mobile networks
play

Mobile Networks Prof. Jean-Pierre Hubaux EPFL The Playground of - PDF document

1 Mobile Networks Prof. Jean-Pierre Hubaux EPFL The Playground of Communication Networks Control Information server Cellular network server Information MSC BS server MSC Control server Control IP network server PSTN Router


  1. 1 Mobile Networks Prof. Jean-Pierre Hubaux EPFL

  2. The Playground of Communication Networks Control Information server Cellular network server Information MSC BS server MSC Control server Control IP network server PSTN Router Switch WLAN MSC: Mobile services Switching Center Station BS: Base Station WLAN: Wireless LAN Gateway PSTN: Public Switched Telephone NW 2

  3. Wireless communication and mobility g Aspects of mobility: i user mobility : users communicate “anytime, anywhere, with anyone” i device portability : devices can be connected anytime, anywhere to the network g Wireless vs. mobile Examples � � stationary computer (desktop) � � laptop connected to an Ethernet network � � desktops in historic buildings � � Personal Digital Assistant (PDA) g The demand for mobile communication creates the need for integration of wireless networks or mobility mechanisms into existing fixed networks: i telephone network � cellular telephony (e.g., GSM) i local area networks � Wireless LANs (e.g.,IEEE 802.11) i Internet � Mobile IP

  4. Wireless systems: development over the last 20 years cordless satellite wireless cellular phones phones networks LAN, wireless 1980: 1981: CT0 PAN NMT 450 1982: 1984: Inmarsat-A 1983: CT1 AMPS 1986: NMT 900 1987: 1988: CT1+ Inmarsat-C 1991: 1991: 1989: CDMA 199x: D-AMPS CT 2 (USA: IS-95) proprietary 1991: 1992: 1992: DECT Inmarsat-B GSM 1995/96/97: Inmarsat-M 1993: IEEE 802.11, PDC HIPERLAN 1994: DCS 1800 1998: Iridium analog 2001 - ?: IEEE 802.15 2002 - ?: digital Bluetooth UMTS/IMT-2000 CDMA-2000 (USA) NMT: Nordic Mobile Telephone DECT: Digital Enhanced Cordless Telecom. AMPS: Advanced Mobile Phone System (USA) DCS: Digital Cellular System CT: Cordless Telephone PDC: Pacific Digital Cellular UMTS: Universal Mobile Telecom. System PAN: Personal Area Network

  5. Mobile devices Laptop Wireless sensors • functionally eq. to desktop • Limited proc. power • standard applications • Small battery Mobile phones • voice, data • simple text displays RFID tag Pager • A few thousands of logical gates • receive only • Responds only • tiny displays PDA • simple text to the RFID reader • simple graphical displays messages requests (no battery) • character recognition • simplified WWW performance performance 5

  6. Upcoming wireless networks g Personal communications g Vehicular networks g Sensor networks g RFID 6

  7. Upcoming wireless networks: personal communications g Proliferation of small operators g Operators in shared spectrum g Mesh networks g Hybrid ad hoc networks g Mesh networks TAP 3 TAP 2 TAP 1 Internet HS TAP 7 TAP 5 TAP 4 7 TAP 6

  8. Preventing greedy behavior at the MAC layer in WiFi hotspots The access point is trusted The access point is trusted Well-behaved node Cheater For more information: http://domino.epfl.ch 8

  9. Wireless Operators in a Shared Spectrum game G = ( Players , Strategy , Utility function ) g operators → players g radio range → strategy g utility: useful coverage of their pilot signal: g = − γ ⋅ coverage interference U i i i i : maximum power range (regulator) R MAX : minimum power range (for coverage) R MIN Static game – Pareto-optimal Nash equilibria γ = = Small : r r R A B MAX γ = = Large : freely roaming users r r R g A B MIN power control of the pilot signal g users attach to the base station with the best g Repeated game – pilot signal: A Nash equilibrium based on R MIN is enforceable using punishments ⎡ ⎤ where the channel gain: ⋅ ⎢ ⎥ P g max i iu ⎢ ⎥ 1 ∑ http://winet-coop.epfl.ch/ + ⋅ = N P g P g ⎢ ⎥ i 0 9 j ju 2 ⎣ ⎦ iu d j iu

