Home // ICWMC 2011, The Seventh International Conference on Wireless and Mobile Communications // View article


A Smart Predictive Link Layer Trigger Algorithm to Optimize Homogenous/Heterogeneous Networks WiFi Handover Decisions

Authors:
Chunmei Liu
Christian Maciocco

Keywords: predictive handover trigger; station collision probability; WiFi networks; intra- and inter-technology handover

Abstract:
In traditional wireless networks, link layer metrics used to trigger handover are mainly signal quality based, such as Received Signal Strength (RSS). While signal quality is one major reason for poor performance in traditional wireless networks, WiFi networks are special in that there is another major reason for poor performance, which is collision. While there exist some metrics to reflect collisions and trigger handover when necessary, such as channel load, this paper explains why these existing metrics, such as channel load, cannot capture the actual collision situation in the network and that one station is experiencing. Based on this observation, this paper proposes a new metric, called station collision probability, as an additional handover trigger metric, and develops a prediction algorithm for this metric. Specifically, for WiFi networks, station collision probability is the probability that a packet being transmitted by a station incurs a collision. A prediction algorithm is developed for station collision probability on unlicensed WiFi networks, which takes the number of collisions between two successful transmissions on the channel as the measurement and predicts the station collision probability by solving a developed equation. The algorithm does not require the station to send any traffic, and applies to real time decisions, including predictive handover decisions to initiate and prepare the handover to reduce latency and service interruption for the end-users. This paper focuses on defining an optimal collision estimation algorithm and the simulation results validates that the algorithm predicts station collision probability and adapts well to the change of network traffic. The predicted station collision probability can then be integrated with the existing signal quality based trigger metrics to trigger handover, which is beyond the scope of this paper and will be the next steps

Pages: 48 to 52

Copyright: Copyright (c) IARIA, 2011

Publication date: June 19, 2011

Published in: conference

ISSN: 2308-4219

ISBN: 978-1-61208-140-3

Location: Luxembourg City, Luxembourg

Dates: from June 19, 2011 to June 24, 2011