A network is a communication system with two or more stations that can communicate with one another. When it is desired to have each computer communicate with two or more additional computers, the interconnections can become complex. As Fig. 12-1 indicates, if four computers are to be interconnected, there must be three links to each PC.
The number of links L required between N PCs (nodes) is determined by using the formula
L = N(N – 1)/2
Assume, e.g., there are six PCs. The number of links is
L = 6(6 – 1)/2 = 6(5)/2=30/2 =15.
The number of links or cables increases in proportion to the number of nodes involved. The type of arrangement shown in Fig. 12-1 is obviously expensive and impractical. Some special type of network wiring must be used, a combination of hardware and software that permits multiple computers to be connected inexpensively and simply with the minimum number of links necessary for communication.
An office with 20 PCs is to be wired so that any computer can communicate with any other. How many interconnecting wires (or links, L) are needed?
Types of Network
Each computer or user in a network is referred to as a node. The interconnection between the nodes is referred to as the communication link. In most computer networks, each node is a personal computer, but in some cases a peripheral device such as a laser printer or an embedded controller built into another piece of equipment can be a node. There are four basic types of electronic networks in common use: wide-area networks (WANs), metropolitan-area networks (MANs), local-area networks (LANs), and personal-area networks (PANs). Let’s take a brief look at each.
Wide-Area Network (WANs)
A WAN covers a significant geographic area. Local telephone systems are WANs, as are the many long-distance telephone systems linked together across the country and to WANs in other countries. Each telephone set is, in effect, a node in a network that links local offices and central offices. Any node can contact any other node in the system. Telephone systems use twisted-pair wire and coaxial cable, as well as microwave relay networks, satellites, and fiber- optic cabling.
There are also WANs that are not part of the public telephone networks, e.g., corporate LANs set up to permit independent intercompany communication regardless of where the various subsidiaries and company divisions, sales offi ces, and manufacturing plants may be. The special communication, command, and control networks set up by the military are also WANs.
The nationwide and worldwide fiber-optic networks set up since the mid-1990s to carry Internet traffic are also WANs. Known as the Internet core or backbone, these high-speed interconnections are configured as either direct point-to-point links or large rings with multiple access points. WANs make it possible for any PC or other Internet-enabled device such as a cell phone to access the World Wide Web or any entity connected to the Internet.
Metropolitan-Area Network (MANs)
MANs are smaller networks that generally cover a city, town, or village. Cable TV systems are MANs. The cable TV company receives signals from multiple sources, including local TV stations, as well as special programming from satellites, and it assembles all these signals into a single composite signal that is placed on fiber-optic and coaxial cables. The cables are then channeled to each subscriber’s home. The cable TV channel selector boxes are all nodes in the system. Most existing cable systems are simplex or one-way transmission systems; however, many cable companies now incorporate two-way communication capability.
Another type of MAN carries computer data. MANs are usually fiber-optic rings encircling a city that provide local access to users. Businesses, governments, schools, hospitals, and others connect their internal LANs to them. MANs also connect to local and long-distance telephone companies. The MANs, or metro networks as they are typically called, also provide fast and convenient connections to WANs for global Internet connectivity.
Local-Area Network (LANs)
A LAN is the smallest type of network in general use. consists primarily of personal computers interconnected within an offi ce or building. LANs can have as few as 3 to 5 users, although most systems connect 10 to several thousand users. Small LANs can be used by a company to interconnect several offices in the same building; in such cases, wiring can be run between different floors of the building to make the connection. Larger LANs can interconnect several buildings within a complex, e.g., large companies with multiple buildings, military installations, and college campuses. Some LANs consist of multiple PCs that are linked both to each other and to a minicomputer or mainframe. This allows each user on the LAN, to have access to the big computer as well as continue to operate independently.
Home networks of two or more PCs are also LANs, and today most home LANs are fully wireless or incorporate wireless segments.
Personal-Area Network (PANs)
A PAN is a short-range wireless network that is set up automatically between two or more devices such as laptop computers, peripheral devices, or cell phones. The distance between the devices is very short, no more than about 10 m and usually much less. PANs are referred to as ad hoc networks that are set up for a specific single purpose, such as the transfer of data between the devices as required by some application. For example, a laptop computer may link up with a printer, or a smartphone may need to download data from a PC. Most PANs just involve two nodes, but some have been set up to handle up to eight nodes and sometimes more.
