What is a wide area network (WAN)?

As we know, local area networks (LANs) are used to connect devices that are close together. The data transmission speed in local networks is therefore often quite high. WANs, on the other hand, connect devices that are geographically distant from each other and therefore WAN technology is also different from LAN technology.

WAN uses different transmission methods, hardware, and protocols than LAN. The data transmission speed in WAN is also much lower than that of LAN. We will study an overview of WAN technologies from several perspectives.

Learn about WAN

• History of WAN
• What is the difference between WAN and LAN?
• WAN Architecture
• Overview of WAN communication
• Types of connections in WAN
• Wide Area Network Services
• WAN Hardware
• Interface protocols
• Data Encapsulation Methods in WAN
• WAN Management and Optimization
• WAN Security
• Interplanetary Internet

History of WAN

WANs have been around since the early days of computing. WANs were based on switched telephone lines and modems, but today's connectivity options also include leased lines, wireless, MPLS, broadband Internet, and satellite.

As technology changes, so do transmission speeds. The 2400bps modems of the early days have evolved into today’s 40Gbps and 100Gbps connections. These speed increases have allowed more devices to connect to the network, enabling the explosion of connected computers, phones, tablets, and smaller Internet of Things devices.

Additionally, improvements in speed have allowed applications that require a lot of bandwidth to be transmitted over the WAN at super-fast speeds. This has allowed businesses to implement applications such as online meetings and large file backups. No one would ever think of conducting an online meeting over a 28kbps modem, but now employees can sit at home and participate in company meetings via video from around the world.

Many WAN links are provided through provider services, where a customer's traffic travels through facilities shared by other customers. Customers can also purchase dedicated links that are used only for a single customer's traffic. These are often used for latency-sensitive or priority-sensitive applications that require high bandwidth, such as video conferencing.

What is the difference between WAN and LAN?

WANs are often contrasted with local area networks, or LANs. LANs are networks that are typically confined to a small building or campus. They are private to an organization or even a single person, and can be created with relatively inexpensive equipment. Your home WiFi network is a LAN.

The technologies and protocols that make LANs easy to set up cannot scale beyond a certain limited distance or handle a truly massive number of endpoints. The purpose of a WAN is to accommodate that scale by connecting one or more LANs. The network technologies and protocols that WANs use to transmit information are different from those used in LANs.

Strictly speaking, the Internet is a WAN. However, when we talk about WANs, we usually mean private or semi-private networks that combine remote LANs. For example, branch offices in different cities might share private internal company resources over a WAN.

WAN Architecture

While LANs are typically maintained by an organization's own IT staff, WANs are often at least partially dependent on physical connections provided by telecommunications service providers. Deciding on what type of connection or communication protocol to use and how to deploy it will set the stage for creating your WAN architecture.

Overview of WAN communication

WANs use the transmission infrastructure of a third-party service provider, typically a telephone company, to provide long-distance connectivity. The most common configuration of a WAN consists of the following components. A message is initiated by the customer and sent by a device called a DTE to the WAN service provider. DCE devices in the service provider’s central office “push” the packet onto the WAN, where it passes through switches to reach its destination. Similar devices at the receiving end complete the journey.

Data Terminal Equipment (DTE): A device at the edge of a WAN link that sends and receives data. The DTE is located at the subscriber's location, and is the connection point between the subscriber's LAN and the service provider's WAN. The DTE is usually a router, but in some cases it can be a computer or a multiplexer. The DTEs on one end will communicate with the corresponding DTE device on the other end.

Damarcation Point: The point of connection between the telephone company's telephone lines and the subscriber's lines. The damarcation point is also known as the network interface or point of presence. Typically, the customer is responsible for all equipment within the damarcation point and the telecommunications company is responsible for all equipment on the other side.

Last Mile Cable (Local Loop ): The cable that connects the Demarcation Point to the Telephone Company's Central Office. It is typically twisted pair (UTP) cable, but can also be a combination of twisted pair, fiber optic cable, and other types of transmission media.

