Washignton School District
                           WAN Protocols  
WAN Design

WANs are used to interconnect LANs that are separated by a large geographical distance, utilizing the services of RBOCs.  WANs are predominately operates at Layer 1 and 2 of the OSI model. 

There are several common Layer 2 encapsulations:

  • HDLC supports both point-to-point and multipoint configurations.  However, it might not be compatible between different vendors.  Additionally, it doesn't support multiple LAN protocols on a single link because it doesn't have a way of indicating which protocol it's carrying.  Although, Cisco HDLC has a proprietary type field that acts as a protocol field.
  • Frame Relay  can transmit data very rapidly, but it has no error correction mechanism.  Consequently, it relies on upper layer protocols for this.
  • PPP  has a protocol field that helps to identify different Layer 3 protocols. 
  • ISDN  is a set of digital services that transmits voice and data over existing phone lines.

Data communication over WANs interconnect DTEs so they can share resources over a wide area.  The WAN path between the DTEs is called the link, circuit, channel, or line.  Typically, a DTE is the router and a DCE is the device that converts data from a DTE into a form acceptable to the WAN service's facility.

All types of encapsulation methods share a common frame format:

  • Flag- Indicates the beginning of the frame and is set to a hexadecimal pattern of 7F.
  • Address- A one- or two-byte field to address the end station in multidrop environments.
  • Control- Indicates whether the frame is an information, supervisory, or unnumbered type frame   It also contains specific function codes.
  • Data- The encapsulated data.
  • FCS-The frame check sequence.
  • Flag- The trailing 7E flag identifier.

There are two types of options available for WANs:

  • Switched connections or
  • Dedicated lines  

Switched connections may be circuit switched or packet/cell switched.  Dedicated lines provide fulltime service.  Consequently, they're used to provide core connectivity between major sites or LAN-to-LAN connectivity.  They're mainly used to carry data, voice, and occasionally video.

When designing a WAN, it's usually too expensive to design a full-mesh WAN core.  Consequently, a partial mesh is used that provides multiple equal cost paths between sites.  Ideally, two sites would be directly connected to minimize the amount of delay.  However, multiple hop paths can be placed between sites that generate less traffic.

Another alternative involves utilizing a switched network.  Specifically, Frame Relay, a cost effective alternative that connects every other site by a virtual circuit.  Each router needs only one physical interface to the carrier  Additionally, latency is only one hope from site to site.

PPP

PPP is the most widely used WAN protocol.  It performs the 4 following functions to allow data to be moved across a WAN with reliability:

  • Link Establishment
  • Link Quality Determination
  • Network Layer Protocol Configuration
  • Link Termination

PPP is able to provide router-to-router and host-to-network connectivity over both synchronous and asynchronous circuits.  Additionally, PPP supports a number of network layer protocols.  Among those include IPX, TCP/IP, and AppleTalk.

PPP utilizes LCP to negotiate and set up control options on the WAN data link  On another front, it uses NCP to encapsulate multiple protocols.

The following is how PPP establishes a connection:

  • An originating PPP node first sends LCP frames to configure and test the data link. N ext, the link is established, and facilities are negotiated.  The originating PPP node then sends NCP frames to choose and configure network layer protocols. T he chosen network layer protocols, such as TCP/IP, Novell IPX and AppleTalk, are configured, and packets from each network layer protocol are then sent.
  • When PPP is configured, two authentication methods are available, PAP and CHAP.   PAP provides a simple method for a remote node to establish its identity using a two-way handshake.  While CHAP is used at the startup of a link, and periodically to verify the identity of the remote node using a three-way handshake.  CHAP is the preferred method of authentication because it provides protection  against playback attacks.

ISDN

ISDN was specifically designed to solve the issue of small offices or dial-up users that required more bandwidth than traditional dial-up users.  It was developed to utilize existing the telephone wiring system yet increase speeds up to 128Kbps.  Not surprisingly,  ISDN works very much like your telephone.  When you want to make a data call, the WAN link is brought up.  Then it stays up for the duration of the call.  After the call has been completed, the WAN link is taken down again.

Another feature and advantage of ISDN is its ability to concurrently deliver voice, data, text, graphics, music, video, and information to users.  Consequently, you can simultaneously surf the Web and talk to another person at the same time.

A popular method of connecting via ISDN is the Remote Node method.  In this method, users connect to a local LAN at the central site for the duration of the call.  This connection is typically through an access server, a device that combines the function of a modem and a router.  Consequently, the remote user is then able to see the same environment that a local user sees.

The following are the advantages of utilizing ISDN:

  • The ability to carry a variety of user-traffic feeds. ISDN provides access to all-digital facilities for video, packet-switched data, and telephone network services.
  • Much faster call setup using out-of-band (D, or delta, channel) signaling than modem connections. For example, a duration of less than one second can be sufficient to make some ISDN calls.
  • Much faster data transfer rate using bearer (B) channel services at 64 kbps per channel as opposed to common modem alternatives of 28.8 kbps. With multiple bearer B channels, ISDN offers users more bandwidth on WANs (for example two B channels equals 128 kbps) than they receive with a leased line at 56 kbps in North America or 64 kbps in much of the rest of the world.

In order to implement ISDN, the following equipment are needed:

  • Terminal Equipment 1 (TE1) - Designates a device that is compatible with the ISDN network.  A TE1 connects to a Network Termination of either Type 1 or Type 2.
  • Terminal Equipment 2 (TE2) - Designates a device that is not compatible with ISDN and requires a Terminal Adapter.
  • Terminal Adapter (TA) - Converts standard electrical signals into the form used by ISDN so that non-ISDN devices can connect to the ISDN network .
  • Network Termination Type 1 (NT1) - Connects 4-wire ISDN subscriber wiring to the conventional 2-wire local loop facility.
  • Network Termination Type 2 (NT2) - Directs traffic to and from different subscriber devices and the NT1. The NT2 is an intelligent device that performs switching and concentrating.

While utilizing ISDN, two encapsulation methods are available:

  • HDLC
  • PPP . 

By itself,  ISDN defaults to utilize HDLC ; However, PPP be initiated in order to reap the benefits of CHAP although ISDN interfaces allow only one encapsulation type.

Frame Relay

Frame Relay is a streamlined data link technology that provides high performance and efficiency that operates at Layer 1 and 2 of the OSI model.  Consequently, it relies on upper layer protocols for error correction.

One concern when implementing Frame Relay is the CIR (committed information rate).  This is the rate in bits per second that the Frame Relay switch agrees to transfer data.  Another important concern is the DLCI (data link connection identifier).  The DLCI is used to identify Frame Relay virtual circuits.  However, DLCI s only have local significance.  Consequently, DLCI values are not unique in the Frame Relay WAN.  

LMI (link management interface) is a Frame Relay signalling specification .  It's purpose is to find out the status of various PVC s that are known by the router.  Additionally, it's also responsible for transmitting keepalive packets to insure that PVCs stay up.  There are two methods of encapsulation available for LMI, the IETF and the Cisco encapsulation type.

                                                                                              Back to Top