Page 1
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
1SLIDEBen-Gurion Univ. Jan 26, 2016
Periodic and a-periodic on-off coded waveforms for
non-coherent RADAR and LIDAR
Nadav Levanon
Tel Aviv University, Israel
Radar SymposiumBen-Gurion Univ., 25 January 2016
With contributions from:
Itzik Cohen, Tel Aviv univ.; Avi Zadok and Nadav Arbel, Bar-Ilan univ.; J. Mike Baden, GTRI
Page 2
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
2SLIDEBen-Gurion Univ. Jan 26, 2016
With the knowledge accumulated in coherent radar, we enrich the branch of non-coherent radar.
Page 3
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
3SLIDEBen-Gurion Univ. Jan 26, 2016
Non-coherent pulse compression
• On-off keying transmitted signal +.
Can utilize saturation amplifier, pulsed oscillator (e.g., magnetron) and especially laser +.
Random phase of each sub-pulse is acceptable and even advantageous.
• Basic compression codes+:
Bipolar a-periodic or periodic codes (Barker, Ipatov, Legendre, m-sequences …)
• Simple envelope detection in receiver +.
Doppler tolerance + (but no Doppler information -).
SNR loss due to non-coherency -.
Sensitivity to multi-scatterer targets -.
• Non-coherent integration of many pulses (summing) is much
simpler + than coherent integration (which requires FFT), but less efficient -.
• Low-sidelobe response requires mismatched processor.
Additional SNR loss because of the mismatch -.
Page 4
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
4SLIDEBen-Gurion Univ. Jan 26, 2016
Receiver block diagram
BPF | |p
bn b2
∑Output
Rectified output b1
p =1 Magnitude detector
p =2 Magnitude square detector
Page 5
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
5SLIDEBen-Gurion Univ. Jan 26, 2016
PERIODIC NON-COHERENT WAVEFORMS
Page 6
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
6SLIDEBen-Gurion Univ. Jan 26, 2016
Periodic on-off waveforms (Example: Laser range finder)
Classical approach: Time Of Flight (TOF); Amplitude (power) limited (Amax)
time
A
0 tpTr
Target illumination time,i rT M T
1max 2
, unambiguousrangerR CT
, rangeresolutionpR t
2 2
average max2p r pP A t T A t C R
2 2 2 2 2 2
on target max
Assumes the beam apperture on target is smaller than the target
4p r p i r p iP MA t T A t T T A t TC R
2 2 2
on target max4p iP A t TC R
Implies a conflict between power-on-
target and unambiguous range.
Increasing A:
• Complicates hardware
• Increases probability of intercept
Possible solution:
Add more pulses within Tr without
reducing the unambiguous range.
Page 7
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
7SLIDEBen-Gurion Univ. Jan 26, 2016
Example of periodic on-off waveform: Barker 13 !!
Perfect Periodic Cross-Correlation (PPCC)
S = 1 1 1 1 1 0 0 1 1 0 1 0 1
R = 1 1 1 1 1 b b 1 1 b 1 b 1 , b=-2
Page 8
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
8SLIDEBen-Gurion Univ. Jan 26, 2016
N 3 4 5 7 11 13
b -1 -2 -3 -1 -1 -2
N = 4 5 13
Code type m-seq & Legendre m-seq & Legendre Legendre
22 1 4 1 1 32 1 4 2 1 4 3 1
All unipolar Barker codes can exhibit PPCC (with the proper b in their reference)
All unipolar m-sequences exhibit PPCC, with their corresponding bipolar reference (b = -1).
m-sequences are available at lengths , N. Takeuchi, et. al. “Random modulation
cw lidar,” Applied Optics, Vol. 22, No. 9, 1 May 1983, pp 1382-1386.
2 1, 2,3,4,...mN m
m-sequences (shift register sequences)
Legendre sequences
All unipolar Legendre sequences exhibit PPCC, with their corresponding bipolar reference (b = -1).
Legendre sequences are available at lengths 4 1, 1,2,3,4,... ; is a primeP k k P
All unipolar modified Legendre sequences exhibit PPCC, with their corresponding bipolar reference
(b = -1).
