Você está na página 1de 24

Performance Study of Hybrid

DS/FFH Spread-Spectrum
Systems in the Presence of
Multipath Fading and MultipleAccess Interference
Mohammed M. Olama
Teja P. Kuruganti
Steven F. Smith
Computational Sciences & Engineering Division
Oak Ridge National Laboratory
Xiao Ma
Electrical Engineering and Computer Science Dept.
University of Tennessee
IEEE CQR 2012

Outline

Introduction

Types of SS Systems

Multipath Fading Channel

Direct Digital Synthesizers (DDS)

Performance Evaluation of DS/FFH

Numerical Results

Conclusion

Managed by UT-Battelle
for the Department of Energy

ORNLs Solution: Hybrid Spread Spectrum


Novel technique (3 Patents); adaptive, programmable.
HSS: a synergistic combination of DS, FH, and TH.
Advantages:

Adaptive Hybrid Spread-Spectrum (HSS) modulation format


combines DSSS and frequency/time hopping in a multidimensional, orthogonal signaling scheme.
Capable of excellent LPI, LPD & security properties
(programmable).
Adaptive, robust protocol for high QoS applications.
Can be operated in burst mode for very low power drain.
Superior resistance to multipath and jamming (high process
gain).
Easily deployed with modern chip technology.
Compliant with existing FCC/NTIA/ETSI rules for ISM bands.
Ideally implemented via modern FPGA-based electronics, ASICs,
and SDR techniques.
Managed by UT-Battelle
for the Department of Energy

Typical HSS Applications

Managed by UT-Battelle
for the Department of Energy

DIRECT-SEQUENCE SPREAD-SPECTRUM
Local PN Clock

PN Clock
PN Sequence
Generator

Carrier

DS
Signal

1
Wide
BP Filter

Data

Power
Spectral
Density

Local
Carrier

PN Sequence
Generator

Power
Spectral
Density

Narrow
BP Filter

Phase
Demod

fc

Narrow spectrum at
output of modulator
before spreading

Managed by UT-Battelle
for the Department of Energy

Frequency

Data
Clock

Power
Spectral
Density

Spread
RFI

RFI

Frequency

Data

fc

Spectrum has wider bandwidth


and lower power density after
spreading with PN sequence
(PN Rate >> Data Rate)

Frequency

fc

Original narrowband, high


power density spectrum is
restored if local PN sequence is
same as and lined up with
received PN sequence

FREQUENCY-HOPPING SPREAD-SPECTRUM
Local PN Clock

PN Clock
PN Sequence
Generator

Data

PN Sequence
Generator

FH
Signal

RF
Frequency
Synthesizer

Wide
BP Filter

Power
Spectral
Density

Power
Spectral
Density

RF
Frequency
Synthesizer

RFI

t4 t2 t6 t0 t3 t5 t1
Frequency

fc

Narrow spectrum at
output of modulator
while hopping

Managed by UT-Battelle
for the Department of Energy

Frequency

fc

Spectrum has same bandwidth


and power density after hopping
with PN sequence (PN Rate <<
Data Rate for standard FHSS)

Narrow
BP Filter

Data
Demod.

Data
Data
Clock

RFI
causes
errors

Data
Output

t0 t1 t2 t3 t4 t5 t6
Time (hops)

Original narrowband, high


power data stream is restored if
local PN sequence is same as
and lined up with received PN
sequence

SLOW HYBRID SPREAD-SPECTRUM (DS/SFH)


Local PN Clock

PN Clock
PN Sequence
Generator

PN Sequence
Generator

DS/FH
Signal

RF
Frequency
Synthesizer

Data

Wide
BP Filter

Power
Spectral
Density

Power
Spectral
Density

RF
Frequency
Synthesizer

RFI

t4 t2 t6 t0 t3 t5 t1
Frequency

fc

DS (wide) spectrum at
output of modulator
while hopping

Managed by UT-Battelle
for the Department of Energy

Frequency

fc

Spectrum has same bandwidth


and power density after hopping
with PN sequence (PN Rate <<
Data Rate for standard HSS;
>> for FastHSS)

Narrow
BP Filter

Data
Demod.

