WSN_energy.pptx
- 1. Energy Conservation in wireless sensor networks Kshitij Desai, Mayuresh Randive, Animesh Nandanwar
- 2. Basic Design • Sensor Network Architecture Internet Sensor Network Sink
- 3. Architecture of a Sensor Node • Ref: Energy Conservation in Wireless Sensor Networks – a Survey
- 4. Observations • Communication Sub-system consumes more energy than computation sub-system • Energy to transmit one bit = Energy for execution 1000 . instructions • Radio component requires same order of energy for reception, transmission and idle states • Sensing sub-system might also require significant amount of energy based on the type of sensor node.
- 5. Three Main enabling Techniques • Duty-cycling • Data-Driven approaches • Mobility
- 6. Duty-cycling • Topology Control • Power Management • Sleep/Wake Protocols • On-demand, scheduled rendezvous and Async • MAC Protocols with low Duty-cycle • TDMA, Contention-based and hybrid
- 7. Data-driven approaches • Data reduction • In-Network Processing • Data-Compression • Data-prediction • Stochastic, Time-series Forecasting and algorithmic approaches • Energy-efficient data acquisition • Adaptive Sampling • Hierarchical Sampling • Model-Driven active sampling
- 9. ATPC: Adaptive Transmission Power Control for Wireless Sensor Networks
- 10. Main Points • What is this paper about? • Power saving for wireless communication • Paper style? • Empirical study + a little theory work • What is the contribution? • Study of spatial-temporal impact on communication • Mechanism to adaptively achieve an optimal transmission power consumption
- 12. Motivation 12 TP1 TP1 T1 T2 The minimum transmission power level to save energy and maintain specified link quality TP2 T2
- 13. DesignGoals • Achieve energy efficiency • The minimum transmission power • Maintain Link Quality • Reliable links • In runtime systems, dynamic environments • Spatial impact • Temporal impact 13
- 14. Roadmap Data Analysis Empirical Observation Algorithm Design Algorithm Evaluation PART 1 PART 2
- 15. Part1-TransmissionPowervs. LinkQuality • Link Quality Metrics • RSSI (Received Signal Strength Indication), LQI (Link Quality Indication), and PRR (Packet Reception Ratio) • Transmission Power Level Index (3~31) • Experiments on Spatial Impact • 5 pairs of motes, 3 environments • 100 packets at each transmission power level • RSSI/LQI/PRR measured at different distances 15
- 16. Part1-InvestigationofSpatialImpact -95 -90 -85 -80 -75 -70 -65 -60 -55 -50 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Transmission Power Level Index RSSI (dbm) 2 ft 6 ft 12 ft 18 ft 24 ft 28 ft -95 -90 -85 -80 -75 -70 -65 -60 -55 -50 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 Transmission Power Level Index RSSI (dbm) 3 ft 6 ft 12 ft 18 ft 24 ft 30 ft 16 (a) RSSI measured on a grass field -95 -90 -85 -80 -75 -70 -65 -60 -55 -50 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Transmission Power Level Index RSSI (dbm) 3 ft 6 ft 12 ft 18 ft 24 ft 30 ft (c) RSSI measured in a parking lot 1. Different shapes at the same distance in different environments 2. Different degree of variation in different environments 3. Approximately linear (b) RSSI measured in a corridor
- 17. InvestigationofTemporalImpact • Experiment on Temporal Impact • In brushwood where human activity is rare, over 72 hours • 9 MicaZ motes in a line, 3 feet apart • A group of 20 packets at each power level every hour -95 -93 -91 -89 -87 -85 -83 -81 -79 -77 -75 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Transmission Power Level Index RSSI (dbm) 9am 1st Day 10am 1st Day 11am 1st Day 12pm 1st Day 1pm 1st Day 2pm 1st Day 17 -95 -93 -91 -89 -87 -85 -83 -81 -79 -77 -75 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Transmission Power Level Index RSSI (dbm) 0am 1st Day 8am 1st Day 4pm 1st Day 0am 2nd Day 8am 2nd Day 4pm 2nd Day 1. Vary gradually but noticeably over time 2. Approximately parallel (a) RSSI measured every 8-hour (b) RSSI measured every hour
- 18. Part1-LinkQualityThreshold 18 Binary link quality thresholds Slight different in different environments (a) RSSI Threshold on a grass field (b) LQI Threshold on a grass field 0 20 40 60 80 100 120 -95 -90 -85 -80 -75 -70 RSSI (dbm) PRR (%) 0 20 40 60 80 100 120 50 60 70 80 90 100 110 LQI (Reading from MicaZ) PRR (%)
- 19. Part2-ModelDesignofATPC • Use a linear model to approximate a non-linear correlation • rssi(tp) = a · tp +b • Least-square approximation • Dynamic model • a and b vary from time to time 19
- 20. Part2-ATPC Overview 0.4TP+32 6 4 0.8TP+49 27 3 0.5TP+23 12 2 Control Model Power Level NodeID 0.4TP+32 6 4 0.8TP+49 27 3 0.5TP+23 12 2 Control Model Power Level NodeID ATPC Table at Node 1 20 Initialization Phase: build models from linear approximation Node 3 Node 5 Node 4 Node 1 Node 2 Transmission Range Runtime Tuning Phase: pairwise closed loop control Packet with Transmission Power Level 12 Notification 25 8
- 22. Part 2- ExperimentSetup 22 • Current transmission power control algorithms – A node-level non-uniform solution (Non-uniform) – Network-level uniform solutions » Max transmission power level (Max) » The minimum transmission power level over nodes in a network that allows them to reach their neighbors (Uniform) • A 72-hour continuous experiment with MicaZ – A spanning tree of 43 nodes, 24 leaf nodes – Leaf nodes send 32 packets to the base every hour
- 23. Part2-Experimental Setup 23 (a) Weather Conditions over 72 Hours (b) Spanning Tree Topology (c) Experimental Site
- 24. Part2-PacketReceptionRatio 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1 0 6 12 18 24 30 36 42 48 54 60 66 72 Time (hours) End-to-end PRR ATPC Max Uniform Non-Uniform 0 10 20 30 40 50 60 70 80 90 100 0 6 12 18 24 30 36 42 48 54 60 66 72 Time (hours) PRR (%) Link with Static Transmission Power Link with ATPC 24 (a) E2E packet reception ratio Max ~ 100% ATPC ~ 98.3% Uniform ~ 98.3% Non-Uniform ~ 58.8% (b) PRR at a chosen link ATPC ~ constantly 100% Static transmission power ~ vary from 0% to 100%
- 25. Part2-TransmissionEnergyConsumption 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1 6 12 18 24 30 36 42 48 54 60 66 72 Time (hours) Relative Transmission Energy Consumption ATPC Max Uniform Non-Uniform 25 Max ~ 100% ATPC ~ 58.3% (1% control overhead) Uniform ~ 68.6% Non-Uniform ~ 43.2% Relative energy consumption
- 26. ConclusionsandFutureWork • Benefits of ATPC lie in three core aspects: • ATPC maintains above 98% E2E PRR over time • ATPC achieves significant energy savings • 53.6% of the transmission energy of Max • 78.8% of the transmission energy of Uniform • ATPC accurately adjusts the transmission power • Adapting to spatial and temporal factors • Towards reliable and energy-efficient routing • Spatial reuse for concurrent transmissions 26
- 27. Questions? 27 Thank you very much!