Transcript Dr Iman Samizadeh Seminar Presentation

Applying Particle Swarm Optimization to transmit video
over wireless ZigBee network
Intelligent Systems Research Centre Seminar
Dr. Iman Samizadeh, December 11th, 2013 - School of Computing
Content
•
•
•
•
•
•
•
Problem
ZigBee – IEEE 802.15.4
MPEG
Swarm Intelligence/Particle Swarm Optimization
Methodology
Results
Conclusion
Problem
Transmitting video
over IEEE
802.15.4
Overview of ZigBee
•
•
•
•
•
Low power consumption
Low cost
Intended for WPAN (Wireless Personal Area Network)
Data rates of 250 kb/s, 40 kb/s and 20 kb/s.
Frequency Bands of 2.4GHz industrial, scientific and
medical (ISM*), 915MHz and European 868MHz band.
• Supporting Star or Peer-to-Peer network.
• Support for low latency devices.
ZigBee Applications
•
•
•
•
•
Smart-home networking
Automotive and industrial networks
Interactive toys
Remote metering/utility meter readers
Sensor networks
ZigBee Stack
ZigBee Network Topologies
Peer-to-peer
Star
Cluster Tree
Y – Power consumption
ZigBee VS Other Wireless Technologies
X – Bandwidth
ZigBee VS Bluetooth
•
Standard
Bluetooth: based on IEEE 802.15.1 (WPAN)
ZigBee: based on IEEE 802.15.4 (WPAN)
•
Maximum network speed:
Bluetooth: 1 Mbit/s
ZigBee: 250 kbit/s
•
Typical network join time
Bluetooth: 3 seconds
ZigBee: 30 milliseconds
•
Protocol stack size
Bluetooth: 250 Kbyte
ZigBee: 4-32 Kbyte
Battery
Bluetooth: Intended for frequent recharging
ZigBee: batteries will last for up to 10 years
Price per chip
Bluetooth: $30
ZigBee: $2
•
•
Picture Expert Group (MPEG)
• MPEG-1(Good for storage on digital media such as video
CDs), MPEG-2 (DVD), MPEG-3 (HDTV) and MPEG-4
(offers transparent information)
High Level Application Architecture
Conventional Methods:
• Constant bit-rate (CBR) - method guarantees
traffic at a constant rate and is commonly used in
typical voice, video and audio, which require more
bandwidth
• Variable bit-rate (VBR ) - method is for the
applications that require buffering. VBR is
typically used to support compressed voice and
video
MPEG Group Of Pictures
Short GOP (DVD): I-B-B-B-B-P-B-B-B-B-I-B-B-B-B-P-B-B-B-B-I
Long GOP (MPEG-4): I-B-B-B-B-B-B-B-B-B-B-P-B-B-B-B-B-B-B-B-BB-I-B-B-B-B-B-B-B-B-B-B-P-B-B-B-B-B-B-B-B-B-B-I
Methodology
• Adaptive Rate Control over IEEE 802.15.4
using Particle Swarm Optimization
MPEG-4 encoding process
Adaptive Systems
• Swarm intelligence
• Cities
• The brain
• The immune system
• Ecosystems
• Computer models
Swarm Intelligence (SI)
Origins: How can birds or fish exhibit such a coordinated
collective behaviour?
• It is an artificial intelligence technique based around the
study of collective behaviour in decentralized, selforganized systems.
• It is made up of a population of simple agents interacting
locally with one another and with their environment.
Normally no centralized control structure dictating how
individual agents should behave, local interactions between
such agents often lead to the emergence of global
behaviour. i.e. in ant colonies, bird flocking, bacteria
modelling and fish schooling
Particle Swarm Optimization (PSO)
Invented by James Kennedy and Russell Eberhart in 1995
They have included the ‘roost’ in SI, so that:
•
•
•
•
Each agent was attracted towards the location of the roost.
Each agent remembered where it was closer to the roost.
Each agent shared information with its neighbours about its
closest location to the roost while learning from their own experience.
Each agent as the population members gradually move into better
regions of the problem space.
James and Russell suggested that the velocities and accelerations of
swarm are more appropriately applied to particles.
PSO Applications
PSO can be tailor-designed to deal with specific real-world
problems. For example problems with continuous, discrete, or
mixed search space, with multiple local minima.
• Computer numerically controlled milling optimization
• Battery pack state-of-charge estimation
• Real-time training of neural networks
• Moving Peaks (multiple peaks dynamic environment)
• Oil industry
PSO propose to fly the
solution in to the problem in
order to resolve the problem
Quantization
• A process of representing a large -possibly infinite – set
of values with a much smaller set
• One of the simplest and most general idea in lossy
compression
Scalar Quantization
Q-Scale
• This is the Quantization scale value
• MPEG’s Q-scale values has a significant affect on amount
of compression.