Data
Data
Clock

Data
Output

RFI
Reduced
by DS
PG

t0 t1 t2 t3 t4 t5 t6
Time (bits)

Original narrowband, high


power data stream is restored if
local PN sequence is same as
and lined up with received PN
sequence

FAST HYBRID SPREAD-SPECTRUM (DS/FFH)


FastHSS

PN Clock
PN Sequence
Generator

PN Sequence
Generator

DS/FH
Signal

RF
Frequency
Synthesizer

Data

t11 t20
t40 t61

Frequency

fc

DS (wide) spectrum at
output of modulator
while hopping

RF
Frequency
Synthesizer

Wide
BP Filter

Power
Spectral
Density

Power
Spectral
Density

Local PN Clock

Frequency

RFI

t21
t60

t00 t30
t31 t41

Narrow
BP Filter

Managed by UT-Battelle
for the Department of Energy

Data
Data
Clock

Data
Output

t01
t50

t10
t51

fc

Spectrum has same bandwidth


and power density after hopping
with PN sequence (PN Rate >>
Data Rate for FastHSS; e.g., as
above, 2 hops per bit).

RFI
Reduced
by DS
PG!

t00 t01t10 t11 t20 t21 t30 t31 t40 t41t50 t51t60 t61

Time (chips)

Original narrowband, high


power data stream is restored if
local PN sequence is same as
and lined up with received PN
sequence

Gp(FH/DS) dB = Gp(FH) dB + Gp(DS)dB = 10 log (no. of hopping channels) + 10 log (BWDS/Rinfo)


8

Data
Demod.

HSS is a Multidimensional Signal


HSS can be defined in 3 axes (code, frequency,
and time).
Each dimension is orthogonal with the others.
Permissible signal spaces along an axis may also be ~
Frequency
orthogonal.
Codes
Frequencies
Time slots

(C4,F2,t0)

Code
(C2,F1,t0)

Easily adaptable to exploit


many degrees of freedom to

(C3,F0,t1)
(C0,F0,t0)

(C1,F2,t2)

meet system requirements.


Some signal overlaps may be orthogonal.
9

Managed by UT-Battelle
for the Department of Energy

Time

MULTIPATH PROPAGATION

LONG-PATH REFLECTION

TRANS.

RCVR.
SHORT-PATH REFL.

Differential delays are approx. 1 ns per foot


Short-path delays (indoors) are typically < 0.1s
Spread-spectrum modulation can largely cancel long-path effects

ORNL is developing new techniques to mitigate short-path degradations!


10 Managed by UT-Battelle
for the Department of Energy

Direct Digital Synthesizers (DDS)


Stored Phase
Phase
Increment

Adder

Delay

Adder

Adder

Phase

Lookup

Quantization

Table

NCO

Phase

Phase

Offset

Quantization
Error

Dither

The implementation of a DDS has two distinct parts:


A phase accumulator accumulates the phase
increment and adds in the phase offset.
The DDS output is then calculated by quantizing the
results of the phase accumulator section and using
them to select values from a lookup table.
11 Managed by UT-Battelle
for the Department of Energy

RF Coexistence Multi-User, Jamming,


and Multipath Fading Performance
The error probability of one hop is:
K

Number of users

Number of FH channels

Number of hops per bit

K 1

Ph = Pk ( j users )
j =0

T
T = b Duration of each hop
L
T
Tc = b Chip duration for PN sequence
NL

K 1

{P ( j users, no jam) + P ( j users, ful jam) + P ( j users, partial jam)}


k

j =0

K 1

= {P( j users, no jam) P k ( | j users, no jam)

Period of the

WDS

Bandwidth of DS waveform

+ P( j users, ful jam) P k ( | j users, ful jam)

WJ

Bandwidth of the wideband jamming

+ P( j users, partial jam) P k ( | j users, partial jam)}

Number of hopping channels corrupted by jamming

WJp

Part of the bandwidth of the channel partially corrupted by jamming

PN sequence

j =0

The error probability of one bit is:


Pb
=
12 Managed by UT-Battelle
for the Department of Energy

L
L+1 d ( Ph )d (1 Ph ) Ld

d=
L

RF Coexistence Multi-User, Jamming,


and Multipath Fading Performance
P

Transmitted signal power

Number of FH channels

Number of hops per bit

From the problem formulation, we have

Tb
Duration of each hop
L
T
Tc = b Chip duration for PN sequence
NL
T=

Period of the

WDS

Bandwidth of DS waveform

WJ

Bandwidth of the wideband jamming

Number of hopping channels corrupted by jamming

WJ

PN

sequence

K 1 1
1
P( j=
users, no jam)
1
j M M
j

K 1 1
1
=
P( j users
, full jam)
1
j M M
j

Noise-to-signal ratio

JSR = N J / 2 PT

Jamming-to-signal ratio

Fraction of the channel jammed

Partial Jamming

W

M

K 1 j

1

M

No Jamming

1
P ( | j users, no jam) = Q
NSR / 2 + I k
j

Interference to signal ratio introduced by the other users hopping in user ks channel
q = WJp / WDS