Storage-Area Network (SANs)
SANs are an outgrowth of the massive data storage requirements developed over the years thanks to the Internet. These networks usually attach to a LAN or Internet server and are designed to store and protect huge data files. The SAN also provides users on the network access to massive data files stored in mass memory units, called redundant arrays of independent disks (RAIDs). RAIDs use many hard drives interconnected to the network. RAIDs have been available for years, but they had to be located close to the computers they served to provide adequate access speed. Today, with high-speed fiber-optic links, the RAIDs may literally be located anywhere, even across the country, since access can be via the Internet or a fi ber-optic WAN or MAN.
Fig. 12-2 shows a highly simplified view of how LANs, MANs, and WANs are interconnected. LANs inside a building is usually connected to a MAN that may be a local telephone central office or a special MAN set up by the organization itself or one managed by a company that leases lines to the organization. The MANs connect to the WANs, which may be a long-distance telephone network or a special optical WAN set up for Internet access or other data transmission applications. Some WANs are hierarchies of rings and direct connection points. MANs and WANs are virtually all fiber-optic networks. The interconnection points of the networks may be special computers called servers, routers, or switching equipment such as an add-drop multiplexer (ADM) that allows data to be added to or extracted from a ring network.
The topology of a network describes the basic communication paths between, and methods used to connect, the nodes on a network. The three most common topologies used are a star, ring, and bus. These topologies apply to the LAN, MAN, and WAN.
A basic star configuration consists of a central controller node and multiple individual stations connected to it (Fig. 12-3). The resulting system resembles a multi-pointed star. The central or controlling PC often referred to as the server, is typically larger and faster than the other PCs and contains a large hard drive where shared data and programs are stored. Any communication between two PCs passes through the server. The server and its software manage the linkup of individual computers and the transfer of data between them.
A star-type LAN is extremely simple and straightforward. New nodes can be quickly and easily added to the system. Furthermore, the failure of one node does not disable the entire system. Of course, if the server node goes down, the network is disabled, although the individual PCs will continue to operate independently. Star networks generally require more cable than other network topologies, and the fact that all communication must pass through the node does place some speed restrictions on the transfer of data.
In a ring configuration, the server or main control computer and all the computers are simply linked in a single closed-loop (Fig. 12-4). Usually, data is transferred around the ring in only one direction, passing through each node. Therefore, there is some amplification and regeneration of the data at each node, permitting long transmission distances between nodes. The ring topology is easily implemented and is low in cost. Expansion is generally simple since a new node can be inserted in the ring at almost any point. The computer that has a message to send identifi es the message with a code or address identifying the destination node, places it on the ring, and sends it to the next computer in the loop. Essentially, each PC in the ring receives and retransmits the message until it reaches the target PC. At that point, the receiving node accepts the message.
The downside of a ring network is that a failure in a single node generally causes the entire network to go down. It is also somewhat difficult to diagnose problems on a ring. Despite these limitations, the ring configuration is widely used.
A bus is simply a common cable to which all the nodes are attached. Fig. 12-5 shows the bus configuration. The bus is bidirectional in that signals can be transmitted in either direction between any two of the nodes; however, only one of the nodes can transmit at a given time. A signal to be transmitted can be destined for a single node or broadcast to all nodes simultaneously.
The primary advantage of the bus is that it is faster than any of the other topologies. The wiring is simple, and the bus can be easily expanded.
A mesh network is one in which each node is connected to all other nodes. You saw this in Fig. 12-1. This is called a full mesh in that every node can talk directly to any other node. Of course, this leads to major costs and complications as the number of nodes increases. This problem is somewhat mitigated by the use of wireless interconnections between nodes where there is no expensive wiring and the attendant routing and maintenance problems. A variation of the full mesh is the partial mesh in which all nodes can communicate with two or more other nodes. This reduces the number of interconnections, making it more practical.
The primary value of the mesh network is that there are multiple paths for data to take from one node to another. This provides for redundancy that can provide a continuous connection when one or more of the links are broken. The lack of one link does not keep the data from reaching its destination by another path. This redundancy provides increased network reliability. Most partial mesh networks are implemented with wireless links.
There are many variations and combinations of the basic topologies discussed above. For example, the daisy chain topology is simply a ring that has been broken. Another variation is called the tree topology. This topology is simply a bus design in which each node has multiple interconnections to other nodes through a star interconnection. In others, the network consists of branches from one node to two or more other nodes.
