1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks...
-
Upload
tamsyn-henry -
Category
Documents
-
view
219 -
download
1
Transcript of 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks...
![Page 1: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/1.jpg)
1/30
Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor
Networks
Wireless and Sensor Network Seminar
Dec 01, 2004
![Page 2: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/2.jpg)
2/30
From
SenSys’04• November 03 ~ 05, 2004, Baltimore, Maryland, USA.
University of Southern California• Authors: Karim Seada, Marco Zuniga, Ahmed Helmy,
Bhaskar Krishnamachari
![Page 3: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/3.jpg)
3/30
Outline
Background of Geographic (Position-based) Routing Problem in Lossy Wireless Sensor Networks Proposed Optimal Forwarding Strategies Analysis and Performance Comparison
![Page 4: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/4.jpg)
4/30
Geographic Routing
A commonly used paradigm in wireless ad-hoc sensor networks when location information is available • Advantage: Conserve energy and bandwidth since discover
floods only within a single hop
• Main component: A greedy forwarding mechanism – each node forwards a packet to the neighbor that is closest to the destination.
![Page 5: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/5.jpg)
5/30
Greedy Forwarding Strategies(Basic geographic routing algorithm)
Forward the packet to the neighbor that makes the most progress towards (is closest to) the destination
![Page 6: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/6.jpg)
6/30
Problem
Assumptions for greedy routing:(i) sufficient network density
(ii) accurate localization
(iii) high link reliability independent of distance within the physical radio range
![Page 7: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/7.jpg)
7/30
Problem (Cont.)
Wireless links in real sensor networks can be extremely unreliable, deviating to a large extent from the idealized perfect-reception within-range models used in simulation
This weak-link problem and the related distance hop trade-off, whereby energy efficient geographic forwarding must strike a balance between shorter, high-quality links, and longer lossy links.
![Page 8: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/8.jpg)
8/30
Sample of a Realistic Analytical Link Loss Model
![Page 9: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/9.jpg)
9/30
Contributions
Detailed study of geographic routing in the context of lossy wireless sensor networks:• A mathematical analysis of the optimal forwarding distance
for both ARQ and No-ARQ scenarios is presented,
Detailed simulation and analysis in several novel blacklisting and neighbor selection / geographic forwarding strategies.
![Page 10: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/10.jpg)
10/30
Key Result
The product of the packet reception rate (PRR) and the distance traversed towards destination is the optimal forwarding metric for the ARQ case, and is a good metric even without ARQ.
![Page 11: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/11.jpg)
11/30
Analytical Link Model
![Page 12: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/12.jpg)
12/30
A Realistic Channel Model for LossyWireless Sensor Networks
where d is the transmitter-receiver distance, γ is the signal to noise ratio (SNR), ρ is the encoding ratio and f is the frame length.
Packet Reception Rate (PRR)
![Page 13: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/13.jpg)
13/30
Metrics
In order to evaluate the energy efficiency of different strategies, the following metrics are used:• Delivery Rate (r): percentage of packets sent by the source which
reached the sink.
• Total Number of Transmissions (t): total number of packets sent by the network, to attain the delivery rate described above.
• Energy Efficiency (Eeff ): number of packets delivered to the sink for each unit of energy spent by the network.
![Page 14: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/14.jpg)
14/30
Assumptions in This Analysis
The analysis is based on the following assumptions:• Nodes know the location and the link’s PRR of their
neighbors.
• Nodes are randomly distributed (i.e. the number of neighbors of certain node is constant).
![Page 15: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/15.jpg)
15/30
Total amount of energy consumed by the network for each transmitted packet is:
Energy Efficiency Model
rerxtxtotal eneee )1( etx and erx: amount of energy required by a node to transmit
and receive a packet ere: the energy used to read only the header of the packet (for
early rejection). n: number of neighbors (considered as a constant)
![Page 16: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/16.jpg)
16/30
Energy Efficiency Model (Cont.)
ktrpE src
eff
psrc is the number of packets sent by the source
k is a constant which includes etotal and a conversion factor for energy units.
![Page 17: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/17.jpg)
17/30
Analysis for ARQ Case
ktpE src
effARQ
Assume no a-priori constraint on the maximum number of retransmissions:
• Infinite retransmissions can be performed, therefore, r is equal to 1
![Page 18: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/18.jpg)
18/30
Analysis for ARQ Case (Cont.)
Thus, in order to maximize EeffARQ, we need to maximize the PRR×Distance product. Hence, we define the metric Xd as PRR(d)×d.
![Page 19: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/19.jpg)
19/30
PMF of Optimal Forwarding Distance in ARQ
![Page 20: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/20.jpg)
20/30
Different Forwarding Strategies
PRR×DIST: the packet is forwarded through neighbors that have the highest PRR×distance metric.
BR: the packet is forwarded through the neighbor that has the highest PRR. If several nodes have the same PRR, the node closest to the sink is chosen.
CONN: the packet is forwarded through the node in the connected region that is closest to the sink.
GR (1%): greedy forwarding through nodes that have PRR above 1%.
![Page 21: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/21.jpg)
21/30
Relative Energy Efficiency for Different Forwarding Strategies
![Page 22: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/22.jpg)
22/30
Analysis for No-ARQ Case
Because in systems without ARQ, the distance between the source and the sink influences the election of the optimal forwarding distance.
h is hop count
![Page 23: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/23.jpg)
23/30
PMF of Optimal Forwarding Distance in No-ARQ
![Page 24: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/24.jpg)
24/30
Relative Energy Efficiency for Different Forwarding Strategies No-ARQ
![Page 25: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/25.jpg)
25/30
Useful Insights
First, the optimal forwarding decision based on the PRR×distance metric in the ARQ case and in the No-ARQ case, strikes a balance between the two extremes of forwarding to the farthest (likely worst reception) neighbor and the nearest (likely best reception) neighbor.
Second, the best candidate neighbor for forwarding often lies in the transitional region.
![Page 26: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/26.jpg)
26/30
Simulation and Comparison
First, show the results for the blacklisting strategies: distance-based, absolute reception based, and relative reception-based at an entire range of blacklisting thresholds and different densities.
• The goal is to identify the optimum thresholds and their characteristics.
Then, compare the different strategies by picking the optimum blacklisting threshold for each density and
include also the original greedy, best reception policy, and the best PRR×distance policy in the comparison.
![Page 27: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/27.jpg)
27/30
Performance of Greedy Forwarding With and Without ARQ at Different Network Sizes
![Page 28: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/28.jpg)
28/30
Performance of PRR×Distance With and Without ARQ at Different Network Sizes
![Page 29: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/29.jpg)
29/30
Real Experiments with Motes
![Page 30: 1/30 Energy-Efficient Forwarding Strategies for Geographic Routing in Lossy Wireless Sensor Networks Wireless and Sensor Network Seminar Dec 01, 2004.](https://reader030.fdocuments.net/reader030/viewer/2022032612/56649edb5503460f94bea97a/html5/thumbnails/30.jpg)
30/30
Conclusion
A mathematical analysis of the optimal forwarding distance for both ARQ and No-ARQ scenarios is presented
The common greedy forwarding approach would result in very poor packet delivery rate.
An important forwarding metric that arose from our analysis, simulations and experiments, is PRR×Distance, particularly when ARQ is employed.