Design of Interference-Aware Communication Systems

Report
Wireless Networking and Communications Group
Design of Interference-Aware
Communication Systems
Prof. Brian L. Evans
Cockrell School of Engineering
24 Mar 2011
WNCG “Dallas or Bust” Roadtrip
Completed Projects – Prof. Evans
2
System
SW release
Prototype
Companies
equalization
MATLAB
DSP/C
Freescale, TI
MIMO testbed
LabVIEW
LabVIEW/PXI
Oil & Gas
Wimax/LTE resource allocation
LabVIEW
DSP/C
Freescale, TI
Camera
image acquisition
MATLAB
DSP/C
Intel, Ricoh
Display
image halftoning
MATLAB
C
HP, Xerox
video halftoning
MATLAB
fixed point conv.
MATLAB
ADSL
CAD tools
Contribution
DSP Digital Signal Processor
MIMO Multi-Input Multi-Output
LTE
PXI
17 PhD and 8 MS alumni
Qualcomm
FPGA
Intel, NI
Long-Term Evolution (cellular)
PCI Extensions for Instrumentation
On-Going Projects – Prof. Evans
3
System
Contributions
SW release
Prototype
Companies
Powerline
Comm.
noise reduction;
testbed
LabVIEW
LabVIEW and
C/C++ in PXI
Freescale,
IBM, SRC, TI
Wimax/WiFi RFI mitigation
MATLAB
LabVIEW/PXI
Intel
RF Test
noise reduction
LabVIEW
LabVIEW/PXI
NI
Underwater
Comm.
MIMO testbed;
space-time meth.
MATLAB
Lake Travis
testbed
Navy
CAD Tools
dist. computing.
Linux/C++
Navy sonar
Navy, NI
DSP Digital Signal Processor
MIMO Multi-Input Multi-Output
PXI
RFI
8 PhD and 4 MS students
PCI Extensions for Instrumentation
Radio Frequency Interference
Radio Frequency Interference (RFI)
4
(Wimax Basestation)
(Microwave)
(Wi-Fi)
(Wimax)
antenna
(Wi-Fi)
(Wimax Mobile)
Wireless
Communication Sources
• Closely located sources
• Coexisting protocols
Non-Communication Sources
Electromagnetic radiation
baseband processor
•
•
•
Computational Platform
Clock circuitry
Power amplifiers
Co-located transceivers
Wireless Networking and Communications Group
(Bluetooth)
RFI Modeling & Mitigation
5



Problem: RFI degrades communication performance
Approach: Statistical modeling of RFI as impulsive noise
Solution: Receiver design
Listen to environment
 Build statistical model
 Use model to mitigate RFI


Goal: Improve communication
10-100x reduction in bit error rate (done)
 10x improvement in network throughput (on-going)

Project began January 2007
Wireless Networking and Communications Group
RFI Modeling
6
Ad hoc and
cellular networks
•Single antenna
•Instantaneous
statistics
• Sensor networks
• Ad hoc networks
• Dense Wi-Fi networks
• Cellular networks
• Hotspots (e.g. café)
Femtocell
networks
•Single antenna
•Instantaneous
statistics
• In-cell and out-of-cell
femtocell users
Symmetric Alpha
Stable
Wireless Networking and Communications Group
• Cluster of hotspots
(e.g. marketplace)
• Out-of-cell
femtocell users
Gaussian Mixture Model
RFI Mitigation
7
Interference + Thermal noise
Pulse
Shaping
Matched
Filter
Pre-filtering
Detection
Rule
Communication performance

