PROJECT REPORT1 (1) new

Adaptive Moving Target Indicator

CHAPTER 1

INTRODUCTIONSince conception of radar just prior to Second World War many different types of radar have

been developed. Although the original radars were developed to extract information about the

position of aircraft, the realization soon came that by transmitting different waveform and

applying different forms of processing to the received waveform, much more detailed

information could be obtained. It was realized that not only could more information about targets

be obtained, but that information could also be gained about operating environment of radar.

Example is weather radar, and sideways looking airborne radars.

Radar (radio detection and ranging) is designed to detect and locate targets within the specified

range. Targets are in general defied as planes and other aircraft. The radar emits an

electromagnetic pulse, and each object that becomes illuminated by the pulse reflects a small

replica of the pulse back to the radar. From this echo signal the radar then attempts to detect and

locate the reflecting object. The radar pulse is reflected by targets, but may also be reflected by

other objects, for instance the ground, high mountains and rain. Such reflections are called

clutter, and they are regarded as noise since they reduce the radars ability to detect and locate

targets. Targets are concentrated spatially, whereas clutter tends to occupy a larger area. The

random clutter process is therefore characterized by its spatial and temporal statistic.

Moving target indication (MTI) is a mode of operation of radar to discriminate a target against

clutter. The most common approach takes advantage of the Doppler Effect. For a given sequence

of radar pulses, the moving target will change its distance from the radar system. Therefore the

phase of the radar reflection that returns from the target will be different for successive pulses.

This differs from a stationary target (or clutter) which will cause the reflected pulses to arrive at

the same phase shift.

Clutter variation requires the use of adaptive cancellers that sense the clutter characteristics and

adjust their weights accordingly. Weights are the complex, than real value and there by allow the

nulls to be steered in Doppler frequency to cancel clutter as appropriate. Adaptive filters have

been proposed as an answer to the problem of clutter suppression in spatially and time varying

Dept of TCE, BNMIT Page 1


clutter environments. The performance of general adaptive filters is reduced when the statistical

properties of the clutter process vary with radial distance, as the filter weights are estimated from

neighboring range bins. An alternative strategy is to make the MTI filter adaptive.

Radar can be broken down in 2 main categories those which can be transmit and receive

continuously called CW (continuous wave radar) and those that transmit for a short period of

time, and then received whilst the transmitter is turned off, called pulse radar. CW radar is quite

severely restricted in amount of information that they can provide hence much more attention is

given to pulse radar. The difficulty of isolating a power full transmitter from a sensitive receiver

tends to limit amount power transmittable and hence limits useful range of radar.

Whatever the type of radar that is employed certain principle always applies. One of these

overriding principles is that in absence of interference the delectability of targets increases as the

energy transmitted increases. This implies that the only limitation on the detection process is the

presence of thermal noise in front end of receiver. Thus from a delectability point of view the

optimum situation is to transmit as much power as possible for as long as possible.

Unfortunately, this conflict with requirement of resolution: there would be little use in defining

the presence of target, if its whereabouts couldn’t be accurately established. Its well known in

Fourier transform theory that in order to resolve something accurately in time, the bandwidth of

transmitted signal has to be large, that to resolve something accurately in frequency, the time

duration of signal has to be large. This is simply a consequence of time frequency duality.

It might be thought that at first that the approximate time-frequency relationship given for simple

pulses would prevent the accurate simultaneous resolution of both time and Doppler frequency (

i.e. velocity) of radar target , but this need not to be so. It is possible to transmit pulses of a fairly

complicated structure that can resolve well in both time and frequency. This technique is

commonly referred to as pulse compression technique. the basic idea of any pulse radar is to

repetitively transmit a pulse, which need not necessarily be the same pulse each time, the time

interval between pulses needed not be same either. Usually it’s convenient to transmit same

pulse each time, and it is also convenient to make the inter pulse period either same each time or

to make periods related to each other by some simple ratio. A little thoughts shows that

transmitting a pulse train, instead of longer more complicated pulse, as a signal duration

increases, the bandwidth doesn’t decrease hence pulse repetition can be viewed as convenient



way of increasing signal duration, without a proportionate decrease in bandwidth. This enables

pulse train to have good resolution properties in time in frequency. Using pulse trains, it’s

possible to design signals of extreme complexity and yet still use simple equipment. In particular

if the transmitted pulse stream is coherent that is burst of carrier maintains the correct phase

relationship with the last burst, then it is possible to get even higher resolution. This means that

coherent pulse radar can give a good performance even in dense target environment. This way

Problems of ambiguity are encounter both in frequency and time processing.

1.1 General Block diagram

Input from the receiver is series of pulses which is nothing but the reflected echoes from

different objects which is in the range of the radar. These echoes consists various information

about the object like distance, density, temperature. Each radar is designed to perform specific

function that it could detect either weather conditions or distance etc and any other information

obtained is considered as clutter to that radar.

Pulse compression is a signal processing technique mainly used in radar, sonar and echography

to increase the range resolution as well as the signal to noise ratio. This is achieved by

modulating the transmitted pulse and then correlating the received signal with the transmitted

pulse. The simplest signal pulse radar can transmit is a sinusoidal pulse of amplitude, A and

carrier frequency, f0, this pulse is transmitted periodically.

An adaptive system performs the processing by using an architecture having time-varying

parameters on the received signals which accompanies with clutters. An adaptive moving target


INPUT FROM RECEIVER

DISPLAY DETECTION

PULSE COMPRESSION

ADAPTIVE MTI CFAR


detector has been designed to meet the challenges of target detection amidst various levels of

clutter environments. The approach has been used that is able to overcome the inherent

limitations of conventional systems (e.g. Moving Target Indicator, Fast Fourier Transform etc.).

Constant false alarm rate (CFAR) detection refers to a common form of adaptive algorithm used

in radar systems to detect target returns against a background of noise, clutter and interference.

In the radar receiver the returning echoes are typically received by the antenna, amplified, down-

converted and then passed through detector circuitry that extracts the envelope of the signal. ,

unwanted clutter and interference sources mean that the noise level changes both spatially and

temporally. In this case, a changing threshold can be used, where the threshold level is raised and

lowered to maintain a constant probability of false alarm. This is known as constant false alarm

rate (CFAR) detection.

The combination of this group of trolleys and big swells present an excellent opportunity to

practice with our radar. In the image above, sea and rain clutter are turned off. Gain is turned up

to 50 - where we leave it offshore (this is too high for inshore work where there are close targets

surrounding us). Notice the almost solid return from the sea clutter within quarter of a mile of the

boat. ARPA vectors are set to true here, to show us the actual courses of these targets.