  10. Vehicular communications: why? g Combat the awful side-effects of road traffic i In the EU, around 40’000 people die yearly on the roads; more than 1.5 millions are injured i Traffic jams generate a tremendous waste of time and of fuel g Most of these problems can be solved by providing appropriate information to the driver or to the vehicle 10

  11. Example of attack : Generate “intelligent collisions” SLOW DOWN The way Similar attack: stop a highway is clear For more information: http://ivc.epfl.ch 11

  12. European Project SeVeCom • SeVeCom: Secure Vehicular Communications • Started January 2006; Duration: 3 years; Total budget: 3 MEuros • Research topics: key management, secure communication, privacy, tamper-proof device • http://www.sevecom.org • A small, well-balanced consortium, with a precise goal • Partners: 12

  13. Sensor network for environmental monitoring 13 http://www.commonsense.ch

  14. Reference model Application Application Transport Transport Network Network Network Network Data Link Data Link Data Link Data Link Physical Physical Physical Physical Radio link 14

  15. Layer model i security g Application layer i service location i Signal processing i new applications, multimedia i adaptive applications g Transport layer i congestion and flow control i quality of service g Network layer i addressing, routing, mobility management i hand-over i authentication g Data link layer i media access i multiplexing i media access control i encryption i modulation g Physical layer i interference i attenuation i frequency allocation 15

  16. Overlay Networks - the global view Integration of heterogeneous fixed and mobile networks with varying transmission characteristics wide area vertical hand-over metropolitan area campus-based horizontal hand-over in-house 16

  17. The classical solution for mobile networks g 2 nd generation (GSM, IS-41,…) deployed, 3 rd generation (UMTS,…) recently deployed g Huge, expensive fixed infrastructure g Operational responsibility: network operators (telcos, ISPs) 17

  18. The new paradigm: mobile ad hoc networks g Terminal and node merge g Everything is potentially mobile g Initial applications: communication in the battlefield (Packet Radio Networks, in the 70’s) g The network is self-organized when it is run by the users themselves 18 g Similar trend at the application layer (Napster � Gnutella)

  19. Upper bound for the throughput of ad hoc networks If we have: - identical randomly located nodes n - each capable of transmitting bits/s W λ Then the throughput ( ) obtainable by each node n for a destination is randomly chosen ⎛ ⎞ W λ = Θ⎜ ( ) ⎟ n ⎜ ⎟ log ⎝ n n ⎠ Ref: P. Gupta, P. Kumar, The Capacity of Wireless Networks IEEE Transactions on Information Theory, March 2000 19

  20. Intuition behind the upper bound N nodes (users) O(N) transmissions from left to right over O( ) transmission links N mean 1 O( ) capacity per attempted transmission N Ways to improve scalability: O(N) users O(N) users Ways to improve scalability: • Directional antennas • Directional antennas N Cut set ~ • Locality of the traffic • Locality of the traffic • Hybrid system • Hybrid system 20

  21. Terminal + Node = Terminode www.terminodes.org g All network functions (packet forwarding, flow control, error control,…) and terminal functions (coding/decoding, A/D and D/A, storage, ciphering,…) are embedded in the terminode Destination g A communication must be relayed by intermediate terminodes g The network is self - organized : it is operated by its users g All terminodes are wireless and battery-operated g All terminodes are potentially Source mobile g There exist gateways to other networks (e.g., Internet and 21 cellular networks)

  22. References g J. Schiller: Mobile Communications , Addison-Wesley, Second Edition, 2004 g B. Walke: Mobile Radio Networks , Wiley, Second Edition, 2002 g T. Rappaport: Wireless Communications , Prentice Hall, Second Edition, 2001 g M. Schwartz: Mobile Wireless Communications, Cambridge University Press, 2004 22

Recommend


More recommend