Central Office: The nearest switchboard station, which is also the WAN service point closest to the subscriber. The central office provides the entry point for calls into the “WAN cloud” and provides the exit points for calls from the WAN cloud to the telephone user. In addition, it acts as a network switching point to forward data packets to other central offices. It also provides stable DC power to the last mile cabling system to establish the circuit.

Data Circuit-terminating Equipment (DCE)

A device that communicates with both the DTE and the WAN cloud. A DCE is typically a service provider router that forwards data between the customer and the WAN cloud. In a narrow sense, a DTE is any device that provides a clock signal to the DTE. A DCE can also be a device similar to a DTE (usually a router) except that each type of device plays a different role.

WAN cloud: A series of trunks, switchboards, and central offices that make up the telephone company's transmission infrastructure. It is shown as a cloud because the physical structure is constantly changing and only those responsible for WebTech360 know where the data is going at the switchboards. For the customer, the important thing is that the data has traveled across the wire to its destination.

Packet-switching exchange: Switching exchanges on a telecommunications company's packet-switching network. PSEs are intermediate points in the WAN cloud.

Data transmitted over a LAN is primarily sent from one digital device (computer) to another digital device via a direct connection. Whereas, because some WANs use the existing analog telephone network, data transmission may use one or a combination of the following methods:

Analog signal transmission

Analog signals are usually represented as waveforms. The intensity and frequency of an analog signal vary continuously, so it can accurately represent continuous motion or sound or multi-state motion. The intensity and frequency of the signal increase and decrease in response to the pitch and intensity of the sound. Analog signals are often used to represent real-time data. Radio, telephone, and other communication media often use analog signals.

Digital Signal Transmission

Instead of a continuously changing stream, digital signals use only two states, 0 and 1, to represent bits of data. This is the ideal method of transmitting signals for computer networks. Computers would need modems, devices that convert the computer's digital signals into analog signals to transmit data over analog telephone lines.

Note : In the past, the PSTN was an entirely analog network. Analog signals from the telephone arrived at the telephone company and were then routed through analog systems to their destination. Today, telephone systems use a combination of both methods. Most switched networks that connect telephone companies are digital, but the last mile that connects most homes and some businesses is still analog. The diagram below shows how two digital computers can be connected over a WAN that has both digital and analog components. When a computer sends a signal over the WAN, the modem converts the digital signal into analog for transmission to the telephone company. The telephone company's modem converts the data back into digital form for transmission over the switched network. The signal is converted back into analog at the telecommunications company's end to be transmitted to the modem of the computer receiving the data. Finally, this modem converts the analog signal into digital form for the computer.

Types of connections in WAN

As a message travels across a WAN cloud, the way it moves from one point to another along its path will vary depending on the physical connection and the protocol used. WAN connections are typically classified into the following types:

Dedicated Connection

This is a permanent connection, connecting one device directly to another. Dedicated connections are stable and fast but can be very expensive. Renting a line from a WAN service provider means you pay for the connection even if you are not using it. Furthermore, because dedicated lines establish a direct connection between only 2 points, the number of lines required increases exponentially as the number of locations connected increases. For example, if you want to connect 2 locations, you need one line, but if you want to connect 4 locations, you will need 6 lines.

Features of dedicated connection:

• Always available
• Use a WAN service provider's leased subscriber line
• More expensive than other WAN solutions
• Use separate connections between points

Use a dedicated connection when:

• There is high volume of data flowing through the LAN.
• Need to connect frequently
• Few locations need to be connected to each other

circuit-switched network

Circuit switching gives you an alternative to leased lines (dedicated connections), allowing you to use shared lines. Circuit switching works bidirectionally, allowing both dial-in and dial-out connections to be established.

When you use a switched network:

1. The computer sends dial data onto the line and a connection is established.
2. The computer receiving the data sends confirmation and locks the line.
3. The computer sending the data transmits the data over the established connection.
4. Once the data transfer is complete, the connection is released to other users.