Fig 1.1: real time display of radar


http://en.wikipedia.org/wiki/Antenna_(radio)

http://en.wikipedia.org/wiki/Clutter_(radar)

http://en.wikipedia.org/wiki/Noise_(radio)

http://en.wikipedia.org/wiki/Radar

http://en.wikipedia.org/wiki/Adaptive_algorithm


1.2 Objective and Scope of the Project

The main object of our project is to design an Adaptive mti filter with improved performance in

non stationary clutter environment Non stationary clutter processes are characterized by a

space/time dependent statistics. An estimator for pulse to pulse amplitude variations of the clutter

process is therefore introduced. This estimator is then implemented in adaptive mti filter, for the

purpose of adapting filter to the non stationary behavior of clutter process. Results of applying

this adaptive mti filter on real radar data demonstrates that the filter gives an extensive reduction

of clutter detections in radar image. Therefore this design can be used in defence, by aircrafts,

security surveillance etc



CHAPTER 2

LITERATURE SURVEYRadar is an object-detection system that uses radio waves to determine the range, altitude,

direction, or speed of objects. It can be used to detect aircraft, ships, spacecraft, guided missiles,

motor vehicles, weather formations, and terrain. The radar dish (or antenna) transmits pulses of

radio waves or microwaves that bounce off any object in their path. The object returns a tiny part

of the wave's energy to a dish or antenna that is usually located at the same site as the transmitter.

Radar was secretly developed by several nations before and during World War II. The term

RADAR was coined in 1940 by the United States Navy as an acronym for Radio Detection And

Ranging. The term radar has since entered English and other languages as a common noun,

losing all capitalization. The modern uses of radar are highly diverse, including air and terrestrial

traffic control, radar astronomy, air-defense systems, antimissile systems; marine radars to locate

landmarks and other ships; aircraft anti-collision systems; ocean surveillance systems, outer

space surveillance and rendezvous systems; meteorological precipitation monitoring; altimetry

and flight control systems; guided missile target locating systems; and ground-penetrating radar

for geological observations. High tech radar systems are associated with digital signal processing

and are capable of extracting useful information from very high noise level.

Fig 2.1: radar



The performances of the studied designs are investigated by comparing the frequency response

characteristics and the average signal-to-clutter suppression capabilities of the filters with respect

to a number of defined performance measures. Two further approaches are considered to

increase the signal-to-clutter suppression performance. First approach is based on a modified

min-max filter design whereas the second one focuses on the multiple filter implementations.

There has been IEEE paper published on adaptive filter in past year.

The algorithm developed to meet our objective on adaptive mti is different from those mentioned

in IEEE paper for adaptive mti. As adaptive mti will involve non-stationary clutters with change

in radar’s velocity which will make performance of adaptive complex.

CHAPTER 3



OVERVIEW OF THE SYSYTEM

3.1 Moving target indicator

Moving target indication (MTI) is a mode of operation of radar to discriminate a target against clutter. In

contrast to another mode, stationary target indication, it takes advantage of the fact that the target moves

with respect to stationary clutter. The most common approach takes advantage of the Doppler Effect. For

a given sequence of radar pulses, the moving target will change its distance from the radar system.

Therefore the phase of the radar reflection that returns from the target will be different for successive

pulses. This differs from a stationary target (or clutter) which will cause the reflected pulses to arrive at

the same phase shift. Radar MTI may be specialized in terms of the type of clutter and environment:

airborne MTI (AMTI), ground MTI (GMTI), etc., or may be combined mode: stationary and moving

target indication (SMTI).In radar (radio detection and ranging) oriented system which is used for varies

application such as In our project we are trying to locate moving target using mti in this process clutter is

eliminated received at consecutive prf(pulse repetitive frequency).When the wavelength is large

compared to the objects dimension, scattering follows a Rayleigh distribution. When the wavelength is

small compared to the objects dimension, it is in the optical region, where the target can be detected. The

received signal for radar system with an input noise that is Gaussian can be modeled as a Rayleigh

distribution. Moving target indicator (MTI) begins with sampling two successive pulses. Sampling begins

immediately after the radar transmits pulse ends. The sampling continues until the next transmit pulse

begins. Sampling is repeated in the same location for the next transmit pulse, and the sample taken (at the

same distance) with the first pulse is rotated 180 degrees and added to the second sample. This is called

destructive interference. If an object is moving in the location corresponding to both samples, then the

signal reflected from the object will survive this process because of constructive interference. If all

objects are stationary, the two samples will cancel out and very little signal will remain. High power

microwave devices, like crossed-field amplifier, are not phase-stable. The phase of each transmit pulse is

different from the previous and future transmit pulses. This phenomenon is called phase jitter.

3.2 Clutters

Radar clutter is unwanted echoes from the natural environment, which “clutter” the radar and

challenge target detection. Clutter includes echoes from land, sea, weather, and animals. Clutter

is spatially distributed and larger in physical size than the radar resolution cell. Manmade objects

(i.e. buildings), are “point,” or discrete, clutter echoes, that produce large backscatter. Large

clutter echoes can mask echoes from desired targets. When clutter is greater than receiver noise,



clutter signal processing dominates. Echoes from land or sea are examples of surface clutter.

Echoes from rain and chaff are examples of volume clutter. The echo magnitude from distributed

surface clutter is proportional to the area illuminated. To independently model the clutter echo of

the illuminated area, the clutter cross section per unit area, σ0 , is utilized: σ0 = σc Ac (9) where

σc is the radar cross section of the clutter occupying an area Ac. σ0 is a dimensionless quantity,

expressed in decibels with a reference value of one, and is also known as the scattering

coefficient, differential scattering cross section, normalized radar reflectivity, backscattering

coefficient, and normalized radar cross section (NRCS).The spectrum of the ground clutter is

assumed to have narrow width and is centered on zero Doppler velocity

3.2.1 Characteristics of clutter

There are 2 types of clutter. Surface and volume clutter. In our project it is important to know

different characteristics of clutter in order to design our notch filter according to it and to

eliminate clutter at that particular region.

a) Surface clutter

In designing surface clutter models, we develop different models based on the grazing angle

which can be effective to model land and sea clutter. Figure Elevation geometry shows the

extent of the surface illuminated by the radar pulse, (top) plan view showing the illuminated

clutter resolution cell consisting of individual, independent scatters Low Grazing Angle. Figure

depicts radar illuminating the surface at a small grazing angle, ψ. For low grazing angles the

range is determined by the radar pulse width τ. The cell width in the cross-range dimension is

determined by the azimuth beam widthθB and the range R. From the simple radar equation, the

received echo power Pr is Pr = Pt G Ae σ (4π) 2 R4

where Pt = transmitter power,

W G = antenna gain

Ae = antenna effective aperture

m2 R = range m



σ = radar cross section of the scatter

we let Pr = S (received target signal power)

σ = σ t, (target cross section).

