• Ei tuloksia

Telecommunication Software

N/A
N/A
Info
Lataa
Protected

Academic year: 2022

Jaa "Telecommunication Software"

Copied!
50
0
0

Kokoteksti

(1)

Telecommunication Software

Lecture 3 , Octo ber 8, 2002

(2)

Map of last and today lectures

• Last time:

– We overviewed various types of networks – We gave a map to networks

• Today’s plan:

– First ½ we dig into details for some special networks – Second ½ we start reviewing network applications

(3)

Network taxonomy

• Criterion: transmission technology

– Broadcast links – Point-to-point links

• Criterion: scale (geographical extent)

– Very local area networks (personal area network) – Local area networks LAN

– Metropolitan area networks MAN – Wide area networks WAN

– Internetworks

• Criterion: organization

– Private networks – Public networks

• Other criterions

– Wireless networks – Home networks

(4)

LANs

• Privately-owned networks within a single

building/company/campus up to a few km in size

• Widely used to connect computers and workstations in company offices and factories to share resources and exchange information

• Restricted in size => the worst case transmission time is bounded and known in advance

• Transmission technology allows

– speeds of 10 Mbps to 100 Mbps

– low delays (microseconds, nanoseconds) – very few errors

– Newer LANs up to 10 Gbps

(5)

LANs 2

• IEEE specified most of today’s LANs (IEEE 802 committee)

(6)

IEEE 802.3 (CSMA/CD): Ethernet

• Medium access control method: CSMA/CD

• All stations share the same transmission medium

• 1970s: XEROX PARC -> Experimental Ethernet, 3 Mbps

• Early 1980: DEC, Intel, XEROX (DIX) -> DIX Ethernet, 10 Mbps

• 1985: IEEE 802.3 standard

• 1995: IEEE 802.3u (Fast Ethernet), 100 Mbps

• 1997: IEEE 802.3x -> full duplex mode doubling the transmission speed (both directions)

• 1998: IEEE 802.3z (Gigabit Ethernet), 1-2 Gbps

• Newer LANs: up to 10 Gbps

(7)

Medium Access Method

• Half-duplex operation mode

– CSMA/CD

– Waiting a random time <--> truncated binary exponential backoff algorithm

– The algorithm is used by all collision-detecting stations to calculate their individual retransmission delay (backoff delay)

• Full-duplex operation mode

– Two stations share the physical medium

– We assume the medium is capable of having simultaneous bidirectional transmission w/o interference

– No contention possible => no CSMA/CD needed

(8)

Ethernet frame format

• Preamble field

– 7 bytes used by the receiving station to establish bit synchronization

• Start Frame Delimiter (SFD) field

– bit sequence 10101011, enables the receiving station to detect the beginning of a frame and locate its first bit

• Destination and Source address fields

– identify the producing station and the station(s) for which the frame is intended

– 2 or 6 bytes long, chosen a priori for a particular network

(9)

Ethernet frame format 2

• Length field

– the current number of bytes transported in the Data field

• Pad field

– Used to fill up the frame with extra bytes for obtaining a frame with minimum length of 64 bytes, w/o preamble and SFD

• Checksum field

– The Frame Check Sequence (FCS) is used to convey a 4-byte Cycle Redundancy Check (CRC) value for detecting bit

transmission errors

– Polynomial (international standard for CRC):

X32+X26+X23+X22+X16+X12+X11+X10+X8+X7+X5+X4+X2+X+1

– CRC produced at transmitting station as function of destination, source, length, data, and pad fields

– Receiving station computes a CRC sum of the same fields and compares it with the checksum value; if a transmission error occurred, it is highly detectable

(10)

(Truncated) binary exponential backoff algorithm

• CSMA/CD

• used by all collision-detecting stations to calculate their individual retransmission delay (backoff delay)

– after 1st collision each station waits 0 or 1 slot times before trying again; if each station picks the same random number, they will collide again

– after 2nd collision each station picks 0,1,2, or 3 slot times before trying again, and waits that number of slot times

– if a 3rd collision occurs, the next time number of slots to wait is chosen randomly by each station from the interval 0 to 23-1

– after i collisions, a random number between 0 to 2i-1 is chosen, and that number of slots is skipped

– after 10 collisions the randomization interval is frozen at a maximum of 1023 slots

