liveliness evaluation of a cooperation and accounting
play

Liveliness Evaluation of a Cooperation and Accounting Strategy in - PowerPoint PPT Presentation

Liveliness Evaluation of a Cooperation and Accounting Strategy in Hybrid Networks Attila Weyland, Thomas Staub and Torsten Braun ASWN 2004, Boston August 9, 2004 University of Bern Institute of Computer Science and Applied Mathematics


  1. Liveliness Evaluation of a Cooperation and Accounting Strategy in Hybrid Networks Attila Weyland, Thomas Staub and Torsten Braun ASWN 2004, Boston August 9, 2004 University of Bern Institute of Computer Science and Applied Mathematics Computer Networks and Distributed Systems Research Group (RVS)

  2. 8/9/2004 Table of Contents 2 Liveliness Evaluation of a Cooperation and Accounting Strategy in Hybrid Net. - Attila Weyland • Introduction • Motivation • Concept • Architecture • Operation • Simulation Scenario & Parameters • Results • Summary & Outlook

  3. 8/9/2004 Introduction 3 Liveliness Evaluation of a Cooperation and Accounting Strategy in Hybrid Net. - Attila Weyland • Multi-hop Cellular Networks – Combine dynamics of mobile ad hoc networks and reliability of infrastructured wireless networks – Compared to single-hop • Increased coverage area • Dynamic adaptation of network topology Provider’s Gateway Backbone

  4. 8/9/2004 Motivation 4 Liveliness Evaluation of a Cooperation and Accounting Strategy in Hybrid Net. - Attila Weyland • New context to deal with weaknesses of mobile ad hoc networks such as – Routing – Security – Cooperation • Stimulate cooperation without threat of punishment • Make cooperation a rewarding alternative to selfishness

  5. 8/9/2004 CASHnet Concept 5 Liveliness Evaluation of a Cooperation and Accounting Strategy in Hybrid Net. - Attila Weyland • Every time a node wants to transmit a self- generated packet, it has to pay with Traffic Credits (TC) • Every time a node forwards a packet, it gets Helper Credits (HC) • Traffic Credits can be bought for real money or traded for Helper Credits at service stations +HC Service Provider’s +HC Gateway Station Backbone -TC +HC

  6. 8/9/2004 Architecture 6 Liveliness Evaluation of a Cooperation and Accounting Strategy in Hybrid Net. - Attila Weyland • Assumptions – Tamper resistant device which allows safe execution of CASHnet functions and maintains two accounts – Distance (in hop counts) to gateway provided by routing protocol – Sufficient processing power on the node • Security mechanisms are based on public key cryptography – Nodes authenticate themselves using certificates with short life time issued by the provider – Transmitted messages are digitally singed ensuring non-repudiation (data integrity and data origin authentication)

  7. Smart 8/9/2004 Operation Card CASH net 7 Gateway Provider’s Liveliness Evaluation of a Cooperation and Accounting Strategy in Hybrid Net. - Attila Weyland Backbone Service -4 TC +1 HC +1 HC +1 HC Station -20 HC - 10 RM Provider’s +20 TC Backbone Gateway +1 HC +1 HC -3 TC

  8. 8/9/2004 Simulation Scenarios 8 Liveliness Evaluation of a Cooperation and Accounting Strategy in Hybrid Net. - Attila Weyland A B Mobile Node Gateway Service Station C D Parameter Value Initial Traffic Credits account state 100 TC Initial Real Money account state 500 Traffic Helper Credits exchange rate 1:1 Exchange threshold at Service Stations 10 HC Distance threshold to Service Stations 50 m

  9. 8/9/2004 Simulation Parameters 9 Liveliness Evaluation of a Cooperation and Accounting Strategy in Hybrid Net. - Attila Weyland Parameter Value Area 1500 m x 800 m Number of nodes 40 Transmission range 250 m Mobility model random waypoint Speed u. d. between 1 and 10 m/s Pause time u. d. between 0 and 20 s Packet generation rate 1, 0.2, 0.1 pkt/s Number of Service Stations 1, 2, 9, 12 Simulation time 900 s