The signal power returned from a target is then

S = Pt G Aeσt (4π) 2 R4

When the echo is from clutter, the cross section σ becomes σc = σ0 Ac, where the area Ac of the

radar resolution cell is Ac = R θB (c τ / 2) sec ψ (12) with

θB = two-way azimuth beamwidth,

c = velocity of propagation,

τ = pulsewidth,

ψ = grazing angle (defined with respect to the surface tangent).

The area Ac in range resolution, is (c τ / 2), where the factor of 2 in the denominator accounts

for the two-way propagation of radar. With these definitions, the radar equation for the surface-

clutter echo-signal power C is C = Pt G Ae [θB(c τ / 2) sec ψ] (4π) 2 R3.When the echo from

surface clutter is large compared to receiver noise, the signal-to-clutter ratio is S C =σt σ0 R θB

(c τ / 2) sec ψ. If the maximum range, Rmax, corresponds to the minimum discernible signal-to-

clutter ratio (S/C)min, then the radar equation for the detection of a target in surface clutter at

low grazing angle is Rmax = σt (S/C)min σ0 θB (c τ / 2) sec ψ (15)

Variation of Surface Clutter with Grazing Angle. Figure shows the general form of surface

clutter as a function of grazing angle. There are three different scattering regions. At high

grazing angles, the radar echo is due mainly to reflections from clutter that can be represented as

individual planar facets oriented so that the incident energy is directed back to the radar. The

backscatter can be quite large at high grazing angles. At the intermediate grazing angles,

scattering is somewhat similar to that from a rough surface. At low grazing angles, back

scattering is influenced by shadowing (masking) and by multipath propagation. Shadowing of

the trough regions by the crests of waves prevents low-lying scatters from being illuminated.



Figure General nature of the variation of surface clutter as a function of grazing angle, showing

the three major scattering regions. High Grazing Angle The surface clutter area viewed by the

radar is determined by the antenna beam widthsθB and φB, in the two principal planes. The

clutter illuminated area Ac in is (π/4) R θB R φB / 2 sin ψ, where ψ = grazing angle and R =

range. The factor π / 4 accounts for the elliptical shape of the illuminated area, and the factor of 2

in the denominator is necessary since in this case θB and φB are the one-way beam widths.

Substituting σ = σ0 Ac, letting Pr = C (the clutter echo power), and taking G = π2 θBφB, the

clutter radar equation in this case is C = π PtAe σ0 128 R2 sin ψ.The clutter power is seen to vary

inversely as the square of the range. This equation applies to the echo power received from the

ground by a radar altimeter or the remote sensing radar known as a scatter meter One

complicating factor in the study of clutter is that it means different things in different situations.

For example, to an engineer developing a missile to detect and track a tank, the return from

vegetation and other natural objects would be considered to be “clutter”. However, a remote

sensing scientist would consider the return from natural vegetation as the primary target. Clutter

is thus defined as the return from a physical object or a group of objects that is undesired for a

specific application. Clutter may be divided into sources distributed over a surface (land or sea),

within a volume (weather or chaff) or concentrated at discrete points (structures, birds or

vehicles). The magnitude of the signal reflected from the surface back to the receiver is a

function of the material, roughness and angle. There are three primary scattering types into

which Clutter is generally classified. These are secular, retro and diffuse as shown in the figures

Below.



Fig 3.1: The reflections from the aircraft totally reflected and partially reflected



Fig 3.2 clutter signals being diffracted in all directions

b) Ground Clutter

Because of the statistical nature of clutter, the mean reflectivity is most often quoted. A

Convenient mathematical way to describe this mean value for surface clutter is the

Constant γ model in which the surface reflectivity is modeled as fluctuating proportion is

reflected back to the radar

σ = γ sinψo , (9.18)

Where σo – Reflectivity (cross section per unit area m2

ψ - Grazing angle at the surface (rad),

γ - Parameter describing the scattering effectiveness.



Figure 3.3: Effect of grazing angle on clutter reflectivity for different clutter types

It can be seen from the figure that, at low grazing angles, the measurements fall below the Model

because of propagation-factor effects. At high grazing angles the measured Reflectivity rises

above the value predicted by the model because of quasi specular

Reflections from surface facets.

For different surface types, the following are typical:

• Values for γ between –10 and –15dB are widely applicable to land covered by crops, bushes

and trees.

• Desert, grassland and marsh are more likely to have γ near –20dB

• Urban or mountainous regions will have γ near –5dB

These values are almost independent of wavelength and polarization, but they only apply to

modeling of mean clutter reflectivity.



Thus we could conclude that clutter amplitude is less for sea clutter compared to surface clutters

due to the facts that surface would have many sharp edges like in buildings and vehicles. These

sharp edges will have major scattering whereas the sea clutter would not have sharp edges as

compared to the surface. In surface clutter homogenous clutter occurs in the forest region this is

due to the reflections from the tress in the forest which are evenly spread, thus getting same kind

of echoes. Whereas the echoes from the cities consisting of different kind of vehicles and

buildings would have non homogenous type of scattering.Some of the examples for clutter are

rain and dust.

a) Rain

The graphs below show the theoretical values for the reflectivity as a function of rainfall rate at

different frequencies.

Figure 3.4: Theoretical raindrop reflectivity vs rainfall rate using Marshall Palmer drop size.

This data is determined using the relationship between the reflected and incident power on small

spherical targets as discussed earlier in the section on the RCS of a sphere. Though a given

rainfall rate does not imply a specific drop-size distribution, the trend that the drops get bigger as

the rainfall rate increases, generally holds true. In the Rayleigh region (πD/λ < 1), the RCS is

given by the following formula



Distributions refl = = , (9.28) 2 2 5 4Rσ π πε K , (9.29) D KSinc1− = ε2+64λ

Where ε - Relative dielectric constant of the material,

D – Diameter of the scattering object,

λ - Wavelength 279

When πD/λ > 10 the equation for RCS reduces to the geometric optics form 2 πσ = . (9.30) 4

These equations can be combined with the density of particles in the medium to determine the

total reflectivity, η. Nησ . (9.31) ∑==ii1

b) Dust

The volume of dust that can be supported in the atmosphere is extremely small, and so the

reflectivity can often be neglected for EM radiation with wavelengths of 3mm or more.

