CHAPTER 1

INTRODUCTION

1.1 GENERAL

Advanced Signal Processing is used to set up large applications which

fluctuate from picture, discourse, sound, biomedical signal preparing to apply autonomy
and instrumentation. As of late, the interest of upgraded digital signal handling
frameworks and arrangements developed radically. DSP calculations and procedures are
basic for day by day life devices, for example, digital TVs, advanced cameras, cell
phones, and digital set-top boxes, and modems, electronic sound and video players. The
traditional analog signal processing has been replaced by digital signal processing (DSP)
technique due to the advantages like programmability, repeatability, stability and
flexibility. To calculate defective parts of the signal is the capacity of a filter in signal
handling. The block diagram of the filter is illustrated in
Figure 1.1. The unfiltered signal is given into the filter, and the signal gets filtered.

Unfiltered Filtered
signal signal
Filter

Figure 1.1 Block diagram of Filter

Digital and analog are the two primary category of the filter, An analog filter
utilizes simple electronic circuits which is made up of segments, for example, Op-amp,
resistors and capacitors to deliver the required performance. Analog filter circuits are
generally used in the applications like less noisy disturbance, video signal improvement,
and realistic equalizers. In a computerized digital filter, the signal is characterized by a
grouping of numbers, instead of a voltage or current. The accompanying diagram
demonstrates the essential setup of such a system. The figure 1.2 shows the analog to
digital converter basic setup block. Here the unfiltered analog signal is converted in to
DAC.

Unfiltered ADC Processing DAC

analog
signal Sampled Digitally Filtered

digitized filtered analog

signal signal signal

Figure 1.2 Basic block diagram of analog to digital filter converter

The fundamental advantages of digital over analog filters give the following
list

1. An advanced digital filter is programmable, i.e. its activity is controlled

by a program put away in the processor's memory. This implies the
computerized digital filter can undoubtedly be changed without
influencing the hardware (equipment). By upgrading the filter circuit, an
analog filter must be replaced.

2. Digital filters are effortlessly planned, executed and tested on a

universally useful PC or workstation.

3. The qualities of analog filter circuits (especially those containing active

segments) are liable to drift and are temperature dependent. The
computerized digital filter doesn't experience the ill effects of these
issues, as are very steady with deference both to time and temperature.

4. Unlike their analog counterpart, an advanced digital filter can deal with
low-frequency signals precisely. As the speed of DSP innovation keeps
on expanding, an advanced digital filter is being connected to high
 recurrence signals in the RF space, which in the past was selective save of
analog innovation.

5. Digital filters particularly increase the flexibility to process signals in

variety of ways.

1.2 FILTER

Filters are a huge subsystem in numerous electronic systems and especially in

the communication field. Filters can be found in regular existence from cell phones to
satellite receiver and power supplies or in internet associations and nearly in all advanced
electronic instrumentations. Additionally, they are utilized as a sound, noise reduction,
multiplexing, picture preparing and signal recognition. Giving a formal meaning of the
word filtering is hard. However, numerous designers characterized it as a signal
processing activity that alters or reshapes the recurrence area attributes of a given signal.
Then, filter takes the input signal and produces the required output as per the filter
concept. In other ways, filtering is the procedure of selectivity tolerating a predefined
band of frequencies to pass and weakening frequencies outside that band. For example, a
radio collector utilizes shifting to choose one of the few AM (Amplitude Modulation), or
FM (Frequency Modulation) communicates stations that are possible in an explicit
geological zone.

In many cases, the target of filtering is to enhance the signal to noise ratio by
removing the noise signal from the ideal one. There are various approaches to arrange
filters, a portion of these are: Passive and Active filters, Analog and computerized digital
filters, software filters or by passband qualities. Passive filters are utilizing different
mixes of separate components, for example, resistors, inductors and capacitors. Dynamic
filters, then again, are using the active components, for example, a transistor or an
operational enhancer just as separate components. Simple shifting is performed on
constant time signals, and the range of frequency is vast, where digital filter performs the
discrete-time signals and the frequency range is infinite. Software filtering can be done
by downloading the program in the hardware by utilizing PC to eliminate the unwanted
signal. At last, filters can be arranged by their recurrence reaction attributes as a low pass,
high pass, band pass and band stop filters.

