DETECTION OF RADAR SIGNALS IN NOISE

UNIT - 5

www.manaresults.co.in
www.manatutor.com
 INTRODUCTION

• The two basic operations performed by radar are

1. detection of the presence of reflecting Objects

2. Extraction of information from the received

waveform to obtain such target data as position,
velocity, and perhaps size.

• Noise ultimately limits the capability of any radar.

www.manaresults.co.in
www.manatutor.com
 MATCHED FILTER RECEIVER
• A network whose frequency-response function
maximizes the output peak-signal-to-mean-noise
(power) ratio is called a matched filter.

• The frequency-response function, denoted H(f),

expresses the relative amplitude and phase of the output
of a network with respect to the input when the input is
a pure sinusoid.

• If the bandwidth of the receiver pass band is wide

compared with that occupied by the signal energy,
extraneous noise is introduced by the excess bandwidth
which lowers the output signal-to-noise ratio.
www.manaresults.co.in
www.manatutor.com
Contd…,
• If the receiver bandwidth is narrower than the
bandwidth occupied by the signal, the noise energy
is reduced along with a considerable part of the
signal energy. The net result is again a lowered
signal-to-noise ratio.

• Thus there is an optimum bandwidth at which the

signal-to-noise ratio is a maximum.

• The receiver bandwidth B should be approximately

equal to the reciprocal of the pulse width τ valid
only for pulsed waveforms but not to other types of
waveforms. www.manaresults.co.in
www.manatutor.com
Contd…,
• The second detector and video portion of the well-
designed radar super heterodyne receiver will have
negligible effect on the output signal-to-noise ratio if
the receiver is designed as a matched filter.
• Narrow banding is most conveniently accomplished
in the IF.
• For a received waveform s(t) with a given ratio of
signal energy E to noise energy No (or) noise power
per hertz of bandwidth, North showed that the
frequency-response function of the linear, time-
invariant filter. www.manaresults.co.in
www.manatutor.com
• The noise that accompanies the signal is assumed
to be stationary and to have a uniform spectrum
(white noise). It need not be gaussian.
• The frequency-response function of the matched
filter is the conjugate of the spectrum of the
received waveform except for the phase shift
exp (- j2∏ftl). This phase shift varies uniformly
www.manaresults.co.in
with frequency. www.manatutor.com
Contd…,
• The frequency spectrum of the received signal may
be written as an amplitude spectrum│S(f)│ and a
phase spectrum exp [- jφs(f)].

• The matchedfilter frequency-response function

may similarly be written in terms of its amplitude
and phase spectra │H( f )│ and exp [-jφm( f )].

www.manaresults.co.in
www.manatutor.com
Contd…,
• The amplitude spectrum of the matched filter is the
same as the amplitude spectrum of the signal, but
the phase spectrum of the matched filter is the
negative of the phase spectrum of the signal plus a
phase shift proportional to frequency.
• The matched filter may also be specified by its
impulse response h(t), which is the inverse Fourier
transform of the frequency-response function.

www.manaresults.co.in
www.manatutor.com
Contd…,

• Since S*(f) = S(-f), we have

• the impulse response of the matched filter is the

image of the received waveform.

www.manaresults.co.in
www.manatutor.com
 Matched filter Characteristics and
Correlation function
• The output of the matched filter is not a replica of
the input signal
• The output of the matched filter may be shown to be
proportional to the input signal cross-correlated with
a replica of the transmitted signal, except for the
time delay t1.
• The crosscorrelation function R(t) of two signals
y(λ) and s(λ), each of finite duration, is defined as

www.manaresults.co.in
www.manatutor.com
• The output yo(t) of a filter with impulse response
h(t) when the input is yin(t) = s(t) + n(t) is

• If the filter is a matched filter, then h(λ) = s(t1 - λ)

and Eq. above becomes

• Thus the matched filter forms the cross correlation

between the received signal corrupted by noise and
a replica of the transmitted signal.
• The replica of the transmitted signal is "built in" to
the matched filter via the frequency-response
www.manaresults.co.in
function. www.manatutor.com
• If the input signal yin(t) were the same as the signal
s(t) for which the matched filter was designed, the
output would be the autocorrelation function.
• The autocorrelation function of a rectangular pulse
of width τ is a triangle whose base is of width 2τ.

www.manaresults.co.in
www.manatutor.com
 DETECTION CRITERIA
• The detection of weak signals in the presence of
noise is equivalent to deciding whether the receiver
output is due to noise alone or to signal-plus-noise.