M W 1

K 1 j

K 1 1
1
=
P ( j users
, partial jam)
1
j M M

Part of the bandwidth of the channel partially corrupted by jamming

NSR = N 0 / 2 PT

K 1 j

Full Jamming

1
P k ( | j users, full jam) = Q
NSR / 2 + JSR / 2 + I k
j

P k ( | j users, partial jam) q P k ( | j users, partial jam, corrupted portion) + (1 q ) P k ( | j users, partial jam, uncorrupted portion)
13 Managed by UT-Battelle
for the Department of Energy

Hybrid DS/FFH Multi-user Simulation


-2

10

Severe (~worst-case)
channel conditions:

Represents user
flooding in rough
terrain.
14 Managed by UT-Battelle
for the Department of Energy

-4

10

Error Probability

K = 20-120 users, 100%


duty cycle
L = 5 hops/bit
M = 20 channels
N = 127 PN code length
DS PG = 21 dB
JNR = 13 dB
Rician coeff. = 0.1
Chan. covariance = 10
Jammed chan. = 5
Channel portion
partially corrupted = 0.4

-3

10

-5

10

-6

10

-7

10

20 users
40 users
60 users
80 users
100 users
120 users

-8

10

15

10
SNR(dB)

Performance of a hybrid DS/FFH


system: Effect of different number of
users (multi-user interference).

20

Hybrid DS/FFH Multi-user Simulation


Severe (~worst-case)
channel conditions:

Represents user
flooding in rough
terrain.
15 Managed by UT-Battelle
for the Department of Energy

-1

JNR=16 dB
JNR=14.7 dB
JNR=13 dB

Error Probability

K = 100 users, 100%


duty cycle
L = 5 hops/bit
M = 20 channels
N = 127 PN code length
DS PG = 21 dB
JNR = 10-16 dB
Rician coeff. = 0.1
Chan. covariance = 10
Jammed chan. = 5
Channel portion
partially corrupted = 0.4

10

10

10

10

JNR=10 dB

-2

-3

-4

15

10
SNR (dB)

Performance of a hybrid DS/FFH


system: Effect of different jamming-tonoise ratios (JNRs).

20

Hybrid DS/FFH Multi-user Simulation


Severe (~worst-case)
channel conditions:

Represents user
flooding in rough
terrain.
16 Managed by UT-Battelle
for the Department of Energy

-1

8 channels fully jammed


6 channels fully jammed
4 channels fully jammed
2 channels fully jammed

Error Probability

K = 100 users, 100%


duty cycle
L = 5 hops/bit
M = 20 channels
N = 127 PN code length
DS PG = 21 dB
JNR = 13 dB
Rician coeff. = 0.1
Chan. covariance = 10
Jammed chan. = 2-8
Channel portion
partially corrupted = 0.4

10

10

10

10

-2

-3

-4

15

10
SNR (dB)

Performance of a hybrid DS/FFH


system: Effect of different number of
fully-jammed channels.

20

Hybrid DS/FFH Multi-user Simulation


Severe (~worst-case)
channel conditions:

Represents user
flooding in rough
terrain.
17 Managed by UT-Battelle
for the Department of Energy

10

Error Probability

K = 100 users, 100%


duty cycle
L = 1-7 hops/bit
M = 20 channels
N = 127 PN code length
DS PG = 21 dB
JNR = 13 dB
Rician coeff. = 0.1
Chan. covariance = 10
Jammed chan. = 5
Channel portion
partially corrupted = 0.4

10

10

10

10

1 hop/bit
3 hops/bit
5 hops/bit
7 hops/bit

-1

-2

-3

-4

15

10

20

SNR (dB)

Performance of a hybrid DS/FFH


system: Effect of different numbers of
frequency hops per bit.

