17EC652 Adaptive Signal Processing syllabus for EC

Adaptive systems:

Definitions and characteristics - applications – propertiesexamples - adaptive linear combiner input signal and weight vectors - performance function-gradient and minimum mean square error - introduction to filteringsmoothing and prediction - linear optimum filtering-orthogonality - Wiener – Hopf equation-performance surface(Chapters 1& 2 of Text). L1, L2

Searching performance surface-stability and rate of convergence:

Learning curvegradient search - Newton's method - method of steepest descent - comparison - Gradient estimation - performance penalty - variance - excess MSE and time constants – mis-adjustments (Chapters 4& 5 of Text). L1, L2

LMS algorithm convergence of weight vector:

LMS/Newton algorithm - properties - sequential regression algorithm - adaptive recursive filters - random-search algorithms - lattice structure - adaptive filters with orthogonal signals (Chapters 6 & 8 of Text). L1, L2, L3

Applications-adaptive modeling and system identification:

Multipath communication channel, geophysical exploration, FIR digital filter synthesis. (Chapter 9 of Text). L1, L2, L3

Inverse adaptive modeling:

Equalization, and deconvolution adaptive equalization of telephone channels-adapting poles and zeros for IIR digital filter synthesis (Chapter 10 of Text). L1, L2, L3

Course Outcomes:

At the end of the course, students should be able to:

• Devise filtering solutions for optimising the cost function indicating error in estimation of parameters and appreciate the need for adaptation in design.
• Evaluate the performance of various methods for designing adaptive filters through estimation of different parameters of stationary random process clearly considering practical application specifications.
• Analyse convergence and stability issues associated with adaptive filter design and come up with optimum solutions for real life applications taking care of requirements in terms of complexity and accuracy.
• Design and implement filtering solutions for applications such as channel equalisation, interference cancelling and prediction considering present day challenges.

Text Book:

Bernard Widrow and Samuel D. Stearns, ―Adaptive Signal Processing‖, Person Education, 1985.

Reference Books:

1. Simon Haykin, ―Adaptive Filter Theory‖, Pearson Education, 2003.

2. John R. Treichler, C. Richard Johnson, Michael G. Larimore, ―Theory and Design of Adaptive Filters‖, Prentice-Hall of India, 2002.