New Document
Computer Science
Computer Catlog
Comm Network Catlog

Network Components
Network Types
The OSI Model
Protocol Notations
Physical Layer
Modulation
Transmission Media
Multiplexing
Digitization and Synchronization
Physical Layer Standards
DataLink Layer
Error Checking
Retrans - Flow Control
Sliding Window Protocol
Data Link Layer Standards
Network Layer
Switching Methods
Routing
Congestion Control
Internetworking
Network Sub layers
Transport Layer
Transport Protocol
Transport Layer Standards
Session Layer
Session Layer Role
Session Protocol
Presentation Layer
Abstract Syntax Notation
Application Layer
Common Application
Specific Application
Message Handling
LAN
IEEE 802 Standards
ANSI FDDI Standard
ISDN
Frame Relay
Broadband ISDN & ATM

Digitization


    Digitization is essentially the opposite of modulation. Whereas in modulation a digital signal is modulated over an analog signal for transmission, in digitization an analog signal is converted into digital format through a process of sampling. For example, the analog signal resulting from human speech can be sampled and converted into digital data, transmitted over digital lines, and converted back to analog signal at the other end. These two functions are performed by a device called codec (coder/decoder).

    It is worth noting that, unlike modulation (which is an exact process since the digital signal at the source and the digital signal received at the destination are identical), digitization is only an approximate process because of sampling. Here the time interval for each sample is one millisecond. Each sample (denoted by a small black box) is a real value which is in turn represented by an integer in the range 0-255 so that it can be represented in one byte of data. This process (of representing a continuous value with a discrete value) is called quantization. The relatively small loss of information inherent in the process is called quantization error.

    The coding process generates the sample data from the analog signal. The decoding process regenerates an approximation of the original signal by fitting a smooth curve to the sampled points. The quality of the regenerated signal can be improved by increasing the sampling rate (i.e., reducing the sampling interval), but up to a limit dictated by the Nyquist's theorem. This limit is exercised by a popular digitization technique called Pulse Code Modulation (PCM) which uses a sampling rate twice that of the original signal frequency. For example, a 4 kHz speech signal is sampled at a rate of 8000 samples per second.The main advantage of digitization is that, due to its resistance to distortion, it is much easier to reliably transmit a digital signal over a long distance than an analog signal.


Synchronization


    When two devices are about to communicate, the transmitter should somehow notify the receiver as to when to expect to receive data. This allows the receiver to prepare itself for receiving the data. Furthermore, such notifications should occur frequently enough so that both devices maintain an agreement about the exact distribution of data over time. This process is called synchronization.

    There are two basic methods of synchronization: synchronous transmission and asynchronous transmission. In synchronous transmission, a clock signal is used as a common source of reference by both the transmitter and the receiver. By tying the data signal to the clock signal, either device can look at the clock signal to know where data bits may begin or end. The clock signal may be provided on a separate line, or be embedded in the data signal itself. Because having a separate clock line increases the costs, it is only used for covering very short distances (e.g., for connecting personal computers).

    In asynchronous transmission, the beginning and end of each byte of data is marked by start and stop bits. This enables the receiver to work out the byte boundaries. Because of its simplicity, asynchronous transmission is cheaper to implement and is therefore more widely used.