Honggang Zhang - Cognitive Green Communications: From Concept to Practice

Transcript

1. 1 UEB - Université Européenne de Bretagne & Supélec Cognitive Green Communications: From Concept to Practice Honggang ZHANG International Chair - CominLabs Université Européenne de Bretagne (UEB) & Supélec/IETR Supélec SCEE Seminar March 21, 2013 – Rennes, France
2. 2 UEB - Université Européenne de Bretagne & Supélec Outline  Part I – the Concept: Energy-efficient Cognitive Green Radio Communications  Part II – the Practice: Cognitive Green Communications for Achieving Energy Saving within Cellular Mobile Networks ACKNOWLEDGEMENT: This presentation is supported by the International Chair Program, CominLabs Excellence Center, Université Européenne de Bretagne (UEB) and SUPELEC/IETR. (GREAT: Green Cognitive Radio for Energy-Aware wireless communication Technologies evolution) Also, thanks to Prof. Jacques Palicot (SUPELEC), Dr. Tao Chen (VTT), Dr. Xianfu Chen (VTT), Mr. Rongpeng Li (ZJU), and Mr. Xuan Zhou (ZJU) for their supporting materials.
3. 3 UEB - Université Européenne de Bretagne & Supélec UEB - Université Européenne de Bretagne
4. 4 UEB - Université Européenne de Bretagne & Supélec Global Warming – The Most Dangerous Threat ?
5. 5 UEB - Université Européenne de Bretagne & Supélec Terrible Climate Change: Trans-Arctic Shipping Routes Navigable 21st-midcentury Source: Laurence C. Smith and Scott R. Stephenson, “New Trans-Arctic Shipping Routes Navigable by Midcentury,” PNAS, January 2013.
6. 6 UEB - Université Européenne de Bretagne & Supélec Data Explosion - Exponential Traffic Growth
7. 7 UEB - Université Européenne de Bretagne & Supélec Data Explosion - Exponential Traffic Growth (2) Source: http://bigdatadiary.com/networks-strain-to-keep-pace-with- data-explosion/internetminute/
8. 8 UEB - Université Européenne de Bretagne & Supélec Part I: Green Communications Paradigm Change from Coverage- & Capacity- Driven to Energy-Efficiency Driven Era
9. 9 UEB - Université Européenne de Bretagne & Supélec Source: Prof. T. Aoyama, Keio University, ISCIT 2010 Keynote Speech. Energy Crisis and Challenges
10. 10 UEB - Université Européenne de Bretagne & Supélec ICT Sector Commitments to Targets and Deadlines for CO2 and Greenhouse Gas Emissions and Energy Efficiency/Consumption (European Commission 2009/03/12) Energy Crisis and Challenges (2)
11. 11 UEB - Université Européenne de Bretagne & Supélec ICT Sector Commitments to Targets and Deadlines for CO2 and Greenhouse Gas Emissions and Energy Efficiency/Consumption (European Commission 2009/03/12) Energy Crisis and Challenges (3)
12. 12 UEB - Université Européenne de Bretagne & Supélec Architecture of Telecommunication Networks
13. 13 UEB - Université Européenne de Bretagne & Supélec Mobile Telecommunications Networks Power Consumption Breakdown Energy consumption composition in Vodafone (Source: Vodafone)
14. 14 UEB - Université Européenne de Bretagne & Supélec Energy Consumption in Radio Access Networks
15. 15 UEB - Université Européenne de Bretagne & Supélec Energy Consumption Reference Model for Base Station 2400 500 300 150 110 Source: Tao Chen, et al., “Network Energy Saving Technologies for Green Wireless Access Networks”IEEE Wireless Communications Magazine, 2011.
16. 16 UEB - Université Européenne de Bretagne & Supélec Energy Consumption Reference Model for Base Station (2) Note: Values in italic are power consumption figures in GSM system.
17. 17 UEB - Université Européenne de Bretagne & Supélec Network-wide Energy Saving Strategies & Techniques Increasing bandwidth can also save energy, depending on context Source: Tao Chen, et al., “Network Energy Saving Technologies for Green Wireless Access Networks”IEEE Wireless Communications Magazine, 2011.
18. 18 UEB - Université Européenne de Bretagne & Supélec Cognitive Green Communications Intelligence with Adaptation, Balancing & Optimization for Network Energy Saving
19. 19 UEB - Université Européenne de Bretagne & Supélec Features & Key Functionalities of Cognitive Radio (Cognitive Cycle) Source: Gurkan Gur and Fatih Alagoz, “Green Wireless Communications via Cognitive Dimension: An Overview”, IEEE Network, March 2011.
20. 20 UEB - Université Européenne de Bretagne & Supélec Embedded Intelligence in a General Cognitive Radio Transceiver Cognitive Radio Node PHY Layer MAC Layer Network Layer Application Layer Source: Xianfu Chen, Zhifeng Zhao, and Honggang Zhang, “Stochastic Power Adaptation with Multi-agent Reinforcement Learning for Cognitive Wireless Mesh Networks,” IEEE Transactions on Mobile Computing, Q4 2012. Xianfu Chen, Zhifeng Zhao, Honggang Zhang, and Tao Chen, “Conjectural variations in multi-agent reinforcement learning for energy-efficient cognitive wireless mesh networks,” in Proceedings of IEEE WCNC 2012, Paris, France, Apr. 2012.