– after 16 collisions a failure is reported and recovery is up to higher layers

(11)

(Truncated) binary exponential backoff algorithm 2

• The algorithm dynamically adapts to the number of stations trying to send

• If randomization interval for all collisions was 1023

– the chance for two stations to collide for a second time: negligible – the average wait after a collision: hundreds of slot times => delay

• If each station always delayed for 0 or 1 slots

– stations will collide again and again

• The algorithm ensures a low delay when only few stations collide

• Also ensures collision is resolved in reasonable interval when many stations collide

• Truncating the backoff at 1023 keeps the bound from growing

too large

(12)

Ethernet retrospective

• Used for over 2 decades, enters its 3

rd

: rare phenomenon

• Simple and flexible => reliable, cheap, easy to maintain

• Interworks easily with TCP/IP, the dominant network protocol (both connectionless)

• Able to evolve in certain essential ways

– Speeds have gone up by several orders of magnitude – Hubs and switches have been introduced

– The above changes did not require changing the software

(13)

IEEE 802.11 (Wireless LAN)

• Ethernet is getting competition

• Standard specifying a LAN based on wireless technology

• Specifies connectivity for portable and mobile stations

Portable station: may be moved to different locations but used in a network only while stationary

Mobile station: may be used in a network while in motion

• Different properties wrt wired LANs

– Limited physical range

– Vulnerability to security attacks – Significantly higher error rates – Dynamic topology

(14)

Wireless LAN 2

Backbone segment

WSeg WSeg WS

WS WS WS

APS APS

Distribution Segment (DS)

- DS enables stations to move transparently between different wireless Segments (WSeg)

- DS not specified by 802.11 (can be implemented based on many technologies including wired LAN)

- each WSeg has one station connected to DS, functioning as Access Point station (APS)

- APS enables wireless stations located in the respective WSeg areas to communicate with stations in different WSegs

(15)

Wireless LAN 3

• Stations of a wireless LAN possess functions to provide the following services

– Association service – Disassociation service – Authentication service – Privacy service

• APS have additional functions to provide the following services

– Distribution service – Integration service

– Re-association service

(16)

Minimum wireless LAN topology

• Minimum LAN 802.11: one WSeg and two WSs

– Requires only authentication, privacy, association and disassociation services

• LAN 802.11 w/o backbone segment: AD-HOC

NETWORK

(17)

Transmission media in Wireless LAN

• 2.4 GHz band frequency hopping spread spectrum (FHSS)

– Frequency changed within a specified band in a pseudo-random fashion, known only to transmitters and receivers

– w/o knowing the frequency sequence and change interval, eavesdropping is impossible

• 2.4 GHz band direct sequence spread spectrum (DSSS)

– A spreading code is used to spread and despread the transferred data

– Each wireless station has its own spreading code

• Baseband infrared

– Infrared technique (for remote controls of TV sets)

– Infrared signals cannot penetrate walls, so cells in different rooms are well isolated from each other

(18)

Frequency Hopping Spread Spectrum

• Uses 79 channels, each 1 MHz wide, starting at the low end of the 2.4 GHz band

• Uses a pseudorandom number generator to produce the sequence of frequencies hopped to

• If all stations use same seed to the generator and stay synchronized in time => they will hop to same

frequencies simultaneously

• The time interval spent at each frequency: dwell time

– Adjustable parameter, but < 400 ms

• Intruder not knowing the hopping sequence & dwell time

cannot eavesdrop on transmissions

(19)

IEEE 802.12 (Demand-Priority)

• Specifies a 100 Mbps LAN controlled by a repeater

• Its physical star topology can be enlarged by cascading multiple repeaters => tree topology

• Cascadable repeaters have a dedicated cascade port reserved for connection to repeaters only

• Ports used to connect stations are called local ports (>=2)

• For interconnection with other LANs bridges may be linked to local ports performing the required media and service adaptation

• Bridges are transparent to repeaters (treated as normal

stations)

(20)

Possible IEEE 802.12 LAN Topology

Repeater

Repeater Station

Station Station Station

Station Station

(21)

Basic operation

• Stations first send a transmission request to associated repeater

• Repeater returns a transmission grant

• Requests: normal priority (NP) or higher priority (HP)