However, under certain circumstances, if the dust density is very high (such as in rock crushers)

or if the propagation path through dust is very long (in dust storms), then it can be useful to

determine the reflectivity and the total attenuation.



Figure 3.5: Backscatter from dust after explosion (a) at 10GHz and (b) 35GHz

3.3 Doppler Effect

The Doppler Effect was first recognized by Christian Johann Doppler, who observed that the

color of a luminous body and the pitch of a sounding body are changed by the relative motions of

the body and observer. The Doppler Effect (or Doppler shift) is the change in frequency of a

wave (or other periodic event) for an observer moving relative to its source. It is named after the

Austrian physicist Christian Doppler, who proposed it in 1842 in Prague. It is commonly heard

when a vehicle sounding a siren or horn approaches, passes, and recedes from an observer.

Compared to the emitted frequency, the received frequency is higher during the approach,

identical at the instant of passing by, and lower during the recession.



When the source of the waves is moving toward the observer, each successive wave crest is

emitted from a position closer to the observer than the previous wave. Therefore, each wave

takes slightly less time to reach the observer than the previous wave. Hence, the time between

the arrival of successive wave crests at the observer is reduced, causing an increase in the

frequency. While they are travelling, the distance between successive wave fronts is reduced, so

the waves "bunch together". Conversely, if the source of waves is moving away from the

observer, each wave is emitted from a position farther from the observer than the previous wave,

so the arrival time between successive waves is increased, reducing the frequency. The distance

between successive wave fronts is then increased, so the waves "spread out". For waves that

propagate in a medium, such as sound waves, the velocity of the observer and of the source are

relative to the medium in which the waves are transmitted. The total Doppler effect may

therefore result from motion of the source, motion of the observer, or motion of the medium.

Each of these effects is analyzed separately. For waves which do not require a medium, such as

light or gravity in general relativity, only the relative difference in velocity between the observer

and the source needs to be considered. A very common example is the change in pitch (not the

frequency) of an approaching or receding vehicle with respect to you. The pitch rises as the

oncoming vehicle gets nearer, goes to zero as the vehicle passes (i.e. there is a zero relative

velocity) and then starts to fall as it recedes. Light from moving objects will appear to have

different wavelengths depending on the relative motion of the source and the observer. Observers

looking at an object that is moving away from them see light that has a longer wavelength than it

had when it was emitted (a red shift), while observers looking at an approaching source see light

that is shifted to shorter wavelength (a blue shift). In many radar applications there is a relative

movement between the radar and the target to be detected. Examples include, Air Traffic

Control, Battlefield Surveillance, and Weapon Locating, all Airborne Radars, SAR and ISAR as

well as many others. Consider the example of Air Traffic Control radar. This has to detect

incoming and outgoing aircraft in the presence of a clutter background. We have already seen

that clutter can be distinguished from receiver noise by virtue of its narrower, low frequency

spectrum.



Fig 3.6: Doppler variation

Targets can be distinguished from background clutter by virtue of their motion. This enables a

track history to be built up. More useful, however, is exploitation of the Doppler effect which

enables moving targets to be filtered such that clutter is rejected based upon the differing

velocities of the two received signal components. A processor that distinguishes moving targets

from clutter by virtue of differences in their spectra is called a Moving Target Indicator or MTI.

MTI processors can take a number of forms.

Consider stationary ground based radar observing an approaching aircraft. As the aircraft

approaches each radar pulse travels a shorter and shorter distance, consequently the phase of the

signal is constantly changing with each pulse or target position. The faster the aircraft

approaches the radar the faster the rate of change of the phase of the reflected signal. Thus the

rate of change of the measured phase to the approaching aircraft is relative to the velocity of the

aircraft.



Fig 3.7: The phase represented by the two-way path from radar to target is Φ=2π (2r/λ)

The Doppler shift is -2vfo/c (- sign because an increase in path length represents a phase lag)

Fig 3.8: doppler w.r.t aircraft



3.3.1 Derivation of Doppler frequency formula

The phase shifting φ of an electromagnetic wave from the radar antenna to the aim and back

results from the ratio of the covered distance and the wavelength of the transmitted energy

multiplied with the scale of the full circle (2·π):

This means: In practice the Doppler- frequency occurs twice at radar. Once on the way from the

radar to the aim, and then for the reflected (and already afflicted by a Doppler-shift) energy on

the way back. In the radar signal processing the Doppler frequency will be divided by the actual

transmitted frequency to eliminate the influence of different transmitter’s frequencies. Now the

Doppler frequency is a measure of the radial speed only and is called “normalized”.



Fig 3.9: doppler spread delay

Doppler spread in the spectrum is the delay occurred of the signal echoes returned, which should

have been returned at the actual time. Doppler spread is mainly due to 2 methods

1) Temporal variation: The time taken by the first echo is different from the time taken by the

other echoes. This is due to the wind, dust, moisture between the target and the radar.

2) Spatial variation: This variation occurs due to the change in the position of object with respect

to radars. It may be because the target is moving or may be because the radar platform moving or

may be the relative motion between the radar and the target.

3.4 Blind speeds

It can be seen that the frequency response has zeroes at Doppler shifts corresponding to



These can also be thought of in terms of target movement such that the pulse-to-pulse phase

changes by 2pi, in other words that the target range changes by l/2 in one PRI.

This also says that to keep the first blind speed above the highest likely target velocity, it is

necessary to use a high PRF. This conflicts with the requirement to use a low PRF in order to

avoid range ambiguities. This is an example of the tradeoffs involved in the choice of PRF in the

design of the radar modulation scheme, and is covered in more detail when considering the

ambiguity function.

