Description

The Gaussian channel is the most important continuous alphabet channel, modeling a wide range of communication channels. Optimal Channel switching strategy is proposed with the presence of average power and cost constraints. Here there are channel switching is performed between single channel, two channel and three channels. Ratio of the power levels are dependent by the larger noise variance. Cost values are dependent to the average noise powers, so that average cost limit are easily specified between two channels. The Error Probabilities are estimated for logarithmic cost functions which are investigated as a function of power and cost. For optimal single channel strategy, only one channel is employed and channel switching is not allowed. For optimal channel switching strategy more than one channel are employed, where the switching of channels are performed between two channels or among three channels. Efficient channel switching strategy is performed at most three channels which operates at the average power and average cost limits. In the case of channel switching between two channels is proved that the ratio of the optimal power levels and ratio of the noise variance estimated under certain conditions. The optimal power level is upper bounded by the noise variance to the smaller one under the same conditions. The error probability for convexity properties have been estimated based on the power and cost functions. At last the optimal channel switching strategy is optioned at most two channels.