Evaluation of Delay/Disruptive Tolerant Network Solutions in Networks under Intentional Attack
Upcoming SlideShare
Loading in...5
×

Like this? Share it with your network

Share

Evaluation of Delay/Disruptive Tolerant Network Solutions in Networks under Intentional Attack

  • 347 views
Uploaded on

Presented during 7th NATO CAX Forum. Hotel Ergife, Rome, 24th September 2012

Presented during 7th NATO CAX Forum. Hotel Ergife, Rome, 24th September 2012

  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
    Be the first to like this
No Downloads

Views

Total Views
347
On Slideshare
346
From Embeds
1
Number of Embeds
1

Actions

Shares
Downloads
2
Comments
0
Likes
0

Embeds 1

http://www.docshut.com 1

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide

Transcript

  • 1. DITEN NATO M&S COE University of GenoaEvaluation of Delay/Disruptive TolerantNetwork (DTN) Solutions in Networksunder Intentional Attack Speaker: Authors: LT(N) Eng. Alessandro CIGNONI PhD Marco CELLO LT(N) Eng. Alessandro CIGNONI Prof. Mario MARCHESE Page 1
  • 2. DITEN NATO M&S COE University of Genoa Outline• Introduction to Delay/Disruptive Tolerant Network (DTN)• DTN as a Strategy for Information Assurance and Infrastructure Network Reliability• Cyber Hyper-Domain• M&S for Cyber Battle Labs and CAX Page 2
  • 3. DITEN NATO M&S COE University of Genoa DTN concept• The DTN architecture embraces the concepts of occasionally-connected networks• The basis for this architecture lies on the Interplanetary Internet• Various operational environments, including those subject to disruption and disconnection and those with high-delay;• Deep space is one specialized example• Other networks to which DTN architecture applies Page 3
  • 4. DITEN NATO M&S COE University of Genoa DTN concept DTN solution applies when End-to-end connection is : not permanently guaranteed; intentionally and not intentionally interrupted; operating with very large delays; operating intermittently Page 4
  • 5. DITEN NATO M&S COE University of Genoa DTN application scenarios Emergency operations, interventions in hazardous areas,… Page 5
  • 6. DITEN NATO M&S COE University of GenoaDTN Architecture Bundle Protocol DTN architecture and Bundle Protocol– DTN architecture based on the introduction of an overlay above transport or other lower layer protocols– The essential point is that in such an overlay, delays and disruptions can be handled at each DTN ‘hop’ in a path between sender and dest– Nodes on the path can provide storage– The DTN architecture does not require contemporaneous end-to- end connectivity Page 6
  • 7. DITEN NATO M&S COE University of GenoaDTN architecture Bundle protocol– The basic unit of data in the Bundle Protocol is a “bundle” which is a message that carries application layer protocol data units– The BP can interface with different lower layer (usually transport) protocols through “Convergence Layer Adapters”, (CLAs) Application Application Bundle Bundle Bundle CLA x CLA x CLA z CLA z Lower Layer x Lower Layer x Lower Layer z Lower Layer z (e.g. transport,…) (e.g. transport,…) (e.g. transport,…) (e.g. transport,…) Other Layers Other Layers Other Layers Other Layers Network x Network x Network z Network z Network x Network z Page 7
  • 8. DITEN NATO M&S COE University of Genoa DTN as an overlay solution– DTN architecture is suited for acting as overlay on top of a heterogeneous network– By installing a Bundle Protocol Agent (BPA) on end-points and nodes at the border of homogeneous segments, the end-to-end path can be divided into many DTN hops.– On each DTN hop different CLAs can be used Page 8
  • 9. DITEN NATO M&S COE University of Genoa Information storage Int Nodes Another important difference between DTN and traditional TCP/IP networking is related to information storage In standard networks information is persistently stored only at end nodes This may not be the case in challenged networks. In DTN networks information is persistently (long-term) stored at intermediate DTN nodes Page 9