3.5 Adaptive moving target indicator

The Adaptive Moving Target Indicator (AMTI) is an adaptive filter automatically suppressing

the unwanted reflection echo (clutter) from the ocean surface, clouds, and rain contained in the

radar receive signal. In this paper, with a view to assuring the suppression of the clutter in the

case where the AMTI is applied to search radar that can acquire only several samples of received

signals with assuring coherence, a burst averaging type AMTI is proposed. In this AMTI, the

received signals without coherence from the same direction intermittently obtained by beam

scanning are used for weighting control. From the performance evaluation by computer

simulation, it is proven that the clutter suppression capability can be improved by about 3 dB in

comparison with the case without burst averaging for a steady single peak clutter. It is also

clarified that the clutter suppression capability can be improved by 0.3 to 1.8 dB by burst

averaging with two to five scans even in the case where the center frequency of the clutter

changes for each beam scan as long as the frequency variation width is less than 12% with



respect to the pulse repetition frequency. In many radar applications there is a relative movement

between the radar and the target to be detected. Examples include, Air Traffic Control,

Battlefield Surveillance, and Weapon Locating, all Airborne Radars, SAR and ISAR as well as

many others. Consider the example of Air Traffic Control radar. This has to detect incoming and

outgoing aircraft in the presence of a clutter background. We have already seen that clutter can

be distinguished from receiver noise by virtue of its narrower, low frequency spectrum.



CHAPTER 4

FILTERSFilters are a basic component of all signal processing and telecommunication systems. The

primary functions of a filter are one or more of the followings: (a) to confine a signal into a

prescribed frequency band or channel for example as in anti-aliasing filter or a radio/TV channel

selector, (b) to decompose a signal into two or more sub-band signals for sub band signal

processing, for example in music coding, (c) to modify the frequency spectrum of a signal, for

example in audio graphic equalizers, and (d) to model the input-output relation of a system such

as a mobile communication channel, voice production, musical instruments, telephone line echo,

and room acoustics.

Filters are widely employed in signal processing and communication systems in applications

such as channel equalization, noise reduction, radar, audio processing, video processing,

biomedical signal processing, and analysis of economic and financial data. For example in a

radio receiver band-pass filters, or tuners, are used to extract the signals from a radio channel. In

an audio graphic equalizer the input signal is filtered into a number of sub-band signals and the

gain for each sub-band can be varied manually with a set of controls to change the perceived

audio sensation. In a Dolby system pre-filtering and post filtering are used to minimize the effect

of noise. In hi-fi audio a compensating filter may be included in the preamplifier to compensate

for the non-ideal frequency-response characteristics of the speakers. Filters are also used to

create perceptual audio-visual effects for music, films and in broadcast studios.

The primary functions of filters are one of the followings:

(a) To confine a signal into a prescribed frequency band as in low-pass, high-pass, and band-pass

filters.

(b) To decompose a signal into two or more sub-bands as in filter-banks, graphic equalizers, sub-

band coders, frequency multiplexers.

(c) To modify the frequency spectrum of a signal as in telephone channel equalization and audio

graphic equalizers.

(d)To model the input-output relationship of a system such as telecommunication channels,

human vocal tract, and music synthesizers.



4.1 Non-Recursive or Finite Impulse Response (FIR) FiltersA non-recursive filter has no feedback and its input-output relation is given by

y(m)=∑k=0

M

bk x(m−k )

Fig 4.1: direct-form finite impulse response (FIR)

The output y (m) of a non-recursive filter is a function only of the input signal x(m). The

response of such a filter to an impulse consists of a finite sequence of M+ 1 sample, where M is

the filter order. Hence, the filter is known as a Finite-Duration Impulse Response (FIR) filters.

Other names for a non-recursive filter include all-zero filter, feed-forward filter or moving

average (MA) filter a term usually used in statistical signal processing literature.

4.1.1 FIR filter types which can be used to remove clutters are:

a) Low pass filterLow pass filters can be used for many applications. One area in which these filters can be

used is on the output of digital to analogue converters where they are able to remove the

high frequency alias components. However they can be used in many other areas where it

is necessary to pass the low frequency components of the signal, but remove the

unwanted high frequency elements. Active low pass filters are capable of providing a

relatively high level of performance for a small number of components.



Fig 4.2: Low pass filter basic response curve

The shape of the curve is of importance with features like the cut-off frequency and roll off being

key to the operation. The cut-off frequency is normally taken as the point where the response has

fallen by 3dB as shown. Another important feature is the final slope of the roll off. This is

generally governed by the number of 'poles' in the filter. Normally there is one pole for each

capacitor inductor in a filter. When plotted on a logarithmic scale the ultimate roll-off becomes a

straight line, with the response falling at the ultimate roll off rate. This is 6dB per pole within the

filter.

Consider the design of a low-pass linear-phase digital FIR filter operating at a sampling rate of

Fs Hz and with a cutoff frequency of Fc Hz. The frequency response of the filter is given by

The impulse response of this filter is obtained via the inverse Fourier integral as



b) High pass filterAs the name implies, a high pass filter is a filter that passes the higher frequencies and rejects

those at lower frequencies. In other words, high-frequency signals go through much easier

and low-frequency signals have a much harder getting through, which is why it's a high pass

filter. This can be used in many instances, for example when needing to reject low frequency

noise, hum, etc. from signals. This may be useful in some audio applications to remove low

frequency hum, or within RF to remove low frequency signals that are not required.

Fig 4.3: High pass filter basic response curve

The shape of the curve is of importance. One of the most important features is the cut-off

frequency. This is normally taken as the point where the response has fallen by 3dB.

Consider the design of a high-pass linear-phase digital FIR filter operating at a sampling rate of

Fs Hz and with a cutoff frequency of Fc Hz. The frequency response of the filter is given by




c) Band pass filter As the name implies a band pass filter is one where only a given band of frequencies is allowed

through. All frequencies outside the required band are attenuated. There are two main areas of

interest in the response of the filter. These are the pass-band where filter passes signals and the

stop-band where signals are attenuated. As it is not possible to have an infinitely steep roll off,

there is an area of transition outside the pass-band where the response is falling but has not

reached the required out of band attenuation. Band pass filters are used in many areas of

electronics. They are particularly widely used for RF applications where tuned circuits are used.

However for lower frequencies, active band pass filters provide an effective means of making a

filter to pass only a given band if frequencies. For these filters the most widely used active

element is an operational amplifier, or op amp. These op amp band pass filters are easy to design

and construct, requiring only a minimum of components. In addition to this, these active band

pass filters provide a very effective level of performance.

Fig 4.4: frequency response of band pass filter



Consider the design of a band-pass linear-phase digital FIR filter operating at a sampling rate of

Fs Hz and with a lower and higher cutoff frequencies of FL and FH Hz. The frequency response

of the filter is given by


d) Band stop filter A band stop filter is a circuit that ideally filters out signals with frequencies in a certain range.

