SAR05: Difference between revisions

From
Jump to navigation Jump to search
No edit summary
 
(4 intermediate revisions by 2 users not shown)
Line 8: Line 8:


== Termine ==
== Termine ==
* 26.04. Herr Neukirchen: Hyacinth
* 26.04. D. Neukirchen: Hyacinth
* 03.05. Herr Stigge und Herr Furche: Multi-Channel Networks : Mesh-Networks with Multi-Interface nodes
* 03.05. M. Stigge und G. Furche: Multi-Channel Networks : Mesh-Networks with Multi-Interface nodes
* 10.05. Herr Sperling: Verteidigungsvortrag zur Studienarbeit; Nachbereitung des Themas vom 3.5.
* 10.05. R. Sperling: Verteidigungsvortrag zur Studienarbeit; Nachbereitung des Themas vom 3.5.
* 17.05. Herr Zubow und Herr J. Mueller: Bericht ueber die Implementationsarbeiten zum BRN
* 17.05. A. Zubow und J. Mueller: Bericht ueber die Implementationsarbeiten zum BRN
* 24.05. Herr Daenicke: Sicherheit in selbst-organisierenden Netzen (z.B. PAN)
* 24.05. P. Reinecke: VPN für das Berlin Roof Net
* 31.05. T. Daenicke: Sicherheit in selbst-organisierenden Netzen (z.B. PAN)
* 07.06. B. Wiedemann: Bericht über den Stand der Diplomarbeit (Softwareverteilungsplattform für Berlin Roof Net)
* 14.06. J. Müller: Nsclick - Integration von NS2 und Click
* 21.06. M. Stigge: Performance impact of traffic patterns on routing.
* 28.06. D. Neukirchen: Security Problems in Routing ... [[Image:Vortrag.secure.routing.pdf||slides]]
* 05.07. G. Furche: Network Deformantion
* 12.07. A. Zubow: Multi-Channel Routing (60 Min); T. Daenicke: Abschlussbericht zur Studienarbeit (30 Min)


= Berlin Roof Net Themen =
= Berlin Roof Net Themen =
Line 206: Line 213:
*Eigentlich eine interessante Thematik: A comparison of TCP performance over three routing protocols for mobile ad hoc networks (2001) [http://portal.acm.org/citation.cfm?id=501425&coll=Portal&dl=ACM&CFID=45379374&CFTOKEN=99502037]. Das Problem ist das hierbei verwendete triviale "Traffic Pattern":
*Eigentlich eine interessante Thematik: A comparison of TCP performance over three routing protocols for mobile ad hoc networks (2001) [http://portal.acm.org/citation.cfm?id=501425&coll=Portal&dl=ACM&CFID=45379374&CFTOKEN=99502037]. Das Problem ist das hierbei verwendete triviale "Traffic Pattern":
**"We simulated the steady-state conditions of a network with various background traffic loads generated by 10 and 40 constant bit rate (CBR) connections. The CBR packet sizes were fixed at 512 bytes. After a warm-up time of 100 seconds, one or more TCP connections were established over each of which an FTP file transfer was conducted for 900 seconds. The TCP packet size was 1460 bytes, and the maximum size of both the send and receive windows was 8. Since routing protocol performance is sensitive to movement patterns, 50 different mobility patterns (scenarios) were generated."
**"We simulated the steady-state conditions of a network with various background traffic loads generated by 10 and 40 constant bit rate (CBR) connections. The CBR packet sizes were fixed at 512 bytes. After a warm-up time of 100 seconds, one or more TCP connections were established over each of which an FTP file transfer was conducted for 900 seconds. The TCP packet size was 1460 bytes, and the maximum size of both the send and receive windows was 8. Since routing protocol performance is sensitive to movement patterns, 50 different mobility patterns (scenarios) were generated."

*Automatically Inferring Patterns of Resource Consumption in Network Traffic [http://citeseer.ist.psu.edu/622665.html]
**the authors present AutoFocus, a tool to " ... automatically group traffic into minimal clusters of conspicuous consumption."
**mainly for Network Admins, but also a way to approch "traffic patterns"


== For Your Information ==
== For Your Information ==

Latest revision as of 11:54, 12 July 2005

Selbstorganisierende Systeme: Projekt-Seminar

Zielstellung: Literatur (Prior Art) finden, vorstellen und diskutieren, die uns bei den derzeit ungelösten Fragestellungen in folgenden beiden Projekten weiterhelfen:

  • Berlin Roof Net
  • Security in PAN (Personal Area Networks)