The common denominator of all LANs is the communication of information. Nodes are linked to transmit and receive information, software, databases, personal messages, or anything else that can be put in binary form.
The earliest LANs were developed primarily as a way to control costs. For example, in an office with multiple PCs, the cost of permitting each computer to have its own printer was prohibitive. Networking is an inexpensive way to link all the computers in the office to a single computer to which the printer is attached. Users share the printer by transmitting information to be printed to the server, which controls print operations for all the office computers.
Another example is the sharing of large files and databases. Networks can be used to connect all the PCs in an office to a central server with a large hard drive containing the data or software to be shared so that any computer can access it. Today, networks are used for many applications other than centralizing and sharing expensive peripherals such as printers and for permitting many users access to a large database. By far the most popular network applications are e-mail, Internet access, and the sharing of common software.
All LANs are a combination of hardware and software. The primary hardware devices are the computers themselves and the cables and connectors that link them. Additional pieces of hardware unique to networks include network interface cards, repeaters, hubs and concentrators, bridges, routers, gateways, and many other special interfacing devices. This section provides an overview of the specific types of hardware involved in networking and especially Ethernet.
Most LANs use some type of copper wire cable to carry data from one computer to another via baseband transmission. The digital data stored in the computer is converted to serial binary data, and voltages representing binary 1s and 0s are transmitted directly over the cable from one computer to another. The three basic cable types used in LANs are coaxial cable, twisted-pair, and fi ber-optic cable. Most local-area networks started out using coaxial cable, but today twisted-pair cable dominates. Fiber- optic cable is used in higher-speed, secure networks, which are not as widespread.
Coaxial cable is a superior medium because its extremely wide bandwidth permits very high-speed bit rates. Although loss is generally high, attenuation is usually offset by using repeaters that boost the signal level and regenerate the signal waveshape. The major benefit of coaxial cable is that it is completely shielded, so that external noise has little or no effect on it. Coax was once widely used in LAN cabling but today has been primarily replaced with twisted-pair cable.
Coaxial cable is shown in Fig. 12-6. It consists of a thin center conductor surrounded by an insulating material that is, in turn, completely encircled by a shield. The shield can be crisscrossed wire braid or solid metal foil. Surrounding the shield is an outer sheath, usually made of PVC.
Twisted-pair cable is the most widely used network cable. Twisted-pair, as the name implies, is two insulated copper wires twisted together loosely to form a cable [Fig. 12-7(a)]. Telephone companies use twisted-pair cable to connect individual telephones to the central offi ce. The wire is solid copper, 22-, 24-, or 26-gauge. The insulation is usually PVC. Twisted-pair cable by itself has a characteristic impedance of about 100 V, but the actual impedance depends on how tightly or how loosely the cable is twisted and can be anywhere from about 70 to 150 V.
There are two basic types of twisted-pair cables in use in LANs: unshielded (UTP) [Fig. 12-7(a)] and shielded (STP) [Fig. 12-7(b)]. UTP cables are highly susceptible to noise, particularly over long cable runs. Most twisted pairs are contained within a common cable sheath along with several other twisted pairs, and crosstalk, the coupling between adjacent cables, can also be a problem; again, this is especially true when the transmission distance is great and switching speeds are high.
STP cables, which are more expensive than UTP cables, have a metal foil or braid shield around them, forming a third conductor. The shield is usually connected to the ground and, therefore, provides protection from external noise and crosstalk. STP cables are, therefore, routinely used for long cable runs.
STP cables are also typically grouped together, and two, four, or more pairs can be contained within a single cable housing [Fig. 12-7(b)]. When multiple lines must be run between computers and connecting devices, multiple cables are the linking medium of choice.
Twisted-pair cable is available in standard types and sizes that are specified and regulated by the standards organizations the American National Standards Institute (ANSI), the Electronic Industries Alliance (EIA), and the Telecommunications Industry Association (TIA). The most often used standard is TIA 568A/568B. This standard defi nes several categories of twisted-pair cable. These are summarized in Fig. 12-8. Categories 2 and 4 are rarely used. The bandwidth is an indicator of maximum bit or baud rate.
The most widely used UTP is category 5 (CAT5). It can carry baseband data at rates up to 100 Mbps at a range up to 100 m. It contains four twisted pairs within the cable and is usually terminated in RJ-45 modular connectors. A newer enhanced version (CAT5e) is also available. It has improved performance and a data rate up to 155 Mbps. The category 6 and category 7 cables are now widely deployed as network speeds have steadily increased. Maximum cable lengths of CAT6 and CAT7 are usually less than 100 m, and an improved RJ45 connector is used to handle the very high frequencies involved. A newer, faster CAT8 cable is under development.