1.3 DIGITAL AND ANALOG FILTERS

1.3.1 Analog Filters

Analog filters work on analog signals, which are known as actual signals.
Differential equations are used to illustrate the analog signal. Analog to Digital Converter
(ADC) is used to sample and digitalized the analog signal. The design of analog filters is
defined in the frequency domain. The performances of analog filters are synchronized by
electrical impedance of inductors and capacitors. Shunt leg acts as reactive impedance, in
a voltage divider circuit.

 The op-amp active filters reduces the output impedance, elevated open
loop gain, and elevated input impedance. Furthermore, op-amp have an
intrinsic buffering property that means a general transfer function of
operational-amplifier steps is the product of the single step transfer
functions, disregarding the loading effect of subsequent steps. This
significantly clarifies the conceptual execution. Passive filters do not offer
this kind of clarification, and total transfer function must be executed as
one non-separable whole.

 Digital filters are much difficult than analog filters, since analog filters
will not change the analog signal into a digital signal. There is no
requirement for ADC or DAC. The signal reminds in its accurate analog
form throughout the operation.
1.3.2 Application of Analog Filter

In an analog filter the electronic circuits are made up of RLC components to

generate the necessary filtering effect. By using analog filters, the separation of an audio
signal before application to bass, mid-range and tweeter loudspeakers is done. Analog
filters are widely used in the range of the desired radio station in a radio receiver and
much more applications. To overcome the problem in a frequency response of audio
system and loudspeaker the analog audio filter is used. The analog filters can generate a
response that is the inverse of a system response or an acoustic loudspeaker response so
that when two are summed, the product is almost flat. Difference between the variations
of a output response, loudspeaker response and summed result is equal to the target. In
each case, the final response is more accurate. Cell phone and PDA filter responses often
have annoying peaks that reduce intelligibility. This is compensated with the responses of
a biquad filter. The compensated responses are much more satisfying and intelligible than
the original.

1.3.3 Advantages of Analog Filter

Analog signal has infinite levels. It requires more power and lesser immunity.
It works on continuous time signals. Analog signals are easy processing. It can be used to
filter, translate in frequency, phase shift, detect and many of the other functions. Its
advantages are that it can be low latency, high speed, higher frequency and many times it
handles partial or residual signals better. While converting the analog signal to a digital
signal, the noise will be added. This reduces by getting a higher resolution ADC, but
that's more expensive and slower. Of course, analog signal still gives the thermal and shot
noise, but can get a better SNR with higher gain. To analyze something in the digital
area; it requires to be sampled, and if the incoming signal encloses frequencies more than
half of the sampling rate get an aliasing effect. Analog area representation offers an
aliasing effect. The main benefit of analog switched capacitor filter seems to have over
the digital filter is potential to reach a high dynamic range, as the signal does not have to
quantize

1.3.4 Disadvantages of Analog Filter

At high frequencies, the active analog filtering is not promising due to

distortion restrictions, filtering necessities and op-amp bandwidth. In addition to, it gives
less accuracy, drift due to component variations, not easy to design as well as simulate.
Because of this disadvantage, it prefers digital filter over an analog filter. The main
difficulty of analog signaling is the noise in the system i.e, randomly discarded the
deviation. Though the signal is copied and re-copied, or transmitted over long distances,
these random deviation alters become dominant. Electrically, these losses can be
decreased by shielding, good connections, and some cable verities such as coaxial or
twisted pair. The effects of noise generate signal loss and distortion. This is not possible
to recover because increasing the signal to recover attenuated parts of the signal enlarges
the noise (distortion/interference) as well. Even if the resolution of an analog signal is
higher than a digital signal, the difference can be overshadowed by the noise in the
signal. Digital filter analysis gives more complexity, and implementation results will be
accurate. In terms of size, accuracy, bandwidth and reliability considerations, analog
filters fail to provide more advantage.

1.4 DIGITAL FILTER DESIGN

Filters can be characterized into digital and analogue relying upon the signal
preparing or on the output and input design. Simple filtering is performed in continuous
time signals and incorporates the simple circuit example capacitor and resistors.
Moreover, simple filters experience the ill effects of nonlinearities, the absence of
adaptability, mistakes caused by varieties in part esteems, affect the ability to clamor, and
lacking repeatability. Computerized sifting is one critical capacity that can be actualized
in the DSP. Compared with the analog filter, a digital filter is performed on discrete-time
signs, and it is favored in various applications as a result of the accompanying
advantages:

 Digital filter has a straight stage reaction, and its recurrence reaction can
be adjusted on the off chance that it is executed utilizing a programmable
processor.