Termine

  • 26.04. D. Neukirchen: Hyacinth
  • 03.05. M. Stigge und G. Furche: Multi-Channel Networks : Mesh-Networks with Multi-Interface nodes
  • 10.05. R. Sperling: Verteidigungsvortrag zur Studienarbeit; Nachbereitung des Themas vom 3.5.
  • 17.05. A. Zubow und J. Mueller: Bericht ueber die Implementationsarbeiten zum BRN
  • 24.05. P. Reinecke: VPN für das Berlin Roof Net
  • 31.05. T. Daenicke: Sicherheit in selbst-organisierenden Netzen (z.B. PAN)
  • 07.06. B. Wiedemann: Bericht über den Stand der Diplomarbeit (Softwareverteilungsplattform für Berlin Roof Net)
  • 14.06. J. Müller: Nsclick - Integration von NS2 und Click
  • 21.06. M. Stigge: Performance impact of traffic patterns on routing.
  • 28.06. D. Neukirchen: Security Problems in Routing ... File:Vortrag.secure.routing.pdf
  • 05.07. G. Furche: Network Deformantion
  • 12.07. A. Zubow: Multi-Channel Routing (60 Min); T. Daenicke: Abschlussbericht zur Studienarbeit (30 Min)

Berlin Roof Net Themen

Der Vollständigkeit halber: Ad-Hoc Networks

Themenkomplex 1: Multi Channel (Mixed) Network

HYACINTH - New York (Dirk Neukirchen) Folien der Präsentation

State University of New York,Stony Brook (sunysb) - Computer Science (cs) - Experimental Computer Systems Lab (ecsl)

Das HYACINTH Projekt hat als Ziel eine WMN (Wireless Mesh Network) Struktur so effizient zu nutzen, dass diese als echter Backbone anstelle fest verdrahteter Strukturen genutzt werden kann, bzw. als performanter Ersatz für die letzte Meile. Desweiteren wird von überwiegend stationären Knoten ausgegangen. Mobile Knoten sollen mit einem Protokoll ähnlich MobileIP angebunden werden. Es werden beliebige Routing Protokolle unterstützt. Da Routing entscheidungen Traffic verursachen, der zum channel load balancing führt.

Projekthomepage: http://www.ecsl.cs.sunysb.edu/multichannel/index.html

University of Illinois (Martin Stigge)

Universitiy of Illinois at Urbana-Champaign Wireless Group Homepage

Paper zum Vortrag: Routing in Multi-Channel Multi-Interface Ad Hoc Wireless Networks (Dies ist eine stark erweiterte Version des recht oberflächlichen WCNC2005-Papers Routing and Interface Assignment in Multi-Channel Multi-Interface Wireless Networks.)

Folien zum Vortrag: http://www.informatik.hu-berlin.de/~mstigge/mcr.pdf

Weiteres interessantes (?) Material


Kommerzielle Anbieter und Projekte

Einige Firmen haben im Bereich Mesh Networking bereits Patente beantragt. Patent Pending Der Markt des Multi-Channel Networkings als Backbonetechnologie scheint sich zu vergrößern. Bei den Geräten handelt es sich um Outdoor Geräte hoher Schutzklassen (IP gegen Regen, Staub).

http://www.tropos.com/

Outdoor Geräte mit 2 NICs kosten ca. 3000$. Ausfallsicherheit durch integrierte Batterie (4-12h).

http://www.meshdynamics.com

Die (Outdoor) Geräte können bis zu 4 NICs besitzen. Die Geräter werden ab Januar 2005 vertrieben, zu Kosten von 3900$ (3 NICs). Quelle

http://www.belairnetworks.com

Wurde als Startup von T-Mobile Venture Fund unterstützt. Die Kosten pro (Outdoor) Gerät liegen bei 4000$ bis 9000$ in Abhängigkeit von der Zahl der NICs (bis zu 4 802.11a NICs). Quelle: BelAir: Meshing Quite Nicely

Distributed Systems Group (Shoshin), University of Waterloo

"Multi-Interface Multi-Channel Wireless Mesh Networks," M.A.Sc., 2004 http://etd.uwaterloo.ca/etd/mamunawa2004.pdf


Hardware für Multi Interface Networks

Netgear WGT634U MiniPCI Steckplatz USB Kosten: ca. 120€

Meshcube (altes Modell des 4G Cube) Wiki MiniPCI Slot Kosten: 199€ (1 Karte),299€ ? (2 Karten)

4G Cube Homepage der Hamburger Firma UMTS/GPRS Karte (optional?) MiniPCI Slot USB Kosten: 449€

Das Board des 4G/Meshcube kann bis zu 8 MiniPCI karten Unterstützen. Ein Bild von 6 MiniPCI Karten.