Switched networks use switched virtual circuits (SVC). A dedicated data path is established at the beginning of a communication process by a series of electronic switches. This dedicated path remains until the end of the communication process.

The public telephone system is a circuit-switched network. When you make a call, the PSTN uses switches to create a physical, direct, dedicated connection for the duration of the call. When you end the call, the switches release the line for other users. Computers connected to the network work in a similar way. When a computer dials into the network, a path through the network is first established, and then the data is sent over this temporary dedicated path.

Packet-switched network

Packet-switched networks do not require leased lines or temporary lines. Instead, the path of a message is established dynamically as the data moves across the network. Packet-switched connections are always on. This means you do not have to worry about setting up a connection or keeping the line dedicated. Each packet includes all the information needed to reach its destination.

Packet switching networks have the following characteristics:

• Messages are divided into small units, called packets.
• Packets are transmitted independently over the internet (possibly on different paths)
• Packets are reordered back to their original order at the destination.
• The sending and receiving devices default to seeing the connection as persistent (no dial-up required)

Packet-switched networks use permanent virtual circuits (PVCs). Although PVCs resemble dedicated, direct connections, the path each packet takes through the internetwork can be different.

Wide Area Network Services

PSTN

The public switched telephone network is the oldest and largest network available for WAN communications. Features of the PSTN include:

• This is a circuit-switched network, with global coverage.
• The interface to the PSTN is analog, so computers use modems to connect to the PSTN.
• Speeds on the PSTN are typically limited to 56 Kbit/s.
• You can use PSTN on demand or rent a dedicated circuit.

Figure 8: PSTN telephone network

Leased Line

For some companies, the benefits of a leased line may outweigh the costs. Leased lines are independent and offer higher speeds than regular PSTN lines. However, they are expensive and are typically only used by larger companies. Other features of leased lines include:

• Provide regular, stable quality connections
• You can pay extra to upgrade your leased line.

X.25

X.25 was introduced in the 1970s. Its original purpose was to connect mainframes to remote terminals. The advantage of X.25 over other WAN solutions is that it has built-in error checking. Choose X.25 if you have to use analog lines or if the line quality is not high.

X.25 is an ITU-T standard for packet-switched WAN communications over the telephone network. The term X.25 is also used for the Physical Layer and Data Link Layer protocols that make up an X.25 network. Originally designed to use analog lines to create a packet-switched network, X.25 networks can also be built on a digital network. Today, the X.25 protocol is a set of rules that define how to establish and maintain connections between DTEs and DCEs in a public data network (PDN). It specifies how DTE/DCE devices and Packet-swiching exchanges (PSEs) will transmit data.

• You need to pay a subscription fee when using the X.25 network.
• When using an X.25 network, you can create a connection to the PDN over a dedicated line.
• X.25 networks operate at 64 Kbit/s (over analog lines)
• The packet size (called frame) in X.25 network is not fixed.
• The X.25 protocol has a very strong error checking and correction mechanism so it can work relatively stable on low quality analog telephone lines.
• X.25 is currently widely used in many countries around the world where the numbers are not yet popular and the line quality is still low.

Frame Relay

Frame Relay is more efficient than X.25 and is gradually replacing it. When using Frame Relay, you pay to lease a line to the nearest node on the Frame Relay network. You send data over your line and the Frame Relay network routes it to the node closest to the recipient and forwards the data down the recipient's line. Frame Relay is faster than X.25

Frame Relay is a standard for packet-switched WAN communications over high-quality digital lines. A Frame Relay network has the following characteristics:

• There are many similarities to deploying an X.25 network.
• There is an error checking mechanism but no error correction mechanism
• Data transfer rate can be up to 1.54 Mbit/s
• Allows for different packet sizes
• Can be connected as a backbone connection to a LAN
• Can be deployed over a variety of connection types (56K, T-1, T-3)
• Operates at the Physical Layer and Data Link Layer in the OSI model.

