All computer systems produce binary data. For these data to be understood by both the sender and receiver, both must agree on a standard system for representing the letters, numbers, and symbols that compose messages. The coding scheme is the language that computers use to represent data.
Coding
A character is a symbol that has a common, constant meaning. A character might be the letter A or B, or it might be a number such as 1 or 2. Characters also may be special symbols such as ? or &. Characters in data communications, as in computer systems, are represented by groups of bits that are binary zeros (0) and ones (1). The groups of bits representing the set of characters that are the "alphabet" of any given system are called a coding scheme, or simply a code.
A byte is a group of consecutive bits that is treated as a unit or character. One byte normally is composed of 8 bits and usually represents one character; however, in data communications, some codes use 5, 6, 7, 8, or 9 bits to represent a character. For example, representation of the character A by a group of 8 bits (say, 01 000 001) is an example of coding.
There are three predominant coding schemes in use today. United States of America Standard Code for Information Interchange (USASCII, or, more commonly, ASCII) is the most popular code for data communications and is the standard code on most microcomputers. There are two types of ASCII; one is a seven-bit code that has 128 valid character combinations, and the other is an eight-bit code that has 256 combinations. The number of combinations can be determined by taking the number 2 and raising it to the power equal to the number of bits in the code because each bit has two possible values, a 0 or a 1. In this case 27 = 128 characters or 28 = 256 characters.
A second commonly used coding scheme is ISO 8859, which is standardized by the International Standards Organization, ISO 8859 is an eight-bit code that includes the ASCII codes plus non-English letters used by many European languages (e.g., letters with acents). If you look closely at Figure 2.20, you will see that HTML often uses ISO 8859.
Unicode is the other commonly used coding scheme. There are many different versions of Unicode. UTF-8 is an eight-bit version which is very similar to ASCII. UTF-16, which uses 16-bits per character (i.e. two bytes, called a "word"), is used by Windows. By using more bits, UTF-16 can represent many more characters beyond the usual English or Latin characters, such as Cyrillic or Chinese.
We can choose any pattern of bits we like to represent any character we like, as long as all computers understand what each bit pattern represents. Figure 3.15 shows the eight-bit binary bit patterns used to represent a few of the characters we use in ASCII.
Transmission Modes
Parallel Parallel transmission is the way the internal transfer of binary data takes place inside a computer. If the internal structure of the computer is eight-bit, then all eight bits of the data element are transferred between main memory and the central processing unit simultaneously on eight separate connections. The same is true of computers that use a 32-bit structure; all 32 bits are transferred simultaneously on 32 connections.
Figure 3.16 shows how all eight bits of one character could travel down a parallel communication circuit. The circuit is physically made up of eight separate wires, wrapped in one outer coating. Each physical wire is used to send one bit of the eight-bit character. However, as far as the user is concerned (and the network for that matter), there is only one circuit; each of the wires inside the cable bundle simply connects to a different part of the plug that connects the computer to the bundle of wire.
|
Character |
ASCII |
|
A |
01000001 |
|
B |
01000010 |
|
C |
01000011 |
|
D |
01000100 |
|
E |
01000101 |
|
a |
01100001 |
|
b |
01100010 |
|
c |
01100011 |
|
d |
01100100 |
|
e |
01100101 |
|
1 |
00110001 |
|
2 |
00110010 |
|
3 |
00110011 |
|
4 |
00110100 |
|
! |
00100001 |
|
$ |
00100100 |
Figure 3.15 Binary numbers used to represent different characters using ASCII
Figure 3.16 Parallel transmission of an 8-bit code
Figure 3.17 Serial transmission of an 8-bit code
Serial Serial transmission means that a stream of data is sent over a communication circuit sequentially in a bit-by-bit fashion as shown in Figure 3.17. In this case, there is only one physical wire inside the bundle and all data must be transmitted over that one physical wire. The transmitting device sends one bit, then a second bit, and so on, until all the bits are transmitted. It takes n iterations or cycles to transmit n bits. Thus, serial transmission is considerably slower than parallel transmission—eight times slower in the case of 8-bit ASCII (because there are 8 bits). Compare Figure 3.17 with Figure 3.16.