0
10
Correlation Receiver
Bayesian Detection
Myriad Pre-filtering
Vector Symbol Error Rate
-1
Symbol Error Rate
-1
10
10 – 100x reduction
in bit error rate
~ 20 dB
-2
10
10
-3
-3
10
10
-40
-35
-30
-25
-20
-15
-10
-5
Signal to Noise Ratio (SNR) [in dB]
Single carrier, single antenna (SISO)
Wireless Networking and Communications Group
~ 8 dB
-2
10
-10
Optimal ML Receiver (for Gaussian noise)
Optimal ML Receiver (for Middleton Class A)
Sub-Optimal ML Receiver (Four-Piece)
Sub-Optimal ML Receiver (Two-Piece)
-5
0
5
10
15
20
SNR [in dB]
Single carrier, two antenna (2x2 MIMO)
RFI Modeling & Mitigation Software
8


Freely distributable toolbox in MATLAB
Simulation of RFI modeling/mitigation
RFI generation
 Measured RFI fitting
 Filtering and detection methods
 Demos for RFI modeling and mitigation


Example uses
Snapshot of a demo
System simulation (e.g. Wimax or powerline communications)
 Fit RFI measurements to statistical models

Version 1.6 beta Dec. 2010: http://users.ece.utexas.edu/~bevans/projects/rfi/software
Wireless Networking and Communications Group
Voltage Levels in Power Grid
High-Voltage
Source: Électricité Réseau
Dist. France (ERDF)
Medium-Voltage
Low-Voltage
Concentrator
“Last mile” powerline communications on low/medium voltage line
9
Powerline Communications (PLC)
10

Concentrator controls medium
to subscriber meters


Plays role of basestation
Applications
Automatic meter reading (right)
 Smart energy management
 Device-specific billing
(plug-in hybrid)


Goal: Improve reliability & rate
Mitigate impulsive noise
 Multichannel transmission

Source: Powerline Intelligent Metering
Evolution (PRIME) Alliance Draft v1.3E
Noise in Powerline Communications
11

Superposition of five noise sources [Zimmermann, 2000]

Different types of power spectral densities (PSDs)
Colored Background
Narrowband
Noise:
Noise:
Can
be lumped
together
asAsynchronous
Periodic
Impulsive
Noise
Synchronous
to Main:
Asynchronous
Impulsive
Noise:
Periodic
Impulsive
Noise
to Main:
•
PSD decreases
• with
Sinusoidal
frequency
with
modulated
amplitudes
• Background
50-100Hz, ShortNoise
duration •impulses
Caused by switching transients
•
50-200kHz
Generalized
•
Superposition• of numerous
Affects several
noise
sources
• subbands
PSD decreases with frequency
•
Arbitrary interarrivals with micro•
•
Caused by switching power supplies
with lower intensity
•
Caused
by medium
andnarrowbands
shortwave
•
Caused
by power convertors millisecond durations
•
Approximated
by
Time varying (order
broadcast
of minutes
channels
and hours)
•
50dB above background noise
Broadband Powerline Communications: Network Design
Powerline Noise Modeling & Mitigation
12



Problem: Impulsive noise is primary
impairment in powerline communications
Approach: Statistical modeling
Solution: Receiver design
Listen to environment
 Build statistical model
 Use model to mitigate RFI


Goal: Improve communication
10-100x reduction in bit error rate
 10x improvement in network throughput

Wireless Networking and Communications Group
Preliminary Noise Measurement
Power Spectral Density Estimate
-75
Power/frequency (dB/Hz)
-80
-85
-90
-95
-100
-105
-110
-115
-120
-125
0
10
20
30
40
50
Frequency (kHz)
13
60
70
80
90
Preliminary Noise Measurement
Power Spectral Density Estimate
-75
Power/frequency (dB/Hz)
-80
-85
-90
-95
-100
-105
-110
-115
-120
-125
0
Colored Background
Noise
10
20
30
40
50
Frequency (kHz)
14
60
70
80
90
Preliminary Noise Measurement
Power Spectral Density Estimate
-75
Narrowband Noise
Power/frequency (dB/Hz)
-80
-85
-90
-95
-100
-105
-110
-115
-120
-125
0
Colored Background
Noise
10
20
30
40
50
Frequency (kHz)
15
60
70
80
90
Preliminary Noise Measurement
Power Spectral Density Estimate
-75
Narrowband Noise
Power/frequency (dB/Hz)
-80
-85
-90
-95
-100
-105
-110
-115
-120
-125
0
Colored Background
Noise
10
20
30
40
50
Frequency (kHz)
16
Periodic and
Asynchronous Noise
60
70
80
90
Powerline Communications Testbed
17