This range can be quite large, depending on inherent characteristics of the circuit. The smaller

the range of frequencies the circuit filters, the higher the Q factor it is said to have. Band stop

filters with high Q Factors are also called notch filters. The band pass filter passes one set of

frequencies while rejecting all others. The band-stop filter does just the opposite. It rejects a band

of frequencies, while passing all others. This is also called a band-reject or band-elimination

filter. Like band pass filters, band-stop filters may also be classified as (i) wide-band and (ii)

narrow band reject filters. The narrow band reject filter is also called a notch filter. Because of its

higher Q, which exceeds 10, the bandwidth of the narrow band reject filter is much smaller than

that of a wide band reject filter.



Fig 4.5: frequency response of band stop filter

Consider the design of a band-stop linear-phase digital FIR filter operating at a sampling rate of

Fs Hz and with a lower and higher cutoff frequencies of FL and FH Hz. The frequency response

of the filter is given by




4.1.2 Filter using window techniques

The window method is most commonly used method for designing FIR filters. The simplicity of

design process makes this method very popular. A window is a finite array consisting of

coefficients selected to satisfy the desirable requirements. This provides a few methods for

estimating coefficients and basic characteristics of the window itself as well as the result filters

designed using these coefficients. The point is to find these coefficients denoted by w[n].

When designing digital FIR filters using window functions it is necessary to specify:

A window function to be used; and

The filter order according to the required specifications (selectivity and stop band

attenuation).

These two requirements are interrelated. Each function is a kind of compromise between the two

following requirements:

The higher the selectivity, i.e. the narrower the transition region; and

The higher suppression of undesirable spectrum, i.e. the higher the stop band attenuation.

The table 4.1 below gives the equations for different window types.

Window Type Weight Equation

Rectangular

Bartlett

Hanning



Hamming

Blackman

Frequency response and weight values of different window types

Fig 4.6 frequency response and weight values



4.2 Recursive or Infinite Impulse Response (IIR) FiltersA recursive filter has feedback from output to input, and in general its output is a function of the

previous output samples and the present and past input samples as described by the following

equation

y(m)=∑k =1

N

ak y (m−k )+∑k=0

M

bk x (m−k )

Fig 4.7: direct-form pole-zero IIR filter

Fig shows a direct form implementation of IIR. In theory, when a recursive filter is excited by an

impulse, the output persists forever. Thus a recursive filter is also known as an Infinite Duration

Impulse Response (IIR) filters. Other names for an IIR filter include feedback filters, pole-zero

filters and auto-regressive-moving-average (ARMA) filter a term usually used in statistical

signal processing literature.



4.3 Comparison of FIR and IIR filters

FIR filter uses only current and past input digital samples to obtain a current output sample

value. It does not utilize past output samples. Simple FIR equation is mention below.

y(n)= h(0)x(n) + h(1)x(n-1) + h(2)x(n-2) + h(3)x(n-3) + h(4)x(n-4)

IIR filter uses current input sample value, past input and output samples to obtain current output

sample value. Simple IIR equation is mention below.

y(n)= b(0)x(n) + b(1)x(n-1) + b(2)x(n-2) + b(3)x(n-3) + a(1)y(n-1) + a(2)y(n-2) +a(3)y(n-3)

Transfer function of FIR filter will have only zeros, need more memory, while transfer

function of IIR filter will have both zeros and poles and will require less memory than

FIR counterpart.

FIR filters are preferred due to its linear phase response and also they are non-recursive.

Feedback is not involved in FIR, hence they are stable. IIR filters are not stable as they

are recursive in nature and feedback is also involved in the process of calculating output

sample values.

FIR filter consume low power and IIR filter need more power due to more coefficients in

the design.

IIR filters have analog equivalent and FIR have no analog equivalent.

FIR filters are less efficient while IIR filters are more efficient.

FIR filters are used as anti-aliasing, low pass and baseband filters. IIR filters are used as

notch (band stop), band pass functions.



FIR filter need higher order than IIR filter to achieve same performance. Delay is more

than IIR filter. It has lower sensitivity than IIR filter. These are disadvantages of FIR

filters.

4.4 FIR Notch Filter

Digital signal processing (DSP) techniques have rapidly developed in the recent years due to

advances in digital computer technology and integrated circuit fabrication [3], [26], [27]. The use

of digital circuits yields high speed as well as high reliability, and also permits us to have

programmable operations. DSP techniques and applications in a variety of areas such as speech

processing, data transmission on telephone channels, image processing, instrumentation, bio-

medical engineering, seismology, oil exploration, detection of nuclear explosion, and in the

processing of signals received from the outer space, besides others. Various types of digital

filters, such as Low pass (LP), High-pass (HP), Band-pass (BP), Band-stop (BS), and Notch

filters (NF), and various types of digital operations such as Differentiation, Integration and

Hilbert transformation, to mention a few, are invariably used in many of the applications just

mentioned. In this review paper, we focus our attention on the design and performance analysis

of notch filters.

The FIR notch filter equation is given by which we are using in our code to eliminate the clutter

is

H (n) = 1- 2cos (w0)e− jw+e−2 jw

4.4.1 Notch filter characteristicsA notch filter highly attenuates/ eliminates a particular frequency component from the input

signal spectrum while leaving the amplitude of the other frequencies relatively unchanged. A

notch filter is, thus, essentially a band stop filter with a very narrow stop band and two pass

bands. The amplitude response, H1 (w), of a typical notch filter (designated as NF1) is shown in

Fig. 4.8 and is characterized by the notch frequency, wd (in radians) and 3-dB rejection

bandwidth, BW. For an ideal notch filter, BW should be zero, the pass band magnitude should be

unity (zero dB) and the attenuation at the notch frequency should be infinite.



Fig 4.8: The amplitude response H1 (w) of notch filter

4.5 Power PC

Power PC 8641 or 8641D is the hardware we have used in our project. It is a 32 bit RISC

processor and 128 bit active engine. The power PC consists of 12 8641D processors in number.

The radar data is too huge that must be processed, in order to process this huge data they are fed

to all these 12 processors. There are 2 types of feeding the huge data to the processors

1) Data slicing

Out of all the huge data received, a single output which is taken in the form single

matrix is sliced into different parts. Each of these parts of matrix is fed to all the

processors to perform the operations. After this is processed now the processor is

ready to take the next set of signal in the form of matrix.

2) Pipelining

Out of all huge data received each single output is fed to single processor rather

than slicing it. The received signal in the form of single matrix is given to one of

the processor. The next signal in the form of single matrix is given to the next

processor and so on. Thus each processor receives different set of data to be

processed. This method helps fast computation of data.