Themenkomplex 2: nsclick

University of Colorado at Boulder http://systems.cs.colorado.edu/Networking/nsclick/

für ns2 homepage Eine Übersicht der ns2 Erweiterungen: http://www.isi.edu/nsnam/ns/ns-contributed.html

Themenkomplex 3: click - CLACK !?

http://citeseer.ist.psu.edu/557554.html (verschiedene Formate) <-PDF Clack: An active network platform (u.a. ICT'03) Research Group Homepage Homepage eines Miterstellers des Papers(Prof.) The University Of New SouthWales ( CS ) , Australia 2003

Sowohl Probleme im Netzwerk, als auch das Echtzeit Experimentieren können zu Situationen führen in denen Änderungen verschiedener Netzwerkalgorithmen notwendig ist. CLACK ist ein Programm, mit dem sich click Elemente live steuern lassen. (so würde ich das rauslesen)

Themenkomplex 4: Adressvergabe im selbstorganisierenden Netz

ZAL Zero-Maintenance Address Allocation (beides das gleiche Dokument)

http://citeseer.ist.psu.edu/718066.html

ZAL Zero-Maintenance Address Allocation

Themenkomplex 5: Routing Protokolle Übersicht speziell: Zone Routing

Routing in Multi-Radio, Multi-Hop Wireless Mesh Networks (2004) [1]

  • see next paper as a basis
  • "We find that in a multi-radio environment, our metric significantly outperforms previously-proposed routing metrics by making judicious use of the second radio."


A high-throughput path metric for multi-hop wireless routing (2003) [2]

  • " This paper presents the expected transmission count metric (ETX), which finds high-throughput paths on multi-hop wireless networks."
  • "For long paths the throughput improvement is often a factor of two or more, suggesting that ETX will become more useful as networks grow larger and paths become longer."


Regional gossip routing for wireless ad hoc networks (2005) [3]

  • "Gossip based routing method has been used and re-investigated to reduce the number of messages in both wired networks and wireless ad hoc networks."
  • "We propose a regional gossip approach, where only the nodes within some region forward a message with some probability, to reduce the overhead of the route discovery in the network."
  • "We expect that the improvement should be even more significant in larger networks."


Papers from MC2R open call: Centralized channel assignment and routing algorithms for multi-channel wireless mesh networks (2004) [4]

  • "In this paper, we propose and evaluate one of the first multi-channel multi-hop wireless ad-hoc network architectures that can be built using standard 802.11 hardware by equipping each node with multiple network interface cards (NICs) operating on different channels."
  • "A detailed performance evaluation shows that with intelligent channel and bandwidth assignment, equipping every wireless mesh network node with just 2 NICs operating on different channels can increase the total network goodput by a factor of up to 8 compared with the conventional single-channel ad hoc network architecture."

Das MC2R Paper ist dem HYACINTH Projekt zugehörig und befasst sich größtenteils mit denselben Problemen. Auf der HYACINTH Seite (im wiki bzw. Projekt Homepage) ist der MC2R Artikel verlinkt (hyacinth-mc2r.pdf)


Design and performance analysis of a proxy-based indirect routing scheme in ad hoc wireless networks (2003) [5]

  • "In this paper, we propose a novel scheme, called the Applicative Indirect Routing (AIR), to control network traffic congestion and refine route availability by coping with unreliable links quickly."
  • "We further show that, in terms of packet delivery ratio, AIR is also a competitive protocol compared with such reactive protocols as Ad hoc On Demand Distance Vector (AODV) and Dynamic Source Routing (DSR)."


MR2RP: the multi-rate and multi-range routing protocol for IEEE 802.11 ad hoc wireless networks (2003) [6]

  • "By analyzing the medium access delay of the IEEE 802.11 medium access control (MAC) protocol, the proposed MR2RP is capable of predicting the transfer delay of a routing path and finding the best one, which has the minimum transfer delay from source to destination." (VoIP?)


On-demand multicast routing protocol in multihop wireless mobile networks (2002) [7]

A dynamic core based multicast routing protocol for ad hoc wireless networks (2002) [8]


Transmission-efficient routing in wireless networks using link-state information (2001) [9]

  • "Simulation results show that STAR is an order of magnitude more efficient than any topology-broadcast protocol proposed to date and depending on the scenario is up to six times more efficient than the Dynamic Source Routing (DSR) protocol, which has been shown to be one of the best performing on-demand routing protocols."