Modified Legendre sequences are available at lengths 4 3, 2,3,4,... ; is a primeP k k P
Modified Legendre sequences
Ipatov sequences
Page 9
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
9SLIDEBen-Gurion Univ. Jan 26, 2016
function [s,r] = perfect_periodic_Legendre_on_off( N )
% Generates a periodic coded signal using Legender and modified Legendre sequences
% The signal exhibits perfect periodic crosscorrelation with its reference
% N is any odd prime
Nspt=sprintf('%g element Legendre on-off waveform ',N);
if isprime(N)==0
disp('Not a prime')
return
end
s=ones(1,N);
if rem((N+3)/4,1)==0 % modified Legendre
s(mod((1:N-1).^2,N)+1)=0;
r=s*2-1;
s(1)=0;
else % Legendre
s(mod((1:N-1).^2,N)+1)=0;
r=s*2-1;
end
d=abs(ifft(fft(s).*conj(fft(r))));
figure, plot(d, 'k')
title(['Periodic cross-correlation of ' Nspt ]);
end
Page 10
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
10SLIDEBen-Gurion Univ. Jan 26, 2016
On-off signal based on Legendre 19
>>[s,r]=perfect_periodic_Legendre_on_off(19);
>> disp([s' r'])
1 1
0 -1
1 1
1 1
0 -1
0 -1
0 -1
0 -1
1 1
0 -1
1 1
0 -1
1 1
1 1
1 1
1 1
0 -1
0 -1
1 1
On-off signal based on modified Legendre 17
>>[s,r]=perfect_periodic_Legendre_on_off(17);
>> disp([s' r'])
0 1
0 -1
0 -1
1 1
0 -1
1 1
1 1
1 1
0 -1
0 -1
1 1
1 1
1 1
0 -1
1 1
0 -1
0 -1
Page 11
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
11SLIDEBen-Gurion Univ. Jan 26, 2016
Example of periodic on-off waveform: m-sequence 15
S = 0 1 1 1 1 0 0 0 1 0 0 1 1 0 1
R = b 1 1 1 1 b b b 1 b b 1 1 b 1, b=-1
Perfect Periodic Cross-Correlation (PPCC)
Page 12
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
12SLIDEBen-Gurion Univ. Jan 26, 2016
Interference between two targets (simulations: m-sequence 127, Legendre 131)
In CW waveforms the returns from two targets always coincide, even if the delay difference is large.
The on-off non-coherent case is especially sensitive to the resulting interference.
The simulation included different random phase of each sub-pulse out of the 650,000 subpulses.
(code-length * number of periods processed * number of averages 650,000 ).
No additive noise.
Page 13
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
13SLIDEBen-Gurion Univ. Jan 26, 2016
Page 14
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
14SLIDEBen-Gurion Univ. Jan 26, 2016
Page 15
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
15SLIDEBen-Gurion Univ. Jan 26, 2016
The modulo-2 sum of an m-sequence and another phase (i.e. time-delayed version)
of the same sequence yields yet a third phase of the sequence.
Page 16
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
16SLIDEBen-Gurion Univ. Jan 26, 2016
18 averages using the same code 18 averages of using 18 different codes
Page 17
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
17SLIDEBen-Gurion Univ. Jan 26, 2016
Performances with noise (simulation), Ipatov 121
Periodic Cross-correlation
output. The reference
contained 3 periods.
Section of signal ( 60 elements),
5 samples per code element
Page 18
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
18SLIDEBen-Gurion Univ. Jan 26, 2016
Laser range finder experiment setup outside the Music Department at Bar-Ilan campus.
Page 19
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
19SLIDEBen-Gurion Univ. Jan 26, 2016
Experimental results with a laser range finder (4003 element on-off Legendre signal)
R2=R0
R
R1=R0 -1.139m
R3=R0 +1.229m
R0 = 273m
R 1.2m
Optical transmitted power = 22.5dBm.
Sub-pulse width = 1ns dR=15cm
Output after averaging 1000
consecutive measurements.
Target illumination time = 0.2s
(= 10-9 4003 49 1000)
Unambiguous range = 600m
Target (white paper) relative forward
power reflectivity = 0.07 .
