Introductory Concept

Due to the skyrocketed increase of internet users and services, high-speed communication is required to to fulfill enormous demand of bandwidth. In this direction, all optical networks with wavelength divisional multiplexing (WDM) technology have become essential to develop such high-speed communication. This book deals with the principles and fabrication of optical network devices such as wavelength router [1-6], WDM [7-11], add/drop multiplexer (ADM) [12-17], photonic switch [18-23], Erbium-doped fiber amplifier (EDFA) [24-26], and EDFA gain equalizer [27-30].

Basic Communication Model

Before discussing optical networks, one should know the basic communication model. Figure 1.1 shows the general block diagram of a communication system having a source system, destination system and transmission media. The source system has a source device that generates raw signals such as data, voice/video and information, and a transmitter that transforms and encodes raw signals in such a way as to produce electromagnetic signals that can be transmitted through a transmission system, which is a complex network connecting source and destination. The destination system has

a receiver that receives signal from a transmission system and converts it into a raw signal, and from the raw signal, the information/message is recovered. In this section, we provide an overview of different communication networks.

Local Area Network

Local area network is a communication network that covers a small geographical area (typically a building or a cluster of buildings) and provides a means for information exchange among the devices/nodes attached to it. The communication between different nodes of the network is mainly based on open system interconnection (OSI) model or transport control protocol/internet protocol (TCP/IP).

OSI Model

Since the origin of communication, its operation varies from vendor to vendor. So standards are needed to promote interoperability among vendor equipment and to encourage economics of scale. Because of the complexity of communication tasks, no single standard will be sufficient. It is better to form a framework for standardization rather than breaking the operation into manageable parts. In 1977, the International Standard Organization (ISO) had started to establish a subcommittee for developing the architecture of the framework. As a result, OSI has been developed [31,32]. The OSI model is a seven-layer architecture in which each layer performs definite functions, namely physical layer, data link layer, network layer, transport layer, session layer, presentation layer and application layer.

  • 1. Physical layer: It permits interconnection with different control procedures such as V.24 and V.25 for various physical media.
  • 2. Data link layer: It controls data transmission through the system having high error rate (i.e., an error rate not acceptable for a great majority of applications). It works in the framework of high-level data link control (HDLC).

It is just above the physical layer.

  • 3. Network layer: It selects a connection path or provides a rout (where the intermediate nodes may be present) for data transmission from one node to the other.
  • 4. Transport layer: It controls successful transportation of data from the source to the destination node. It provides totality of transmission service and ensures that data are delivered error-free, in sequences with no loss and delicacy.
  • 5. Session layer: It provides synchronization or organization dialog between the source and the destination before data transmission. It does function above the transport layer. It provides a mechanism for recovery and permits backup.
  • 6. Presentation layer: It does general interest functions related to representation and manipulation of structured data just before the application layer. It defines the format of the data to be exchanged between different applications.
  • 7. Application layer: It performs management functions and generally useful mechanisms that support distributed applications.

Figure 1.1 shows how data are transmitted in an OSI architecture with the use of a protocol data unit (PDU). When user A has a message to send to user B, it transfers these data to the application layer, where a header is added to the data making it A-PDU. Then, it is passed to the presentation layer. In the same way, these PDU goes through the layers as per the figure (by using HDLC format) to a data link layer. The data link layer unit, also called as a frame, is then passed to a communication path/ link in the network by using a physical link. When the frame is received in the destination node/target node, a reverse process occurs. As the PDU ascends, each layer strips off the outermost header, acts on the protocol information contained therein, and passes the remainder up to the next layer.

TCP/IP Protocol

Since 1990, TCP/IP has become more popular than the OSI model because of its simplicity and interoperability over different networks, thus providing different services through its IP layer. In an OSI model, protocols at the same level of hierarchy have certain features in common. In this direction, TCP/IP architecture is better than that of the OSI model. The TCP/IP has five layers [33,34] - physical layer, network access layer, internet layer, transport layer and application layer.