21. 21 UEB - Université Européenne de Bretagne & Supélec Machine Learning Why Reinforcement Learning?
22. 22 UEB - Université Européenne de Bretagne & Supélec Basics of Reinforcement Learning  Policy: What to do  Reward: What is good  Value: What is good because it predicts reward  Model: What follows what Policy Reward Value Model of environment
23. 23 UEB - Université Européenne de Bretagne & Supélec Processes of Reinforcement Learning
24. 24 UEB - Université Européenne de Bretagne & Supélec Workflow of Energy Saving Mechanism Enabled by Cognitive Process/Cycle Source: Oliver Blume, et al. “Energy Savings in Mobile Networks Based on Adaptation to Traffic Statistics,” Bell Labs Technical Journal 15(2), 77–94 (2010).
25. 25 UEB - Université Européenne de Bretagne & Supélec Once upon a Time – What was Cognitive Radio, Really? Joe Mitola’s Cognitive Radio (1999) Simon Haykin’s Cognitive Radio (2005) DySPAN’s Cognitive Radio (2007) Cognitive Radio (G. Gur and F. Alagoz, 2011)
26. 26 UEB - Université Européenne de Bretagne & Supélec Once upon a Time – What was Cognitive Radio, Really? (2)
27. 27 UEB - Université Européenne de Bretagne & Supélec Part II: The Practice – Energy Saving for Greener Cellular Mobile Networks “Tidal Effect” of Cellular Networks’ Traffic Flow & Loads
28. 28 UEB - Université Européenne de Bretagne & Supélec Representative Patterns of Traffic Loads during One Day (Cellular Networks)
29. 29 UEB - Université Européenne de Bretagne & Supélec Normalized load of three different cell sectors over 3 weeks. The moving average of each cell over one second has been plotted. The cells show high load (Top), varying load (Middle), and low load (Bottom). Source: Daniel Willkomm et al., “Primary User Behavior in Cellular Networks and Implications for Dynamic Spectrum Access”. Representative Patterns of Traffic Loads during 3 Weeks (Cellular Networks)
30. 30 UEB - Université Européenne de Bretagne & Supélec E-commerce website: 292 production web servers over 5 days. (Traffic varies by day/weekend, power doesn’t.) Representative Patterns of Traffic Load during 5 Days (Core Networks/Internet)
31. 31 UEB - Université Européenne de Bretagne & Supélec Base Stations’ Traffic Loads Measurement Campaigns in Zhejiang (China) Source: Xuan Zhou, Zhifeng Zhao, Rongpeng Li, Yifan Zhou, and Honggang Zhang, “The Predictability of Cellular Networks Traffic,” IEEE ISCIT2012, October 2012.  Traffic records from 9 MSCs and SGSNs with about 7000 BSs with coverage of 780 km2  Both GSM and UMTSBSs traffic from January to December in 2012, serving about 3 million subscribers
32. 32 UEB - Université Européenne de Bretagne & Supélec Measured Traffic Loads Variation Patterns (One Week)
33. 33 UEB - Université Européenne de Bretagne & Supélec Typical Examples of Measured Base Stations’ Traffic Loads in Zhejiang (China) Source: Rongpeng Li, Zhifeng Zhao, Yan Wei, Xuan Zhou, and Honggang Zhang, “GM-PAB: a grid-based energy saving scheme with predicted traffic load guidance for cellular networks,” in Proceedings of IEEE ICC 2012, Ottawa, Canada, Jun. 2012.
34. 34 UEB - Université Européenne de Bretagne & Supélec Sensing and Prediction of Cellular Networks’ Traffic Flows & Loads ,t ,t ,t x x y 3 2 1   BS1 BS3 BS2 Router route 1 route 3 route 2 link 2 link 1 link 3            6 , 3 6 , 2 6 , 1 5 , 3 5 , 2 5 , 1 4 , 1 3 , 3 2 , 3 4 , 1 3 , 2 2 , 2 4 , 1 3 , 1 2 , 1 1 , 3 1 , 2 1 , 1 x x x x x x x x x x x x x x x x x x X Interpolation: fill in the missing data from incomplete and/or indirect measurements of the Traffic Matrices Future Anomaly Missing
35. 35 UEB - Université Européenne de Bretagne & Supélec Sensing and Prediction of Cellular Networks’ Traffic Flows &Loads (2) Source: Rongpeng Li, Zhifeng Zhao, Xuan Zhou, and Honggang Zhang, “Energy savings scheme in radio access networks via compressive sensing-based traffic load prediction,” European Transactions on Emerging Telecommunications Technologies (ETT), Nov. 2012.