Hybrid DS/FFH Multi-user Simulation


Severe (~worst-case)
channel conditions:

Represents user
flooding in rough
terrain.
18 Managed by UT-Battelle
for the Department of Energy

-1

10 hopping channels
20 hopping channels
30 hopping channels

Error Probability

K = 100 users, 100%


duty cycle
L = 5 hops/bit
M = 10-40 channels
N = 127 PN code length
DS PG = 21 dB
JNR = 13 dB
Rician coeff. = 0.1
Chan. covariance = 10
Jammed chan. = 5
Channel portion
partially corrupted = 0.4

10

10

10

10

40 hopping channels

-2

-3

-4

10

15

20

SNR (dB)

Performance of a hybrid DS/FFH


system: Effect of different numbers of
available hopping channels.

Hybrid DS/FFH Multi-user Simulation


Severe (~worst-case)
channel conditions:

Represents user
flooding in rough
terrain.
19 Managed by UT-Battelle
for the Department of Energy

-3

10

-4

10

-5

10

BER

K = 100 users, 100%


duty cycle
L = 5 hops/bit
M = 20 channels
N = 4-64 PN code length
DS PG = 21 dB
JNR = 13 dB
Rician coeff. = 0.1
Chan. covariance = 10
Jammed chan. = 5
Channel portion
partially corrupted = 0.4

-2

10

-6

10

-7

4 PN Code Length

10

8 PN Code Length
16 PN Code Length

-8

10

32 PN Code Length
64 PN Code Length

-9

10

5
SNR

Performance of a hybrid DS/FFH


system: Effect of different DS PN code
lengths.

10

Hybrid DS/FFH Multi-user Simulation


Severe (~worst-case)
channel conditions:

Represents user
flooding in rough
terrain
20 Managed by UT-Battelle
for the Department of Energy

Error Probability

K = 100 users, 100%


duty cycle
L = 5 hops/bit
M = 20 channels
N = 127 PN code length
DS PG = 21 dB
JNR = 13 dB
Rician coeff. = 0.1-0.7
Chan. covariance = 10
Jammed chan. = 5
Channel portion
partially corrupted = 0.4

10

10

10

-1

-2

-3

gamma=0.7
gamma=0.5
gamma=0.3
gamma=0.1

10

-4

10

15

20

SNR (dB)

Performance of a hybrid DS/FFH


system: Effect of different Rician fading
channel parameters.

Hybrid DS/FFH Multi-user Simulation


Severe (~worst-case)
channel conditions:

Represents user
flooding in rough
terrain.
21 Managed by UT-Battelle
for the Department of Energy

No Hit
2% Hit
4% Hit
6% Hit
10% Hit
-4

10

BER

K = 100 users, 100%


duty cycle
L = 5 hops/bit
M = 20 channels
N = 127 PN code length
DS PG = 21 dB
JNR = 13 dB
Rician coeff. = 0.1
Chan. covariance = 10
Jammed chan. = 5
Channel portion
partially corrupted = 0.4

-3

10

-5

10

-6

10

5
SNR

10

Performance of a hybrid DS/FFH


system: Effect of different hitting rates.

Hybrid DS/FFH Comparative Simulation


Severe (~worst-case)
channel conditions:

-1

10

-2

10

-3

10

BER

Two-path Rayleigh
chan.
P1: = 0, gain = 0.7
P2: = 0.3 s, gain =
0.4
Equal-bandwidth cases
DS SF = 16
FH: 16 ch., 4 b/hop
DS/FH: SF = 16, 4
hopping freqs.

10

-4

10

-5

10

-6

10

-7

10

Represents the fixedbandwidth advantage


of DS/FFH format over
other modulations.
22 Managed by UT-Battelle
for the Department of Energy

2-Path Rayleigh BPSK


DS
SFH
FFH
Hybrid DS-SFH
Hybrid DS-FFH

5
SNR

10

Comparative performance of a hybrid


DS/FFH versus other forms in Rayleigh
fading.

Conclusions

The performance of a hybrid DS/FFH system was


analytically evaluated in a worst-case use scenario.

We derived the average BER for a hybrid DS/FFH


system that includes the effects from wide-band and
partial-band jamming, multi-user interference and/or
varying degrees of Rician/Rayleigh fading.

Numerical results exploring the parameter space of the


HSS system have been presented to demonstrate its
effectiveness under different conditions and scenarios.

The detailed performance and security aspects of HSS


signals will be further analyzed in a future paper.

23 Managed by UT-Battelle
for the Department of Energy

24 Managed by UT-Battelle
for the Department of Energy

Você também pode gostar