3.

Evolution of network technologies Presentation Transcript

1. 3. Evolution of network technologies 3.1. Evolution of transport technologies (backbone transport - switching/routing and transmission systems) 3.2. Evolution of access networks technologies to broadband (xDSL, CATV, Broadband Wireless Access) 3.3. Evolution of mobile networks (to 3G and beyond) 2. 3.1. Evolution of transport technologies A. Public Network Principles These 3 techniques will be discussed next Transport (Core/ Backbone) Network Transmission Network Terminations Access Gateway Wireless Technologies Access Network Twisted Pair Cable/Coax Powerline Optical Fiber Switching/ Routing 3. 4. Years 1840 1900 1950 1975 1980 1990 2000 Telegraph Manual switching Electro-mechanics Analog Digital Hand telegraph Operator Cr-B 55 QE 70 DE-1 PABX-1 PABX-2 PABX-NG (IP) Ethernet Gbit Ethernet Private Public ISDN DE-NG (IP) ATM 10 Gbit Ethernet 1884 Self-dial 1935 (B-ISDN) IP /X25/SMDS FR Cellular radio GSM UMTS/IMT-2000 DE-2 NMT B. Evolution of switching technologies G-MPLS MPLS 5. Switching technologies (Cntd) CS (PSTN) FR (FS, 70-s, DN) IP ( PS-DG, 60s, Internet) .25 ( PS-VC, 60-s, DN) MS (Tlg) ( CS, 80-s, B-ISDN) Connection-oriented technologies Connectionless-oriented technologies 6. Transport technologies in network backbones ATM IP OB BACKBONE OPTIONS MPLS 7. C. Transport technologies in network backbones - ATM ATM IP OB BACKBONE OPTIONS MPLS 8. ATM and the IETF model ATM Layer 1/2 Quality of Service (QoS) Multimedia Transport Constant Bit Rate (CBR) - Voice Variable Bit Rate (VBR) - WWW Available Bit Rate (ABR) E-mail Unspecified Bit Rate (UBR) Application Transport Network Data Link Physical 9. Putting ATM to work Voice Delay Delay Variation Loss Data Delay Delay Variation Loss Video Delay Delay Variation Loss Multimedia Delay Delay Variation Loss 1 2 3 4 5 10. ATM QoS Constant Bit Rate for switched TDM traffic (AAL1): Access Aggregation (TDM for GSM/GPRS, ATM for UMTS) Digital Cross-Connect Backbone Voice Transport - Basic Real-time Variable Bit Rate for bursty, jittersensitive traffic: Backbone Voice Transport Advanced (AAL2) Optional for Packetized Access Transport & Aggregation (3G UTRAN, 2G CDMA) Non realtime Variable Bit Rate for bursty high priority data traffic: 2.5G data services Unspecified Bit Rate+ with Minimum B/W Guarantee for internal data : Operations, Admin & Maintenance (element management, stats collection, network surveillance, ) Billing data Internal LAN traffic (email, web, file sharing, ) between operators business offices LINE RATE (LR) CBR nrt-VBR ABR UBR UBR+ rt-VBR

11. ATMs role in the networks segments Premise LAN/Desktop Campus Backbone Access Low Speed (56/64) Medium Speed (E1) High Speed (>E1 to SDH) Integrated Access Backbone Voice Data Video Multimedia 1 2 3 4 5 12. ATM and the Competition Premise LAN/Desktop - Ethernet, HS Ethernet, Gigabit Ethernet Campus Backbone - HS Ethernet, Gigabit Ethernet Access Low Speed (56/64) - ISDN, ADSL Medium Speed (E1) xDSL, E1 High Speed (>E1 to SDH) - SDH Integrated Access - E1, xDSL, SDH Backbone Voice Traditional Telephony, IP Backbones Data Optical Backbones, IP Backbones Video Optical Backbones, IP Backbones Multimedia Optical Backbones, IP Backbones 13. ATM Summary Multimedia Not used much on Premise Present use in Backbone Predictable Performance/Guaranteed QoS 14. ATM IP OB BACKBONE OPTIONS D. Transport technologies in network backbones - IP MPLS 15. IP and the IETF Model Network Layer (Layer 3) End-to-End Addressing/Delivery Best Effort Service Physical Data Link Network Transport Application IP 16. Putting IP to work Voice Delay Delay Variation Loss Data Delay Delay Variation Loss Video Delay Delay Variation Loss Multimedia Delay Delay Variation Loss 1 2 3 4 5 17. IPs Role in the networks segment Premise LAN/Desktop Campus Backbone Access Low Speed (56/64) Medium Speed (E1) High Speed (>E1 to SDH) Integrated Access Backbone Voice Data Video Multimedia 1 2 3 4 5 18. IP and the Competition Premise LAN/Desktop No Real Competition Campus Backbone No Real Competition Access Low Speed (56/64) ISDN Medium Speed (E1) xDSL, non-channelized E1 Integrated Access E1, multiple E1, Frame Relay, SDH Backbone Voice Traditional Telephony Data Optical Backbones Video Optical Backbones Multimedia Optical Backbones, ATM Backbones 19. Why use IP? Wide acceptance Internet popularity Global reach - IP Standards Mature standards Interoperability IP Protocol characteristics Simple protocol Good general purpose protocol Best Effort Protocol 20. IP summary Globally popular Originally developed for data Mature standards Interoperability Best Effort Protocol Voice over IP gaining popularity 21. We need a better Internet Reliable as the phone Next Generation Networks Powerful as a computer Mobile as a cell phone and Working right away as a TV set 22. Main directions of improvement 1. Scalability 2. Security 3. Quality of service 4. Mobility IPv6 23. ATM IP OB BACKBONE OPTIONS E. Transport technologies in network backbones - MPLS MPLS

