Gaussian mixture model-expectation maximization based signal strength prediction for seamless connectivity in hybrid wireless networks

Authors:
Prashant K. Wali;M. N. Anil Prasad;N. C. Shreyas;N. C. Chaithanya;P. K. Kuttaiah
Affiliations:
PES School of Engineering, Bangalore, India;PES School of Engineering, Bangalore, India;PES School of Engineering, Bangalore, India;PES School of Engineering, Bangalore, India;PES School of Engineering, Bangalore, India
Venue:
Proceedings of the 7th International Conference on Advances in Mobile Computing and Multimedia
Year:
2009

Citing 2
Cited 0

An Effective Vertical Handoff Scheme Supporting Multiple Applications in Ubiquitous Computing Environment

ICESS '05 Proceedings of the Second International Conference on Embedded Software and Systems
Efficient mobility management for vertical handoff between WWAN and WLAN

IEEE Communications Magazine

Quantified Score

Hi-index	0.00

Visualization

Abstract

Signal strength is the predominant factor when providing seamless connectivity to a mobile user in hybrid wireless networks besides factors like bandwidth, cost and quality of service. Primarily, the availability of signal strength is considered to connect to a network or to disconnect from a network before taking into account other factors. A hand off to an alternative network has to be initiated if the signal strength in the current network is expected to go below a threshold value. This calls for a signal strength prediction scheme which considers the randomness in the signal strength behavior and the user movements. To that end, this paper proposes a scheme based on Gaussian Mixture Model (GMM) and Expectation Maximization (EM) for signal strength prediction to make decisions for hand off between networks. The prediction results for the test data shows the effectiveness of the proposed scheme.