• The Q-scale values in MPEG-4 can be set for I, P, and Bframes separately.
• The scale can be from 1 to 31, (Larger numbers will result in
better compression but at the expense of worse quality).
• Increasing the Q-scale affected the amount of compression
the most.
Comparing VBR, CBR and PSO
ANOVA Test for 240 Frames
Kruskal–Wallis – Nonparametric Box plot
for 20 GOPs
Peak Signal-to-noise-Ratio
VBR
CBR
PSO
Method
Frame rate
PSNR
VBR with gaussian noise
379kbps
20.9351 dB
CBR with gaussian noise
266kbps
17.8369 dB
PSO with gaussian noise
209kbps
20.8851 dB
Conclusion
• The computer simulation results confirm that use of Particle
Swarm Optimization to develop an adaptive rate control,
improves the quality of picture whilst reducing data loss and
communication delay, when compared to conventional
MPEG video transmissions. Also, achieve an optimum level
of quality of picture whilst maintaining the ZigBee target
bitrate, increases the available bandwidth and reducing the
data loss.
That's all I have for now,
Thank You!
References
1.
2.
IEEE Computer Society, (August 31, 2007). IEEE Standard 802.15.4a (2007)
J. Xie, L.-T. Chia, Enhancement layer rate control for high bitrate SNR scalable video coding. J. Vis. Commun. Image R.
16, pp. 159–179
3. D. Isovic and G. Fohler, “Quality aware MPEG-2 stream adaptation in resource constrained systems,” in 16th Euromicro
Conference on Real-time Systems (ECRTS 04), Catania, Sicily, Italy, July 2004.
4. D. Isovi´c, G. Fohler, L. Steffens, Real-time issues of MPEG-2 playout in resource constrained systems. Journal of
embedded copmuting, issue 3, Jun 2004, pp. 239 - 256 (2005)
5. A. Zainaldin, I. Lambadaris, B. Nandy, Adaptive Rate Control Low bit-rate Video Transmission over Wireless Zigbee
Networks. Communications, 2008. ICC '08. IEEE International Conference on May 2008, pp. 52–58 (2008)
6. M. Shafei, M. Rezaei, S. Tavakoli, F. Mohanna, A Fuzzy Bit Allocation Algorithm for Variable Bit Rate Video Coding.
Machine Vision and Image Processing (MVIP), 2011 7th Iranian, pp. 1–4 (2011)
7. J. Kennedy, R.C, Eberhart, Y. Shi, Swarm Intelligence, Morgan Kaufmann Publishers, San Francisco, CA (2004)
8. J. Kennedy, R.C, Eberhart, Particle swarm optimization, Proc. IEEE Int'l Conf. on Neural Networks, IV, pp. 1942–1948.
Piscataway, NJ: IEEE Service Center (1995)
9. A. Owen, I. Harvey, Adapting Particle Swarm Optimization for Fitness Landscapes with Neutrality, 2007 IEEE Swarm
Intelligence Symposium (2007)
10. C. Ji, Y. Zhang, M. Tong, S. Yang, Particle Filter with Swarm Move for Optimization, Proceedings of the 10th
international conference on Parallel Problem Solving from Nature: PPSN X, pp. 909–918 (2008)
11. R. Hassan, B. Cohanim, O. D. Weck, G. Venter., A Comparison Of Particle Swarm Optimization And The Genetic
Algorithm. Structures, Structural Dynamics, and Materials Conference 2005, pp. 1—13, (2005)
References
12. H. Azarkish, H. Farahat, S. Masoud, S. H. Sarvari, Comparing the Performance of the Particle Swarm Optimization and
the Genetic Algorithm on the Geometry Design of Longitudinal Fin. World Academy of Science, Engineering and
Technology. vol. 1, issue 1, February 2012, pp. 1--5, (2012)
13. S.N. Sivanandam, S.N. Deepa, A Comparative Study Using Genetic Algorithm and Particle Swarm Optimization for
Lower Order System Modelling. International Journal of the Computer the Internet and Management. Date 2002; pp. 1-10, (2002)
14. I.E.G. Richardson, I.E.G.: H.264 and MPEG-4 Video Compression, John Wiley & Sons Ltd, Chichester, England (2003)
15. I. E. G. Richardson, H.264 and MPEG-4 Video Compression Video Coding for Next-Generation Multimedia: Video
Coding for Next-generation Multimedia - ISBN-10: 0470848375 (2003)
16. M.E.H. Pedersen, SwarmOps for Matlab - Numeric & Heuristic Optimization, November 2010, http://www.hvasslabs.org/projects/swarmops/matlab/ (2010)