24. Routers that handle MPLS and IP are called Label Switch Routers (LSRs) LSRs at the edge of MPLS networks are called Label Edge Routers (LERs) Ingress LERs classify unlabelled IP packets and appends the appropriate label. Egress LERs remove the label and forwarding the unlabelled IP packet towards its destination. All packets that follow the same path ( LSP- Label Switched Part ) through the MPLS network and receive the same treatment at each node are known as a Forwarding Equivalence Class (FEC). MPLS Model LSP FEC A B LER LSR LSR LER 25. E. Switching Technologies - Summary Driving forces (mid of 80th) Common platform for different types of traffic ISDN is not suitable (N-ISDN low bit rates, circuit switching) ATM will not become as the most important switching technology since 2000s Main competitors (Performance/Price) # Ethernet (LANs) # xDSL (Access) # IP /MPLS ( Backbones ) 26. ATM IP OB BACKBONE OPTIONS F. Transmission technologies in network backbones - OB MPLS 27. Stated data rates for the most important end-user and backbone transmission technologies -1 28. Stated data rates for the most important end-user and backbone transmission technologies -2 29. Stated data rates for the most important end-user and backbone transmission technologies -3 30. Stated data rates for the most important end-user and backbone transmission technologies -4 31. Evolution of transmission technologies Years 1900 1970 1980 1990 2000 Frequency modulation, FDM PDH 1935 Time multiplexing, TDM Wavelength multiplexing Transmission media Modulation methods Frequency modulation systems SDH WDM Copper cable Copper cable Radio Coax Coax Fiber Optics Satellite radio Radio all optical 32. Technological limitations of different transmission media Optical fibers are the only alternative at high bandwidth and distances Fiber Coax Cellular Wireless* *Capacity in Mbit/s/sq_km, Bandwidth 500 MHz 250 Copper Twisted Pair 33. Optical systems move from backbone to access Entry process of optical systems into access occurs very slowly ... Prognosis 10-15 years, reason: exchange of copper cables and maturity of technologies yesterday today tomorrow 5 Years 10-15 Years Access Metro Backbone Copper Optical ISDN POTS Fiber optics and laser Copper Optical ADSL Optical additional: color filter and optical amplifier additional: optical switch, color converter 34. Today optical transmission system consists mainly of electronics and passive optical components SDH and WDM process signals most of the time only electronically Amplifiers are the only active optical elements in the network SDH networks: WDM networks: Signal Multiplexer Cross connector Optical fiber Amplifier TDM MUX TDM MUX, Cross- connect, control Electrical signal Opto- electronics Active optics Passive optics Electronics Optical fiber TDM MUX WDM MUX, Cross- connect Electrical signal Active optics Passive optics

Electronics WDM MUX Passive optics: - lenses - prisms - grating Control Passive optics: - lenses - grating - mirrors Optical signal 35. Day after tomorrow: All-optical switching and multiplexing All-optical systems process signals only optically Electronics disappear Nortel (03/2002): large scale stand-alone optical switches are likely for longer term market requirements Optical fiber Switch Matrix Aktive Optik Passive optics WDM MUX Passive optics: - lenses - prisms - grating Control Active optics: - Switch color converter - amplifier Optical signal Signal Multiplexer Switch Amplifier 36. Future photonic switches Optics are good for transport Electronics are good for switching Electronics as far as possible Evolution instead of Revolution at least, 5 years for first all-optical systems in backbone and metro area 37. G. Concluding remarks - growth of network capacity and Death of distance phenomenon Growth of network capacity reduction of information transmission costs New generation of transmission systems new ratio Cost of transmission/Bandwidth PCM SDH/SONET DWDM Bandwidth becoming a less dominating factor in cost of connection Cost of one-bit-transmission has an obvious tendency to become very close to zero in long distance communications systems Flattened networks Death of distance phenomenon (F. Cairncross, 1997) Challenges for operators 38. 39. Bandwidth using 32 terrestrial carriers connecting to the New York metropolitan area have a combined potential capacity of 818.2 Terabits per second. Of that, only 22.6 Terabits per second -- 2.8 percent -- of network bandwidth is actually in use Int'l IP Using City Bandwidth, Bandwidth, Gbit/s Gbit/s London 550.3 9,5 Paris 399.4 9,3 Frankfurt 320.2 10,3 Amsterdam 267.1 8,2 40. Development of costs for IC sector Source: Economist Cost of information processing $ per instruction per second Cost of a three-minute telephone call from New York to London, $ to be continued to be continued