Integrate ideas from multiple standards (e.g. PRIME)
Quantify communication performance vs complexity tradeoffs
 Extend our existing real-time DSL testbed (deployed in field)

GUI
GUI

Adaptive signal processing methods


Channel modeling, impulsive noise filters & equalizers
Medium access control layer scheduling

Effective and adaptive resource allocation
Thank you for your attention!
18
Backup
Designing Interference-Aware Receivers
20
Guard zone
Statistical Modeling of RFI
• Derive analytically
• Estimate parameters at receiver
Physical (PHY) Layer
• Receiver pre-filtering
• Receiver detection
• Forward error correction
Medium Access Control (MAC) Layer
• Interference sense and avoid
• Optimize MAC parameters
(e.g. guard zone size, transmit power)
RTS / CTS: Request / Clear to send
Example: Dense WiFi Networks
Wireless Networking and Communications Group
Statistical Models (isotropic, zero centered)
21

Symmetric Alpha Stable [Furutsu


Characteristic function
Gaussian Mixture Model [Sorenson & Alspach, 1971]


& Ishida, 1961] [Sousa, 1992]
Amplitude distribution
Middleton Class A (w/o Gaussian component) [Middleton, 1977]
Wireless Networking and Communications Group
Validating Statistical RFI Modeling
22
Validated for measurements of radiated RFI from laptop
0.4
Symmetric Alpha Stable
Middleton Class A
Gaussian Mixture Model
Gaussian
0.35
Kullback-Leibler divergence

0.3
Radiated platform RFI
• 25 RFI data sets from Intel
• 50,000 samples at 100 MSPS
• Laptop activity unknown to us
0.25
0.2
Smaller KL divergence
• Closer match in distribution
• Does not imply close match in
tail probabilities
0.15
0.1
0.05
0
0
5
10
15
20
Measurement Set
Wireless Networking and Communications Group
25
Turbo Codes in Presence of RFI
23
Return
Parity 1
Systematic Data
Decoder 1
-

Gaussian channel:

Parity 2
Decoder 2
-

1
Middleton Class A channel:
Extrinsic
Information
Leads to a 10dB improvement at
BER of 10-5 [Umehara03]
Independent of
channel
statistics
Wireless Networking and Communications Group
A-priori
Information
Depends on
channel
statistics
Independent
of channel
statistics
RFI Mitigation Using Error Correction
24

Return
Turbo decoder
-
Parity 1
Decoder 1
Systematic Data
-
Interleaver
Interleaver
-
Parity 2


Decoder 2
-
Interleaver
Decoding depends on the RFI statistics
10 dB improvement at BER 10-5 can be achieved using
accurate RFI statistics [Umehara, 2003]
Wireless Networking and Communications Group
Extensions to Statistical RFI Modeling
25

Extended to include spatial and temporal dependence
Statistical Modeling of RFI
Single Antenna
Instantaneous statistics
• Symbol errors

Spatial Dependence
• Multi-antenna receivers
Multivariate extensions of
Symmetric Alpha Stable
 Gaussian mixture model

Wireless Networking and Communications Group
Temporal Dependence
• Burst errors
• Coded transmissions
• Delays in network
RFI Modeling: Joint Interference Statistics
26
Cellular networks
Multivariate Symmetric Alpha Stable
Multivariate Gaussian Mixture Model
Throughput performance of ad hoc networks
Network Throughput (normalized)
[ bps/Hz/area ]

