1. Grid: Scalable Ad-Hoc Wireless Networking Douglas De Couto http://pdos.lcs.mit.edu/grid

2. Goal: Networks out of Chaos AFDBECGJIH

3. Ad hoc Applications Temporary, fast setup Emergencies & events Rooftop networks No wires, trenches, etc. Developing communities Cheap, incremental, automatic

4. Direct Contact Scales Badly AFDBECGJIH “Hello J!”

5. Solution: Multi-hop Forwarding AFDBECGJIH “A to J: Hello!”

6. Design Challenges Finding routes Cope with mobile nodes Conserving battery power Coping with malicious/faulty nodes Scaling to large networks

7. Completed Research Scalable routing: Geographic forwarding Distributed P2P location database Low-power forwarding Understanding capacity limits Avoiding malicious nodes Current research: link selection

8. System Status Software distributions for Linux, BSD PC, iPaq Works with unmodified Internet software Two Grid nets deployed In-building network Rooftop network

9. LCS Grid Net 5 5 5 5 5 5 5 555 5 6 6 6 6 6 6 17 static nodes on 5 th /6 th floors A dozen iPaq hand-helds wired gateway

10. Roof-Top Grid Net LCS 5 4 3 1 2 6

11. Geographic forwarding (GF) Packets addressed to  id G,location G  Next hop is chosen from neighbors to move packet geographically closer to destination location Per-node routing overhead constant as network size (nodes, area) grows Requires location service, which adds overhead A B C D F C’s radio range E G

12. A E H G B D F C J I K L Each node has a few servers that know its location. 1. Node D sends location updates to its servers (B, H, K). 2. Node J sends a query for D to one of D’s close servers. “D?” Grid Location Service (GLS) overview

13. level-0 level-1 level-2 level-3 All nodes agree on the global origin of the grid hierarchy GLS’s Spatial Hierarchy

14. 3 servers per node per level n s s s ss s s s s s is n’s successor in that square. (Successor is the node with “least ID greater than” n ) sibling level-0 squares sibling level-1 squares sibling level-2 squares

15. Queries search for destination’s successors Each query step: visit n’s successor at increasing levels, until location server found n s s s s s s s s s3 x s2 s1 location query path

16. Geographic forwarding is less fragile than source routing. DSR queries use too much b/w with > 300 nodes. Fraction of data packets delivered successfully Number of nodes DSR Grid GF + GLS performs well Biggest network simulated: 600 nodes, 2900x2900m (4-level grid hierarchy)

17. GLS properties Spreads load evenly over all nodes Degrades gracefully as nodes fail Queries for nearby nodes stay local Per-node storage and communication costs grow slowly as the network size grows : O(log n), n nodes More details: Li et al, Mobicom 2000

18. Does Grid Find Useful Paths? AFDBECGJIH

19. Mistake: Shortest-Path Routes AFDBECGJIH A’s max range

20. Link Quality Isn’t Bi-modal

21. Route metrics How to select good routes? Compare metrics Good metric: expected total packet transmissions Want to mimimize Route metric = sum of link metrics Fight strong bias towards shortest paths While penalizing longer paths

22. Obstacles to Better Routing Want to detect and avoid lossy links, but… Loss rate masked by 802.11 re-sends Changes quickly with time, motion

23. How to find loss rate? Signal strength?

24. Current Work Trying to directly measure loss rates Route broadcast packets Long time constants 802.11 protocol beacons? Requires driver integration

25. Grid Summary Grid routing protocols are Self-configuring Easy to deploy Scalable http://www.pdos.lcs.mit.edu/grid

26. End Of Talk Demo

27. Application: Smart Devices Internet Access Point Print E-Mail Share Remote Control

28. Application: Rooftop Nets Game server School/Homework Server Internet Access

29. Application: Disaster Services Disaster may have damaged phone system &c Want to avoid N 2 plans for N services to communicate

30. Topology Distribution Scales Badly 1. “C can reach A and B.” ABCDF 3. Data from F to B. 2. “D can reach A, B, and C.” G

31. Geographic Forwarding Scales Well Longitude Latitude AFDBECG “Send towards lat G / lon G.”

32. Location Database Longitude Latitude AFDBECG DB 1. “G is at lat G / lon G” 2. “Where is G?”

33. Distributed Location Database Each node is DB for a few other nodes How to find a node’s location server(s)? Every node has an unchanging ID hash(ID) maps ID to position in unit square

34. G’s Location Server is a Point G hash(G) = 0.1,0.9 x (0,0) H I

35. Spatial Grid Hierarchy All nodes agree on the global origin of the Grid hierarchy

36. Multiple Servers per Node G c ba

37. Lookups Expand in Scope G c ba A ?

38. Grid Protocol Overhead Grows Slowly Protocol packets include: Grid update, Grid query/reply. Number of nodes Protocol Overhead (packets per second)