WAN Tutorial – Basic WAN Switching Concept Explained

This tutorial explains basic concepts of WAN Networking including WAN Switching (Circuit, Cell and Packet) and WAN Terminology (CPE, DP, Local Loop, CO, Toll network, CSU, DSU, DTE, DCE, HDLC, PPP, PPPoE, ISDN, LAPD, LAPB, Frame Relay, MPLS, ATM, DSL, Cable, Synchronous and Asynchronous connection) in detail.

What is WAN?

WAN is the term that is used to refer all those technology and hardware which involves in connecting multiple network segments together. These network segments can be located far apart. WAN is the abbreviated form of Wild Area Network.

Difference between LAN and WAN

LAN (Local Area Network) represents a small segment of network that can be span in limited geographical area such as your home network, university campus, cyber café and office building.

WAN represents a large part of network that is not bounded by geographical location. It can be spanned in several locations. Basically WAN is made up from multiple LANs.

For example, an example company with an office in Delhi has 50 computers all connected together. It would be considered as a LAN. Now suppose company expanded its branch and opened new office in Jaipur. The network in Jaipur office would also be considered as LAN until we connect it with Delhi office in order to share information and data. Once both offices (LANs) are connected with each other, they would be considered as WAN.

Another major difference between LAN and WAN is that we own LAN, but we lease WAN. In LAN we have our own space, so we can install cables, switches, routers and other connecting devices as connections require. But in WAN, connections physically pass through the other people’s property where we don’t have right to put our cables and other networking devices. So what options left for us?

We need to lease or rent connection from cable companies. Let’s understand this in other way. Suppose you are in Delhi and you want to go Jaipur by car. Will you build your own road for transportation?

No you do not build your own road for transportation. Simply you use the road built by government and pay toll tax for use. Same mechanism works behind the cable network. A cable network company builds its network and we use it. Cable Network Company charges for using its network. There are several cable network companies available in market. Think about your ISP. It is also a cable network company. In short how a cable network company connects two LAN is described in WAN.

Basic WAN Terminology

Before we understand WAN networking concepts in detail, let’s be familiar with some basic WAN Network terminology.

CPE

CPE (Customer premises equipment) is the device that is located in subscriber’s premise such as modem, switches, CSU/DSU, DTE router and NT1.

Demarcation point

DP (Demarcation point) is the spot, from where responsibility of customer and Cable Network Company divides. Usually it is a device installed by telecommunication (Cable Network) company. In simple terms, it’s a delivery point where LAN delivers or receives data packet from Telecommunication Company.

Local Loop

Local loop is the connection between DP and CO (central office).

Central office

This is the nearest office of Telecommunication Company. You get connection from this office.

Toll network

Toll network is the internal infrastructure of Telecommunication Company for transporting your data.

A Telco (Telecommunication Company) uses several thousand kilometers of network cables and specialized networking devices to create its own network. This network provides a service that works like a crossover cable between two demarcation points but physically hidden from customer. Based on its network and customer requirement a Telco provides several types of WAN connections.

CSU/DSU

A CSU/DSU (Channel Service Unit/Data Service Unit) is a device that converts data signal between LAN network and WAN network. LAN network and WAN network use separate communication technology. A CSU/DSU understands both technologies. DSL and cable modems are the example of CSU/DSU.

DTE

DTE (Data Terminal Equipment) is a device (usually a router or PC) that converts data frame into signals and reconvert received signals in data frame. DTE device communicates with DCE device.

DCE

DCE (Data circuit terminating equipment) is a device (usually modem or CSU/DSU) that provides clock rate and synchronization.

WAN Connection Types

Nearly all WAN connections fall under one of following four categories:-

1. Leased Line Connection
2. Circuit Switched Connection
3. Packet Switched Connection
4. Cell Switched Connection

Leased Line Connection

Leased line connection is a dedicate connection between two LANs. It simulates a single Ethernet crossover cable between local LAN and remote LAN. Just like Ethernet connection you can transmit data any time without any setup procedures. It’s an always on connection that provides guaranteed bandwidth and minimal delay. With all goodies leased line has following disadvantage:-

• Usually it is available for shorter distance.
• This is the most expensive WAN solution.
• Whether you use it or not, you will be charged for entire time.
• Each line requires a separate interface on your router.

