1 / 27

Voice Communication in the Delay-Tolerant Networks

Master’s Thesis Presentation Md. Tarikul Islam. Voice Communication in the Delay-Tolerant Networks. Supervisor: Prof. Jörg Ott 18.11.2009 Otaniemi. Thesis Contents. Motivation Objective Delay-Tolerant Networking System Architecture Implementation Performance Evaluation Future Works.

lana
Download Presentation

Voice Communication in the Delay-Tolerant 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. Master’s Thesis Presentation Md. Tarikul Islam Voice Communication in the Delay-Tolerant Networks Supervisor: Prof. Jörg Ott 18.11.2009 Otaniemi

  2. Thesis Contents • Motivation • Objective • Delay-Tolerant Networking • System Architecture • Implementation • Performance Evaluation • Future Works

  3. Motivation • Traditional Push-to-talk services • Are mainly infrastructure based. They may fail to operate when the mobile users travel in the infrastructure-less environment. • Rely on the traditional Internet protocols that may not work in the challenged scenarios. • Requires sufficient node density to establish end-to-end path; thus fails to communicate in sparse mobile ad-hoc environments. • Moreover the phenomenon of attenuation and interference in the wireless networks may lead packet losses and longer delays, which badly affects the overall user experience.

  4. Objective • To develop a voice communication system (DT-Talkie) which enables both individual and group users to communicate over infrastructure-less and other challenged networking environments in the walkie-talkie fashion. • To implement the DT-Talkie for Maemo based Nokia Internet Tablets, leveraging the DTN reference implementation developed in the DTN Research Group. • To carry out a set of simulations using the ONE simulator to evaluate the performance of the DT-Talkie using several DTN routing protocols in different mobility scenarios.

  5. Delay-Tolerant Networking (1) • DTN is an approach which enables communication in the environments characterized by intermittent connectivity, long or variable delay, high error rates, low data rates and frequent network partitions. • In the DTN architecture, messages are forwarded using store-carry-forward mechanism. • The DTN architecture does not rely on end-to-end path at any point of time for communication. • A delay-tolerant network operates as an overlay on top of a divergent set of regional networks, including Internet.

  6. Delay-Tolerant Networking (2) • The Bundle Protocol provides common overlay service to interoperate among the wide range of network types. • The convergence layers performs mapping between the bundle protocol and network-specific lower layers. A bundle node classification DTN protocol hierarchy

  7. System Architecture (1) • Voice messages are encoded after capturing. • Application level framing mechanism is applied to send other contents along with the voice messages. MIME is used for this purpose. • Bundle Protocol services are used to transport the voice messages as bundles in the mobile DTNs. DT-Talkie general processing steps

  8. System Architecture (2) Sample MIME message generated in a DT-Talkie session

  9. System Architecture (3) • Bundle Addressing • All the endpoints in the DTN domain are identified by URI-like endpoint identifiers (EID), which has the general form of <scheme name>:<scheme-specific part>. • The “dtn:” scheme is adopted as a default scheme in the DTN reference implementation using the structure of “dtn://node-id/application-id”. • In the DT-Talkie “<host>.dtn” is used as node-id and “dttalkie” is used as application-id (e.g., dtn://nokia-n810.dtn/dttalkie). • Group Communication • The same concept of one-to-one communication is applied with the exception that the voice messages are destined to a multicast EID. • dtn://<group_name>.dtn/dttalkie (e.g., dtn://netlab.dtn/dttalkie).

  10. System Architecture (4) • Voice Message Fragmentation • Generally large messages lead to longer transfer times and the contact duration in the opportunistic DTN environment may be too short to reliably transmit a large single message. • Fragmentation is applied in the application layer to speed-up the DT-Talkie session interactivity with the assumption of better link connectivity. • Silence suppression is used to separate talk-spurts from a voice message. Signal representation of a voice message

  11. System Architecture (5) MIME encapsulation of a voice message fragment

  12. System Architecture (6)Codec Negotiation (First Approach) MIME encapsulation of the same voice message using three codecs

  13. System Architecture (6)Codec Negotiation (Second Approach) MIME encapsulation of uncompressed audio and XML content

  14. Implementation (1) • GTK+, Hildon - Used for GUI • Gmime - To create and parse MIME messages • DTN2 (V 2.6.0) - To send and receive DTN bundles • Gstreamer - To capture and playback voice messages High-level architecture of the DT-Talkie application

  15. Implementation (2) DT-Talkie application state flow diagram

  16. Implementation (3) • DT-Talkie fragmentation • Built-in Voice Activity Detection (VAD) feature of the G.729 codec is used in the one-to-one communication scenario. Fragments creation through separating talk-spurts

  17. Implementation (4) DT-Talkie screenshot

  18. Performance Evaluation (1) • ONE simulator • An agent-based discrete event simulation engine • Combines movement modeling, routing simulation, visualization and reporting in a single program • Simulation scenarios • Simple RWP and more realistic WDM mobility models • Epidemic and Spray-and-Wait routing protocols • Performance metrics • Delivery probability and delivery delay • Message and fragmentation modes • Simulation parameters • RWP • 100 nodes, 1000 X 1000 m2 simulation area (open space) • 6 hours simulation time, 2 hours warmup period and 2 hours cooldown period • WDM • 323 nodes, Helsinki city area • 1 day simulation time, 6 hours warmup period and 2 hours cooldown period

  19. Performance Evaluation (2)

  20. Performance Evaluation (3)

  21. Performance Evaluation (4) Selection of a destination node when communication radius is specified corresponding to a particular source node

  22. Performance Evaluation (5)

  23. Performance Evaluation (6)

  24. Performance Evaluation (7) A voice session with three interactions

  25. Performance Evaluation (8)

  26. Future Works • Integration of codec negotiation feature. • Fragmentation support with the capability of link adaptivity. • Enabling security services (authenticity and confidentiality).

  27. Thank You

More Related