IIR Based Digital Filter Design and Performance

Analysis

May 2019

1 Introduction
Signal processing is the leading part for any type of communication system.
Signal processing is compulsory to avoid unwanted signals. For processing in-
formation, analog and digital signal processing can be used. Digital signal pro-
cessing offers more stability and better accuracy control compared to analog
signal processing. To process the signal, filters are applied.Some selected au-
dio signals have been used to observe different outcome from digital filters, and
the filters output was also plotted and played. The aims of this research are
to figure out how digital filters perform better. To accomplish this goal, the
various parameters of digital filters need to be optimized.Parameters such as
magnitude response, phase delay, phase response, group delay, pulse response,
pole / zero plots and step response were used to optimize the outcome and
determine the better output. The IIR Butterworth filter was used to get the
better output.Butterworth filter is the name of the coordination between at-
tenuation and phase response. Occasionally, because it has no flow in the stop
band or the pass band, this is called a maximally flat filter. At the expense of a
relatively wide evolution section from pass band to stop band, the Butterworth
filter obtains its flatness.All simulation and results were analyzed and verified
using MATLAB’s FDA tool to analyze different parameters. In practice, a tool
is developed to filter audio signals.

2 Lierature review
For signal processing, a lot of research has been done on digital filter. A.K.M.
Fazlul Haque worked with FDATool to extract enviable parameters and help
increase digital filter performance[3 ].Using FDATool[4 ], Mohammad Saiful Is-
lam, Shaila Shanjada and Mithun Banik worked to improve digital filter extrac-
tion aspects.An analysis of digital IIR filters was conducted by Yaduvir Sing,
Sweta Tripathi and Manoj Pandey with LabVIEW[5 ]. ZHANG Chengliang and
WANG Aihong worked on MATLAB research and simulation of the IIR digital
filter design[6 ]. Ethan Elenberg and others. Worked on the design of digital

1
audio filters[7 ].Er. Daljit Singh Bajwa, Er.Karamjeet Singh and Navpreet Kaur
Chahal worked on survey paper of digital IIR filter design [8]. Ekta Yadav and
Rupali worked on digital IIR filter design [9].All of them worked on analyzing
and designing the IIR digital filter in the research papers above. For these
analysis purposes, different types of design methods were used such as window
function method, impulse invariance method, and some of them used LabVIEW
software for their research. This paper contains various forms of filters such as
high-pass filters, low-pass filters, band-stop filters and band-pass filters.The per-
formance of the digital filter was examined. For this design purpose, the IIR
Butterworth design method is used. A tool for filtering audio signals is also
being developed.

3 METHODOLOGY
In this paper, the actions of different digital filters and the responses have been
studied and analyzed based on different parameters. Also, the filter output of
a selected audio signal has been examined.MATLAB FDA (Filter design and
analysis) tool has been used for analysis part and MATLAB GUIDE (Graphi-
cal user interface development environment) tool for the experiment of filtering
audio signal.

Why IIR filter?

There are mainly two digital filter categories. Infinite Reaction to Impulse (IIR)
and Finite Reaction to Impulse (FIR). Filters that usually have an impulse re-
sponse that contains only a limited number of values are typically called FIR
filters[10]. FIR filters are involved in filtering problems where within the pass-
band there is a requirement for linear phase personalityFIR filters are involved
in filtering problems where within the filter passband there is a requirement for
linear phase personality. FIR filter design focuses on a direct approximation
of the specific magnitude response, with the frequently added requirement that
the phase response be linear[11].On the other hand, infinite impulse response
(IIR) is called the filters that have infinite dimension impulse response. Only
IIR filters have been applied in this paper to analyze various parameters. In
contrast to FIR filters, IIR filters have more efficient filtering characteristics.