Leased line is also known as point to point or dedicated connection. Leased line uses HDLC and PPP data link layer encapsulation protocol.

Circuit Switched Connection

Circuit switched connection is just like a phone call. Whenever you have data to transmit, open the circuit, transmit the data and close the circuit. Thus in positive side, you only need to pay when you actually use it. And in negative side, you have to establish an end to end connection every time whenever you have data to transmit. Circuit switched connection is available it two types:-

Asynchronous serial connection

It uses analog modem and standard telephone system to establish dial-up connection. This is the cheapest WAN services. On another side this is most unreliable and the slowest WAN service. In this service top connection speed in USA is 53Kbps.

Synchronous serial connection

It uses digital ISDN line, BRI (Basic Rate Interface) and PRI (Primary Rate Interface) for dial-up connection. It’s an expensive service that provides guaranteed bandwidth.

Typically Circuit Switched connections are used for secondary backup solution in offices as well as for temporary low speed connection in home network. As it charges per-minute, it’s not suitable for primary connection, depending on the destination: the more data we have to transmit, the more time it will take, and the more money it will cost.

Packet Switched Connection

Packet switched connection is the cost effective solution of leased line connection. People, who cannot afford leased line, can use this. It allows us to share bandwidth with others to save money. It looks like a leased line but charges like circuit switched line.

Yep you read it right; it looks like a leased line, not acts like a leased line. After all it’s a shared line, you will get full bandwidth if no other is not sending data at that time. But if others are also sending data at the same time then you will get congested bandwidth. Packet switched connection is the best suitable for bursty type data transmission. It is not suitable for constantly type data transmission.

Both leased line and circuit switched connection use a physical circuit path to connect two sites. While leased line uses same circuit path every time, circuit switched connection builds path every time a data call is made. Probable you may get same or new path for every data call in circuit switched network.

Packet switched connections use a different approach, instead of physical path, it builds logical path over the physical path and uses that to connect two sites. These logical circuits are called virtual circuits (VCs). VCs are not tied up with any specific physical circuit. They can be built from any available physical connection. We can create multiple logical circuits over the same physical circuit. Therefore we can use single router interface to connect multiple sites.

Cell Switched Connection

This is the enhanced version of packet switched connections. It could provide guaranteed bandwidth, minimal delay, limited number of errors and Quality of Services. This service uses fixed length (53 bytes) packets known as cell to transmit the data. If you want to connect multiple remote sites with a single router interface then Packet Switched or Cell switched connections are the best options.

Based on data transmission technology, above wan connections can be divided in two types:-

Asynchronous connection

Asynchronous connection is a one way communication. Sender PC does not need to coordinate with receiver PC before sending data. Sender system simply sends data. Receiver system has to look in incoming data for decoding method and coordinate.

Synchronous connection

Synchronous connection is a two way communication. Sender PC and Receiver PC synchronize the connection before sending any user data. Once connection is established, sender system sends separate signals with every data packet regarding transmission setting (decoding method, clock rate, coordinate etc ). Receiver system also updates sender about what has been received so far with its signal.

Now we have basic understanding of WAN terms, in remaining part of this tutorial we will have quick look on most popular WAN solution.

WAN Protocols, services and encapsulation methods

HDLC

HDLC (High-Level Data-Link Control) protocol works at Data Link layer. In comparison with other WAN protocol it creates very little overhead. It is derived from SDLC (Synchronous Data Link Control) protocol. HDLC protocol doesn’t carry any network layer information in its header. For this reason any vendor who wants to use HDLC protocol has to implement his own method to identify the Network Layer protocol. This method may not work with other vendors. Thus if we want to use HDLC protocol, we have to by all devices from same vendor.