• HP: real time voice, video, data transmission

• Two queues (for NP and HP) maintained to store incoming requests

• While HP queue has items to process, it serves them before serving requests from NP queue (FIFO queues)

• Timers monitor the pending time of NP requests to avoid their starvations

– At about 250 ms a NP request is upgraded to a HP request

(22)

Processing a request

• Repeater (R) sends a transmission grant to relevant station

• R sends an idle-down signal to remaining stations

• R issues an incoming signal to remaining stations

– To be prepared to receive the frame

• Frame sent by transmitting station is forwarded to addressed receiving stations

• During frame transmission, new requests can be accepted

• These new requests will never interrupt ongoing

transmissions, even if they are HP

(23)

Station and Repeater architecture

• Stations and repeaters possess the same physical layer functionality, divided into 2 sublayers:

– physical medium independent sublayer (PMI): provides basic frame transmission by applying certain data scrambling and encoding techniques

– physical medium dependent sublayer (PDI): provides signal generation and recognition, and clock recovery

• MII: medium-independent interface

– between PMI and PDI

– specifies signaling time and electrical interface characteristics

• MDI: medium-dependent interface

– between PDI and medium

specifies mechanical, electrical, optical and transmitted signal interface requirements

• Both interfaces may be implemented physically or logically

(24)

Station and Repeater architecture 2

• MAC layer is different for stations and repeaters

• Stations sublayer supports 2 well defined MAC formats and interfaces (IEEE 802.3 and IEEE 802.5)

– Only one format and interface supported at the same time within an IEEE 802.12 network

• Repeaters sublayer provides arbitration and control

of frame forwarding between different ports of a

repeater

(25)

Wide Area Networks (WANs)

• Span a large geographical area (country, continent)

• Contain a collection of machines (hosts) intended for running user (application) programs

• Hosts are connected by a (communication) subnet

• Hosts are owned by customers, subnets are typically

owned and operated by a phone company or an Internet Service Provider

• Subnet carries messages from host to host

• Separation of pure communication aspects of the

network (subnet) from application aspects (hosts) greatly

simplifies the complete network design

(26)

WAN Subnet

• Subnet consists of transmission lines and switching elements (routers)

– Transmission lines move bits between machines; made of copper wire, optical fiber, radio links

– Routers are specialized computers connecting >= 3 transmission lines

(27)

Packet-switched WANs

(28)

Satellite WANs

• Each router has an antenna through which it can send and receive

• All routers can hear output from the satellite

• In some cases they can also hear upward transmissions of their fellow routers to the satellite

• In some cases routers are connected to a substantial point-to-point subnet with only some of them having a satellite antenna

• Satellite networks are inherently broadcast => most

useful when broadcast property important

(29)

Example WANs

• Telex networks

• Telephone networks

• Packet-switched data networks

– Intended for computer data transmission

• Television distribution networks

• Radio distribution networks

• Integration of these networks and their services into a single Integrated Services Digital Network (ISDN) is studied by the International Telecommunication Union Standardization

Sector (ITU-T)

(30)

Addressing

• Common to all networks: they require means to address connected stations

• Different addressing techniques and formats exist:

– IEEE address format – ITU-T address format – OSI address format

(31)

Applications

• Applications in a communication network provide a variety of services

• We overview

– Classical applications – Distributed applications – Multimedia applications – Real-time applications

• Some of these applications may belong to several categories above

• We study their various perspectives

(32)

Classical applications

• Designed for and applied in today’s

telecommunications and data communication networks

• We review

– File transfer – Virtual terminal – Electronic mail – Remote job entry – Telephone

– Telefax

– Telex and Teletex

(33)

File transfer

• Computing nodes in a network may maintain their own file (storage) system

• A networked application providing file transfer services:

enables remote users to copy files between storage devices located at different nodes

• FTP (File Transfer Protocol): one of the 1

st

and still widely used file transfer application

– Specified in RFC959 (1985)

– Based on client-server paradigm

– Requires control connection and data connection

– For a reliable end-to-end file transmission TCP is used

(34)

FTP

• Data transfer processes (DTP)

– Server and client used to read from and write to the local file systems and control data connections while transferring files

• Protocol interpreters (PI)

– Server and client used to process and exchange FTP commands and replies via the control connection

• Client interface

– May additionally exist at the client side to enable human users to interact with the client PI in a user-friendly way