 The execution of digital filter does not fluctuate with ecological changes,
for example, warm varieties since it is reliant on numerical computations,
not mechanical attributes of the segments.

 Unlike analog filters, the computerized filter can be utilized at low

frequencies.

 It can be shifted various information motions by one digital filter without

the need to repeat the equipment parts. In this manner, it repeatable from
unit to unit.

 Digital filter is adaptable, do not suffer from the internal noise, high
accuracy, and stable;

Most fundamentally, an advanced digital filter can be achieved in two principle

procedures relying upon applications: equipment and programming. In the equipment
approach, the main components are adders, multipliers, move registers and defer that can
be executed a few activities, for example, expansion, augmentation and move task
contingent upon the method dictated by the exchange capacity of the digital filter. In any
case, in the product approach, the advanced filter is connected as a product program
running in the memory of a computerized flag processor, it is exceptionally valuable on
account of the filter system can be substantially balanced by adjusting the product
program with unmistakable enhancements. Using optimal methods is the design of FIR
digital filter. The nature of their impulse response and forms of response are the two
groups of categorization of digital filters.
Impulse Response

FIR: This term implies that the advanced digital filter has a limited number of samples in
their impulse response, or the length of the filter drive reaction is limited.

IIR: This term implies that the digital filter has a tremendous number of tests in their
drive reaction, or the span of the filter impulse response is vast.

Realization

Recursive Realization: This term implies that the present output of the digital
filter y(n) is dependent on past filter yields {y(n-1),… } just as the past and current filter
inputs {x (n), x (n-1)… }

Non-Recursive Realization: This term implies that the present output of the computerized
Filter y(n) is dependent only on the present and past filter inputs {x(n), x (n-1)… },
however without utilizing past outputs.

The filter configuration process comprises two sections, the approximation and
the implementation. The approximation section manages the decision of parameters or
coefficients in the filter's transfer function. The implementation part manages with a
structure to realize the transfer function. The estimate stage can be classified into four
steps:

 An ideal or desired response is chosen

 A class of Filters is determined (for instance, FIR versus IIR).

 A design criterion is picked (minimum square).

 To Design the transfer function by selecting an algorithm.

The implementation stage can likewise be isolated into four stages:

  Chosen a set of structures.

 Comparing various implementations is chosen for a criterion.

 The structure is chosen best, and its parameters are determined from the
transfer function.

 The structure is actualized in both hardware and software.

Digital filters process are used to digitize or else sample the signals. This filter
verifies a quantized time-domain representation for the convolution of sampled input
time function and a description of the weighting function of the filter. To enrich a certain
aspect of the signals mathematical operations were done on sampled and discrete time
signals. The simulation of analog LC passive ladder filters was done to enhance the
performances of digital filters.

To comprise the digital filters ADC block is to alter the analog form of signal to digital
form of signal. The sampled data are executed by arithmetic operations. These methods
are used in the digital processor. This type of processor is used in a microprocessor, PC
or DSP chip. To perform high performances an operation like filtering, the FPGA was
used instead of general purpose processor or specialized DSP based parallel design. Also,
the arithmetic calculation was carried out on sampled data. The input values are
multiplied by constants, and the product values are added together the calculations are
involved. The data are stored in the memory. At last, the conversion of the processed
digital signal to a corresponding analog signal is done by D-A converter.

In the digital filter, the signal is represented as a sequence of the number than a
voltage or else current. Figure 1.3 shows the basic block diagram of the digital filter. It
consists of pre-amplifier, Anti-aliasing filter, Anti-imaging filter, Analog-to-digital
converter and the digital-to-analogue converter. To diminish the noise in discrete time
signals, the pre-amplifier is used in front of the digital filter. An anti-aliasing filter is
exploited to restrict the bandwidth of a signal to approximately prove the sampling
 theorem. Similarly, an Anti-imaging filter is used in the output side for getting a
horizontal analog signal from the digital signal. In the input side, after anti-aliasing filter,
A/D converter is used to converting analog signal to discrete time signal.

