COMPRESSED CHANNEL SENSING AND ESTIMATION

Abstract

A Trustworthy wireless communications requires precise know-how of the essential multipath channel which generally includes channel probe using a famous waveform(training) ,then after input probe is processed linearly and afterwards impulse response is approximated by arriving at the output of the channel. A large number of communication channels in practicality have a very less no of non -zero coefficients in their impulse responses.LS method or minimum mean squares estimation method or other methods to estimate channel used traditionally can't be used in these sparse channels as they are low dimensional. As opposed to above, this report proposes investigating the application of the theory and methodology of Compressed Sensing to solve the problem of estimating multipath channels. Convex, linear programming or greedy pursuit algorithm based methods to estimate sparse channels, and presentation of simulation results that show the performance gains which can be achieved by using each of the compressive sensing algorithm, are put forward in this thesis