1. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Multi-user Data Sharing System in Radar Sensor Networks Ming Li, Tingxin Yan, Deepak Ganesan, Eric Lyons, Prashant Shenoy, Arun Venkataramani, and Michael Zink Department of Computer Science University of Massachusetts, Amherst

2. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Emerging Rich Sensor Networks Richer energy Tethered power High data rate Many MB/second Diverse users/applications needs E.g. First responders, Commuters, Insurance, for traffic monitoring Radar Sensor Network Camera Sensor Network

3. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science CASA Radar Sensor Networks Densely monitoring the lower troposphere Tornado, storm, flood, … High rate sensor streams 300MB per radar scan every 30 seconds Stream-based system Data processing is done on proxy Wide-area wireless mesh network Multiple, diverse user needs Emergency personnel, meteorologist, other… Internet Proxy Emergency Personnel Meteorologist Normal User Tornado Detection Reflectivity Overview Precipitation Data Stream Data Processing Query

4. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Challenges in Multi-user WSNs A B C D A B C D Tornado Detection 3D Assim Wind Dir Estimation NWS Emergency Personnel Researcher Limited network resources Bandwidth << Data needs Diverse end user query needs Diverse data quality metrics Tornado: location error. Wind direction: direction error Different spatial areas of interest Wind direction: overlapping area between radars Different data fidelity needs Tornado detection > 3D assimilation Different priorities and deadlines Priority:NWS > Em. Mgr Deadline: Em. Mgr < NWS

5. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Problem Statement A B C D A B C D Tornado Detection 3D Assim Wind Dir Estimation NWS Emergency Personnel Researcher How to design next generation wireless radar sensor networks to: Jointly optimize for different data quality metrics and different priorities and deadlines of different users Share bandwidth and data across different users Adapt gracefully to bandwidth dynamics Prioritize important data during critical weather events.

6. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Key Ideas in Multi-User Data Sharing Utility-driven transmission scheduling to prioritize data transmission and maximize overall utility Progressive compression to minimize bandwidth usage and adapt to bandwidth fluctuation Global transmission control to prioritize data transmission among radars

7. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Outline Motivation Key Ideas Progressive Compression Utility-driven Transmission Scheduling Global Transmission Control Evaluation Summary

8. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Radar Utility-based Scheduler Progressive Compressor … Progressive Compressor SPIHT algorithm [set partition in hierarchical trees] Wavelet-based encoder, preserves important features of interest for meteorologist Adapts to bandwidth fluctuation Most important data is transmitted first Raw Data NWS Emergency Personnel Researcher Err Data Size Decode error of SPIHT

9. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Progressive Compressor Operation of Progressive compressor: Split scan into spatial regions, each with a set of queries associated with it. Generate a separate stream for each spatial region. Radar Utility-based Scheduler Progressive Compressor … Raw Data NWS Emergency Personnel Researcher Tornado Detection Assimilation Wind Dir Estimation

10. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Utility-based Scheduler Decides which packet offers greatest improvement to overall utility Key questions How to determine utility of packet to a query? How to aggregate utilities across diverse queries? How to schedule packets based on their utilities? Radar Utility-based Scheduler Progressive Compressor … Raw Data NWS Emergency Personnel Researcher Tornado Detection Assimilation Wind Dir Estimation

11. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Utility of a Packet to a Query What does scheduler have Marginal Data MSE What does scheduler need Marginal query quality How to map data MSE to query quality Train a mapping function a priori using sample data sets Distance between detected tornado and the actual one Intensity of detected tornadoIntensity of actual tornado SPIHT Data Stream Error Trace Application level Networking level Tor Err Data MSE Training

12. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Aggregate Utility across Diverse Queries Aggregated utility is a weighted combination of utilities for each query Weight query =f(query_priority, query_deadline) Util Agg =sum(Weight query *Util query ) Utility 1 Utility 2 Tornado Utility U tor, W 1 U asm, W 2 Utility Query1 Tornado Detect Weight 1 Assim Utility Query2 Assimilatio n Weight 2

13. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Schedule Packets based on Utility Schedules packet with the highest utility Optimal if utility function as a function of data size is concave Scheduler P1P1 P2P2 P3P3 U 1 >U 2 >U 3 P1P1 Utility Data Size

14. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Outline Motivation Key Ideas Progressive Compression Utility-driven Transmission Scheduling Global Transmission Control Evaluation Summary

15. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Global Transmission Control Problem: Radar with critical data may not get sufficient bandwidth Solution: Proxy pauses streams that are achieving low/no utility gain Proxy Tornado Detect Assim High Util Low Util Global Transmission Control

16. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Outline Motivation Key Ideas Progressive Compression Utility-driven Transmission Scheduling Global Transmission Control Evaluation Summary

17. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Evaluation Testbed 13 MacMini as adhoc mesh nodes 3-hop topology Data Sets Real data traces from Oklahoma radar testbed Simulated data by ARPS(Advanced Regional Prediction of Storms) Query Pattern Tornado Detection, 3D assimilation and Wind Direction assimilation queries arrive in a round robin manner. Deadlines are chosen from a Poisson distribution with mean at 30 seconds.

18. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Determining the Utility Function Tornado detection needs more accurate data than 3D assimilation.

19. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Performance of Utility-driven Scheduler Compare utility-driven scheduler to random scheduler 2x The utility-based scheduler achieves 2 times higher utility than the random scheduler

20. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Scalability Demonstrates that our system as a whole scales well with network size

21. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Scalability Baseline System Averaging compression Bandwidth estimation Estimate bandwidth for next epoch (30 secs) based on history of bandwidth from previous epochs. Conservative estimate to ensure that compressed scan can be transmitted in the 30 seconds.

22. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Scalability Demonstrates that our system as a whole scales well with network size 3x 38% 30% 10x 4% Our system achieves more than 10 times higher utility than the baseline system

23. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Related Work Multi-query optimization in sensor network SQL Queries and simple aggregates: Trigoni, et. al [DCOSS 2005] We have more complex data processing requirements. Utility-driven network design in sensor networks and Internet SORA [NSDI05], Kelly et al [JORS98] Does not address application-level data quality metrics and data sharing between users Global transmission control Conflict Graph – Jain et al [Wireless Network 05], Rate control – Rangwala et al [Sigcomm06] We use application level utility of data to control transmissions. Radar sensor networks Schedules radar scanning strategy to satisfy end-user needs Zink et al [EESR05].

24. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Summary Illustrates new challenges in next-gen radar sensor networks Design and implementation of a multi-user data sharing system that: Gracefully degrades utility under bandwidth fluctuations by using progressive compression Utility-driven packet scheduling based on end-user data quality metrics, priorities, and deadlines. Globally prioritizes data transmission across radars. Results show one order of magnitude improvement in application utility over existing radar data transmission system.

25. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science The End Questions?

26. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Future work Joint radar sensing, bandwidth and energy optimization Extend system to other types of WSNs like camera sensor networks. Design a hop-by-hop bulk transfer protocol that optimizes radar data transfer Explore rate control and bandwidth allocation for global transmission control

27. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Radar MUDS Overview Utility-based Scheduler Progressive Compressor …

28. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Utility-based Scheduler Schedules packets between different streams based on the overall utility of each packet to the queries Each stream is shared by multi-queries and each application has different data needs How to determine utility of packet to an application? How to aggregate utilities across diverse queries? How to schedule packets based on their utilities? Radar Utility-based Scheduler Progressive Compressor … Raw Data NWS Emergency Personnel Researcher Tornado Detection Assimilation Wind Dir Estimation

29. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Utility-based Scheduler Schedules packets between different streams based on the overall utility of each packet to the queries Each stream is shared by multi-queries and each application has different data needs How to determine utility of packet to an application? How to aggregate utilities across diverse queries? How to schedule packets based on their utilities? Radar Utility-based Scheduler Progressive Compressor … Raw Data NWS Emergency Personnel Researcher Tornado Detection Assimilation Wind Dir Estimation

30. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Adaptation to Bandwidth Fluctuation Compare progressive SPIHT to non-progressive SPIHT 4x Progressive SPIHT achieves up to 4 times lower data MSE than the non-progressive scheme Bandwidth Estimator SPIHT Data Stream SPIHT Compressed Data Progressive Non-Progressive Bandwidth Trace Moving Window

31. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Scalability Baseline System Bandwidth estimation Averaging compression Scan 1 Scan 2Scan 3 … Transmit 1 Transmit 2Transmit 3 … Sense Transmit b/w estimate b/w estimate b/w estimate CDF Bandwidth 5%

32. U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Radar Utility-based Scheduler Progressive Compressor … Progressive Compressor SPIHT algorithm [set partition in hierarchical trees] Wavelet-based encoder, preserves important features of interest for meteorologist Adapts to bandwidth fluctuation Most important data is transmitted first Raw Data NWS Emergency Personnel Researcher Err Data Size Decode error of SPIHT Tornado Detection Assimilation Wind Dir Estimation