  • 1. Physical layer: It defines the characteristics of transmission medium, signaling rate and encoding scheme.
  • 2. Network access layer: It makes a logical interface between an end system and a subnetwork w'here a connection path is selected.
  • 3. Internet layer: It does the function of routing data from the source node to the destination host through one or more networks connected by routers.
  • 4. Host-to-host transport layer.
  • 5. Application layer.

Figure 1.2 show's the transport layer through different layers in TCP/IP protocol [34]. When user A has a message to send to user В via different applications, as given in Figure 1.3a, and transfers these data to the application layer, a header of TCP is added to the data. In the same w'ay, these user data go through the layers as per Figure 1.2, adding with different header files in these layers. In the IP layer, IP header files are added. Finally it is passed to the communication path/link in the network-I and via router having NAPI and II and IP interface, and the frame is received in the destination node/target node, w'here a reverse process occurs. The PDU ascends, and each layer strips off the outermost header, acts on the protocol information contained therein, and passes the remainder up to the next layer. Finally, it goes to user B. Figure 1.3 show's the different applications provided by the TCP/IP architecture, in w'hich simple mail transfer protocol (SMTP), hypertext transfer protocol (HTTP), file transfer protocol (FTP), TELNET and broader gateway protocol (BGP) make use of TCP via IP layer to get connection to another host of corresponding module, w'hereas user datagram protocol (UDP), internet control message protocol (ICMP) and open shortest path first (OSPF) make use of IP directly for connection.

(CONTINUED) (b) ATM connection relationship

FIGURE 1.3 (CONTINUED) (b) ATM connection relationship.

Wide Area Network

Wide area network (WAN) has been traditionally considered to be a network that covers large geographical area. It consists of a number of interconnected switched nodes. Here, the transmission from one device to the other is routed through these internal nodes to the specified destination device. For this purpose, a switching facility is used to move the data from one node to the other until it reaches the destination. The WANs are mainly implemented by using two switching technologies - circuit switching [35,36] and packet switching [35,36]. Apart from that, WANs may use asynchronous transfer mode (ATM) and frame relay architectures.

Circuit Switching

In a circuit-switched network [33], a dedicated communication path is established between two stations through the nodes of the network. There are three steps - circuit establishment, data transfer and circuit disconnect after data transfer is over.

  • 1. Circuit establishment: Before signal for data is transmitted, an end-to-end (source-to-destination) circuit must be established. For this establishment, a signal path must be selected by routing. Routing in circuit switching is done in three ways: fixed routing, alternate path routing and adaptive routing. These routing approaches are discussed in the next chapter. After selection of path will be dedicated for transmission of data for this connection.
  • 2. Data transfer: After establishment of connection, data are transmitted through a dedicated path selected for routing. This path cannot be shared by other stations or nodes till the data transfer is over.
  • 3. Circuit disconnect: After data transfer is over, the circuit is disconnected and transferred to another user.

Packet Switching

A different approach is used in packet switching. Here, data are sent out in a sequence of small chunks with destination and route address, and these chunks are called as packets. Each packet is passed through the network from node to node along the path leading from source to destination. At each node, the entire packets is received store briefly and then transmitted to the next node. The links used in the path are shared by other users to send their packets. There are two types of packet switching - virtual circuit packet switching [36] and datagram packet switching [33].

Like circuit switching, in virtual circuit packet switching, there is a requirement of connection call setup, and after connection is set up for a path, all the packets should flow through the same path to the destination. But the links used in the path are shared by the packets of other users.

In datagram packet switching, it does not require to set up connection previously, and the individual datagram packets are routed independently to destination not by single path but maybe by a number of paths. The datagram packet switching is faster than virtual circuit packet switching because of less queue time delay.

Frame Relay

The packet switching relatively exhibits a high bit error rate while it is implemented for long-distance communication. To control these errors, it requires more overload; besides, extra time is required to process these overloads at each intermediate node. This overload is unnecessary and counterproductive. The frame relay networks are used for operating efficiently at user data rates of up to 2 Mbps. The purpose of achieving these high data rates is to strip out most of the overload involved with error control. In this direction, frame relay was developed by reducing overload, with consideration of a smaller number of layers [37].