Pre- Anti- Anti-

Amplifier Aliasing Imaging
Microprocessors/G
Filter Filter
eneral Purpose
Processor/DSP/FP
Input signal Output signal
GA/ASIC

Information Control
Signals Signals

A/D Digital Filter D/A

Converter Converter

Figure 1.3 Block Diagram of Digital Filter

A/D converter consists of three steps. They are Sampling, Quantizing, and
Encoding. The conversion of the analog signal into discrete time signal is implemented
by sampling theorem, which states that, when choosing sampling frequency less than
twice of maximum input frequency, it will cause aliasing to the original spectrum. The
sampled discrete time signals then quantized by either using rounding or truncating
method. After quantization, every sampled data is coded using a suitable encoder. The
output of encoder produces a digital signal.

In general, the frequency of signal increases, then the disparity in efficiency

also increases. It can be implemented in a straight forward manner. The advantages of
digital filter over analog filter are linear phase, very high stopband attenuation, very low
passband ripple, reducing the size and reducing power consumption. For all those
advantages, the FIR filter's response must be programmable or adaptive.
 The DSP processor and digital microprocessor are the important device utilized
in an extensive range of applications and can implement complex digital filters from the
smallest amount of filters to audio range of frequencies.

1.5 IMPLEMENTATION OF DIGITAL FILTERS

The main objective of the digital filter is to find the suitable architecture which
reduces the quantization and numerical errors. The following processing architectures are
few examples for implementation of the digital filter.

 Universal Processor.

 DSP with an instruction set optimized for signal processing.

 FPGA or ASIC is special hardware.

1.5.1 Realization with FPGA

FPGA provide the next generation in the programmable logic devices. The
word field in the name refers to the capacity of the gate array exhibits to be modified for
an explicit capacity by the client rather than by the maker of the devices. The word array
is utilized to show a progression of segments and lines of doors that can be customized by
the end client when contrasted with standard entryway clusters. The field programmable
door exhibits are bigger devices. The programmable logic squares of FPGA are called
Configurable Logic Block (CLB). The FPGA design comprises of three sorts of
configurable components (i) IOBs – a border of info/yield squares (ii) CLBs-a center
exhibit of configurable logic squares (iii) Resources for interconnection The IOBs give a
programmable interface between the inward; cluster of logic squares (CLBs) and the
devices outer bundle pins. CLBs perform client determined logic capacities, and the
interconnect assets convey signals among the squares. A configurable program put away
in inner static memory cells decides the logic capacities and the interconnections. The
configurable information is stacked into the devices amid catalyst reinventing capacity.
FPGA devices are modified by stacking arrangement information into inward memory
cells. The FPGA devices can either effectively perused its setup information out of an
outside sequential or byte-wide parallel PROM (ace modes), or the design information
can be kept in touch with the FPGA devices (slave and fringe modes).FPGA is an
integrated circuit that includes numerous (64 to over 10,000) identical logic cells that be
able to be viewed as standard components. A Look-Up-Table (LUT) is the majority
frequently used type of logic block used within FPGAs. There are two types of FPGAs (i)
SRAM-based FPGAs and (ii) Anti-fuse technology based one-time programmable (OTP).

Every FPGA consists of the following elements:

 Configurable Logic Blocks (CLBs)

 Configurable Input/Output Blocks (IOBs)

 Two layer metal network of vertical and horizontal lines for

interconnecting the CLBs.

Figure 1.4 FPGA Architecture

The block diagram for FPGA architecture is shown in figure 1.4.

1.5.1.1 Configurable Logic Blocks (CLBs)

It typically consists of registers (memories), 4-input Look up Tables (LUT),

multiplexers and flip-flops. LUT is used for the realization of combinational logic.LUT is
used for the realization of combinational logic. It is a 16-bit configurable memory which
is capable of realizing all 4-input combination logical functions. The output of the LUT
can be registered to realize sequential function.

1.5.1.2 Configurable I/O Logic Blocks

CLBs, as well as routing channels, are enclosed by a set of programmable I/Os,

which is a collection of transistors for configurable I/O drivers.

