Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan...
-
Upload
silas-moore -
Category
Documents
-
view
223 -
download
0
Transcript of Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan...
![Page 1: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/1.jpg)
Mitigating Congestion in Wireless Sensor Networks
Bret Hull, Kyle Jamieson, Hari BalakrishnanNetworks and Mobile Systems Group
MIT Computer Science and Artificial Intelligence Laboratory
![Page 2: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/2.jpg)
Congestion is a problem in wireless networks
• Difficult to provision bandwidth in wireless networks– Unpredictable, time-varying channel– Network size, density variable– Diverse traffic patterns
• The result is congestion collapse
![Page 3: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/3.jpg)
Outline
• Quantify the problem in a sensor network testbed
• Examine techniques to detect and react to congestion
• Evaluate the techniques– Individually and in concert– Explain which ones work and why
![Page 4: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/4.jpg)
Investigating congestion
• 55-node Mica2 sensor network
• Multiple hops
• Traffic pattern– All nodes route to
one sink
• B-MAC [Polastre], a CSMA MAC layer 100 ft.
16,076 sq. ft.
![Page 5: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/5.jpg)
Congestion dramatically degrades channel quality
![Page 6: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/6.jpg)
Why does channel quality degrade?
• Wireless is a shared medium– Hidden terminal collisions– Many far-away transmissions corrupt packets
Sender
Receiver
![Page 7: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/7.jpg)
Per-node throughput distribution
![Page 8: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/8.jpg)
Per-node throughput distribution
![Page 9: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/9.jpg)
Per-node throughput distribution
![Page 10: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/10.jpg)
Per-node throughput distribution
![Page 11: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/11.jpg)
Goals of congestion control
• Increase network efficiency – Reduce energy consumption– Improve channel quality
• Avoid starvation– Improve the per-node end-to-end throughput
distribution
![Page 12: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/12.jpg)
Hop-by-hop flow control
• Queue occupancy-based congestion detection– Each node has an
output packet queue– Monitor instantaneous
output queue occupancy
– If queue occupancy exceeds α, indicate local congestion
![Page 13: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/13.jpg)
Hop-by-hop flow control
• Hop-by-hop backpressure– Every packet header has a
congestion bit– If locally congested, set
congestion bit– Snoop downstream traffic
of parent
• Congestion-aware MAC– Priority to congested nodes
01 Pa
cke
t
![Page 14: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/14.jpg)
Rate limiting
• Source rate limiting– Count your parent’s
number of descendents
– Limit your sourced traffic rate, even if hop-by-hop flow control is not exerting backpressure
![Page 15: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/15.jpg)
Related work
• Hop-by-hop flow control– Wan et al., SenSys 2003– ATM, switched Ethernet networks
• Rate limiting– Ee and Bajcsy, SenSys 2004– Wan et al., SenSys 2003– Woo and Culler, MobiCom 2001
• Prioritized MAC– Aad and Castelluccia, INFOCOM 2001
![Page 16: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/16.jpg)
Congestion control strategies
No congestion control Nodes send at will
Occupancy-based
hop-by-hop flow control
Detects congestion with queue length and exerts hop-by-hop backpressure
Source rate limiting Limits rate of sourced traffic at each node
Fusion Combines occupancy-based hop-by-hop flow control with source rate limiting
![Page 17: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/17.jpg)
Evaluation setup
• Periodic workload• Three link-level
retransmits• All nodes route to one
sink using ETX• Average five hops to
sink• –10 dBM transmit
power• 10 neighbors average 100 ft.
16,076 sq. ft.
![Page 18: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/18.jpg)
Metric: network efficiency
• Penalizes– Dropped packets (buffer drops, channel losses)– Wasted retransmissions
Interpretation: the fraction of transmissions that contribute to data delivery.
2 packets from bottom node, no channel loss, 1 buffer drop, 1 received: η = 2/(1+2) = 2/3
1 packet, 3 transmits, 1 received: η = 1/3
![Page 19: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/19.jpg)
Hop-by-hop flow control improves efficiency
![Page 20: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/20.jpg)
Hop-by-hop flow control conserves packets
No congestion control Hop-by-hop flow control
![Page 21: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/21.jpg)
Metric: imbalance
• ζ=1: deliver all received data
• ζ ↑: more data not delivered
Interpretation: measure of how well a node can deliver received packets to its parent
i
![Page 22: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/22.jpg)
Periodic workload: imbalance
![Page 23: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/23.jpg)
Rate limiting decreases sink contention
No congestion control With only rate limiting
![Page 24: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/24.jpg)
Rate limiting provides fairness
![Page 25: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/25.jpg)
Hop-by-hop flow control prevents starvation
![Page 26: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/26.jpg)
Fusion provides fairness and prevents starvation
![Page 27: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/27.jpg)
Synergy between rate limiting and hop-by-hop flow control
![Page 28: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/28.jpg)
Alternatives for congestion detection
• Queue occupancy
• Packet loss rate– TCP uses loss to infer congestion– Keep link statistics: stop sending when drop
rate increases
• Channel sampling [Wan03]– Carrier sense the channel periodically– Congestion: busy carrier sense more than a
fraction of the time
![Page 29: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/29.jpg)
Comparing congestion detection methods
![Page 30: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/30.jpg)
Correlated-event workload
• Goal: evaluate congestion under an impulse of traffic– Generate events seen by all nodes at
the same time
– At the event time each node:• Sends B back-to-back packets (“event
size”)• Waits long enough for the network to drain
![Page 31: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/31.jpg)
Small amounts of event-driven traffic cause congestion
![Page 32: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/32.jpg)
Congestion control slows down the network
![Page 33: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/33.jpg)
Software architecture
• Fusion implemented as a congestion-aware queue above MAC
• Apps need not be aware of congestion control implementation
Application
Routing
Fusion Queue
MAC
CC1000 Radio
![Page 34: Mitigating Congestion in Wireless Sensor Networks Bret Hull, Kyle Jamieson, Hari Balakrishnan Networks and Mobile Systems Group MIT Computer Science and.](https://reader036.fdocuments.net/reader036/viewer/2022062304/56649f285503460f94c4029e/html5/thumbnails/34.jpg)
Summary• Congestion is a problem in wireless
sensor networks
• Fusion’s techniques mitigate congestion– Queue occupancy detects congestion– Hop-by-hop flow control improves efficiency– Source rate limiting improves fairness
• Fusion improves efficiency by 3× and eliminates starvation
http://nms.csail.mit.edu/fusion