PPP

PPP (Point to Point) protocol is wildly used data link layer protocol. It has a field for Network Layer Protocol information in its header. For this reason vendors need not change anything. They can use it as it is. If you have equipment from different vendors then use this protocol to create point to point links.

PPPoE

PPPoE (Point-to-Point Protocol over Ethernet) is the Ethernet solution of PPP protocol. When used with ADSL service it provides a direct connection to Ethernet interface while supporting DSL as well. In upside it provides authentication, encryption, and compression. In downside it has lover MTU (maximum transmission unit) than standard Ethernet has.

ISDN

ISDN (Integrated Services Digital Network) uses circuit switched connection technology. It uses existing phone lines for voice and data transmission. Usually it is used for secondary backup connection. It provides a cost effective upgrade option for users who require a higher speed connection than analog modem connection.

LAPD

LAPD (Link Access Procedure, D-Channel) protocol was developed for ISDN. It is used to satisfy the signaling requirement of ISDN connection.

LAPB

LAPB (Link Access Procedure, Balanced) was created for X.25 as connection oriented protocol at Data Link Layer. It can also be used for simple data link transportation. Due to its windowing and strict time out techniques it creates a lot of overhead.

Frame Relay

Frame Relay uses packet-switched connection technology. It can provide bandwidth of 64Kbps up to 45Mbps. It builds logical circuit over physical circuit to transmit the data. With logical circuit we can connect multiple remote sites from single interface. Beside this it also provides dynamic bandwidth and congestion control features.

MPLS

MPLS (Multi-Protocol Label Switching) is the frame relay based data forwarding technology. Just like frame relay it also uses some features of circuit switched network over the packet switched network. It uses labels (numbers) to forward the packets. Labels are assigned in the edge of MPLS network. Core routers forward packets based on these labels that make forwarding process faster. Gradually companies are moving to MPLS from frame relay.

ATM

ATM (Asynchronous Transfer Mode) is the enhanced version of frame relay. Beside all goodies of frame relay it also provides simultaneous transmission for video, voice and data. ATM uses fixed length (53 bytes) cells for data forwarding. For faster data forwarding it supports external clocking.

DSL

In DSL (Digital subscriber line) service, a connection is setup between CPE (Usually modem) and the DSLAM (Digital Subscriber Line Access Multiplexer). DSLAM is a device that concentrates connections from multiple DSL lines. It is located at CO (Center Office of Telco). It uses twisted pair copper telephone wires for connectivity. It is the same phone line that is used to connect your phone with CO. In CPE side it uses DSL modem that follows DSL physical and data link layer standards to transmit data from Telco. Following figure illustrates a simple DSL service network.

DSL service provides asymmetric speed. In asymmetric speed downloading speed is much faster than uploading speed.

Cable

This service uses existing cable TV network for Internet. It can provide up to 27Mbps download and 2.5Mbps upload bandwidth. In this service, users connected with a certain cable network segment sharing bandwidth. For this reason actual bandwidth a user may get is 256Kbps to 6Mbps.

Key points of DSL and Cable services

• Both services use existing network. DSL uses phone line while Cable uses CATV (Cable TV) cable.
• Both services provide asymmetric speeds where downloading speed is much faster than uploading speed.
• Both services work in always on mode where user can access Internet without taking any action to start Internet connection.
• Both services are used to connect CPE with CO.
• DSL service uses twisted pair cable while Cable service uses coaxial cable.
• In customer side DSL service uses DSL modem while Cable service uses cable modem.
• In company side DSL uses DSLAM device to filter voice and data while Cable service uses router to split data and videos.

Serial WAN Encapsulation method and protocol

Not all of above introduced protocols and methods are available for serial interface. What options are available for serial interface can be listed by encapsulation ? command from serial interface mode. For example following figure illustrates available options for serial interface in router.

Output may be different on different series routers. You may get more options such as atm-dxi, lapb, smds and x25 in real Cisco router. This output is taken from a virtual router running in packet tracer network simulator software.

As output shows we have three encapsulation methods:-

1. HDLC
2. PPP
3. Frame-Relay

That’s all for this part. In next parts of this article I will explain these protocols in details with examples.