Radar Project

Pulse Compression Radar

By: Hamdi M. Joudeh and Yousef Al-Yazji

Supervisor: Dr. Mohamed Ouda

1
Introduction:

Radars can be classified according to

the waveforms:
- Continuous Wave (CW) Radars.
- Pulsed Radars (PR).

We are concerned in Pulsed Radars:

- Train of pulsed waveforms.
- Transmitted periodically.
2
Basic Concepts:
 Target Range: R= cΔt / 2
 Inter pulse period (IPP) and Pulse repetition
frequency (PRF): PRF=fr=1/IPP
 Duty Cycle = dt = t ⁄ T, Pav = Pt × dt.

3
Basic Concepts:

 Range ambiguity:

4
Basic Concepts: Range resolution:

5
Pulse Compression:

 Short pulses are used to increase range

resolution.
 Short pulses = decreased average power.
 Decreased average power=Decreased
detection capability.
 Pulse compression = Increased average
power + Increased Range resolution.

6
Advantages of pulse compression:

Maintain the pulse repetition

frequency (PRF) .
The avoidance of using high peak
power.
Increases the interference immunity.
Increases range resolution while
maintaining detection capability.
7
The concept of pulse compression:

 1- Generation of a coded waveform:

(various types).
 2- Detection and processing of the echo:
(achieved by a compression filter).

 The actual compression process takes

place in the receiver by the matched filter
or a correlation process.
8
Methods of implementation:

 Active generation and processing:

9
Methods of implementation:

 Passive generation and processing:

10
Types of pulse compression:
Linear FM: Advantages
Easiest to generate.
The largest number of generation and
processing approaches.
SNR is fairly insensitive to Doppler
shifts.

11
Linear FM: Disadvantages
 Range-doppler cross coupling.

12
Types of pulse compression:
Linear FM: The process
 LFM the transmitted pulse.

 Receiver: matched filter.

 compression ratio is given by B*T

13
Linear FM: Up and Down Chirp

14
Linear FM: Compression

 Compression Ratio=T/t.
 ∆R = C*t/2.
 Higher Compression Ratio = Better range
resolution.
 Compression Ratio=B*T .
 wideband LFM modulation = Higher
compression ratio.

15
Linear FM: Example
 Overlapped received waveforms:

16
Linear FM: Example

 Detected pulses (output of matched filter)

17
Phase Coded: Introduction

 Long Pulse with duration(T) divided to (N)

coded sub-pulses with duration(t).
 Uncoded pulse (T), ∆R = C*T/2.
 Duration of compressed pulse = duration
of sub-pulse = t.
 Compression ratio = B*T = T/t.
 New ∆R = C*t/2 (better).

18
Phase Coded: Codes used

 binary codes, sequence of either +1 or -1.

 Phase of sinusoidal carrier alternates
between 0° and 180° due to sub-pulse.

19
Phase Coded: Codes used

 Must have a minimum possible side-lobe

peak of the aperiodic autocorrelation
function.

20
Phase Coded: Barker code

 Optimal binary sequence, pseudo-random.

 Pseudo-random = deterministic .
 Pseudo-random has the statistical
properties of a sampled white noise.

21
Phase Coded: Auto correlation
function of the Barker sequence
 Peak = N, 2Δt wide at base.

22
Phase Coded: Detection and
compression
compressed pulse is obtained in the
receiver by correlation or matched
filtering.
compression ratio = N = T/t.
half-amplitude width = t = sub-pulse
width.
∆R = C*t/2.
23
Phase Coded: Auto Correlation
MATLAB example.
 Two un-coded overlapped long pulses.

24
Phase Coded: MATLAB work

 Two barker coded overlapped long pulses.

25
Implementation of Biphase-Coded
System Using MATLAB:

26
Implementation of Biphase-Coded
System Using MATLAB:
 Why I and Q detection?

27
Software steps and approaches:
Waveform Generation:
 Required inputs:
- Barker code sequence.
- Maximum Range. (to calc. IPP).
- Range resolution. (to calc. pulse width).

28
Waveform Generation:

29
Path and Receiver losses:

 Radar equation:
 Modified: L= Radar losses

 RCS of 0.1 and 0.08 m2

 Ranges = 60 and 61 Km
 F = 5.6 GHz,
 G = 45dB
 L= 6dB
30
Path and Receiver losses:

31
Added Noise:
 Implementing AWGN, a major challenge.
 We need the standard deviation, σ2 = No/2.

 K=Boltzmann’s constant, and Te=effective noise temperature.

32
Added Noise:

 Calculate (SNR)I from

Te=290K, Pt=1.5 MW.
 Substitute in
Using the actual E in MATLAB, sum(signal2).
And Bt = #of subpulses.

 MATLAB function randn().

 Noise = σ*randn(# of noise samples)

33
Added Noise:

34
Detection:

 Matched filter, I and Q detection.

35
Correlation:
 Result:

36
Observations:

Calculating the range difference:

 Between the two peeks 130 samples.
 Δt = samples*Ts. Where Ts= 5*10-8 sec.
 ΔR = Δt* C / 2, ΔR = 975m.
 Error of 2.5%

37
Observations:

- For 500m difference:

 ΔR = 520m.
 Error = 4%.

- Error
 ΔR decreases, the error increases.
 Error due to noise and sampling time.

38
References:

 Radar Handbook - 2nd Ed. - M. I. Skolnik.

 MATLAB Simulations for Radar Systems
Design, Bassem R. Mahafza and Atef Z.
Elsherbeni.
 Digital Communications - Fundamentals and
Applications 2nd Edition - Bernard Sklar.
 http://mathworld.wolfram.com/BarkerCode.ht
ml

39
Thank You for your attention

40