నైరూప్య

Analysis of Communication Signals with Nonuniform Sampling Times

P.Rajesh,K.Uma maheswari,S.Srinivasa Rao.

Communication signals plays vital role in the automatic transmission of data over various applications and also in the various logistic and electrical activities of defence services. Analysis of communication signals is essential to ensure spectral efficiency and integrity of information. The main objective of the analysis is to minimize the noise present in the communication signals like voice, telegraphy, teletype writer and facsimile (FAX) etc. Nonuniform sampling occurs in many applications where uniform sampling either not possible or practically not achievable. This case relates to event based sampling in the time domain, such as queue process in routers, data networks and astronomical data processing etc. Non-uniform sampling signal can be estimated using FFT, Several methods to compute an approximate Fourier Transform for posterior analysis in terms of alias suppression and leakage. Though most of the work was focused on the development of posterior analysis of the signal a priori estimation of stochastic properties of the signal may result in better transformation approximation. In this way the frequency resolution of noise spectrum will be minimized. For optimum level minimizing purpose of frequency resolution, non-parametric power spectrum estimation methods are considered. The main focus of this paper is minimization of frequency resolution of noise spectrum and increasing frequency resolution of signal spectrum. For this purpose different non-parametric power spectrum estimation methods are used. In the non-parametric methods, periodogram is modified averaging and smoothing using different window techniques. This leads to the minimization of frequency resolution and variance with the help of non -parametric power spectrum estimation methods.