• This is the type of decision probably made

(subconsciously) by a human operator on the basis
of the information present at the radar indicator.

• When the detection process is carried out

automatically by electronic means the radar
detection process was described in terms of
threshold detection. www.manaresults.co.in
www.manatutor.com
• There are two types of errors that might be made in the
decision process.

• This occurs whenever the noise is large enough to

exceed the threshold level. In statistical detection theory
it is sometimes called a type I error. The radar engineer
would call it a false alarm.

• A type II error is one in which the signal is erroneously

considered to be noise when signal is actually present.
This is a missed detection to the radar engineer.

• The setting of the threshold represents a compromise

between these two types of errors.
www.manaresults.co.in
www.manatutor.com
• Neyman-Pearson observer:
• Most radars utilize the equivalent of the Neyman-
Pearson Observer and operate with a fixed number of
pulses.

• The threshold level was selected so as not to exceed a

specified false-alarm probability; that is, the probability
of detection was maximized for a fixed probability of
false alarm.

• This is equivalent to fixing the probability of a type I

error and minimizing the type II error.

• This type of threshold detector is sometimes called a

www.manaresults.co.in
Neyman-Pearson Observer.www.manatutor.com
• Likelihood-ratio receiver:
• The likelihood ratio is an important statistical tool
and may be defined as the ratio of the probability-
density function corresponding to signal-plus-noise,
Psn(ν),to the probability-density function of noise
alone, pn(ν).

• It is a measure of how likely it is that the receiver

envelope ν is due to signal-plus-noise as compared
with noise alone.
www.manaresults.co.in
www.manatutor.com
• If the likelihood ratio Lr(ν) is sufficiently large, it would
be reasonable to conclude that the signal was indeed
present.

• Thus detection may be accomplished by establishing a

threshold at the out put of a receiver which computes
the likelihood ratio.
• Inverse probability receiver:
• This method has been applied by Woodward and Davies
to the reception of signals in noise.

• It is based upon the application of Bayes' rule for the

probability of causes.
www.manaresults.co.in
www.manatutor.com
• The problem of forming the best estimate of the
cause of the event is a problem in inverse
probability.

• A limitation of the method of inverse probability

based on the application of Bayes' rule is the
difficulty of specifying the a priori probabilities.

• whenever the a priori probabilities are known, the

inverse-probability method may be used with
confidence.

• When the a priori probabilities are not known, the

www.manaresults.co.in
likelihood-ratio test is usually employed.
www.manatutor.com
• Relationship of inverse probability and
likelihood-ratio receivers:

• The likelihood ratio follows from inverse probability

if the assumption is made that the a priori
probabilities are equally likely.

• Both the methods may be implemented by

computing the cross-correlation function between
the received signal and the signal si(t).
www.manaresults.co.in
www.manatutor.com
• ldeal observer:

• Idea Observer was formulated by Siegert.

• The criterion of the Ideal Observer maximizes the

total probability of a correct decision (or) minimizes
the total probability of an error.

• The probability of an error is

www.manaresults.co.in
www.manatutor.com
• These probabilities must be known beforehand in
the Ideal Observer.
• The Ideal Observer applies equal weight to an error
due to a false alarm and to an error due to a miss.
• If the errors are not of equal importance, the
theory of the Ideal Observer may be modified to
take this into account using the concepts of
www.manaresults.co.in
statistical decision theory
www.manatutor.com
• Sequential observer:

• Sequential Observer makes an observation of the

receiver output and, on the basis of this single
observation, decides between one of three choices:

(1)The receiver output is due to the presence of signal

with noise

(2)The output is due to noise alone; or

(3)The available evidence is not convincing enough to

make a decision between choices 1 and 2
www.manaresults.co.in
www.manatutor.com
• The term sequential detection is sometimes used
synonymously with Sequential Observer.

• Sequential detection has also been used to describe a

two-step that can be employed with phased-array
radar.

www.manaresults.co.in
www.manatutor.com
DETECTOR CHARACTERISTICS
• The portion of the radar receiver which extracts the
modulation from the carrier is called the detector.