  10. 8/9/2004 Starvation Periods 10 CASHnet, 5 s packet interval, 2 service stations, run 0 Liveliness Evaluation of a Cooperation and Accounting Strategy in Hybrid Net. - Attila Weyland CASHnet, 5 s packet interval, 2 service stations, run 0 CASHnet, 5 s packet interval, 12 service stations, run 0 Starvation period Starvation period Starvation period 39 39 37 37 35 35 39 33 33 31 31 37 29 29 27 27 35 25 25 23 23 33 21 21 19 19 31 17 17 15 15 29 13 13 11 11 27 9 9 7 7 25 5 5 3 3 23 1 1 0 100 200 300 400 500 600 700 800 900 0 100 200 300 400 500 600 700 800 900 21 Simulation time (s) Simulation time (s) 19 5 s packet interval 5 s packet interval 17 2 service stations 12 service stations 15 13 11 • Starvation: a node is unable to transmit self- 9 7 generated packets because of missing TCs 5 3 1 0 100 200 300 400 500 600 700 800 900 Simulation time (s)

  11. 8/9/2004 Starvation Events/Duration Category 11 CASHnet, 1 s packet interval, 1 service station Liveliness Evaluation of a Cooperation and Accounting Strategy in Hybrid Net. - Attila Weyland CASHnet, 1 s packet interval, 1 service station CASHnet, 1 s packet interval, 12 service stations 20 Mean starvation occurance 20 Mean starvation occurance 20 Mean starvation occurance 18 18 1 s packet interval 1 s packet interval 16 16 18 1 service station 12 service stations 14 14 12 12 16 10 10 8 8 14 6 6 4 4 12 2 2 0 0 0 40 80 120 160 200 240 280 320 360 400 440 480 520 560 600 640 680 720 760 800 0 40 80 120 160 200 240 280 320 360 400 440 480 520 560 600 640 680 720 760 800 10 Starvation length (s) Starvation length (s) CASHnet, 10 s packet interval, 1 service station CASHnet, 10 s packet interval, 12 service stations 20 20 Mean starvation occurance Mean starvation occurance 8 18 18 10 s packet interval 10 s packet interval 16 16 6 1 service station 12 service stations 14 14 12 12 4 10 10 8 8 2 6 6 4 4 0 2 2 0 40 80 120 160 200 240 280 320 360 400 440 480 520 560 600 640 680 720 760 800 0 0 0 40 80 120 160 200 240 280 320 360 400 440 480 520 560 600 640 680 720 760 Starvation length (s) 800 0 40 80 120 160 200 240 280 320 360 400 440 480 520 560 600 640 680 720 760 800 Starvation length (s) Starvation length (s)

  12. 8/9/2004 Results 12 Liveliness Evaluation of a Cooperation and Accounting Strategy in Hybrid Net. - Attila Weyland • Duration and frequency of starvation events correlates with – Number of Service Stations – Location of Service Stations • Simulation results affected by mobility model – Random waypoint movement paths behave centric (2 service stations worse than 1 centered) • Per packet charging lets nodes run out of Traffic Credits/Real Money quickly

  13. 8/9/2004 Summary & Outlook 13 Liveliness Evaluation of a Cooperation and Accounting Strategy in Hybrid Net. - Attila Weyland • Highly decentralized accounting and security architecture • Selfish nodes are allowed, but cooperation is encouraged via rewards • Cost sharing between sender & receiver • Evaluation of starvation property through simulations � Compare with other cooperation schemes � Use different mobility models, e.g. restricted random waypoint � Study effects of possible extensions (e.g. charging for ad hoc only traffic, deposit payment for receiving traffic, increasing granularity) � Specify charging/remuneration relation

  14. 8/9/2004 Implementation 14 Liveliness Evaluation of a Cooperation and Accounting Strategy in Hybrid Net. - Attila Weyland • ns-2 [Vint Project], Wireless and Mobility extensions [Rice] and AODV+ [Hamidian] • Class CashnetNode inherits from MobileNode • Agent at ns2 src/sink does rewarding • Class CMUTrace extended for CASHnet events

Recommend


More recommend