Twisted-pair cable specifications also include attenuation and near-end cross talk figures. Attenuation means the amount by which the cable attenuates the signal. The value for 100 m of CAT5 cable is 26.5 dB at 10 MHz and 222 dB at 100 MHz. The longer the cable, the greater the amount of loss in the cable and the smaller the output. The cable acts as a low-pass filter and also distorts digital signals.
The other key specification is near-end cross talk (NEXT). Cross talk refers to the signal that is transferred from one twisted pair in a cable to another by way of capacitive and inductive coupling. Near-end cross talk is the signal appearing at the input to the receiving end of the cable. Like noise, NEXT can interfere with the received signal on the cable. The NEXT specification indicates the level of cross-talk signal attenuation. In 100 m of CAT5 cable, it can range from 262 dB at 1 MHz to 232 dB at 100 MHz.
Many newer office buildings are constructed with special vertical channels or chambers, called plenums, through which cables are run between floors or across ceilings. The cable used this way, called plenum cable, must be made of fireproof material that will not emit toxic fumes if it catches fire. Plenum cable can be either coaxial or twisted-pair.
A fiber-optic cable is a nonconducting cable consisting of a glass or plastic center cable surrounded by a plastic cladding encased in a plastic outer sheath (Fig. 12-9). Most fiber-optic cables are extremely thin glass, and many are usually bundled together. In a fi ber-optic system, the binary data voltage levels turn a laser off and on to at the transmitter to send the 1s and 0s. The light pulses travel down the fiber and are detected at the receiver by a photodetector diode that translates the light pulses back into voltage levels.
The two basic types of fiber cables are multimode fiber (MMF) and single-mode fiber (SMF). MMF is usually the plastic-type and used in shorter cables because it has greater loss than glass. SMF is glass, more fragile, and thinner, and offers less loss over longer distances. Special fiber-optic connectors are required to attach the cables to the network equipment. Fiber-optic cables are covered in greater detail. Speeds of up to 1 Tbps (terabits per second) are achievable by using fiber optics.
All cables used in networks have special terminating connectors that provide a fast and easy way to connect and disconnect the equipment from the cabling and maintain the characteristics of the cable through the connection.
Coaxial Cable Connectors
Coaxial cables are no longer used in networks. However, if you do find them, they use two types of connectors, N connectors, and BNC connectors. For details.
Most telephones attach to an outlet by way of an RJ-11 connector or modular plug [Fig. 12-10(a)]. RJ-11 connectors are used to connect PC modems to the phone line but are not used in LAN connections.
A larger modular connector known as the RJ-45 connector is widely used in terminating twisted pairs [see Fig. 12-10(b)]. The RJ-45 contains eight connectors, so it can be used to terminate four twisted pairs. Matching jacks on the equipment or wall outlets are used with these connectors. Most LANs today use RJ-45 connectors.
A wide range of connectors is available to terminate fiber-optic cables. Like electrical connectors, these are designed to provide a fast and easy way to attach or remove cables.
Network Interface Controller
A network interface controller (NIC) provides the I/O interface between each node on a network and the network wiring. In older PCs, these were printed circuit cards that plugged into the PC bus. Today the NIC is one or more ICs that are integrated into the motherboard of the PC and provide connectors at the rear of the computer for attaching the cable connectors. NICs perform a variety of tasks. For example, when the PC wishes to transmit information over the network, it takes data stored in RAM to be transmitted and converts it to a serial data format. This serial information is usually stored within RAM on the NIC. Logic circuitry in the NIC then groups the information into frames or packets, the format of which is defined by the communication protocol used by the LAN.
Once the packet or frame has been formed, the binary data is encoded, normally by using the Manchester code, and then sent to a logic-level converter, which generates the proper binary 0 and binary 1 voltage levels that are sent over the coaxial or twisted-pair cables.
Upon reception, the destination NIC recognizes when it is being addressed, i.e., when data is being sent to it. The NIC performs logic-level conversion and decoding, recovering the serial frame or packet of information; performs housekeeping functions such as error detection and correction; and places the recovered data in a buffer storage memory. The data is then converted from serial to parallel, where it is transferred to the computer RAM and used by the software.
