Mathur, SanjaySingh, Vipra2019-11-222019-11-222019-08http://krishikosh.egranth.ac.in/handle/1/5810135943This thesis presents a study on the blind equalization of Bipolar 2-PAM and 4-QAM signals using Non-Linear Autoregressive Exogenous Input (NARX) neural system. The thesis can be divided into two parts, the first part involves the theoretical issues about the blind equalization and neural equalizers, the second part addresses the problem of software implementation. The thesis begins with some basic concepts and theories of the channel equalization, which are fundamentals for designing blind equalizers, and then introduces blind equalization concept. Next presented in the thesis is the neural network architecture that is suitable for blind equalization. Its stability and convergence properties are analyzed in the thesis. Also, the thesis proposes a blind equalization model that combines the concept of decision-feedback equalizer and the NARX neural networks. Simulations with 2-PAM and 4-QAM signals have been carried out to test the performances of the proposed model. The second part of the thesis presents a software equalizer design which has been simulated. Based on the results from the simulation, the performance of the neural equalizer is discussed. Finally, some suggestions for further work are included at the end of the thesis.ennullThesis