1 / 17

Robust Multicast - Dawn Progress Report

Robust Multicast - Dawn Progress Report. Mario Gerla DAWN Review UCSC , Oct 6, 2009. DAWN Research Activities 2008-2009. Network Coding: Dynamic encoding and forwarding rate adjustment at intermediate nodes as function of channel and jamming

Download Presentation

Robust Multicast - Dawn Progress Report

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. Robust Multicast - Dawn Progress Report Mario Gerla DAWN Review UCSC , Oct 6, 2009

  2. DAWN Research Activities 2008-2009 Network Coding: Dynamic encoding and forwarding rate adjustment at intermediate nodes as function of channel and jamming Optimal generation size selection based on topology changes Multi rate, multicast network coding - performance bounds Multicast: Inter-domain RelayCast:Extended delay tolerant Relay Multicast models to inter-domain, inter-coalition scenarios ProbeCast: Call acceptance control of non elastic flows - analytic formulation CoCAST: Ad Hoc Multicast in urban convoys using Cognitive Radios Security: Attribute Based Encryption: add a “fading” function to simplify “situation dependent” attribute updating

  3. CoCast: Ad Hoc Multicast in urban convoys using Cognitive Radios Wooseong Kim, Soon Oh, Mario Gerla, and Joon-Sang Park

  4. Scenario and Problem • Challenges • Primary user protection (local residents, businesses) • Dynamic channel assignment • Scalability of the multicast protocol (multiple sources) • Goal • Urban convoy (eg, platoon) • Multicast the data traffic from sensor-enabled vehicles to the subscribers (patrol cars) by utilizing idle WiFi channels to avoid disturbing the primary service (WiFi access points).

  5. Example: Motorcade • Motorcade through city streets • Platoon with 100’s of vehicles • Surveillance video streams are generated by sensor-enabled escorts and multicast to patrollers • Store and forward within platoon only (security and privacy) • Primary network - WiFi • Dense WiFi access point deployment – must minimize disruption of residential WiFi • Also, minimize interference caused by residential traffic on platoon presidential motorcade

  6. On-Demand Multicast Routing Protocol (ODMRP) • On-demand mesh creation • A source initiates Join Query flooding when it has data to send. • Intermediate nodes relay Join Query after recording the previous hop as backward pointer. • Multicast subscribers send Join Reply messages, following backward pointers to the source. • Upon receiving a Join Reply, a node declares itself as part of the forwarding group. R S mesh R Join Query Join Reply R R R

  7. CoCast (Cognitive MultiCast) Protocol • Extension of ODMRP • Assumes that every node has a single radio interface. • There is a common control channel (CCC) known to all nodes • Arbitrary WiFi channel, rotated to minimize impact • Channel Sensing • Every node scans the spectrum periodically to identify idle channels (scanning is asynchronous; interleaved with data transmissions). • Multicast tree construction • As in ODMRP, Join Query and Join Reply message exchange in CCC • Each Source builds own Multicast Tree • Channel availability piggybacked on control messages. • Periodic Route (mesh) maintenance and refresh

  8. CoCast: Channel Sensing Common Control Channel (CCC): 0 Data Channels: 1,2,3 • Each node periodically senses spectrum and keeps the list of available (idle) channels over history window. • Nodes exchange available channel information (e.g. frequency, bandwidth, modulation, etc.) through the CCC. 3 CH3 CH2 CH1 1,2 1,3 1,2,3

  9. CoCast: Multicast Tree Construction • Each source builds a multicast tree and allocate channels simultaneously. • Works similar to ODMRP. • Source floods Join Query and receives Join Reply from multicast receivers through the CCC. • Join Query includes channel availability information. • Join Reply includes channel selection. available channels Source node Forwarding node S F Receiver node Switching node R F

  10. CoCast: Frequency Switching • Channel frequency switching with a single radio • Switching nodes: nodes that must switch from one channel to another. • Contributions to Channel switching delay (Dswitch): Dswitch = Dhw + Dprotocol Dhw = DRF_Rec + DBB_Rec • Dhw: hardware delay • DRF_Rec: RF switching delay (80μs ~ several ms) • DBB_Rec: baseband reconfiguration delay (0 to hundreds of ms) • Dprotocol: protocol delay (overhead in the underlying MAC/PHY protocol)

  11. CoCast: Frequency Switching(2) • Deafness problem • Switching nodes create deafness problem (i.e. unable to receive packets from upstream ch when tuned on different downstream ch). • A switching node issues a LEAVE message before switching to another channel, and the preceding forwarding node buffers packets until it receives a JOIN message. Source node S Receiver node R Forwarding node F

  12. CoCast: Channel Allocation • Join Query flooding in the CCC • Cognitive nodes compute Active channels (two-way available) after receiving Join Queries that contain Available channels. 6 1, 2, 6 1, 5 R 1,2,4,6 F F 1,5,7 1,2,4,6 1,2,5,6 S 1,2,4,5 1, 2, 6 F R 1,2,5,6 F 2,4,6 1, 2, 5 R 2, 6 7 data channels (1~7), 1 common control channel (0) 2, 6

  13. CoCast: Channel Allocation • Sending back Join Reply • Receiver selects Listen channel from Active channel list, and reports in JR • Forwrd nodes set own TX Channel to Listen channels of downstream node • Goal: minimize # of channel frequency switches along path 6 6 1 6 R 6 2 F 1 F 6 S 6 2 1 F R 6 2 F 2 R 2 2, 6 2 2 2

  14. Simulation Setup • Simulation configuration in QualNet 3.9 • Reference Group Mobility Model • Random way point motion within the platoon • Each area of the field has different WiFi channel occupancies • Channel idle/busy state changes in space and time

  15. Simulation Results • Variable number of source nodes • 2 ~ 10 source nodes within a single multicast group • Comparison between ODMRP and CoCast in terms of delivery ratio and end-to-end delay • Static case vs. mobile case • CoCast scales better with increasing number of source nodes.

  16. Simulation Results • Varying the number of Channels and Sources • 2 to 7 channels • 10 receivers • The benefit of having multiple channels increases as the number of sources increases (ie, as traffic increases).

  17. Conclusions • CoCast extends ODMRP to multihop, multichannel “cognitive” primary/secondary network scenarios. • Opportunistic use of multiple channels in WiFi urban scenario (while deferring to pre-existing WiFi users) improves robustness of multicast in vehicular platoons. Future work: • develop performance bounds using Branch and Bound techniques • Explore use of multiple radios on vehicles

More Related