The NIC is the key hardware component in any LAN. It completely defines the protocols and performance characteristics of the LAN. Today, most NICs have been compressed into a single chip thanks to advanced semiconductor processing. The exception is the transformer and selected discrete components. Because most PCs are networked, NIC chips are built into all PC and laptop motherboards.
When signals from a NIC must travel a long distance over coaxial cables or twisted-pair cables, the binary signal is greatly attenuated by the resistance of the wires and distorted by the capacitance of the cable. In addition, the cable can pick up noise along the way. As a result, the signal can be too distorted and noisy to be received reliably. A common solution to this problem is to use one or more repeaters along the way (see Fig. 12-11). A repeater is an electronic circuit that takes a partially degraded signal, boosts its level, shapes it up, and sends it on its way. Over long transmission distances, several repeaters may be required.
Repeaters are small, inexpensive devices that can be inserted into a line with appropriate connectors or built into other LAN equipment. Most repeaters are really transceivers—bidirectional circuits that can both send and receive data. Transceiver repeaters can receive signals from either direction and transmit them in the opposite direction.
A hub is a LAN accessory that facilitates the interconnections of the cables to the nodes. Whether the network topology used by a LAN is bus, ring, or star, the wiring usually resembles a star. This is so because the cabling for most networks today is permanently installed in walls, ceilings, and plenums. Bus and ring topologies, where cables logically run between individual PCs, are not convenient for plenum wiring, as they do not provide an easy way to modify the network to add or remove nodes in different parts of the office or building.
The device that facilitates such wiring is the hub, a central connecting box designed to receive the cable inputs from the various PC nodes and to connect them to the server (see Fig. 12-12). In most cases, hub wiring physically resembles a star because all the cabling comes into a central point or hub. However, the hub wiring is such that it can logically implement either the bus or the ring configuration. That is, inside the hub the wiring connects the nodes into a miniature ring or bus.
Hubs are usually active devices containing repeaters. Hubs amplify and reshape the signal and transmit it to all connection parts. Hubs are available with 8, 12, 16, 24, 32, and 48 ports. All signals received at the hub are repeated to all nodes connected to the hub.
A bridge is a network device that is connected as a node on the network and performs bidirectional communication between two LANs (see Fig. 12-13).
A bridge can also be used when one LAN becomes too big. Most LANs are designed for a maximum upper limit of nodes. The reason for this is that the greater the number of nodes, the longer and more complex the wiring. Furthermore, when many individuals attempt to use a LAN simultaneously, performance deteriorates greatly, leading to network delay. One way to deal with this problem is to break a large LAN into two or smaller LANs. First, it is determined which nodes communicate with other nodes the most, and then a logical breakdown into individual LANs is made. Communication between all users is maintained by interconnecting the separate LANs with bridges. The result is improved overall performance.
A bridge is generally designed to interconnect two LANs with the same protocol, e.g., two Ethernet networks. However, there are bridges that are able to accomplish protocol conversion so that two LANs with different protocols can converse.
Remote bridges are special bridges used to connect two LANs that are separated by a long distance. A bridge can use the telephone network to connect LANs in two different parts of the country or can connect two LANs on a large campus or the grounds of a big military base through a fiber-optic cable or wireless connection.
A switch is a hublike device used to connect individual PC nodes to the network wiring. Unlike a hub, a switch provides a means to connect or disconnect a PC from the network wiring. Switches have largely replaced hubs in most large LANs because switches greatly expand the number of possible nodes and improve performance.
As a network grows, more and more nodes are connected to the wiring. This has the effect of slowing down data transfers as all nodes must share the media. The longer cables restrict the data rate. And the greater number of users vying for the cable makes access times longer. These problems are overcome with a switch. The switch can be used to divide the LAN into smaller segments. This immediately improves performance.
LAN switches recognize individual node addresses. When transmitting data from one PC to another, the switch detects the address of the receiving PC and connects it to the wiring. Otherwise, a PC is disconnected from the wiring by the switch until it is ready to send or receive data. By reducing the loads imposed by all the unused PCs, the switch allows the network to be significantly faster.
Ethernet switches have become the key component in most LANs. They implement the interconnection between servers, PCs, and other nodes but also speed up the entire network by reducing the number of collisions and allowing one PC to speak directly to another. This arrangement also provides increased network security by preventing intrusions from one node to the others.