– May have a local definable language

• FTP allows clients to set up control connections to 2

different servers to arrange file transfer between them,

remotely

(35)

FTP Options

• While transferring files, representation transformations may be performed

– 3 parameters that are adjustable by clients at each transfer job individually: data type, data structure, transmission mode

• Data type ->

clients specify if file data should be interpreted as 8-bit ASCII, 8-bit EBCDIC, sequence of bits or blocks with size defined by an extra parameter

• Data structure -> defines whether file is organized as a

– File structure (sequence of bits)

– Record structure (list of data records)

– Page structure (set of independent indexed data pages)

• Transmission mode -> denotes the functions performed on the file while transferred

– Stream mode (data transferred uninterpreted as flow of bytes)

– Block mode (file transferred as series of data blocks enabling error recovery and restart at application level)

– Compressed mode (data amount reduced) to get higher throughput

(36)

Virtual terminal

• Human users typically interact with a computing system via a terminal: keyboard and display

• In networks is desirable to use a terminal of any local node to interface with any other remote node

• Variety of terminals with different capabilities exists

– Their heterogeneous interfaces must be supported by each computing system

• To avoid this: VIRTUAL TERMINAL (VT)

– abstract and universal terminal covering a wide range of existing terminals

– The remote node can only provide one terminal type: VT – The node with the actual terminal performs the required

adaptations between VT and its physical terminal characteristics

(37)

TELNET

• Widely used application providing a VT service

– Defined in RFC854 (1983)

– Requires bidirectional and reliable connection between local and remote nodes

– On both sides a Network VT (NVT) exists, providing a 8-bit oriented protocol to provide the VT service

– NVT can be used independently to interact with local terminal device or local application process

– Interface between NVT and terminal/process: display+keyboard – NVT operates generally in line-buffered mode

– TELNET provides the concept of option negotiation (enables sophisticated terminals/processes with new service capabilities)

(38)

X.400 electronic mail

• ITU-T’s X.400 series of recommendations specifies a Message

Handling System (MHS) enabling users to exchange messages in a store-and-forward fashion

• Seen as electronic mail (e-mail)

• Around since 80’s; before 90’s only in academia; after that everybody

• Very informal

– Own jargon: CUL8R, ASAP, BTW, ROTFL, IMHO

(39)

MHS

• MHS messages (msg) consist of an envelope in which information is carried as msg content between users

• Users: persons or computer processes

• Msg is submitted by its originator and conveyed by MHS to one/more recipients

• Each user is attached a dedicated computer process, the User Agent (UA) that interacts on the user’s behalf with the Message Transfer System (MTS)

• MTS -> multiple Message Transfer Agents (MTAs) that cooperate in a store-and-forward fashion to transport and delivery msg

• Basic MTS capabilities

– Submission, Transfer, Delivery of msg between UAs

– Additionally: (non-)delivery notifications to msg originators

(40)

MHS as a basis

• Users connected to different MHSs can interact with a given MHS via Access Units (AUs)

– Located between MTS and the external MHS – Perform the required adaptations between both

• F.410: recommended time targets for

– Msg delivery, transfer, delivery notification

• MHS as foundation on which other application specific msg systems are built

– InterPersonal Messaging (IPM) – Electronic Data Interchange (EDI) – Voice messaging system

(41)

Internet email

• Internet community has its own email system since 1982

• Simple Mail Transfer Protocol (SMTP)

– Used to reliably transfer mail between Mail Transfer Agents (MTAs)

– Used also by user (agents) to post mail

• To retrieve mail: protocols used between UAs and MTAs

– Post Office Protocol (POP)

– Internet Message Access Protocol (IMAP)

• Format of mail msg: RFC822, ’82: ASCII

– Msg envelope (header)

– Unstructured msg body conveying the msg content

(42)

Internet email 2

• MIME (Multipurpose Internet Mail Extensions)

– Allows textual msg headers and text in character sets richer than ASCII

– Provides an extensible set of different formats for non-textual msg bodies (e.g. audio, video, ps)

– Allows multipart msg bodies

• SMTP and POP can convey only RFC822 msg

– MIME msg need to be adequately encoded and decoded before and after transfer

• IMAP supports MIME

(43)

SMTP, ‘82

• Used to reliably transfer RFC822 mail between MTAs

• Used also by UAs to post mail

• If receiving MTA is not the final destination, it relays the msg to the next relevant MTA until final destination

• If multiple recipients defined, sending and receiving agents

negotiate which recipients can be reached through respective MTA

• If some recipients cannot be reached through selected MTA, the sending agent tries another relay agent

• One copy only of the mgs is forwarded to all identified recipients

• Expects a reliable, connection-oriented transport service (such as TCP)

• Transaction-oriented protocol, exchanging requests and responses in ASCII plaintext.