Page 20
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
20SLIDEBen-Gurion Univ. Jan 26, 2016
Unique noise behavior[s‘ r']
1 1
0 -1
1 1
1 1
0 -1
0 -1
0 -1
0 -1
1 1
0 -1
1 1
0 -1
1 1
1 1
1 1
1 1
0 -1
0 -1
1 1
In each period, (N+1)/2 envelope
detected samples (all positive),
are multiplied by 1 and (N-1)/2
samples are multiplied by -1 .
Page 21
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
21SLIDEBen-Gurion Univ. Jan 26, 2016
Sadogursky, A. and Levanon, N. “Performances of Marcum’s (S+N)-N integration scheme
with fluctuating targets”, IEEE Trans. AES, Vol. 50, No.1, Jan. 2014, pp. 319-328.
Page 22
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
22SLIDEBen-Gurion Univ. Jan 26, 2016
Noise only. Correlation
with {1,-1} reference.
Noise only. Correlation
with {1,0} reference.
Field trial at Tel Baruch coast, with a
modified FURUNO magnetron marine
radar (magneton project with Elisra.
Project manager Erez Ben-Yaacov).
Page 23
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
23SLIDEBen-Gurion Univ. Jan 26, 2016
Extending the un-ambiguous range of a simple magnetron radar by a factor of 3
Transmitted non-coherent pulse train: 1 1 0 1 1 0 1 1 0 …... (every 3rd pulse is blocked)
Reference pulse train: 1 1 -1 1 1 -1 1 1 -1 …...
Clutter and targets will
replicate after 3 PRIs,
rather than after 1 PRI.
PRI
Legendre 3 and
its reference
Pulse fluctuations can
slightly distort the
perfect periodic cross-
correlation.
Page 24
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
24SLIDEBen-Gurion Univ. Jan 26, 2016
Field trial near the port of Ashdod, with a
modified FURUNO magnetron marine radar
(magneton project with Elisra. Project
manager Erez Ben-Yaacov).
Page 25
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
25SLIDEBen-Gurion Univ. Jan 26, 2016
Near-clutter
replicating at a
range of 12km
(corresponding
to the PRI)
Ship target at a
range of 13.1 km
Correlation with {1,0} reference
Page 26
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
26SLIDEBen-Gurion Univ. Jan 26, 2016
Correlation with {1,-1} referenceShip target at a
range of 13.1 km
Page 27
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
27SLIDEBen-Gurion Univ. Jan 26, 2016
A-PERIODIC NON-COHERENT WAVEFORMS
Example: Compressing a single long pulse by internal on-off coding
Page 28
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
28SLIDEBen-Gurion Univ. Jan 26, 2016
Manchester Encoded Unipolar Barker 13
1 0 1 0 1 0 1 0 1 0 0 1 0 1 1 0 1 0 0 1 1 0 0 1 1 0
Trans
Ref
Unipolar Barker 13: 1 1 1 1 1 0 0 1 1 0 1 0 1
Page 29
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
29SLIDEBen-Gurion Univ. Jan 26, 2016
A-periodic cross-correlations between a unipolar Barker 13 and a mismatched reference
Unipolar Barker 13: 1 1 1 1 1 0 0 1 1 0 1 0 1
The mismatched filter
(MMF) used, is a
sequence of length
65 (=5*13) optimized
for minimum integrated
sidelobes (SL), with
higher weight given to
the minimization of the
near SL.
The MMF can get any
value, both positive
and negative.
Page 30
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
30SLIDEBen-Gurion Univ. Jan 26, 2016
Reduced duty cycle of the sub-pulses (1) Better range resolution (2) Higher bandwidth
(3) Lower energy in the compressed pulse
Page 31
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
31SLIDEBen-Gurion Univ. Jan 26, 2016
Fig. 1. Flow cytometry
setup implementing spatial
modulated emission.
ON-OFF SIGNAL - USE IN OPTICAL MASKS
Page 32
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
32SLIDEBen-Gurion Univ. Jan 26, 2016
Page 33
Nadav Levanon, Tel-Aviv UniversityOn-Off Waveforms
33SLIDEBen-Gurion Univ. Jan 26, 2016
Thank you