Fig. 12-14 shows how a switch works. If PC D wants to send data to PC C, it does so directly by connecting the two PCs together. The other computers on the network are not involved, so their connections do not load the network circuitry. No collisions occur, so the data transfer is as fast as the network allows. The switch is usually provided with software that allows it to be configured in different ways. For example, a broadcast mode is available where one PC can transmit the same message to all computers on the network. The switch identifies each PC by its media access control (MAC) address. The MAC address is a 48-bit number unique to each PC. The MAC address is made up of 6 bytes or octets identified by its hexadecimal code in the following format:
The first 3 bytes identify the manufacturer of the NIC or PC, such as Intel, Dell, or Cisco; the last 3 bytes are special to the PC or other device. The MAC address is hardwired into each NIC or PC when it is built and is used as the source or destination address in the Ethernet protocol frame. Ethernet switches use the MAC address to route the data from the source to the desired destination. The switch actually “learns” the addresses as the network is used and creates a MAC address lookup table in its memory. It also learns which ports each PC is connected to. The lookup table is updated each time a message is sent or received.
Like bridges, routers are designed to connect two networks. The main difference between bridges and routers is that routers are intelligent devices that have decision-making and switching capabilities.
The basic function of a router is to expedite traffic flow on both networks and maintain maximum performance. When many users access a network at the same time, conflicts occur and speed performance is degraded. Routers are designed to recognize traffic buildup and provide automatic switching to reroute transmissions in a different direction, if possible. If the transmission is blocked in one direction, the router can switch transmission through other nodes or other paths in the network.
Some routers are a combination of a bridge and a router. There are many different types of routers for the wide variety of networks in use. They can switch, perform protocol conversion, and serve as communication managers between two LANs or between a LAN and the Internet.
A gateway is another internetwork device that acts as an interface between two LANs or between a LAN and a larger computer system. The primary benefi t of a gateway is that it can connect networks with incompatible protocols and configurations. The gateway acts as a two-way translator that allows systems of different types to communicate.
Fig. 12-15 shows a typical gateway system, one designed to interconnect one or more PC-based LANs to a mainframe. There are many different types of gateways available depending upon the equipment and protocols involved. Most gateways are computers and are sometimes referred to as gateway servers.
As the number of companies that provide hardware increases, the functions of each device vary and devices labeled as routers may perform the functions of the switch, router, and gateway.
As discussed, modems are interfaces between PCs and communications systems, such as the telephone or cable TV networks. They convert the binary signals of the computer to analog signals compatible with the telephone or cable TV system and, at the other end, convert the analog signals back to digital signals.
Modems are widely used in home networking to connect to an Internet service provider (ISP), which provides services such as Internet access and e-mail.
One of the most complex and expensive parts of any LAN is the cabling. The cables themselves, as well as their installation and maintenance, are expensive, especially when the LAN is being installed in an existing building. In addition, in large, growing organizations, LAN needs change regularly. New users must be added, and the network must be reconfigured during expansion, reorganizations, and moves. One way to avoid the expense and headache of running and maintaining LAN cabling is to use wireless LANs, which communicate via radio.
Each PC in a wireless LAN must contain a wireless modem or transceiver. This device is usually one or more chips built into the motherboard. In any case, the radio modem transceiver converts the serial binary data from the computer to radio signals for transmission and converts the received radio signals back to binary data. Wireless LANs operate as cable-connected LANs in that any node can communicate with any other node. Wireless LANs can have a top speed of up to 1 Gbps or more.
Ethernet LANs| Coaxial Cable | Twisted-Pair Cable | Advance Ethernet ( Network | Topologies | connectors | LAN | WAN | MAN | STAR | RING, BUS )
Protocols | OSI Model | Error Detection | Redundancy | Convolutional ( Network | Topologies | connectors | LAN | WAN | MAN | STAR | RING, BUS )
Spread Spectrum | Wideband Modulation | Broadband ( Network | Topologies | connectors | LAN | WAN | MAN | STAR | RING, BUS )
Multiplexer | Demultiplexer | FDM | TDM | PAM | Applications ( Network | Topologies | connectors | LAN | WAN | MAN | STAR | RING, BUS )
Digital Codes | Hartley’s Law | ASCII | Asynchronous | Encoding ( Network | Topologies | connectors | LAN | WAN | MAN | STAR | RING, BUS )
click here to learn more ( Network | Topologies | connectors | LAN | WAN | MAN | STAR | RING, BUS )
click here to learn ( Network | Topologies | connectors | LAN | WAN | MAN | STAR | RING, BUS )
Reference : Electronic communication by Louis Frenzel