• One form of detector is the envelope detector, which

recognizes the presence of the signal on the basis of the
amplitude of the carrier envelope. All phase information
is destroyed.

• The zerocrossings detector destroys amplitude

information.

• The coherent detector is an example of one which uses

both phase and amplitude.
www.manaresults.co.in
www.manatutor.com
• Envelope detector:
• The function of the envelope detector is to extract
the modulation and reject the carrier. By eliminating
the carrier, all phase information is lost and the
detection decision is based solely on the envelope
amplitude.

• The envelope detector consists of a rectifying

element and a low-pass filter to pass the modulation
frequencies but to remove the carrier frequency.

• In the linear detector the output signal is directly

proportional to the input envelope.
www.manaresults.co.in
www.manatutor.com
• in the square-law detector, the output signal is
proportional to the square of the input envelope.

• Assume that there are n independent pulses with

envelope amplitudes v1, v2, . . . , on available from
the radar receiver.

• The probability-density function for the envelope of

11 independent noise samples is the product of the
probability-density function for each sample.

www.manaresults.co.in
www.manatutor.com
www.manaresults.co.in
www.manatutor.com
• Logarithmic detector:

• If the output of the receiver is proportional to the

logarithm of the input envelope, it is called a
logarithmic receiver.

• The detection characteristics (probability of detection as

a function of the probability of false alarm, signal-to-
noise ratio, and the number of hits integrated) for the
logarithmic receiver have been computed by Green,
following the methods of Marcum.

• It might be used to prevent receiver saturation or to

reduce the effects of unwanted clutter targets in certain
www.manaresults.co.in
types of non-MTI radar receivers
www.manatutor.com
• Zero-crossings detector:
• The information contained in the zero crossings of
the received signal can, be used for detecting the
presence of signals in noise.

• The greater the signal-to-noise ratio the less will be

the average number of zero crossings.
• The average number of zero-crossings per second at
the output of a narrow-bandpass filter of rectangular
shape when the input is a sine wave in gaussian
noise is

www.manaresults.co.in
www.manatutor.com
• Coherent detector:

www.manaresults.co.in
www.manatutor.com
• The coherent detector consists of a reference
oscillator feeding a balanced mixer.

• The input to the mixer is a signal of known

frequency f, and known phase plus its
accompanying noise.

• The reference-oscillator signal is assumed to have

the same frequency and phase as the input signal to
be detected.

www.manaresults.co.in
www.manatutor.com
• The output of the mixer is followed by a low-pass
filter which allows only the d-c and the low-
frequency modulation components to pass while
rejecting the higher frequencies in the vicinity of the
carrier.

• Although the coherent detector may be of superior

sensitivity than other detectors it is seldom used in
radar applications since the phase of the received
signal is not usually known.

www.manaresults.co.in
www.manatutor.com
 AUTOMATIC DETECTION
• The function of the radar operator viewing the
ordinary radar display is to recognize the presence
of targets and extract their location.

• When the function is performed by electronic

decision circuitry without the intervention of an
operator, the process is known as automatic
detection.

• The automatic detector has also been called plot

extractor and data extractor.
www.manaresults.co.in
www.manatutor.com
• Usually there are four basic aspects to automatic
detection:

(1) The integration of the pulses received from the

target.
(2) The detection decision; and the determination of
the target location in.
(3) Range.
(4) Azimuth.

www.manaresults.co.in
www.manatutor.com
• Binary moving-window detector:

www.manaresults.co.in
www.manatutor.com
• it has also been called double-threshold detector, m-
out-of-n detector, coincidence detector, sliding-
window detector, and binary integrator.

• it has the advantage of simplicity.

• The radar video is passed through a threshold

detector, which may be thought of as a bottom
clipper.

• Only those signals whose amplitude exceeds the

preset threshold are allowed to pass.
www.manaresults.co.in
www.manatutor.com
• This is the first of two thresholds, hence the name
double-threshold detector.

• The output of the first threshold is sampled by the

quantizer at least once per range-resolution cell.

• A standard pulse is generated if the video waveform

exceeds the first threshold, and nothing if it does not
.
• These are designated by 1 or 0 respectively.