Performance evaluation of routing protocols for ad hoc wireless networks (2004) [10]

  • "Our results indicate that despite its improvement in reducing route request packets, CBRP has a higher overhead than DSR because of its periodic hello messages while AODV's end-to-end packet delay is the shortest when compared to DSR and CBRP. PAODV has shown little improvements over AODV."

Brainstorming Teil 2

Fassung des Begriffs 'Traffic Pattern' zur Optimierung der Topologiebildung in Multi-channel Multi-Hop Wireless Networks

  • Analogien zu Verkehrsflüssen in "Strassennetzwerken" ?

Erfahrungen mit / Verbesserungen von ETX

  • Comparison of routing metrics for static multi-hop wireless networks (2004) ACM Link:
Comparison of routing metrics ... 
    • "In this paper, we conduct a detailed, empirical evaluation of the performance of three link-quality metrics---ETX, per-hop RTT, and per-hop packet pair---and compare them against minimum hop count."

(stand vorher weiter oben)

Secure Routing in self-organizing networks (related: denial of service in ad-hoc networks).

Themenkomplex 6: Traffic Pattern

  • The performance impact of traffic patterns on routing protocols in mobile ad hoc networks (2004) [11]
    • "These results motivate the need for performance evaluation of MANETs to not only include rich and diverse mobility models as has been done in the past but also include diverse traffic patterns that stress a wide set of protocol design issues."
    • Aber auch dieses Modell eines "Traffic Patterns" ist extrem simpel: "Number of connections per traffic source (X): The number of connections per traffic source is varied to model the simultaneous connections per node in many potential applications in MANETs."
  • Im Zusammenhang mit VoIP: Study of delay patterns of weighted voice traffic of end-to-end users on the VoIP network (2002) [12]
    • "In this paper we study delay patterns of weighted voice traffic of end-to-end users on the Voice over Internet Protocol (VoIP) network. We compare the delay performance of voice traffic which varies with queue management techniques [...]"
  • Eigentlich eine interessante Thematik: A comparison of TCP performance over three routing protocols for mobile ad hoc networks (2001) [13]. Das Problem ist das hierbei verwendete triviale "Traffic Pattern":
    • "We simulated the steady-state conditions of a network with various background traffic loads generated by 10 and 40 constant bit rate (CBR) connections. The CBR packet sizes were fixed at 512 bytes. After a warm-up time of 100 seconds, one or more TCP connections were established over each of which an FTP file transfer was conducted for 900 seconds. The TCP packet size was 1460 bytes, and the maximum size of both the send and receive windows was 8. Since routing protocol performance is sensitive to movement patterns, 50 different mobility patterns (scenarios) were generated."
  • Automatically Inferring Patterns of Resource Consumption in Network Traffic [14]
    • the authors present AutoFocus, a tool to " ... automatically group traffic into minimal clusters of conspicuous consumption."
    • mainly for Network Admins, but also a way to approch "traffic patterns"

For Your Information

Suchseiten

Google scholar: http://scholar.google.com/ Citeseer: http://citeseer.ist.psu.edu/ ACM: http://portal.acm.org/

falls Citeseer offline (wie jetzt)/nicht verfügbar: "site:citeseer.ist.psu.edu" in normale google Suche einfügen

Interessante(?) Suchmaschine: http://clusty.com/ Clustering von Suchergebnissen nach Begriffen

Konferenzen

Ns2 stuff

Simulationsskripte für ns2:

hier:? http://www.monarch.cs.cmu.edu/cmu-ns.html

Zitat: The scenarios we used in preparing MobiCom'98 paper ``A Performance Comparison of Multi-Hop Wireless Ad Hoc Network Routing Protocols, are available as a 16MB tar file cmuns-scenarios.tar.gz that expands to 150MB when uncompressed.

Scratchpad

  • Topologiekontrolle - siehe auch MobiHOC 2005

Topology Control in Heterogeneous Wireless Networks: Problems and Solutions, Ning Li (University of Illinois at Urbana-Champaign), Jennifer Hou (University of Illinois) ?

On the Node-Scheduling Approach of Topology Control in Ad Hoc Networks mobihoc/2005 (future)

  • Blogeinträge

http://www.smallworks.com/archives/00000073.htm
http://www.smallworks.com/archives/00000072.htm

Security in PAN Themen

  • Vergleich von Microsoft Passport und Liberty Alliance

http://java.sys-con.com/read/36823.htm Passport und Liberty Alliance schließen sich nicht aus

  • Methoden zur Authentifizierung, Autorisierung und zum Accounting im PAN
  • wie oben, jedoch einsetzbar wenn lange Zeit offline