uSense A Unified Asymmetric Sensing Coverage Architecture foruSense统一不对称感知覆盖结构.pptVIP

Conclusion A Dynamic Switch-based Forwarding (DSF) scheme for extremely low duty-cycle sensor networks Addressed both sleep latency and lossy radio links Dynamic switching is essential Distributed model for data delivery ratio (EDR), E2E delay (EED) and energy consumption (EEC). Optimal forwarding on these three metrics A generic metrics that converge to ETX (in always-awake networks) and DESS (in perfect-link networks) Yu Gu@SenSys’07 * * * Yu Gu and Tian He Minnesota Embedded Sensor System (MESS) Department of Computer Science Engineering This work is supported by National Science Foundation Sleep Latency in Low Duty-Cycle Sensor Networks Sleep now. Wake up in 35 seconds Sleep now. Wake up in 4 seconds Sleep now. Wake up in 57seconds Sleep now. Wake up in 13 seconds 35s latency 57s latency 4s latency 13s latency A B C D E Yu Gu@SenSys’07 Unreliable Radio Links 90% 95% 50% 70% A B C D E Yu Gu@SenSys’07 State-of-the-art Solutions: ETX (MobiCom’03) 50%, 100s 50%, 100s 40%, 10s 40%, 10s ETX = 1/0.5 + 1/0.5 = 4 ETX = 1/0.4 + 1/0.4 = 5 Expected E2E delay is 400s Expected E2E delay is 50s A B C D Sole link quality based solutions cannot help reduce E2E delay in extremely low-duty cycle sensor networks! ETX only considers link quality Yu Gu@SenSys’07 State-of-the-art Solutions: DESS (INFOCOM’05) 10%, 10s 10%, 10s 100%, 20s 100%, 20s DESS = 10 + 10 = 20s DESS = 20 + 20 = 40s Expected E2E delay is 200s Expected E2E delay is 40s A B C D Sole sleep latency based solutions cannot help reduce E2E delay in extremely low-duty cycle sensor networks! DESS only considers sleep latency Yu Gu@SenSys’07 State-of-the-art Solutions (2) Only Consider impact of link qualities Only Consider impact of Duty Cycling 80 fold performance degradation! 20 fold performance degradation! Intelligent MAC protocols (B-MAC, S-MAC, SCP-MAC …) provide significant performance improvement at the MAC layer. We focus on further performance improvement at the network layer. Yu Gu@SenSys’07