1 / 22

An Information Model for Geographic Greedy Forwarding in Wireless Ad-Hoc Sensor Networks

An Information Model for Geographic Greedy Forwarding in Wireless Ad-Hoc Sensor Networks. Junchao Ma, Wei Lou Dept. of Computing The HK Polytechnic University. Zhen Jiang Comp. Sci. Dept. West Chester University. Jie Wu Dept. of Computer Sci. & Eng. Florida Atlantic University & NSF.

kuniko
Download Presentation

An Information Model for Geographic Greedy Forwarding in Wireless Ad-Hoc Sensor Networks

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. An Information Model for Geographic Greedy Forwarding in Wireless Ad-Hoc Sensor Networks Junchao Ma, Wei Lou Dept. of Computing The HK Polytechnic University Zhen Jiang Comp. Sci. Dept. West Chester University Jie Wu Dept. of Computer Sci. & Eng. Florida Atlantic University & NSF INFOCOM'08

  2. Outline • Introduction • Our Approaches • Experimental Results • Conclusion • Future Work INFOCOM'08

  3. Goal • An efficient path in wireless ad-hoc sensor networks (WASNs) • Fewer hops and detours • Faster data delivery • More energy conserved destination source INFOCOM'08

  4. Problem • Local minimum phenomenon (void) • Sparse deployment • Physical obstacles • Node failures • Communication jamming • Power exhaustion • Animus interference block void stuck node destination source INFOCOM'08

  5. Idea • Information helps routing to • Predict the ‘void’ ahead • Make a slightturn early to sufficiently avoid being blocked • Non-detour routing, i.e., greedy forwarding without perimeter routing • Make a turn only if necessary • To keep the optimality of a straight forwarding INFOCOM'08

  6. Challenge 1 • Identification of the affected area of a void • Relative to the positions of the source and the destination destination destination source destination source source destination affected area affected area Case 1 Case 2 INFOCOM'08

  7. Challenge 2 • Mutual impact of void areas • Global optimization achieved by neighborhood optimizations • No routing table, flooding, or broadcasting • Routing decision at each intermediate node • Neighbor information collection and distribution destination area of mutual impact source INFOCOM'08

  8. Challenge 3 • Unstructured WASNs • Hard to ensure whether the forwarding still achievable ahead destination ? source INFOCOM'08

  9. Illustration What kind of information and how to conduct a forecast? No global information New York (destination) Information exchanged with next light traffic prediction West Chester (source) light control related to travel destination INFOCOM'08

  10. Related Work • “Dead end” model • No optimization • Boundary model • No global optimization • Hull algorithm, or turning angle model • No consideration of the relative positions INFOCOM'08

  11. Our Approaches • Tradeoff between routing adaptivity and structure regularity • Safety information for such a forwarding • Information based forwarding (SLGF routing) INFOCOM'08

  12. Tradeoff • A forwarding with infrastructure in WASNs • LAR2: • Forwarding to a neighbor that is closer to the destination • We adopt LAR1 • Forwarding limited in the so-called request zone destination destination forwarding candidate LAR 1 LAR 2 request zone current node current node INFOCOM'08

  13. Safety Information • Inspired by the safety model in 2-D mesh networks • An unsafe area contains nodes that definitely causing routing detour. • Constructed by a labeling process via information exchanges among neighbors. INFOCOM'08

  14. Details of the Labeling Process • Unsafe node • A node without any neighbor in the request zone • A node without any safe neighbor in the request zone • Unsafe area • Connected unsafe nodes • Estimated as a rectangle at an unsafe node. • 4 Different types of unsafe status • Due to 4 different types of request zones INFOCOM'08

  15. An Example of Type-I Safety Information INFOCOM'08

  16. SLGF Routing • Four phases conducted in the order • Enforced forwarding • Safe forwarding • Perimeter routing (for making a slight turn) • Retreating (in the opposite direction) INFOCOM'08

  17. Simulation • To verify whether LAR1(-) + Safety Info. (-) + Info. based routing (+) is better than boundary info. based routing. • Forwarding routings • GF (LAR2 + boundary information) • LGF (LAR1) • SLGF (LAR1 + safety information) INFOCOM'08

  18. INFOCOM'08

  19. Result Summary • Cost safety information < boundary information • Routing success GF = LGF = SLGF • Routing path LGF < GF << SLGF INFOCOM'08

  20. Conclusion • Unicast routing but neighborcast information construction • Tradeoff between routing adaptivity and information model cost • Mutual impact of void areas • Better forwarding routing to achieve more straight paths INFOCOM'08

  21. Future Work • New balance point of the tradeoff between routing adaptivity and information model cost • More accurate information • Better forwarding routing to achieve more straight path INFOCOM'08

  22. Questions? Thank you! INFOCOM'08

More Related