Asynchronous Transfer Mode

ATM is another approach in which a fixed length packet called cell is transmitted for data transfer [38]. These ATM cells have little overload of error correction. Due to fixed length, the processing time is also reduced. In ATM, multi-virtual channels of fixed length are available for transmission of data. As per the demand for bandwidth, the numbers of virtual channels are set dynamically for maintaining the quality of service. After selection, the virtual channels are put into a virtual path. So, it is required to set virtual channel connections, and finally, a virtual path connection is set. The overload in ATM is less than that of frame relay. As a result, the data rate is more in ATM (10-100 Mbps) than that of frame relay (2 Mbps). Figure 1.3b shows an ATM connection concept, how the virtual channel connections construct virtual paths and finally how groups of virtual paths make a transmission path of ATM network.

VSAT Network via Satellite

Before discussing VSAT network, one should know about satellite microwave. A communication satellite is basically a microwave relay station in which satellite is used as a transponder to connect two or more ground-based microwave transmitter/receivers which are basically very small aperture terminals (VSAT) [33]. The satellite transmits one frequency band named as downlink frequency, whereas it receives transmission on one frequency band named as uplink frequency. For a satellite to get a communication effectively, it is needed to make it stationary relative to its position over earth because offline -of-sight communication with the users/ stations at all times. The coverage area for elevation angle / and altitude H is derived by considering the distance between two users stationed at two extreme points of coverage area, DM written as [39] from Figure 1.4a

(a) Satellite with its coverage and (b) VSAT broadcasting network

FIGURE 1.4 (a) Satellite with its coverage and (b) VSAT broadcasting network.

where Rc = radius of earth. Considering velocity of propagation of signal, C, the maximum propagation delay is written as

The total number of satellites required for global coverage can be written as

where n = minimum number of satellites seen from any point at any time, where n = l for single-fold coverage and n = 2 for double-fold coverage.

There are three types of polar orbit satellites:

  • • Low earth orbit (LEO) - altitude 500-1500km,
  • • Medium earth orbit (MEO) - altitude 5000-15,000km,
  • • Geostationary orbit- altitude 35,784 km.

LEO and MEO satellite networks provide a wide service area where there is less non-telecommunication infrastructure, especially on rural and hilly regions of Asia, Africa, Eastern Europe, South America, and the polar areas [40]. These LEO and MEO satellite networks also cover global coverage to their users, which a typical GEO satellite system cannot accommodate. In this direction, LEO satellite system, which is Motorola’s IRIDIUM system, was employed in May 1998 for global coverage [40].

The IRIDIUM system is the first initiative of global coverage of wireless communication system to provide voice, data, fax, and paging services to the world. At an altitude of 780 km above the earth, 66 satellites derived by using equation (1.1 a) are required in six planes. Each plane has 11 satellites. Planes have a near-circular orbit, with co-rotating planes spaced 31.6° apart and counter-rotating planes (one and six) spaced 22° apart [40]. The minimum elevation angle normally for an earth station is considered to be 8.2°, which maximizes the coverage area of the satellite and improves the link quality compared with lower elevation angles. The average satellite in-view time is approximately 10 minutes.

Further, we can increase the coverage area with MEO satellite, but the propagation time delay increases due to higher altitude. Further, both LEO and MEO satellites have lower lifetime in comparison to GEO satellites

Out of three satellites, GEO satellite is mostly used for VSAT network, mainly because of higher coverage area than other two satellites and high lifetime, although it has a high cost of installation and maintenance. Two satellites using same frequency band will interfere with each other while they come closer. To stay away from the problems between two satellites, 4-degree spacing of one satellite is used for 4/6 GHz and a 3-degree spacing of other satellite for 12/14 GHz. There are two types of transmission of signal via satellite - broadcasting and point-to-point transmission.

In broadcasting transmission, data are transmitted by one user treated as transmitter, whereas other VSAT receives the data as a receiver as shown in Figure 1.4b. Among applications in this direction are television distribution, long-range radio broadcasting and private business broadcasting. In the case of public broadcasting service (PBS), the television programming is distributed by the use of satellite channels.