(44)

POP-version 3, ‘96

• Simple protocol enabling UAs to retrieve msg stored at MTA

• POP connection established between one UA and one MTA

• Requires a reliable connection-oriented transport service (such as TCP)

• Connection-oriented protocol using ASCII for requests/replies

• All mail retrieved must agree with RFC822 (

ASCII presentation

)

• Reason to have UA and MTA with different protocols

– MTA is part of MTS which requires that every node is continuously running to receive/relay mail

– To release terminals from this requirement, they make use of an UA process to retrieve mail on demand

(45)

IMAP-version 4, revision 1,’96

• Allows users to retrieve/manipulate electronic mail msg stored at an MTA’s site

• Permits users to create, delete, control multiple remote msg folders (mailboxes)

• Permits parsing, searching, and selective fetching of msg attributes, texts, and portions

• A UA may work offline and resynchronize with its related MTA later on

• Requires reliable connection-oriented transport service

• Transaction-oriented protocol interacting by requests and responses in ASCII plaintext

• Handles and interprets MIME msg

• Able to process multiple requests in parallel

(46)

A comparison of POP3 and IMAP

(47)

MIME, ‘98

Problem: RFC 822 allows only ASCII msg to be sent, the mail lines must be <= 1000 characters (SMTP)

Solution: Internet community defined a MIME standard compliant and compatible with RFC822 and SMTP

• Adds

– New header fields to the original RFC822msg header to indicate MIME version (1.0)

– The content media type conveyed in the msg body – The content transfer encoding applied

– 2 additional fields carrying the content identifier and content description

• Content media types are defined in RFC2046, currently

divided into 7 categories

(48)

MIME Content Media types

(49)

Remote job entry

• Problem: in a network environment it may be worthwhile migrating particular jobs to remote nodes for processing

• Solution: a general concept to transfer, control and

manipulate jobs in a network specified in ISO8831, ’92

• The users of a Job Transfer and Manipulation (JTM) service have different roles

• Initiation agency

– User that initiates the execution of a job description to a JTM provider

– Upon the issued job description, JTM compiles a work specification containing instructions such as to

• Which files are requires

• Where are they stored

• To which node they must be moved for further processing

• Where the new generated files should be stored finally

(50)

Remote job entry 2

• Source agency

– User storing files that can be requested by the JTM provider depending on the work specification

• Sink agency

– User at which the JTM provider may dispose files (printers/file storage)

• Execution agency

– User that performs functions on input files given by the JTM provider and returns output files to the JTM provider

• It is transparent to JTM provider how source, sink, or

execution agencies process given files or manage

retrievable files

Viittaukset

LIITTYVÄT TIEDOSTOT

The martti desktop application uses a small database file on each of the computer where the application is installed, while the web application uses a centralized database

You should send one PDF file including your solutions to the pen-and-paper problems and the report for the programming problem, and a single compressed file including the code for

These frameworks include iCloud storage services, location services, data storage technologies like SQLite and XML management and the Core Data framework unique to iOS.. The

I will go through the steps that I took when designing the API and provide a working example how to integrate the Amazon S3 cloud file storage and Windows Azure to Qt applications....

FTP (File Transfer Protocol) is used to transmit collected data from the measurement device to the processing server.. Telnet is used to command FMDM program on

To provide an example, if defect prediction is done for the alpha branch of the version 6.2 using CM, then the CM file for 6.1 is used as training data, and the CM file for 6.2 is

The present study attempts to briefly summarize the basic gene transfer vectors, mouse model of atherosclerosis, and vascular endothelial growth factors, which are nowadays widely

Proton-transfer- reaction mass spectrometer (PTR-MS) has become a widely used technique in real-time monitoring of VOCs (Ellis and Mayhew, 2014). With PTR-MS, VOCs are