23-05-2012, 04:51 PM
LMS and RLS algorithm
LMS and RLS algorithm.pdf (Size: 100.73 KB / Downloads: 70)
Project Plan:
The project is organized into modules in such a way that the modules
reduce the project into independent tasks and thus effectively reducing the
complexity involved. Each module involves a brief literature review on the task to
be implemented, mathematical analysis, implementation details and complexities
involved, testing-debugging and finally the demo.
The entire project can be divided into 4main modules:
1. Synopsis and Literature review Phase.
2. Collection of samples.
3. Study of adaptive filters.
a) LMS algorithm.
b) RLS algorithm.
4. Design and implementation.
The above modules are further divided into different modules and the schedule
for the same has been shown in PERT chart that in page. We have referred these
technical papers for our project
Anticipated Bottlenecks :
The bottlenecks those can be anticipated in the final phase of the project. These
are the following bottlenecks:
1. Appropriate adjustment of the delay :
The delay between the input signal and the noise signal must be such
that the noise signal in the input and the noise signal we take, must
be highly correlated. This delay should neither be large nor very
small.
2. Collection of samples:
It is very much necessary that the samples we collect have clear
distinction between the primary signal and the noise as most of the
analysis in the beginning will be carried out on these signals.
Analysis for differentiating these signals from noise is tough job as
both would be much similar in many ways.
3)Anticipated Simulation:
The Simulation involves taking input which contains a primary signal and
noise signal. To extract primary signal we need only noise signal as well. The signal
containing primary signal and noise is given as input to the adaptive filter
designed. The output is expected to be free from the noise signal.
The above procedure is carried out for filter designed using LMS and RLS
algorithm one at a time. A detailed report showing the comparison of these
algorithms is presented.
Progress Report:
So far we have made a thorough Literature survey and finalized the project course
and this plan is clearly outlined in Pert Chart. Now we turn our attention towards
the study of adaptive filters.
An adaptive filter is a filter that self-adjusts its transfer function according to an
optimization algorithm driven by an error signal. Because of the complexity of the
optimization algorithms, most adaptive filters are digital filters. By way of
contrast, a non-adaptive filter has a static transfer function. Adaptive filters are
required for some applications because some parameters of the desired
processing operation are not known in advance. The adaptive filter uses feedback
in the form of an error signal to refine its transfer function to match the changing
parameters.