted in this area, many of the proposed approaches produce unsatisfactory results when exposed to the harsh, uncertain, noisy conditions of a manufacturing environment. An artificial neural network approach is developed in this study to abate the effects of the environmental noise sources and harsh factory conditions on the localization of the wireless sensors. A simulator, imitating the noisy and dynamic shop conditions of manufacturing environments, is employed to examine our proposed neural network. Subsequently, a sensitivity analysis is conducted using design of experiments methods. The results obtained indicate that noise intensity and anchor node topology give the most significant impact on the performance of the proposed localization technique. These results, combined with the inherent distributed and changeable nature of wireless sensor networks, have led us to investigate a multi-agent solution to this problem.
KeywordsWireless sensor networks Localization Artificial neural network Simulation Design of experiments
Unable to display preview. Download preview PDF.
- 3.Wu, H., Wang, C., Tzeng, N.: Novel self-configurable positioning technique for multihop wireless networks. IEEE/ACM Trans. Network 3, 609–621 (2005)Google Scholar
- 5.Moore, D., Leonard, J., Rus, D., Teller, S.S.: Robust distributed network localization with noisy range measurements. In: Proceedings of SenSys 2004, pp. 50–61 (2004)Google Scholar
- 8.Ogawa, T., Yoshino, S., Shimizu, M., Suda, H.: A new in-door location detection method adopting learning algorithms. In: IEEE Computer Society Proceedings of the First IEEE International Conference on Pervasive Computing and Communication (PerCom 2003)Google Scholar
- 9.Intel Corporation, Expanding Usage Models for Wireless Sensor Networks. Technology@Intel Magazine, 4–5 (2005)Google Scholar
- 11.Hecht-Nielsen, R.: Neurocomputing. Addison-Wesley, Reading (1990)Google Scholar
- 12.Gholami, M., Cai, N., Brennan, R.W.: An Artificial Neural Network Approach to the Problem of Wireless Sensors Network Localization. Submitted to Robotics and Computer-integrated ManufacturingGoogle Scholar