Ad hoc networks
10
With RFI Mitigation
Without RFI Mitigation
9
8
Network throughput improved
by optimizing distribution of
ON Time of users (MAC parameter)
~1.6x
7
6
5
4
3
2
2
4
6
8
10
12
Expected ON Time of a User (time slots)
Wireless Networking and Communications Group
14
16
RFI Mitigation: Multi-carrier systems
27

Proposed Receiver
Iterative Expectation Maximization (EM) based on noise model

Communication Performance
10
10
Bit Error Rate

10
10
10
0
OFDM Receiver
Single Carrier
Proposed EM-based Receiver
-1
Simulation Parameters
• BPSK Modulation
• Interference Model
2-term Gaussian Mixture Model
-2
~ 5 dB
-3
-4
-10
-5
0
5
10
15
Signal to Noise Ratio (SNR) [in dB]
Wireless Networking and Communications Group
20
Smart Grids: The Big Picture
28
Real-Time :
Customers profiling
enabling good
predictions in demand =
no need to use an
additional power plant
Long distance
communication :
access to isolated
houses
Micro- production
: better knowledge
of energy produced
to balance the
network
Demand-side
management : boilers
are activatedduring the
night
whenelectricityisavailable
Anydisturbance due to a
storm : action
canbetakeninmediatelybased
on real-time information
Smart building :
significant cost reduction
on energy bill through
remote monitoring
Smart car : charge of
electricalvehicleswhile panels
are producing
Source: ETSI
Security
featuresFireisdetecte
d : relaycanbeswitched
off rapidly
29
Wireless Networking & Comm. Group
Applications
Systems of systems
Networks of networks
Networks of systems
Systems
Networks
Middleware
Compilers
Protocols
Operating systems
Processors
Collaboration
with UT faculty
outside of WNCG
Circuit
design
Data
acq.
Communication
links
Waveforms
Antennas
Wires
Devices
17 faculty
140 grad students
Wireless Networking & Comm. Group
30
Computation
Communications
Networking
Applications
B. Evans
Embedded DSP
J. Andrews
Communication
S. Nettles
Network Design
B. Bard
Security
C. Caramanis
Optimization
A. Bovik
Image/Video
A. Gerstlauer
Embedded Sys
R. Heath
Comm/DSP
S. Shakkottai
Network Theory
G. de Veciana
Networking
S. Sanghavi
Network Science
A. Tewfik
Biomedical
S. Vishwanath
Sensor Networks
L. Qiu
Network Design
T. Rappaport
RF IC Design
T. Humphreys
GPS/Navigation
H. Vikalo
Genomic DSP
Our Publications
31
Journal Publications
• K. Gulati, B. L. Evans, J. G. Andrews, and K. R. Tinsley, “Statistics of Co-Channel
Interference in a Field of Poisson and Poisson-Poisson Clustered Interferers”, IEEE
Transactions on Signal Processing, vol. 58, no. 12, Dec. 2010, pp. 6207-6222.
• M. Nassar, K. Gulati, M. R. DeYoung, B. L. Evans and K. R. Tinsley, “Mitigating NearField Interference in Laptop Embedded Wireless Transceivers”, Journal of Signal
Processing Systems, Mar. 2009, invited paper.
Conference Publications
• M. Nassar, X. E. Lin, and B. L. Evans, “Stochastic Modeling of Microwave Oven
Interference in WLANs”, Proc. IEEE Int. Conf. on Comm., Jun. 5-9, 2011.
• K. Gulati, B. L. Evans, and K. R. Tinsley, “Statistical Modeling of Co-Channel
Interference in a Field of Poisson Distributed Interferers”, Proc. IEEE Int. Conf. on
Acoustics, Speech, and Signal Proc., Mar. 14-19, 2010.
• K. Gulati, A. Chopra, B. L. Evans, and K. R. Tinsley, “Statistical Modeling of Co-Channel