• Thus the output of the quantizer is a series of 1’s and

0’s.

• This is then separated into separate range cells by the

www.manaresults.co.in
range gate. www.manatutor.com
• The number m is the second threshold to be passed
in the double-threshold detector. The two
thresholds must be selected jointly for best
performance.
• Advantages:
1. it is less sensitive to the effects of a single large
interference pulse that might exist along with
the target echo pulses.
2. An analysis of the problem of detection of
www.manaresults.co.in
signals in clutter www.manatutor.com
Figure: Optimum number of pulses m opt (out of a maximum of
n) for a binary moving window detector. A constant
www.manaresults.co.in
www.manatutor.com
(nonfluctuating) target is assumed.
• Tappeddelay-line integrator:

Figure: Pulse integrator using tapped delay line.

www.manaresults.co.in
www.manatutor.com
• This form of integrator is also known as the moving-
window detector or analog moving-window
detector.

• The time delay through the line is made equal to the

total integration time, and the taps are spaced at
intervals equal to the pulse-repetition period.

• The number of taps equals the number of pulses to

be integrated.

• The outputs from each of the taps are tied together

to form the sum of the previous n pulses.
www.manaresults.co.in
www.manatutor.com
• It is similar to the binary moving-window detector.

• It does not suffer the 1.5 to 2 dB loss of the binary

detector.

• It can be employed to estimate the target's angle by

beam splitting.

• The angular location of the target can be estimated from

the output of this detector by taking the midpoint
between the first and last crossings of the detection
threshold, or by taking the maximum value of the
running sum.

• The accuracy of the location measurement is only about

20 percent worse than theoretical.
www.manaresults.co.in
www.manatutor.com
• Recirculating-delay-line integrator:

• The delay-line integrator, or moving-window

detector, described above requires that a number of
pulses, equal to that expected from the target, be
held in storage.

• A simplification can be had by recirculating the

output through a single delay line whose delay is
equal to the pulse repetition period.

• The recirculating delay-line integrator is also called

as feedback integrator, adds each new sweep to the
www.manaresults.co.in
sum of all the previous sweeps
www.manatutor.com
Recirculating-delay-line integrator, or feedback integrator, k = loop
www.manaresults.co.in
gain < 1. (a) single delay loop; (h) double loop: (c) two-pole filter.
www.manatutor.com
• To prevent unwanted oscillations, or "ringing," due
to positive feedback, the sum must be attenuated by
an amount k after each pass through the line.

• The factor k is the gain of the loop formed by the

delay line and the feedback path.

• It must be less than unity for stable operation. The

effect of k < 1 is that the integrator has imperfect
“memory.”

• The optimum value of k depends on the number of

pulses received from the target.
www.manaresults.co.in
www.manatutor.com
• The factor k in the recirculating-delay-line
integrator of Fig. a is equivalent to an exponential
weighting of the received pulses.

• It results in a loss of about 1.0 dB in signal-to-noise

ratio as compared with the ideal postdetection
integrator that weights the received pulses in direct
proportion to the fourth power of the antenna beam
pattern.

• It is 0.5 dB less efficient than the moving window

detector with uniform weights.
www.manaresults.co.in
www.manatutor.com
• The double-loop integrator of Fig.b is a two-pole
filter with a multiple pole.

• It is about 0.3 dB less efficient than the ideal

integrator with optimum weights.

• The double delay-tine configuration of Fig. c is

also a two-pole filter but unlike the double-loop
integrator, the two poles need not be at the same
location.

• Its detection performance is only 0.15 dB less

efficient than the optimum.
www.manaresults.co.in
www.manatutor.com
• The recirculating delay-line integrators can be used
to obtain an estimate of the angular location of the
target.

• The target can be found, as in beam splitting, by

taking the midpoint between the start and end of
the threshold crossing.

• Its relatively good angle-estimating accuracy, good

detection performance, along with the relative
simplicity of a feedback integrator makes the two-
pole filter a good choice as an automatic detector
for scanning radars.
www.manaresults.co.in
www.manatutor.com
 CONSTANT-FALSE-ALA RM-RATE (CFAR)
RECEIVER
• The threshold at the output of a radar receiver, is
chosen so as to achieve a desired false-alarm
probability.

