Lecture02 - Physical Layer

Lecture02 - Physical Layer - 8/31/2009 Objectives CNIT...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 8/31/2009 Objectives CNIT 24000 – Data Communication & Networks Identify the critical components of communication that occur at the Physical Layer Detail their functional purpose(s) Examine their inter-related operations Explore how each of these different aspects affects our ability to send data Physical Layer Concepts A Sample Data Communications Architecture Another Sample Architecture 1 8/31/2009 Data Digitization How do we represent data? We call them bits The smallest single piece of data Because we are representing information with these bits/bytes, they have meaning… Data Digitization – Encoding Standards American Standard Code for Information Interchange (ASCII) Developed by ANSI 7 bit characters + 1 parity bit Conversion process(es) Character encoding – The process of transforming human readable text into machine readable codes. Extended Binary Coded Decimal Interchange Code (EBCDIC) Developed by IBM (remains proprietary) 8 bit characters ISO 10646 (UNICODE) 16 bit characters Backwards compatible with ASCII Encoding Standards cont. Conversion between ASCII & EBCDIC Order of bits is not equivalent (MSB vs LSB) Data Transmission 2 major options: Serial Parallel There is no Series-Parallel as there is in the electrical fundamentals world UNICODE Includes multiple language support (some Asian dialects, Hebrew, Greek, Russian, Sanskrit) Includes symbols & marks not available in other codes Vendors are responsible for enabling support in OS Support in old OSes 2 8/31/2009 Serial Data Transmission A single wire carries all of the data from source to destination Data is sent as bits in a sequential manner What does this mean for data transmission rates? Parallel Data Transmission A single channel carries all of the data from source to destination Cable vs Bus Nearly all LAN & WAN technologies use serial communications Data is sent as bits arranged across eight or more wires simultaneously What does this mean for data transmission rates? Typically utilized internal to computing devices and only over short distances Serial vs Parallel Serial Sequential Transmission of bits Able to send data over long distances Nearly all LAN/WAN technologies utilize serial communications Common Implementations: Telco/ISP links Ethernet SATA Data Transmission - Timing Parallel Simultaneous Transmission of bits Able to send data over short distances Limited by electrical signal interference and degradation by EMI Common Implementations: PCI, memory bus Many HDD interfaces Older Printers & some scanners Communications involves the exchange of detectable signals (data) When do we sample the wire? The mechanism to handle this is controlled in one of two ways: Synchronous Asynchronous 3 8/31/2009 Synchronous Data Transmission Synchronous = Time-based A clock provides the synchronization to handle this communication Internal External Asynchronous Data Transmission No timing source to control when data is sent Utilizes start & stop bits to indicate beginning and end of communication May have 1, 1.5, or 2 stop bits Timing is maintained regardless of presence of data Synchronization must be re-established for every communication cycle Synch vs Asynch Timing and Transmission Serial & Synchronous T-Carrier system, SONET Frame Relay, xDSL, ISDN, USB, Firewire Serial & Asynchronous Most “terminal” applications, keyboard, mice ATM Firewire Parallel & Synchronous Internal buses Printers & Scanners Parallel & Asynchronous Not many communications protocols or standards Reviewed for HPC 4 8/31/2009 Bi-directional communication Simplex vs Duplex Half-Duplex vs Full-Duplex Modulation Many analog systems are still in use Meaning we must be able to convert digital to analog The process of converting digital bits into an analog signal that can be sent over an analog medium is called Modulation The process of converting an analog signal into a digital bit stream is called Demodulation Devices that perform both of these operations are called Modems (modulator/demodulator) Modulation Amplitude Modulation Manipulate the amplitude of the carrier wave in order to indicate data In true AM, only the amplitude changes. The Frequency & Phase remain constant Example? Basic Carrier Wave 5 8/31/2009 Frequency Modulation Manipulate the frequency of the carrier wave in order to indicate data In true FM, only the frequency changes. The Amplitude & Phase remain constant Phase Modulation Manipulate the phase of the carrier wave in order to indicate data In true PM, only the phase changes. The Frequency & Amplitude remain constant Usually referred to as FSK in industry and publications Usually referred to as PSK in industry and publications How are high data-rates keyed? Bit vs Baud Increase # of signaling events Interpret more than one bit per baud # of detectable events 2 4 8 16 32 64 128 # of Also called bits/baud 1 2 3 4 5 6 7 Dibit Tribit Quadbit Quintbit Bit range 0-1 00-11 000-111 0000-1111 0000011111 000000111111 00000001111111 Enhanced PSK Methods Quadrature Phase Shift Keying DQPSK Phase Shift Bit Pattern 0° 90° 180° 270° 00 01 10 11 Quadrature Amplitude Shift Keying (QAM) Ex. 16QAM 6 8/31/2009 Theories for Data Transmission 90° Nyquist’s Theorem 1101 1100 1110 1111 180° 1001 1000 1010 1011 0° 0001 0000 0010 0011 0101 0100 0110 0111 -90° As the number of unique, detectable symbols increases, the data transmission rate of the link increases. C = 2W where C is the max data rate & W is the bandwidth of the channel. C = 2W log2 M where M is the # of detectable events Example: V.34 Modem C = 2 (3200 Hz) log2 (16) = 6400 * 4 = 25,600 bps 802.11 Radio C = 2 (5,000,000 Hz) log2 (16) = 10,000,000 * 4 = 40 Mbps Theories for Data Transmission cont Shannon’s Law The higher the data rate, the more interference is caused by a specified amount of noise in the channel Results in a higher bit error rate Signal to Noise Ratio (S/N) = 10 log (signal power/noise power) Data Compression The process of replacing redundant data patterns with a smaller code that represents the pattern Using advanced compression algorithms, a reduction of up to 75% of the original size can be seen Can be done in either hardware or software Should NOT be done in both, simultaneously C = W log2 (1 + S/N) 7 8/31/2009 Popular Compression Schemes Text Huffman – usually with Lemple-Ziv Shannon-Fano Multiple Access Technologies Multiplexing Concentrating Switching Circuit Switching Packet Switching Image/Video JPEG MPEG WMV Why do we need these approaches? Audio MP3 WAV* AAC Vorbis Switching Allows a temporary channel to be established, maintained, and terminated between a source and a destination Circuit Switching A dedicated channel is defined from source to destination prior to data forwarding Guaranteed resources for duration connection No other data is forwarded along the channel Gives the appearance of a direct connection If network is busy, a call-blocking algorithm is enabled 8 8/31/2009 Packet Switching Individual pieces of the message are sent from source to destination over an minimallyconstrained network A dedicated channel is not defined from source to destination prior to data forwarding Requires control information in addition to data payload No guarantee of resources for any part of the connection Other data is forwarded along portions of the same channel If network is busy, all connections will experience degradation of services Switching cont Overhead Connectionless Small Benefit Re-route data Reliability Set-up Little to none Yes Addressing Global Known as Datagram Error Correction End-user devices By VC Flow Control Enduser devices By VC Connection Oriented Large Local logical channel number Virtual Circuit or Packet Things to Know Review 9 8/31/2009 Questions? 10 ...
View Full Document

Ask a homework question - tutors are online