The MPC8641D uses two high-performance superscalar e600 cores running at up to 1.5 GHz.

This three-issue machine has a compact 7-stage pipeline which is particularly efficient with code

that branches unpredictably. It avoids the extensive delays associated with flushing a long

pipeline on mispredicted branches. Unpredictable branching is typical of code paths driven by

largely random arrival of different types of packets. These processors support up to 8 out-of-

order instructions on the system bus that allows for making forward progress even while waiting

for previous instructions to finish (ie, access to main memory required). The e600 has an on-

board 128-bit vector processor for efficient data movement (useful for copying TCP payloads

from kernel space to user space) and for math functions that rival a DSP. With a large backside

L2 cache for each core, the e600 benefits from high bandwidth and low latency between the

processor and the L2 cache. There are two flexible high-performance I/O ports. Dual 8-lane PCI

Express ports leverage PCI legacy with a high-performance serial point-to-point link that is

commonly used to connect to a variety of other on-board high-performance devices. There are

two flexible high-performance I/O ports. Dual 8-lane PCI Express ports leverage PCI legacy

with a high-performance serial point-to-point link that is commonly used to connect to a variety

of other on-board high-performance devices. There are four Ethernet controllers, supporting 10

Mbps, 100 Mbps, and 1000 Mbps. The Ethernet controllers have advanced capabilities for TCP

and UPD checksum acceleration, QoS support, and packet header manipulation. Each Ethernet

controller can be converted into a FIFO mode for high-efficiency ASIC connectivity. The

MPC8641D supports flexible software implementations: symmetric multiprocessing (SMP) and

Asymmetric multiprocessing. With SMP, one operating system runs on both cores, but from a

programming perspective, it appears that the developer is writing a program for a single core.

With Asymmetric multiprocessing, two instances of the same operating system or two entirely

separate operating systems can be run on the two cores, largely unaware of each other.

The following gives overview of the MPC8641 key feature set:

High-performance, 32-bit superscalar microprocessor that implements the PowerPC ISA

Eleven independent execution units and three register files

Branch processing unit (BPU)

Four integer units (IUs) that share 32 GPRs for integer operands

64-bit floating-point unit (FPU)



Four vector units and a 32-entry vector register file (VRs)

Three-stage load/store unit (LSU)

Fig 4.9: PowerPC processing nodes

Fig 4.10: 8641 D

Chapter 5Dept of TCE, BNMIT Page 39

http://www.google.co.in/url?sa=i&rct=j&q=&esrc=s&frm=1&source=images&cd=&cad=rja&uact=8&ved=0CAcQjRw&url=http://www.imicrosystem.com/RuggedsolutionVXSCPUBoard.html&ei=TUA_VYjDO8STuATg7YGwBQ&bvm=bv.91665533,d.c2E&psig=AFQjCNFg6zLTI12_tQET6FWZjHFwLgu2iw&ust=1430294975774915

http://www.google.co.in/url?sa=i&rct=j&q=&esrc=s&frm=1&source=images&cd=&cad=rja&uact=8&ved=0CAcQjRw&url=http://www.cwcdefense.com/products/dsp-fpga/power-architecture-dsp/champ-av6.html&ei=_UA_VebXA4KNuASig4DwBg&bvm=bv.91665533,d.c2E&psig=AFQjCNFg6zLTI12_tQET6FWZjHFwLgu2iw&ust=1430294975774915


IMPLENTATIONThe beginning flow of the project was done in 4 main stages:

1. To generate a data signal matrix with complex values and noise in Matlab.

2. Adding the target and clutter to the same signal

3. Design a filter with suitable specifications to remove the clutters

4. Pass the signal though the filter.

We have generated the data in Matlab through the noise matrix using the function

normrnd which created random noise with the specifications like mean, variance,order of

the matrix which had to be specified. Each column represents 1 sweep in the prf. Noise in

the matrix has no exact amplitude value, thus we use standard deviation which

determines the amplitude of noise. The noise amplitude is 6-8 times of standard

deviation. Mean defines the range of the random values that has to be generated for the

matrix. The order of the matrix was 128 X 16.

For the matrix which consisted of random noise matrix has to be added with the target

and the clutter. The target could be desired object of interest. The target of sinusoidal

signal with different phase was added to all the 16 columns. The clutter of sinusoidal

signal with same phase was added to all the columns.

The filter which had to design was basically meant to remove the stationary and non

stationary clutters associated. To remove clutters we choose windows, as the windows

could be designed for any order. In order to do this we first designed windows like

Kaiser, Blackman, hamming windows using Matlab to check if it could eliminate the

clutters. But these windows had a very low order and very less bandwidth. Due to the less

bandwidth it could eliminate very few clutters. Because of this reason we choose band

stop and band pass filters, and we could eliminate most of the clutters but in these filters

we needed two frequencies which couldn’t be related with the velocity of target and

clutter. We later came up with a new filter design which overcame all the problems like

the bandwidth, design the filter for any of the order also it could relate the relative

velocity of the clutter and the target. This filter is known as notch filter.



This notch filter which we had designed gave very good results compared to all the other

filters. In this stage we had to pass the signal data generated to the designed filter and

remove the clutters. There are mainly 2 steps to do this.

a) To shift the filter to the position wherever the clutter is present.

b) To shift the data signal to wherever the filter is present

Fig 5.1: represents the 2 cases for removing the clutter.

In implementation of the project where we have to remove 3 clutters from the received data from

the radar. The position of the clutter is generated by hardware which makes use of parameters

like velocity of radar and PRF that is pulse repetitive frequency using these parameters Doppler

of clutter is calculated. As radar is moving stationary clutter in case of MTI behave as non-

stationary clutter shifted at certain Doppler ,Doppler of clutter is found using velocity and prf as

in case of MTI where stationary clutters used to lie near zero Doppler in this case due to adaptive

condition clutter will shift at Doppler frequency . knowing this we can remove clutter at exact

position ,to remove a certain frequency we make use of fir notch filter as its stable as compared

to iir notch filter if used for such applications. To make use of notch filter we design Doppler

dependent filter where the notch frequency is same as Doppler frequency used to remove a

particular clutter at particular frequency or Doppler. In our case the position generated for 3

clutters are at Doppler zero, prf/4, and prf/2. Using these generated Doppler we design notch

frequency at these positions and remove all 3 perfectly. The filter response is generated for

particular number of pulses which depends on size of data received from radar.