1.5.1.3 Programmable Interconnects

They are unprogrammed interconnection on the chip which has channeled

routing with fuse links. Three types of interconnect architectures are available: They are,

 Row-Column architecture

 Island Style architecture

 Sea-of-Gates architecture

Multiplexers in CLB can also be used for the realization of the logic function
with more than four inputs. They allow combining outputs of LUTs CLBs are used for
the realization of main logic in FPGA. It typically consists of 4-input Lookup Tables
(LUT), multiplexers and flip-flops.

Advantages of FPGAs are described as follows.

1. High gate density (i.e.). It offers large gate counts.

 2. No custom masks tooling is required (Low cost).

3. Low risk and highly flexible.

4. Reprogram ability for some FPGAs (design can be altered easily).

5. Suitable for prototyping.

6. Parallelism

7. Allows for system-level extraction of parallelism to match input data at

design time.

8. Huge computational capability.

9. Fast development and dynamic reconfiguration.

10. Updating new pattern rules (no simply rules)

11. The device should not stop while updating new rules

12. Update time for new rules

Because of these advantages, FPGAs are widely used for DSP applications.
For DSP applications, multipliers, the adder with carry chain architecture, etc., are
integrated into FPGAs to increase the speed of computation.

1.5.2 Types of Digital Filters

By criteria, the filters are classified into two main types of digital filters
are,

 Finite Impulse Response filter (FIR filter)

 Infinite Impulse Response filter (IIR filter)

1.5.2.1 Finite impulse response digital filters

The FIR filter is also called a Recursive filter. The FIR filter uses, previous
output values with input values in every stage, hence it is called the recursive filter. In
processor memory, the previous input values are stored.

From above, it is clear that the expression for FIR filter consists of not only
input values but also consists of previous output values. The term digital filter arises
because this filter operates only on discrete time signals. The direct form of Filter of FIR
is shown in Figure 1.5.

Figure 1.5 Direct form FIR filter

Feed-forward means that there is no feedback of past or future outputs to form

the present input, just having only input related terms.

The uniqueness of FIR filter are:

 Linear phase characteristics

 High filter order (more complexity circuits) and

 More stability
1.5.2.2 Infinite Impulse Response (IIR) digital filter

The Infinite Impulse Response or IIR filter be named as the non-recursive

filter. Current and previous input values compute the current output of Infinite Impulse
Response digital filter.

The fundamental characteristics of the IIR filter are:

 Non-linear phase characteristics

 Low filter order (fewer complexity circuits) in addition to

 Resulting digital filter has the potential to become unstable.

The IIR filter has high frequency response than FIR filters of the same order.

1.6 STRENGTH OF FIR FILTER OVER IIR FILTER

A FIR filter is a kind of digital FIR filter. It is finite because it's a response to
an impulse is zero. Due to zero structure characteristics, the FIR Filter has the linear
phase response when the filter coefficients are symmetric, as is the case in most standard
filtering applications. An FIR's implementation noise characteristics are easy to model,
especially if no intermediate truncation is used whereas IIR’s implementation noise
characteristics are difficult to model.

1.7 MERITS OF FIR FILTER

Compared to IIR filter, Filter of FIR has the following advantages:

 FIR filter is simple to implement.

  FIR filter can also be implemented as linear-phase filtering. This means
no phase shift across the frequency. Also, the phase can be corrected the
amplitude independently.

 Coefficients are easy and simple to calculate

 FIR filter calculation can be done by looping a single order.

 They can be implemented using fractional arithmetic.

 They can be realized efficiently in hardware

Unlike IIR filters, it is always likely to implement a FIR filter using

coefficients with the magnitude of a smaller amount than 1.0. To make implementation
easier fixed-point DSPs is considered to be the main concern.

1.8 REAL-TIME APPLICATIONS OF FIR FILTER

A filter is a frequency selective linear time-invariant (LTI) system that is a

system that passes specified frequency components as well as rejects others. A filter is a
frequency selective linear time-invariant (LTI) system that passes specified frequency
components as well as rejects others. The discrete-time filters realization of LTI systems
which have LCCDE representation and are casual. These processes are known as
convolution. The demand for low power digital signal processing and high-speed multi-
standard wireless communications is quickly growing. The unstable growth in portable
multi-media and wireless multi-standard applications has mostly depended on the signal
processing technique. The FIR filter is one of the key features of signal processing
approach. The FIR filter needs two important key factors known as low complexity and
reconfigurability. For a signal and image processing approach, standard wireless
communication and mobile telecommunications applications, improved FIR filter (in
terms of moderate complexity and reconfigurability) are used.
1.9 FIR AND IIR FILTERS