Basic Electricity
TECHNICAL FOCUS
There are two general categories of electrical current: direct current and alternating current. Current is the movement or flow of electrons, normally from positive (+) to negative (-). The plus (+) or minus (-) measurements are known as polarity. Directcurrent (DC) travels in only one direction, whereas alternating current(AC) travels first in one direction and then in the other direction.
A copper wire transmitting electricity acts like a hose transferring water. We use three common terms when discussing electricity. Voltage is defined as electrical pressure —the amount of electrical force pushing electrons through a circuit. In principle, it is the same as pounds per square inch in a water pipe. Amperes (amps) are units of electrical flow, or volume. This measure is analogous to gallons per minute for water. The watt is the fundamental unit of electrical power. It is a rate unit, not a quantity. You obtain the wattage by multiplying the volts by the amperes.
Digital Transmission
Digital transmission is the transmission of binary electrical or light pulses in that it only has two possible states, a 1 or a 0. The most commonly encountered voltage levels range from a low of +3/-3 to a high of +24/-24 volts. Digital signals are usually sent over wire of no more than a few thousand feet in length.
All digital transmission techniques require a set of symbols (to define how to send a 1 and a 0), and the symbol rate (how many symbols will be sent per second).
Figure 3.18 shows four types of digital transmission techniques. With unipolar signaling, the voltage is always positive or negative (like a DC current). Figure 3.18 illustrates a unipolar technique in which a signal of 0 volts (no current) is used to transmit a zero, and a signal of +5 volts is used to transmit a 1.
An obvious question at this point is this: If 0volts means a zero, how do you send no data?For the moment, we will just say that there are ways to indicate when a message starts and stops, and when there are no messages to send, the sender and receiver agree to ignore any electrical signal on the line.
To successfully send and receive a message, both the sender and receiver have to agree on how often the sender can transmit data—that is, on the symbol rate. For example, if the symbol rate on a circuit is 64 Hertz (Hz) (64,000 symbols per second), then the sender changes the voltage on the circuit once every 1/64,000 of a second and the receiver must examine the circuit every V64,000 of a second to read the incoming data.
Figure 3.18 Unipolar, bipolar, and Manchester signals (digital)
In bipolar signaling, the 1′s and 0′s vary from a plus voltage to a minus voltage (like an AC current). The first bipolar technique illustrated in Figure 3.18 is called nonreturn to zero (NRZ) because the voltage alternates from +5 volts (a symbol indicating a 1) and —5 volts (a symbol indicating a 0) without ever returning to 0 volts. The second bipolar technique in this figure is called return to zero (RZ) because it always returns to 0volts after each bit before going to +5 volts (the symbol for a 1) or —5 volts (the symbol for a 0). In Europe, bipolar signaling sometimes is called double current signaling because you are moving between a positive and negative voltage potential.
In general, bipolar signaling experiences fewer errors than unipolar signaling because the symbols are more distinct. Noise or interference on the transmission circuit is less likely to cause the bipolar’s +5 volts to be misread as a —5 volts than it is to cause the unipolar’s 0 volts as a +5 volts. This is because changing the polarity of a current (from positive to negative, or vice versa) is more difficult than changing its magnitude.
How Ethernet Transmits Data
The most common technology used in LANs is Ethernet3; if you are working in a computer lab on campus, you are most likely using Ethernet. Ethernet uses digital transmission over either serial or parallel circuits, depending on which version of Ethernet you use. One version of Ethernet that uses serial transmission requires 1/10,000,000 of a second to send one symbol; that is, it transmits 10 million symbols (each of 1 bit) per second. This gives a data rate of 10 Mbps, and if we assume that there are 8 bits in each character, this means that about 1.25 million characters can be transmitted per second in the circuit.
Ethernet uses Manchester encoding. Manchester encoding is a special type of bipolar signaling in which the signal is changed from high to low or from low to high in the middle of the signal. A change from high to low is used to represent a 0, whereas the opposite (a change from low to high) is used to represent a 1. See Figure 3.18. Manchester encoding is less susceptible to having errors go undetected, because if there is no transition in midsignal the receiver knows that an error must have occurred.