• The false-alarm rate is quite sensitive to the

threshold level.

• If changes in the false-alarm rate are gradual, an

operator viewing a display can compensate with a
manual gain adjustment.
www.manaresults.co.in
www.manatutor.com
• A CFAR may be obtained by observing the noise or
clutter background in the vicinity of the target and
adjusting the threshold in accordance with the
measured background.

www.manaresults.co.in
www.manatutor.com
• The output of the test cell is the radar output.

• The spacing between the taps is equal to the range

resolution.

• The outputs from the delay line taps are summed. This
sum, when multiplied by the appropriate constant,
determines the threshold level for achieving the desired
probability of false alarm.

• Thus the threshold varies continuously according to the

noise or the clutter environment found within a range
interval surrounding the range cell under observation.
www.manaresults.co.in
www.manatutor.com
• This form of CFAR has sometimes been called Adaptive
Video Threshold, or AVT.

• The threshold is determined by whichever of the two sums is

the greater.

• Typically the number of taps used in a cell-averaging CFAR

might vary from 16 to 20.

• If the target echo is large, energy can spill over into the
adjacent range-resolution cells and affect the
measurement of the average background.

• Fur this reason, the range cells surrounding the test cell
are often omitted when averaging the background.
www.manaresults.co.in
www.manatutor.com
• There are several other methods for achieving CFAR
besides the use of cell averaging.

1. One of the first CFAR receivers used is

postdetection integrator to estimate the average
noise.

2. Another CFAR technique is the hard limiter,

sometimes called the Dicke fix.

• Advantage: CFAR is widely used to prevent clutter

and noise interference from saturating the display of
an ordinary radar and preventing targets from being
www.manaresults.co.in
obscured. www.manatutor.com
• Disadvantages:

• It introduces an additional loss compared to

optimum detection, and in some systems the number
of pulses processed needs to be large to keep the
loss low.

• CFAR maintains the false-alarm rate constant at the

expense of the probability of detection. Thus, it
causes targets to be missed.

• The operator is usually given no indication that there

may be missed detections.
www.manaresults.co.in
www.manatutor.com
 Efficiency of non matched filters
• In practice the matched filter cannot always be
obtained exactly. It is appropriate, therefore, to
examine the efficiency of non matched filters
compared with the ideal matched filter.
• The measure of efficiency is taken as the peak
signal-to noise ratio from the non matched filter
divided by the peak signal-to-noise ratio (2E/No)
from the matched filter.
• It can be seen that the loss in SNR incurred by use
of these non-matched filters is small.

www.manaresults.co.in
www.manatutor.com
Fig.1 Efficiency, relative to a matched filter, of a single-tuned
resonant filter and a rectangular shaped filter, when the input signal
www.manaresults.co.in
τ. B =filter bandwidth.
is a rectangular pulse of widthwww.manatutor.com
Table 1 Efficiency of non-matched filters compared
with the matched filter
www.manaresults.co.in
www.manatutor.com
Matched filter with Non-White noise
• In the derivation of the matched-filter characteristic,
the spectrum of the noise accompanying the signal
was assumed to be white; that is, it was independent
of frequency.
• If this assumption were not true, the filter which
maximizes the output signal-to-noise ratio would not
be the same as the matched filter.
• It has been shown that if the input power spectrum
of the interfering noise is given by [Ni (f)]2, the
frequency-response function of the filter which
maximizes the output signal-to-noise ratio is
www.manaresults.co.in
www.manatutor.com
• When the noise is non-white, the filter which
maximizes the output signal-to-noise ratio is called
the NWN (non-white noise) matched filter.
• For white noise [Ni(f)]2 = constant and the NWN
matched-filler frequency-response function of Eq.
above reduces to that of Eq. discussed earlier in
white noise. Equation above can be written as

www.manaresults.co.in
www.manatutor.com
• This indicates that the NWN matched filter can be
considered as the cascade of two filters.

• The first filter, with frequency-response function

l/Ni(f), acts to make the noise spectrum uniform, or
white.

• It is sometimes called the whitening filter.

• The second is the matched filter when the input is

white noise and a signal whose spectrum is
S(f )/Ni(f ).
www.manaresults.co.in
www.manatutor.com