Digital IIR filter has an impulse response with an infinite number of nonzero
samples. Actually, with a finite number of nonzero input values, the IIR filter could have
a vast span of nonzero values of output. The primary explanation behind having the
characteristics of infinite response in the presence of feedback, which implies that IIR
digital filter utilizes a portion of its past yield tests to ascertain the present yield test; this
way, it called as a recursive filter. Like other feedback forms, IIR filter having input can
deliver unsteadiness in the capacity of these filter and influence it to have nonlinear stage
qualities, particularly when the criticism is excessively huge. This will cause motions in
the output of the IIR filter and prompts off a base response which might be difficult to
recognize and correct. To achieve the tasks utilizing a little measure of less memory as
well as computational power

FIR advanced filter has a few alluring properties in connection to IIR Filters.

 FIR filter has precisely linear phase.

 FIR filter are consequently steady

 There are a few adaptable techniques for designing an FIR filter

 FIR filter is convenient to realize

 Linear-stage FIR Filters can have long delay among input and

output.

 If the stage isn't to be direct, at that point, IIR Filters can be significantly
more productive.
1.10 ALTERA QUARTUS II

The Altera Quartus II design software provides a complete, multi-platform

design environment that easily adapts to your specific design needs. It is a comprehensive
environment for system-on-a-programmable-chip (SOPC) design. The Quartus II
software includes solutions for all phases of FPGA and CPLD design (Figure 1.6).

Design
entry

Power
analysis
Synthesis

Debugging
Place and
route
Engineering
change and
management
Timing
Analysis
Timing
Closure
Simulation

Programming
and
configuration

Figure 1.6 Quartus II Design flow

1.11 VHDL

VHDL is generally used to compose content models that portray a logic circuit.
Such a model is prepared by a synthesis program, just in the event that it is a piece of the
logic structure. A program is utilized to test the logic configuration using reproduction
models to speak to the logic circuits that interface to the plan. This accumulation of
reenactment models is regularly called a test seat. A VHDL test system is normally an
event-driven simulator. This implies every exchange is added to an event line for an
explicit planned time. For example on the off chance that a signal task ought to happen
after 1 nanosecond, the occasion is added to the line for time +1ns. Zero deferral is
likewise permitted, yet at the same time should be planned: for these cases Delta delay is
utilized, which speak to an endlessly little time step. The reenactment changes between
two modes: articulation execution, where activated explanations are assessed, and
occasion preparing, where occasions in the line are handled. VHDL has builds to deal
with the parallelism natural in equipment structures. However, these develop (forms)
contrast in language structure from the parallel builds in Ada (assignments). Like Ada,
VHDL is specifically and isn't case delicate. To straightforwardly speak to tasks which
are basic in equipment, there are numerous highlights of VHDL which are not found in
Ada, for example, an all-encompassing arrangement of Boolean administrators including
NAND and NOR. VHDL has document info and yield capacities and can be utilized as a
broadly useful dialect for content preparing. However, records are all the more used
usually by a reenactment test seat for boost or check information. There are some VHDL
compilers which construct executable doubles. For this situation, it may be conceivable to
utilize VHDL to compose a test seat to confirm the usefulness of the structure utilizing
records on the host PC to characterize improvements, to collaborate with the client, and
to contrast results and those normal. Notwithstanding, most architects leave this activity
to the test system. It is generally simple for an unpracticed engineer to deliver code that
recreates effectively yet that can't be blended into a certain device, or is too expensive
ever to be down to earth. One specific trap is the unplanned creation of straightforward
hooks instead of D-type flip-tumbles as capacity elements. One can structure equipment
in a VHDL IDE (for FPGA execution, for example, Xilinx ISE, Altera Quartus, Synopsys
Simplify or Mentor Graphics HDL Designer) to deliver the RTL schematic of the ideal
circuit. From that point forward, the created schematic can be confirmed utilizing
reenactment programming which demonstrates the waveforms of information sources and
yields of the circuit after producing the suitable test seat. To provide an adequate test seat
for a specific circuitry or VHDL code, the sources of info must be characterized
accurately. For instance, for clock input, a circle procedure or an iterative explanation is
required.[12] A last point is that when a VHDL show is converted into the "entryways
and wires" that are mapped onto a programmable logic devices, for example, a CPLD or
FPGA, at that point it is the real equipment being designed, as opposed to the VHDL
code being "executed" as though on some type of a processor chip.

