This section describes Voice over IP (VoIP), and it suggests why you should use VoIP for voice transmission.

 

Voice can be transmitted across a traditional telephony network in two forms: analog and digital.

A typical residential telephone is an analog device that converts your voice into an electrical current. The receiving telephone recreates the initial variation of amplitude and frequency. What the party at the other end of the line hears isn’t really the original speech; it's an analogy of the original speech. This is why the signal is called an analog signal.

Analog signaling can be represented by a continuously variable waveform (illustrated below), and can represent an infinite number of states. Because analog signals tend to degrade over distance, however, they need to be periodically amplified.

Waveform Analog Signal

In the early telephone network, when analog transmission was passed through amplifiers to boost the voice signal, the ambient line noise was amplified as well. Therefore, the sound quality of the transmission was degraded, often resulting in an unusable connection. Typical residential telephones are still based on analog technology.

In the 1970s, to improve quality of the received signal, add features, and handle increased call volume, the telephone network began migrating to digital-loop carrier technology, represented by T1, T3, E1, and E3. T1, used primarily in North America and Japan, is a 24-channel digital system that carries data at 1.544 Mbps through the telephone switching network. E1, used predominantly in Europe, carries data at a rate of 2.048 Mbps using 30 64-Kbps digital channels.

 

In comparing the telephone network to a data network, we could say that the telephone network is a very large circuit-switched network. More than any other network, however, the PSTN is ubiquitous—it exists everywhere, it is dependable and easy to use, and everyday life would be inconceivable without it.

Data networks, on the other hand, are privately owned and administered, so architecture, size, and complexity may vary. Data networks were likely originally designed for a different purpose other than to transport voice traffic, which is more sensitive to delay and can demand more bandwidth than data.

VoIP enables Cisco routers and switches to carry telephony-style voice traffic—that is, live, packetized voice traffic such as telephone calls—over IP-based data networks (intranetworks or internetworks) rather than the PSTN.

To configure an IP network to carry voice, you first need to make sure that the network is well-engineered end-to-end. Although good design practices are essential to VoIP, a complete discussion of network design is beyond the scope of this tutorial.

Cisco routers and switches are equipped to handle the origination, transport, and termination of VoIP traffic. When transmitting voice signals across a data network, the router on the network performs some of the same basic functions as a switch, which are as follows:

  1. The analog signal of the originating call is converted into a digital signal in the digital signal processors (DSPs) in the router.
  2. The voice signal is then segmented into frames, which are then coupled in groups of two and stored in voice packets that conform to the standard IP format.
  3. The IP packets are sent over the Internet in compliance with ITU-T specification H.323, explained in the related document.

At the receiving end, this process is reversed, with the packets being converted to digital signaling, then back to analog to be received and heard on the other end of the call.

The above discussion is a very brief summary of issues relating to analog-to-digital conversion of voice over an IP network. This topic is covered in greater detail later in module 3. Compression techniques, CODECs, and mean opinion scores (MOS) and their relationship to voice quality are discussed in these related documents: Quality of Service Overview, Defining Analog Voice, Waveform Coding Techniques, CODECs, and Reducing Bandwidth Consumption, but are beyond the scope of this course. A future product in the CIM for Voice Internetworking series will cover Quality of Service (QoS) technology topics, configuration, and troubleshooting.

 

VoIP is IP-based. IP is considered a connectionless or best-effort transport when used in conjunction with the User Datagram Protocol (UDP). For this reason, UDP, which is datagram-based (or connectionless), best suits the specific requirements of voice traffic. UDP is preferred as the transport for VoIP, even though TCP, which is connection-oriented, is generally considered as a more ideal transport mechanism because of its built-in reliability.

So the voice packet to be sent over IP is split into two portions: (1) the control signaling and (2) the actual packetized voice traffic. The control signaling then runs on top of TCP, and the real-time voice is sent across UDP.

Consider the underlying reasons for using UDP as the transport for voice traffic:

The following figure shows the VoIP packet.

VoIP Packet Structure

TCP offers reliability in that it guarantees retransmission of lost frames, so is used to carry the control signaling, but this reliable delivery is useless in the internetwork transportation of packetized voice because a frame that arrives late as a result of retransmission is as useful as no frame at all—that is, it has no effect. In other words, retransmission of packets is not meaningful. By the time the resent packet arrives at the end user endpoint, the required delivery time has long been transgressed.

 

Among the many reasons to use VoIP for voice traffic are the following:

 

Implementation of an Internet telephony network in a business can have many benefits. It enables toll bypass, remote PBX presence over WANs, unified voice and data trunking, and POTS-Internet telephony gateways. The primary benefits are significant cost savings because of the following:

Placing a Phone Call from a PC...

How Toll Bypass in a Large Corporation Works

 

 

VoIP is primarily a software feature; however, to use this feature on a Cisco 3600 series router, you must install a voice network module (VNM). The VNM holds either one or two voice interface cards (VICs). Each two-port VIC is specific to a particular signaling type associated with a voice port type, such as FXS, FXO, or E&M analog voice ports.

A Cisco 3640 can house up to three modules with up to a total of six VICs that slide into the voice/fax network modules and provide the interface to the telephony equipment.

FXO, FXS, and E&M Interfaces

Foreign eXchange Office (FXO) and Foreign eXchange Station (FXS) are types of RJ-11 interfaces on Cisco routers that utilize either loop-start or ground-start analog signaling techniques:

E&M interfaces support E&M signaling types I through V, either two-wire or four-wire operation, and support wink-start, delay-start, and immediate-start signaling. E&M connections are frequently used to connect computer telephony connections to a switch, and are also used in trunking arrangements switch-to-switch and switch-to-network.

Two Views of the Cisco 3600 Series Routers

Rules for Voice over Data Networks...

It is important to match the correct interface type with the signaling type required for the application in the VoIP network.

The examples in the first part of this tutorial use a Cisco 3640 router with FXS/FXO and a version of Cisco IOS software that supports VoIP. The next module covers E&M technology and configuration.

The VoIP gateway implementation on the Cisco 3600 series is a subset of the ITU-T standards. Contact the ITU-T for more details or to obtain copies of the standards documents.

 

 

Go on to Configuration of Basic Analog Voice over IP.