When you sign up for Frame Relay service, you are committed to a level of service called the Committed Information Rate (CIR). The CIR is the maximum data transfer rate you are guaranteed to receive on a Frame Relay network. However, when network traffic is low, you can send data at speeds faster than the CIR. When network traffic is high, priority is given to customers with high CIRs.

ISDN (Integrated Services Digital Network)

One of the purposes of ISDN is to provide WAN access to homes and businesses using copper telephone lines. For this reason, the first ISDN deployment plans proposed replacing existing analog lines with digital lines. Today, the transition from analog to digital is taking place strongly around the world. ISDN improves operational performance compared to dial-up WAN access and is less expensive than Frame Relay.

ISDN defines standards for using analog telephone lines for both digital and analog data transmission. The characteristics of ISDN are:

• Allows broadcasting of multiple data types (voice, video, graphics...)
• Higher data transfer rates and connection speeds than traditional dial-up connections

ATM

ATM (Asynchronous Transfer Mode) is an advanced packet switching system that can simultaneously transmit data, voice and digital images on both LAN and WAN networks.

This is one of the fastest WAN connection methods available today, with speeds ranging from 155 Mbit/s to 622 Mbit/s. In fact, it can theoretically support speeds that are much higher than what is currently possible with today's transmission media. However, higher speeds mean higher costs, with ATM being much more expensive than ISDN, X25, or FrameRelay. ATM features include:

Uses small, fixed-size (53 byte) data packets (cells), which are easier to handle than the variable-size packets in X.25 and Frame Relay.

• High data transmission speed, theoretically can reach 1.2 Gbit/s
• High quality, low noise so almost no need for error checking
• Can be used with a variety of physical transmission media (coaxial cable, twisted pair cable, fiber optic cable)
• Multiple types of data can be transmitted simultaneously

WAN Hardware

The WAN hardware you use depends on the WAN service you want to connect to. Each WAN protocol has different specifications and requirements for hardware and transmission media. However, depending on your choice, there are many hardware options that can be compatible with different WAN services.

The WAN service provider is responsible for the WAN and provides the local loop to the Demarc (see Internet Made Simple #2/2004). The local loop is usually copper cable, the same type of cable used for telephone service.

Set up a phone line

Many homes and businesses today use a 4-wire cable consisting of two twisted pairs of copper wires: the first pair is used for telephones and the second pair is used as a backup. This allows new businesses to be ready for a WAN connection without having to install new wiring. An analog line uses two copper wires and a digital line can use two or all four copper wires of the Last Mile Cable, depending on the type of WAN connection. Telephone companies must modify the line switching in the Central Office to carry digital signals over Last Mile Cable.

Copper wiring is classified by bandwidth. Bandwidth, in turn, determines how much data you can send and whether the signal is analog or digital. Below we will look at two methods of classifying bandwidth on copper wiring.

Plain Old Telephone Service (POTS)

The analog phone system sends only one analog signal over each pair of wires: each of these separate signals is considered a channel. Using POTS and a modem to send analog signals gives you a 64Kbit/s channel, of which only 56Kbit/s of bandwidth is available for data transmission. Modems and traditional phone lines are fine for using the Internet for email and other general purposes. However, if you need to send and receive large amounts of data, it can take a long time.

POTS service has the following characteristics:

• Current lines use only two twisted pairs of wires.
• The signal on the Last Mile Cable is an analog signal.
• A modem is needed to convert digital signals into analog signals.
• The effective line speed is limited to 56 Kbit/s.

T-Carries

The physical layer of many WAN systems in the United States is based on the T-Carrier technology developed by Bell/AT&T. T-1 lines use all four copper wires: one pair to send and one pair to receive data. They do not use additional wires but instead establish virtual channels. Fiber optic cables and other types of transmission lines used for Last Mile Cable allow for higher data rates.

T-carries technology has the following characteristics:

• Use two pairs of twisted copper wire cable
• Using digital signals
• Supports multiple 64 Kbit/s channels on a single wire

T-carrier lines are classified based on the number of channels they can support.