1.11.1 Advantages

The key advantage of VHDL, when used for systems design, is that it allowed
the behavior of the necessary system to be described (modeled) and verified (simulated)
before synthesis tools translate the design into real hardware (gates and wires). Another
benefit is that VHDL allows the description of a parallel system. VHDL is a dataflow
language, unlike procedural computing languages such as BASIC, C, and assembly code,
which all run sequentially, one instruction at a time. A VHDL project is multipurpose.
Being created once, a calculation block can be used in many other projects. However,
many formational and functional block parameters can be tuned (capacity parameters,
memory size, element base, block composition and interconnection structure). A VHDL
project is portable. Being created for one element base, a computing device project can be
ported on another element base, for example, VLSI with various technologies. A big
advantage of VHDL compared to original Verilog is that VHDL has a full type system.
Designers can use the type system to write much more structured code (especially by
declaring record types).

1.2 PROBLEM STATEMENT

The major requirement in signal processing is to remove noise or an unwanted

signal in a noisy environment and to extract the information. The existing FIR and IIR
Filter contains complexity in the hardware, delay and power dissipation in the system.
Here, an integrator is used to suppress the noise effects. Once the transfer function is
determined, it can be realized using different structures. For example, one can implement
an IIR filter using the direct form, or a cascade of second-order sections, or one of several
other structures. While they are all equivalent when infinite precision is used, different
structures behave differently when the coefficients and the arithmetic operations are
quantized.

1.13 MOTIVATIONS OF THE RESEARCH

The demand for optimized digital signal processing systems and solutions
grow drastically. In DSP, the versatile means the algorithms can be repeated, forwarded,
and compressed depends on the need. The stability can be achieved by designing FIR
filters at the cost of filter length for the given specifications. The stable IIR filters can
also be designed with most care. The speed, area, and power are the important
optimization goal in any VLSI signal processing system. In previous research works the
main focus is on parallel, pipelining and concurrent processing of filtering operation for
resource utilization but ignored the low design hardware complexity with high
computational speed. So, the main motive of this work is to design an efficient DSP
systems with high performance in terms of less number of logic elements and the less
power dissipation with high computational speed and reduced design complexity.

1.14 RESEARCH AIMS AND OBJECTIVES

The main objective of this work is to design fast IIR and FIR filter which can
be implemented in FPGA device to meet the real time application.

 To design IIR filter using look-ahead pipelining for first order, second
order with level-1 & level-2.
  The FIR filter is designed using MA filters using cascade combination of
CIC which contains both feed forward and feedback sections for 8, 16, 32
& 64 taps.

 Analysis of different FIR filter form such as with direct form, transposed
form and systolic array has been performed.

 The Goertzel algorithm is implemented using Simulink for the efficient

evaluation of one selectable frequency component from a discrete signal

 Finally the filter is implemented using sub filter implementation

technique.

1.15 ORGANIZATION OF THESIS

The thesis contains seven chapters and is organized as follows:

Chapter 1: Introduction. It deals with crucial information regarding digital

filter techniques. Also discusses FIR and IIR technique involved in DSP implementation.
Further advantages of FPGA and main objectives of the current research work are
discussed.

Chapter 2: Literature Review. It deals with the review of the literature

regarding different architectures based Digital FIR filter and IIR filter. Also, it discusses
the issues of traditional techniques and reasons of DSP model.

Chapter 3: FPGA Implementation of Fast Digital FIR and IIR Filters This
chapter discusses the different FIR and IIR techniques of DSP method, using different
filter techniques called MA filter with the help of look ahead arithmetic.

Chapter 4: Efficient Digital Filter Implementation for High-Performance DSP

Systems Application. This chapter discusses the digital filter structures such as direct
form, transposed form and systolic array. The systolic array structure is very often used to
implement parallel processing. The FPGA implementation of a Goertzel algorithm .

Chapter 5: Result and Discussion. In this chapter, the comparative analysis of

all the three methods proposed in this research work with respect to previous studies.

Chapter 6: Conclusion and Future Scope. In this chapter, the conclusion of

research work and scope of the future work are presented.