Interference”, Proc. IEEE Int. Global Comm. Conf., Nov. 30-Dec. 4, 2009.
Cont…
Wireless Networking and Communications Group
Our Publications
32
Conference Publications (cont…)
• A. Chopra, K. Gulati, B. L. Evans, K. R. Tinsley, and C. Sreerama, “Performance Bounds
of MIMO Receivers in the Presence of Radio Frequency Interference”, Proc. IEEE Int.
Conf. on Acoustics, Speech, and Signal Proc., Apr. 19-24, 2009.
• K. Gulati, A. Chopra, R. W. Heath, Jr., B. L. Evans, K. R. Tinsley, and X. E. Lin, “MIMO
Receiver Design in the Presence of Radio Frequency Interference”, Proc. IEEE Int.
Global Communications Conf., Nov. 30-Dec. 4th, 2008.
• M. Nassar, K. Gulati, A. K. Sujeeth, N. Aghasadeghi, B. L. Evans and K. R. Tinsley,
“Mitigating Near-Field Interference in Laptop Embedded Wireless Transceivers”, Proc.
IEEE Int. Conf. on Acoustics, Speech, and Signal Proc., Mar. 30-Apr. 4, 2008.
Software Releases
• K. Gulati, M. Nassar, A. Chopra, B. Okafor, M. R. DeYoung, N. Aghasadeghi, A. Sujeeth,
and B. L. Evans, "Radio Frequency Interference Modeling and Mitigation Toolbox in
MATLAB", version 1.6 beta, Dec. 16, 2010.
Wireless Networking and Communications Group
References
33
RFI Modeling
1. D. Middleton, “Non-Gaussian noise models in signal processing for telecommunications: New
methods and results for Class A and Class B noise models”, IEEE Trans. Info. Theory, vol. 45, no. 4,
pp. 1129-1149, May 1999.
2. K. Furutsu and T. Ishida, “On the theory of amplitude distributions of impulsive random noise,” J.
Appl. Phys., vol. 32, no. 7, pp. 1206–1221, 1961.
3. J. Ilow and D . Hatzinakos, “Analytic alpha-stable noise modeling in a Poisson field of interferers or
scatterers”, IEEE transactions on signal processing, vol. 46, no. 6, pp. 1601-1611, 1998.
4. E. S. Sousa, “Performance of a spread spectrum packet radio network link in a Poisson field of
interferers,” IEEE Transactions on Information Theory, vol. 38, no. 6, pp. 1743–1754, Nov. 1992.
5. X. Yang and A. Petropulu, “Co-channel interference modeling and analysis in a Poisson field of
interferers in wireless communications,” IEEE Transactions on Signal Processing, vol. 51, no. 1, pp.
64–76, Jan. 2003.
6. E. Salbaroli and A. Zanella, “Interference analysis in a Poisson field of nodes of finite area,” IEEE
Transactions on Vehicular Technology, vol. 58, no. 4, pp. 1776–1783, May 2009.
7. M. Z. Win, P. C. Pinto, and L. A. Shepp, “A mathematical theory of network interference and its
applications,” Proceedings of the IEEE, vol. 97, no. 2, pp. 205–230, Feb. 2009.
Wireless Networking and Communications Group
References
34
Parameter Estimation
1. S. M. Zabin and H. V. Poor, “Efficient estimation of Class A noise parameters via the EM
[Expectation-Maximization] algorithms”, IEEE Trans. Info. Theory, vol. 37, no. 1, pp. 60-72, Jan.
1991 .
2. G. A. Tsihrintzis and C. L. Nikias, "Fast estimation of the parameters of alpha-stable impulsive
interference", IEEE Trans. Signal Proc., vol. 44, Issue 6, pp. 1492-1503, Jun. 1996.
Communication Performance of Wireless Networks
1. R. Ganti and M. Haenggi, “Interference and outage in clustered wireless ad hoc networks,” IEEE
Transactions on Information Theory, vol. 55, no. 9, pp. 4067–4086, Sep. 2009.