• T1 (24 channels, used in the US)
• E1 (31 channels, used in Europe)

T-carrier lines are also classified according to the type of data that will be transmitted on the line (e.g. pure data, digital audio, digital video, etc.). Furthermore, users can subscribe to a portion of the T1 line's service and use some of its available channels.

Note: The T-carrier types are used to describe bandwidth, not WAN protocols. For example, ISDN is a WAN service that uses digital transmission over four wires. ISDN bandwidth depends on how much of the T1 line capacity is used.

Basic Rate ISDN (BRI)

Basic Rate ISDN consists of two 64Kbit/s channels (called B channels) and one 16Kbit/s channel (called D channel). Hence it is also known as 2B+D. The B channels carry digital data, voice and video. The D channel is the service channel used for both data and control information. ISDN BRI is ideal for homes and small businesses that need higher data rates than traditional modems.

Below are the two most typical use cases of ISDN BRI:

• One B channel is used for voice, the other is used for data.
• Both channels are used for data transmission at a total rate of 128 Kbit/s.

Note: The total bandwidth of ISDN BRI is 144 Kbit/s (2 B channels and 1 D channel) while the total data transmission rate is 128 Kbit/s (data is sent over only 2 B channels)

Primary Rate ISDN(PRI)

In the US, Primary Rate ISDN uses the entire T1 line, supporting 23 64 Kbit/s B channels and one 64 Kbit/s D channel, so it is called 23B+D. ISDN PRI is used in businesses that require high-speed, constantly on connections.

In Europe, Primary Rate is often referred to as 30B+D because it uses the entire E-1 line to support 30 B channels and 1 D1 channel.

In addition to the line, you need hardware to connect to the WAN and format the signal correctly for the type of connection you are using. For example, the hardware might be modems that convert digital signals to analog signals. You will use one or two of the following types of hardware for all-digital networks.

Multiplexer

As shown in the figure below, multiplexers operate at both ends of a transmission line. At the sending end, a multiplexer is a device that combines signals from two or more devices for transmission over a single line. At the receiving end, a multiplexer with demultiplexing capabilities separates the combined signals into their original individual signals. Many WAN routers have built-in multiplexers.

Statistical multiplexer: Uses separate virtual channels on the same physical line to send different signals simultaneously. (signals are transmitted at the same time on the line).

Time-division multiplexer: Sends packets of different signals at different time intervals. Instead of dividing the physical medium into channels, it allows data streams to use the medium at specific time “slots” (signals take turns using the medium for short periods of time).

CSU/DSU (Chanel Service Unit/Data Service Unit)

This is a device that connects networks to high-speed lines such as T-1. It formats data streams into framing formats and defines line codes for digital lines. Some CSU/DSUs are also multiplexers, or are built into routers. You may also hear of a CSU/DSU as a digital modem, but this is not entirely accurate. Modems convert data from analog to digital and vice versa, whereas CSU/DSUs simply reformat the data from an existing digital format.

The CSU receives the signal and transmits the received signal to the WAN line, reflects the reply signal when the phone companies need to check the equipment and prevents electromagnetic interference.

The DSU acts as a modem between the DTE and the CSU. It converts data frames from the format used on a LAN to the format used on a T-1 line and vice versa. It also handles line management, time division error, and signal regeneration.

Interface protocols

There are different types of “interface” protocols for WAN connectivity. “Interface”, in this context, refers to the format of physical layer frames or the methods of defining bit signals (formatting electromagnetic pulses).

Synchronous Serial Protocols

Synchronous serial protocols use precise clock signals between the DCE and DTE to transmit data in time. In synchronous communication, a large number of data frames are sent when the clock is synchronized and the data transmission rate is pre-established. This is a very bandwidth-efficient method of communication.

Synchronous signaling protocols include:

• V.35
• RS-232 (EAI/TIA)
• X.21
• RS-449
• RS-530

Although each “interface” protocol uses a specific type of connector, most connectors can be used for multiple interfaces. Typically, the type of hardware you have will determine which connector is used. In fact, check the pin number on the connector to make sure it matches the serial port on your device. Common connector types include (the numbers represent the number of pins in the connector): DB60, DB25, DB15, DB9.

