Application of Chaos Theory and Wavelet to Modeling the Traffic of Wireless Sensor Networks

In this paper, nonlinear chaos theory and wavelet transform are applied in modeling and forecasting traffic of wireless sensor network (WSN). Wavelet-weighted local model of chaotic prediction not only takes the advantages of lower and more determinism in chaos theory, but also the advantages of unique multi-resolution analysis in wavelet transform, which can be used to generate a rich set of prediction traces that exhibit similar data structure to the original traffic data of WSN. The computing complexity is also reduced. Numerical simulation verified that the forecasting precision is high, and achieved results are satisfactory.