36. 36 UEB - Université Européenne de Bretagne & Supélec Network Energy Saving through BS Switching on/off (Sleep Mode)
37. 37 UEB - Université Européenne de Bretagne & Supélec Block Diagram of Reinforcement Learning - The learning system and the environment are both inside the feedback loop
38. 38 UEB - Université Européenne de Bretagne & Supélec Reinforcement Learning: Actor-Critic Approach
39. 39 UEB - Université Européenne de Bretagne & Supélec Stochastic BS Switching Operation with Actor-Critic Learning Source” Rongpeng Li, Zhifeng Zhao, Xian Chen, and Honggang Zhang, “Energy Saving through a Learning Framework in Greener Cellular Radio Access Networks,” in Proceedings of IEEE Globecom 2012, Anaheim, USA, Dec. 2012.
40. 40 UEB - Université Européenne de Bretagne & Supélec Stochastic BS Switching Operation with Actor-Critic Learning (2)
41. 41 UEB - Université Européenne de Bretagne & Supélec Base Stations’ Traffic Load State Vector
42. 42 UEB - Université Européenne de Bretagne & Supélec Traffic Loads and BS Power Consumption Model
43. 43 UEB - Université Européenne de Bretagne & Supélec Actor-Critic Learning: Markov Decision Process
44. 44 UEB - Université Européenne de Bretagne & Supélec Actor-Critic Learning: Markov Decision Process (2)
45. 45 UEB - Université Européenne de Bretagne & Supélec Actor-Critic Learning: Markov Decision Process (3)
46. 46 UEB - Université Européenne de Bretagne & Supélec Actor-Critic Learning Scheme for BS Power Saving
47. 47 UEB - Université Européenne de Bretagne & Supélec Parameter description Value Simulation area 1.5km * 1.5km Maximum transmission power Macro BS 20W Micro BS 1W Maximum operational power Macro BS 865W Micro BS 38W Height Macro BS 32m Micro BS 12.5m Intra-cell interference factor 0.01 Channel bandwidth 1.25MHz File requests Arrival rate File size 100kbyte Constant power percentage Numerical Analysis
48. 48 UEB - Université Européenne de Bretagne & Supélec Energy Saving by Actor-Critic Learning (BS Switching & Sleep Mode) Performance comparison between Actor-Critic learning framework (LF) based energy saving scheme and the state-of-the-art (SOTA) scheme (JSAC, Sept. 2012) under various static/variant traffic arrival rates. Source” Rongpeng Li, Zhifeng Zhao, Xian Chen, and Honggang Zhang, “Energy Saving through a Learning Framework in Greener Cellular Radio Access Networks,” in Proceedings of IEEE Globecom 2012, Anaheim, USA, Dec. 2012.
49. 49 UEB - Université Européenne de Bretagne & Supélec Basics and Advantages of Transfer Learning
50. 50 UEB - Université Européenne de Bretagne & Supélec 50 Xuan Zhou, Zhifeng Zhao, R. Li, Y. Zhou, J. Palicot, and Honggang Zhang, “TACT: A Transfer Actor-Critic Learning Framework for Energy Saving in Cellular Radio Access Networks,” arXiv:1211.6616, November 2012. Stochastic BS Switching Operation with Transfer Reinforcement Learning
51. 51 UEB - Université Européenne de Bretagne & Supélec Basics and Features of Transfer Reinforcement Learning
52. 52 UEB - Université Européenne de Bretagne & Supélec Features of Transfer Actor-Critic Learning
53. 53 UEB - Université Européenne de Bretagne & Supélec TACT : The Transfer Learning Framework for Energy Saving Scheme
54. 54 UEB - Université Européenne de Bretagne & Supélec Numerical Analysis
55. 55 UEB - Université Européenne de Bretagne & Supélec Performance impact of the transfer rate factor θ to the TACT scheme when λ = 5× 10−6 Energy Saving by Transfer Actor-Critic Learning (BS Switching & Sleep Mode) Performance comparison among classical AC scheme, TACT scheme and SOTA scheme under various homogeneous traffic arrival rates when the transfer rate θ = 0.1 Xuan Zhou, Zhifeng Zhao, R. Li, Y. Zhou, J. Palicot, and Honggang Zhang, “TACT: A Transfer Actor-Critic Learning Framework for Energy Saving in Cellular Radio Access Networks,” arXiv:1211.6616, November 2012.
56. 56 UEB - Université Européenne de Bretagne & Supélec Summary & Conclusion  Environmental-friendly Green Communications: – A paradigm change from traditional coverage- & capacity-driven to energy-efficiency driven communications and networks (Smart, sustainable, and self-harmonized greener ICT).  Cognitive Green Radio Communications: – Besides spectrum and energy, intelligence is the THIRD kind of resource, but without limitation of scarcity. – Learning and decision making algorithms under green constraint can play a significant role in enabling energy- and spectral- efficient greener future communications. – Effective energy saving can be realized by using various learning approaches in mobile cellular networks. Cognitive Green Communications: From Concept to Reality!
57. 57 UEB - Université Européenne de Bretagne & Supélec Thanks! Email: Honggang.Zhang@supelec.fr