Quick Detection of Changes in Traffic Statistics: Application to Variable Rate Compression
Ivy Hsu, Jean Walr · 1999
In many communication applications, sources generate data at a variable rate. This variation can be captured by a two-level model. At a larger time scale, the source alternates between a finite number of models at the higher level, while each model represents the local fluctuation at the smaller time scale. Typically the data goes through a lossy compression at the encoder, and then is entered into a buffer before being transmitted over a fixed-rate or a regulated-rate (e.g., Leaky Bucket) channel. Since the buffer size is finite, the choice of compression parameters must balance the tradeoff between the distortion due to lossy compression and the distortion caused by buffer overflow. As the optimal compression parameters may be different for each individual model, a method is needed to detect changes from one model to another. We show via examples that for many common traffic models, the problem of detecting changes in traffic statistics can be reduced to the following general change ...