Having used power pc as our hardware we converted the software simulation done on mat lab to

c code which is supported by visual box software used for power pc in lrde. Processing radar

signals require many power pc‘s working parallel, as radar signal obtained consist of large

amount of data which once received is compressed and signal to noise ratio is checked then

further depending of particular application its processes, the application used in our project

consist of finding particular position of clutter depending on Doppler principal and removing it

therefore processing time is less as this is 2-3% of the total processing that happens. For

hardware implementation in c code different software supported function are used for filter

performance and its effectiveness on received signal taking care of parameter used.



5.1 FLOWCHART


READ INPUT DATA FROM RADAR

READ SIZE OF INPUT MATRIX

READ AND STORE SIZE OF COMPLEX MATRIX

DESIGN FILTER FOR REQUIRED NUMBER OF PULSES

GENERATE COMPLEX MATRIX FROM INPUT

PLOT THE OUTPUT

MULTIPLY FILTER RESPONSE WITH INPUT DATA

PLOT INPUT DATA

MAKE DOPPLER DEPENDANT FILTER TO REMOVE REQUIRED

DOPPLER

CALCULATE DOPPLER OF CLUTTER BASED ON VELOCITY

OF RADAR


In our project input consist of clutter and target signals obtained from radar. As input signal is

obtained we first read the signal from hardware and store its size for further processing. The

input signal is in form of matrix therefore its dimension is store for further processing. We get to

know total number of pulses knowing the size of matrix. Now the read data is converted into

complex form as radar signals are complex in nature. Input data is then plotted. Then specific

filter is chosen to remove particular clutter to do so first position of clutter is read from hardware

using Doppler principle and velocity of radar. Now filter is linked or made dependent on these

factors where giving Doppler frequency as read as input of the signals obtained they can be

eliminated. Using doppler frequency we calculate notch frequency which would eliminate that

particular position where clutter lies.

According to Doppler frequency formula that is fd =2*vr*ft/c where vr is velocity of radar, ft as

transmitted frequency. Now omega for Doppler frequency is calculated and given as notch

frequency. The response of notch filter is then plotted for total number of pulses obtained from

input data. After calculating response these are multiplied by input data obtained number of prt

used for different clutter signals. In the end output for each case with different doppler is plotted

to show that at different doppler of clutter it can be removed.



CHAPTER 6

TEST RESULTS

Fig6.1: received signal from radar

This is the matlab plot obtained from the reflections from three different targets which are

located in different positions. These positions are determined by its prf, velocity and the distance

between the radar and the target. The x axis represents the position of the object measured from

the Doppler and the y axis represents the magnitude of the targets. These received signals are

passed through the designed filter to eliminate the undesired targets which is nothing but, called

as clutters.



6.2 Notch filter response at 0 doppler is obtained by using fir notch filter equation

6.3 Output of notch filter performed on received data



The first fig 6.2 represents plot for the notch filter obtained for particular frequency response, in

this case it is at 0 Doppler. The received signal consisting of three targets is passed through this

respective filter. At this specification of the filter the clutter signal at 0th position is eliminated.

The x axis represents the frequency and y axis represents the magnitude. The position of the

filter can be changed according to the prf. PRF is the frequency range beyond which a target

could not be located. This is known as pulse repetitive frequency. PRF is calculated by the time

required by the radar to hit the target and receive back the reflections from the objects.



Fig 6.4: Notch filter response at prf/4 doppler is obtained by using fir notch filter equation

6.5 output of notch filter performed on received data for clutter at prf/4 doppler



The first fig 6.4 represents plot for the notch filter obtained for particular frequency

response, in this case it is at prf/4 Doppler. The received signal consisting of three targets is

passed through this respective filter. At this specification of the filter the clutter signal at prf/4 th

position is eliminated. The x axis represents the frequency and y axis represents the magnitude.

The position of the filter can be changed according to the prf. PRF is the frequency range beyond

which a target could not be located. This is known as pulse repetitive frequency. PRF is

calculated by the time required by the radar to hit the target and receive back the reflections from

the objects.



6.6 notch filter response at prf/2 doppler is obtained by using fir notch filter equation

6.7 output of notch filter performed on received data for clutter at prf/2 doppler



The first fig 6.6 represents plot for the notch filter obtained for particular frequency response, in

this case it is at prf/2 Doppler. The received signal consisting of three targets is passed through

this respective filter. At this specification of the filter the clutter signal at prf/2 th position is

eliminated. The x axis represents the frequency and y axis represents the magnitude. The position

of the filter can be changed according to the prf. PRF is the frequency range beyond which a

target could not be located. This is known as pulse repetitive frequency. PRF is calculated by the

time required by the radar to hit the target and receive back the reflections from the objects.



CHAPTER 7

CONCLUSIONThe MTI system which was used is incapable of detecting targets at low frequency also they

eliminated only stationary clutters which made the moving clutters wrongly interpreted as

targets. It also removed targets associated with stationary clutter which appeared at 0 doppler. As

a result, a new method is developed where the filter is made automatically to recognize the

clutter characteristics and adjust itself to a condition to remove only the moving clutters. This is

known as adaptive filter.

In this project, a simple and efficient technique for adaptive moving target detector using a

notch filter is presented. The implementation is to enhance adaptive filtering capability during

radar target detection in an ever changing environment. The proposed design is highly effective

even in very strong clutter conditions. Since it is very difficult to extract target embedded with

non-homogeneous clutters, using the relative velocity between the target and clutter is

successfully used to remove clutters. The MATLAB simulation results for adaptive MTI shows

that the proposed design works satisfactorily in various clutter environments.

An adaptive MTI filter that is unique, since the filter weights are estimated in the same resolution

cell as is filtered. Additionally the algorithm is very simple and fast. The average Doppler shift

must be estimated separately.



REFERENCES

[1] Bowyer D B, Rajaw: kai.an P K and Oebhart W W. Aahptive Clwer

Autoregtvssive Spe~:&~l Estimation, IBEE Trans. Aerospace and Eleattonic

Sptems, \tol.A@b15 No. 4 July 1979

[2]JArmstrong B C, GriFfith White R G. Perfomurn processots in heten Rada&

Sonar Navig, Vol. 142, No 4 August 1995

[3]Scolnik M I (1990): R& Hundbook, Second Edition, Mdjraw-Hill, ISBN0-07

057913-X

[4]Schleher D C (1991 :): A427 and Pulsed Doppler R h, Artech HOUS~, ISBN0-

89006-320-6.

[5] Fundamentals of radar system: Byron eddy


PROJECT REPORT1 (1) new

Documents