On tractable wireless channel models

We present techniques to capture the behavior of realistic channels with mathematically tractable models, We examine different state-space aggregation techniques to reduce a large number of states of Markov chain to a fewer number of states, The property of strong and weak lumpability is discussed and the stochastic bounding techniques are examined. These techniques are applied to three different types of previously published channel models for mobile VHF, wireless indoor, and Rayleigh fading channels. Results show that our stochastic bounding technique can produce a simpler yet still accurate upper bound the for original channel model. We find that the bounds perform well when the higher-layer error control protocols such as stop-and-go and TCP are considered.