In point-to-point transmission, signal transmissions are in both directions via satellite. There are two types of VSAT network which uses point-to-point transmission - centralized VSAT network and distributed VSAT network. Two users in two different VSATs in centralized network transmit their data via a central HUB as shown in Figure 1.5a. In distributed network, there is no central HUB, and two users in two different VSATs transmit directly without going through via as shown in Figure 1.5b.

The satellite transmission uses a frequency range of 1-15 GHz. In fact, there is significant noise from natural sources including galactic, solar and atmospheric noise and human-made interference from various electronic devices below 1 GHz, whereas above 15 GHz, the signal power is heavily attenuated in atmosphere. There

(a) Centralized VSAT network and (b) distributed VSAT network

FIGURE 1.5 (a) Centralized VSAT network and (b) distributed VSAT network


are two frequency bands used for satellite communications in the range - C band and Ka band. For C band, the satellites provide point-to-point transmission with frequency ranges 5.925-6.425 GHz from earth to satellite (uplink) and 3.7-4.2GHz from satellite to earth (downlink). Since this frequency ranges are saturated due to having so much traffic, there are two frequency ranges beyond 10GHz used in Ka. In Ka band, the uplink and downlink frequency ranges are 14-14.5 and 11.7-12.2GHz, respectively. Due to tremendous demand of bandwidth, there is another band Ku used in satellite communication even after 5 GHz. The ranges for Ku band are 27.5-31 and 17.7-21.2 GHz, respectively. But in the case of Ku band, the bandwidth used for both uplink and downlink is 3.5 GHz and those for other C and Ka band is 0.5 GHz. There are several characteristics of satellite communication to be considered:

  • • A propagation time of one quarter of second is taken for data transmission from one earth station to another station via satellite. Almost the same order of time delay is required for telephonic transmission.
  • • There are problems of error control and flow control that will be discussed later in this book.
  • • Satellite communication is mainly broadcasting in nature, but it can be used as point-to-point bidirectional communication.

Integrated Services Digital Network

Rapid development of communication technologies has resulted in an increasing demand of worldwide public telecommunication networks in which a variety of services such as voice and data (computer communication) are distributed. The Integrated Services Digital Network (ISDN) is a standard network having user interfaces and is also realized as digital switches and paths accommodating a broad range of traffic types and proving value-added processing services [41]. Standards of ISDN are made by ITU-T (formerly CCITT). The ISDN is a single worldwide uniformly accessible network having multiple networks connected within national boundaries. It has both circuit-switching and packet-switching connection at 64 kbps. There are mainly two types of services - voice communication and non-voice (data) communication. Figure 1.5c shows the ISDN architecture in which different services such as voice, PBX signal and data are connected to ISDN through its interface, and then all interfaces are connected to ISDN central office through digital transmitted media/pipe. The central ISDN office is connected to different networks such as circuit-switched network and packet-switched network with digital pipes at a certain bit rate. There are two generations of ISDN - narrowband ISDN (N-ISDN) [41] and broadband ISDN (B-ISDN) [41].

Narrowband ISDN

The first generation of N-ISDN referred to as narrowband ISDN is based on 64 kbps channel as the basic unit of switching and has a circuit-switching orientation. It is developed by using a frame relay concept. All traffic in this system use D channel employing link access protocol-D (LAPD) which has two forms of service to LAPD users: the unacknowledged information-transfer service and acknowledged information-transfer service. The unacknowledged information-transfer service provides for the transfer of frames containing user data with no acknowledgment, whereas the acknowledged information-transfer service is a more common service similar to link access protocol- B (LAP-B) and HDLC [33]. Table 1.1a shows different narrowband channels included for transmission in N-ISDN with their application services.

Broadband ISDN

The second generation of ISDN referred to as В-ISDN provides services to both narrowband and broadband channels having a data speed of more than 100 Mbps. Initially, it uses ATM-based network, and later as the number of services increases, it uses optical network to provide services. We will discuss optical network thoroughly in this book.

Table 1.1b shows different broadband signals included for transmission in B-ISDN with their application services, apart from inclusion of narrowband signals.

< Prev   CONTENTS   Source   Next >