Business data communications 4e chapter 10

Chapter 10: Transmission Efficiency Business Data Communications, 4e Transmission Efficiency: Multiplexing  Several data sources share a common transmission medium simultaneously  Line sharing saves transmission costs  Higher data rates mean more cost-effective transmissions  Takes advantage of the fact that most individual data sources require relatively low data rates Multiplexing Diagram Alternate Approaches to Terminal Support  Direct point-to-point links  Multidrop line  Multiplexer  Integrated MUX function in host Direct Point-to-Point Multidrop Line Multiplexer Integrated MUX in Host Frequency Division Multiplexing  Requires analog signaling & transmission  Total bandwidth = sum of input bandwidths + guardbands  Modulates signals so that each occupies a different frequency band  Standard for radio broadcasting, analog telephone network, and television (broadcast, cable, & satellite) FDM Example: ADSL  ADSL uses frequency-division modulation (FDM) to exploit the 1-MHz capacity of twisted pair  There are three elements of the ADSL strategy  Reserve lowest 25 kHz for voice, known as POTS  Use echo cancellation or FDM to allocate a small upstream band and a larger downstream band  Use FDM within the upstream and downstream bands, using “discrete multitone” Synchronous Time-Division Multiplexing (TDM)  Used in digital transmission  Requires data rate of the medium to exceed data rate of signals to be transmitted  Signals “take turns” over medium  Slices of data are organized into frames  Used in the modern digital telephone system  US, Canada, Japan: DS-0, DS-1 (T-1), DS-3 (T-3),  Europe, elsewhere: E-1, E3, … SONET/SDH  SONET (Synchronous Optical Network) is an optical transmission interface proposed by BellCore and standardized by ANSI  Synchronous Digital Hierarchy (SDH), a compatible version, has been published by ITU-T  Specifications for taking advantage of the highspeed digital transmission capability of optical fiber SONET/SDH Signal Hierarchy STS-1 and STM-N Frames Statistical Time Division Multiplexing  “Intelligent” TDM  Data rate capacity required is well below the sum of connected capacity  Digital only, because it requires more complex framing of data  Widely used for remote communications with multiple terminals STDM: Cable Modems  Cable TV provider dedicates two channels, one for each direction  Channels are shared by subscribers, so some method for allocating capacity is needed\-typically statistical TDM Cable Modem Scheme Transmission Efficiency: Data Compression  Reduces the size of data files to move more information with fewer bits  Used for transmission and for storage  Combines w/ multiplexing to increase efficiency  Works on the principle of eliminating redundancy  Codes are substituted for compressed portions of data  Lossless: reconstituted data is identical to original (ZIP, GIF)  Lossy: reconstituted data is only “perceptually equivalent” (JPEG, MPEG) Run Length Encoding  Replace long string of anything with flag, character, and count  Used in GIF to compress long stretches of unchanged color, in fax transmissions to transmit blocks of white space Run-Length Encoding Example Huffman Encoding  Length of each character code based on statistical frequency in text  Tree-based dictionary of characters  Encoding is the string of symbols on each branch followed String TEA SEA TEN Encoding 10 00 010 011 00 010 10 00 110 Lempel-Ziv Encoding  Used in V.42 bis, ZIP  buffer strings at transmitter and receiver  replace strings with pointer to location of previous occurrence  algorithm creates a tree-based dictionary of character strings Lempel-Ziv Example Video Compression  Requires high compression levels  Three common standards used:  M-JPEG  ITU-T H.261  MPEG MPEG Processing Steps  Preliminary scaling and color conversion  Color subsampling  Discrete cosine transformation (DCT)  Quantization  Run-length encoding  Huffman coding  Interframe compression ... Encoding is the string of symbols on each branch followed String TEA SEA TEN Encoding 10 00 010 011 00 010 10 00 110 Lempel-Ziv Encoding  Used in V.42 bis, ZIP  buffer strings at transmitter and... “Intelligent” TDM  Data rate capacity required is well below the sum of connected capacity  Digital only, because it requires more complex framing of data  Widely used for remote communications. .. Codes are substituted for compressed portions of data  Lossless: reconstituted data is identical to original (ZIP, GIF)  Lossy: reconstituted data is only “perceptually equivalent” (JPEG, MPEG)