2. A. Hasan and J. G. Andrews, “The guard zone in wireless ad hoc networks,” IEEE Transactions on
Wireless Communications, vol. 4, no. 3, pp. 897–906, Mar. 2007.
3. X. Yang and G. de Veciana, “Inducing multiscale spatial clustering using multistage MAC contention
in spread spectrum ad hoc networks,” IEEE/ACM Transactions on Networking, vol. 15, no. 6, pp.
1387–1400, Dec. 2007.
4. S. Weber, X. Yang, J. G. Andrews, and G. de Veciana, “Transmission capacity of wireless ad hoc
networks with outage constraints,” IEEE Transactions on Information Theory, vol. 51, no. 12, pp.
4091-4102, Dec. 2005.
Wireless Networking and Communications Group
References
35
Communication Performance of Wireless Networks (cont…)
5. S. Weber, J. G. Andrews, and N. Jindal, “Inducing multiscale spatial clustering using multistage MAC
contention in spread spectrum ad hoc networks,” IEEE Transactions on Information Theory, vol.
53, no. 11, pp. 4127-4149, Nov. 2007.
6. J. G. Andrews, S. Weber, M. Kountouris, and M. Haenggi, “Random access transport capacity,” IEEE
Transactions On Wireless Communications, Jan. 2010, submitted. [Online]. Available:
http://arxiv.org/abs/0909.5119
7. M. Haenggi, “Local delay in static and highly mobile Poisson networks with ALOHA," in Proc. IEEE
International Conference on Communications, Cape Town, South Africa, May 2010.
8. F. Baccelli and B. Blaszczyszyn, “A New Phase Transitions for Local Delays in MANETs,” in Proc. of
IEEE INFOCOM, San Diego, CA,2010, to appear.
Receiver Design to Mitigate RFI
1. A. Spaulding and D. Middleton, “Optimum Reception in an Impulsive Interference EnvironmentPart I: Coherent Detection”, IEEE Trans. Comm., vol. 25, no. 9, Sep. 1977
2. J.G. Gonzalez and G.R. Arce, “Optimality of the Myriad Filter in Practical Impulsive-Noise
Environments”, IEEE Trans. on Signal Processing, vol 49, no. 2, Feb 2001
Wireless Networking and Communications Group
References
36
Receiver Design to Mitigate RFI (cont…)
3. S. Ambike, J. Ilow, and D. Hatzinakos, “Detection for binary transmission in a mixture of Gaussian
noise and impulsive noise modelled as an alpha-stable process,” IEEE Signal Processing Letters,
vol. 1, pp. 55–57, Mar. 1994.
4. G. R. Arce, Nonlinear Signal Processing: A Statistical Approach, John Wiley & Sons, 2005.
5. Y. Eldar and A. Yeredor, “Finite-memory denoising in impulsive noise using Gaussian mixture
models,” IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing, vol. 48,
no. 11, pp. 1069-1077, Nov. 2001.
6. J. H. Kotecha and P. M. Djuric, “Gaussian sum particle ltering,” IEEE Transactions on Signal
Processing, vol. 51, no. 10, pp. 2602-2612, Oct. 2003.
7. J. Haring and A.J. Han Vick, “Iterative Decoding of Codes Over Complex Numbers for Impulsive
Noise Channels”, IEEE Trans. On Info. Theory, vol 49, no. 5, May 2003.
8. Ping Gao and C. Tepedelenlioglu. “Space-time coding over mimo channels with impulsive noise”,
IEEE Trans. on Wireless Comm., 6(1):220–229, January 2007.
RFI Measurements and Impact
1.
J. Shi, A. Bettner, G. Chinn, K. Slattery and X. Dong, "A study of platform EMI from LCD panels –
impact on wireless, root causes and mitigation methods,“ IEEE International Symposium on
Electromagnetic Compatibility, vol.3, no., pp. 626-631, 14-18 Aug. 2006
Wireless Networking and Communications Group

similar documents