Asynchronous Protocols

Asynchronous transmission protocols add start and stop bits to each packet to make transmission thinner, instead of requiring the sending and receiving devices to use a pre-agreed clock. Asynchronous transmission is often used between two modems. However, this is a costly method of transmission because the extra bits slow down the data transmission rate.

Asynchronous protocols are used to set standards for analog modem communications. A modem you purchase may support one or more different asynchronous communications standards. Asynchronous communications protocols include: V.92, V.45, V.35, V.34, V.32, V.32 bis, V.32 turbo, V.22.

Asynchronous signal transmission using standard telephone lines and jacks. Connectors can be: RJ-11 (2 wires), RJ-45 (4 wires), RJ-48.

Data Encapsulation Methods in WAN

The WAN physical layer protocols define the hardware and method of transmitting bit signals. The data link layer protocols control the following functions:

• Check and fix errors
• Set up link
• Organizing the fields of the data frame
• Point-to-point flow control

Physical link layer protocols also define the method of data encapsulation or the format of the data frame. The method of data encapsulation in WANs is often referred to as HDLC (high-level data link control). This term is both a generic name for Data Link protocols and the name of a protocol within the WAN protocol suite and service. Depending on the WAN service and connection method, you may use one of the following data encapsulation methods:

• Cisco HDLC for synchronous, point-to-point connectivity with other Cisco routers.
• LAPB for X.25 networks
• LAPD, used in conjunction with other protocols for B channels in ISDN networks.
• Cisco/IETF for Frame Relay Networks

The figure shows the most common data encapsulation methods and how they are used for typical WAN connections. As can be seen in the figure, PPP is a flexible method that can be used for many types of WAN connections. In general, which method to use will depend on the type of WAN service, such as Frame Relay or ISDN, and the data encapsulation method used by the network service provider.

WAN Management and Optimization

Because data transmission is still based on physical rules, the greater the distance between two devices, the longer it takes for data to travel between them. Likewise, the greater the distance, the greater the delay. Network congestion and dropped packets can also cause performance issues.

Some of these issues can be solved by using WAN optimization, which makes data transfer more efficient. This is important because WAN links can be expensive, so there are a number of technologies that have been developed to reduce the amount of traffic going over WAN links and ensure that it arrives efficiently. These optimization methods include reducing redundant data (also known as de-duplication), compression, and caching (bringing frequently used data closer to the end user).

Traffic can be shaped to give time-sensitive applications like VoIP higher priority than other, less urgent traffic like email, thus improving the overall performance of the WAN. This can be formalized into Quality of Service (QoS) settings that define traffic classes based on the priority each class receives relative to other classes, the type of WAN connection each type of traffic will travel over, and the bandwidth each type receives.

As a separate category, SD-WAN optimizes the WAN.

WAN Security

Traffic between WAN sites can be protected by a virtual private network (VPN), which provides security to the underlying physical network, including authentication, encryption, confidentiality, and non-repudiation. In general, security is an important part of any WAN deployment, as the WAN connection presents a potential vulnerability that an attacker could use to gain access to the private network.

For example, a branch office without a full-time information security officer may be lax in its cybersecurity practices. As a result, a hacker who breaches the branch network may gain access to the company’s main WAN, including valuable assets that should not have been compromised. In addition to networking features, many SD-WAN services also provide security services, which should be taken into account during deployment.

Interplanetary Internet

WAN technology isn’t limited to Earth. NASA and other space agencies are working to create a reliable “interplanetary internet” that will be used to transmit experimental messages between the International Space Station and ground stations.

The Disruption Tolerant Networking (DTN) program is the first step toward providing an Internet-like structure for communications between space-based devices, including communications between Earth and the Moon or other planets. But barring any significant breakthroughs in physics, network speeds will likely exceed the speed of light.