  • 10. DITEN NATO M&S COE University of Genoa Information storage Int Nodes• This feature differentiates the DTN architecture also from PEPs.• In contrast, bundles can be stored at intermediate nodes for extended durations, and also be saved in persistent memory Page 10
  • 11. DITEN NATO M&S COE University of GenoaDTN as Network Defense Strategy The new idea is to use DTN to increment Infrastructrure Network Resilience, mitigating the effects of an intention attack to network links/nodes; The attack is considered as a bandwidth reduction up until no bandwidth availability Page 11
  • 12. DITEN NATO M&S COE University of GenoaNetwork Resilience and Cyber Defence Page 12
  • 13. DITEN NATO M&S COE University of GenoaNetwork Resilience and Cyber Defence Network Resilience to: Protect Core Infrastructure Assure Information Superiority in the Cyber Battle Field - Sithuation Awarness - Common Operational Picture Page 13
  • 14. DITEN NATO M&S COE University of GenoaDTN as Network Defense Strategy Effect of the attack Attack Bandwidth Cancellation Link Disruption and no service Page 14
  • 15. DITEN NATO M&S COE University of GenoaDTN as Network Defense Strategy If network nodes use the DTN architecture, can this help managing and mitigating the negative effect of the attack? Even if the hypothesis must be deeper verified, preliminary analysis support the idea Page 15
  • 16. DITENNATO M&S COE University of Genoa Preliminary results Page 16
  • 17. DITEN NATO M&S COE University of Genoa Very simple model S 1 2 3 D Links behavior modeled as 4 independent continuous Time Markov Chains πG stationary probability of Good state (no interruption); πB stationary probability of Bad state (interruption) TB sojourn time in Bad state (exponentially distributed with parameter λB) 1/λB average sojourn time in Bad state Tx transmission time Page 17
  • 18. DITEN NATO M&S COE University of Genoa Very simple model Single IP packet generated by S to D; TCP/UDP latency (must wait to complete a path) – (πB/ λB) 4 + 4Tx DTN latency – 4(πB/ λB) + 4Tx Page 18
  • 19. DITENNATO M&S COE University of Genoa Very simple model Comparison of TCP/IP and DTN latency 300 Latency between S and D [s] 250 200 TCP/IP latency 150 DTN latency 100 50 0 1 3 5 7 9 11 13 15 17 19 Sojourn Time in Bad state [s] Page 19
  • 20. DITEN NATO M&S COE University of Genoa M&S Pillar More accurate Protocol Model; Protocol Behaviour Simulation - NS3 Based / OPNET Based Network Infrastracture Simulation – OPNET Based Page 20
  • 21. DITEN NATO M&S COE University of Genoa Cyber Hyper-Domain DTN in Software Defined Networking Hypothesis: Cyber Hyper Dimensional or Cyber Hyper-Domain – Free BSD Jials – Stanford Clean Slate Projects Page 21
  • 22. DITEN NATO M&S COE University of Genoa Cyber Hyper-Domain Hyper-Dimentional Cyber Domain (men-driven and/or autonomus cognitive processes to inter- dimension switch); – Time – Space – Virtual Space – Autonomous Systems Domains /Topology / Routing Strategies and Protocols Different NetworkTopology and Routing Strategy are separated in different Jails. Page 22
  • 23. DITEN NATO M&S COE University of Genoa Cyber Defence CAX Cyber Hyper-Domain M&S Distributed Battle Labs Interconnection Men and Autonomous Agents CAX in the simulated Cyber Hyper-Domain Page 23
  • 24. DITEN NATO M&S COE University of Genoa Cyber Defence CAX Assure Information Superiority in the Cyber Hyper-Domain which directly translates into Power Superiority in the Battlefield Page 24
  • 25. DITEN NATO M&S COE University of GenoaContact :NATO M&S COE: DITEN – University of Genoa:LT(N) Eng. Alessandro CIGNONI Prof. Mario Marchesesesto.natomes.areadottrina02@smd.difesa.it mario.marchese@unige.it PhD Marco Cello marco.cello@unige.it Page 25