Introduction
The integration of voice with data is
not new. For many years, voice has been integrated with data in networks
based on TDM multiplexer technology. However, today, newer technologies are
being used to carry voice too. Frame Relay and ATM have been used, and continue
to be used extensively to transport voice, indeed, ATM is a core part of
the third generation (3G) mobile network architecture carrying both the voice
and data from the radio base stations into the core network. Meanwhile IP
is rapidly becoming the choice today in the form of voice over IP, or IP
telephony.
The use of these technologies for the
support of voice traffic is, unfortunately, not as simple as many would like to
think. It creates many challenges for those responsible for network engineering.
For example, delay, echo, problems with voice compression and incorrect voice
signal levels can all easily result in poor communication even to the extent of
complete voice communication failure. This course addresses all of these issues
and many more, providing an insight into what issues exist and how they may be
resolved.
Who should attend
the course?
This course was originally aimed at
individuals from a data background who need to develop a greater understanding
of voice technologies, from the basics through to the practical "in the
field" issues. It has however also been found to be suitable for those who
already have experience in voice networking, yet wish to consolidate their
knowledge and understand many of the new issues facing voice in communication
networks.
Course Agenda
The following is an outline of the
sections included in the course:
- Introduction
to Voice Communications
- Digital Voice
- The Telephone
- The
Telephone Exchange
- Digital
Voice Transmission
- E1 Digital Voice
Interface
- DS-1 (T1) Digital Voice
Interface
- Signalling
on Analogue Interfaces
- Signalling
on Digital Interfaces
- Overview of
Basic Rate ISDN
- Voice/Speech
Compression
- Speech
Impairments
- Echo and Echo Control
- Voice
and Data Integration
- Voice over Frame Relay
(VoFR)
- Voice and Telephony
Over ATM (VTOA)
- Voice over IP
Course Length
The course length is three days.
1
Introduction to Voice
The objective of this section is to
give a brief introduction to the true fundamentals of voice in terms of how
speech is generated and the importance of certain aspects of speech when applied
to networking communications.
- How speech is generated and the sounds
that make up speech
- The importance of certain frequencies
and levels
- Review of dBs
2
Digital Voice
Today, voice is usually integrated into
networks in a digital format. However, to be compatible with the human user, the
signals must still be in an analogue form at some point, notably the telephone
set. This section looks at how voice is converted from an analogue signal into a
digital format and looks at some of the issues involved in its conversion.
- Comparison of analogue and digital
voice
- PCM encoding: Analogue - Sample - PAM
- Quantise - Compand - PCM (G.711)
- A law and mu law companding - Where
should A law and mu law be used?
- Quantisation Distortion
- Power of the digital signal
3
The Telephone
The telephone is the fundamental
building block of a voice network. While it was originally designed in 1874 and
patented in 1876, its principles have not changed greatly since then. In fact,
many of the techniques used in these early telephones are still used extensively
today. The telephone also has attributes that make it one of the greatest causes
of problems in networks today, that of echo. This section goes into the
operation of the basic 2 wire telephone set, and to give a complete picture,
also looks at other proprietary types of telephone, and finally looking at the
operation of the Basic Rate ISDN set.
- Basic operation of telephone - 2 to 4
wire hybrid, transmit/receive levels etc.
- Loop disconnect and DTMF signalling -
Advantages and disadvantages
- Digital telephone set
4
The Telephone Exchange
This section gives an overview of a
modern digital telephone exchange and looks at how it switches telephone calls,
how it routes calls and how it interfaces to other equipment such as
multiplexers.
- Call routing in a digital PBX
- Principles of switching
- Interfaces available - analogue and
digital
5
Digital Voice Transmission
Today, practically all voice is carried
across networks in a digital form. This section serves as an introduction to the
next two, looking at the basis of the support of multiple PCM voice channels
across standardised primary rate digital interfaces.
- The Channel Bank
- An introduction to the E1 and DS-1/T1
digital voice interfaces
6
E1 Digital Voice Interface
The predominant type of interface used
for voice transport within a Wide Area Network today is the digital interface.
There are two main types of digital interface available on a PBX, those of DS-1
(T1) and E1. DS-1 is used in North America and in some cases in Japan, while E1
is the standard used elsewhere. This section looks at the E1 interface in depth.
- Electrical characteristics - G.703 -
HDB3
- Framing - G.704
- Channel Associated Signalling -
Timeslot 16
- Common Channel Signalling
- E1 Alarms
7
DS-1 Digital Voice Interface
The operation of a DS-1 interface is
somewhat different to an E1, and a detailed understanding of the differences is
key for those involved with networks using both. This sections looks at the same
areas as with E1s, however, as we shall see, the techniques used are quite
different in many respects.
- Electrical characteristics AMI - B8ZS
- Framing - D4 and Extended Superframe
- Channel Associated Signalling - Robbed
bit signalling
- Common Channel Signalling
- DS-1 Alarms
8
Signalling on Analogue Interfaces
Whilst most interfaces between
telephone exchanges and other voice networking equipment are digital today,
there is still a vast demand for analogue interfaces. Typically, analogue
interfaces are used where only a small number of voice channels are required
whereas digital interfaces are used where a larger number are required. This
section discusses a number of different analogue interface types, focussing on a
particularly common type known as the E & M (Ear and Mouth) interface.
- Loop Disconnect
- E & M (Main section)
9
Signalling on Digital Interfaces
Digital signalling systems are used on
digital interfaces to carry such information as call setup details in addition
to more advanced messages for activating PBX features across a network. It is
important to understand the differences between the two main types of digital
signalling i.e. Channel Associated Signalling (CAS) and Common Channel
Signalling (CCS), since they are quite different. CAS tends to be easier to deal
with, while CCS is more common and, typically, provides more feature capability.
This section looks at some examples of CAS yet concentrates on CCS with prime
focus on signalling protocols including Q.931, QSIG and Euro-ISDN.
- Channel Associated Signalling
- Common Channel Signalling
- "Standard" and proprietary
CCS protocols - ISDN/Q.931, Euro ISDN, QSIG, DPNSS, CCITT#7 (CCS7, SS7)
10
Overview of Basic Rate ISDN
While much of this course is orientated
towards primary rate interfaces, this section gives an introduction to Basic
Rate ISDN and looks at its operation and some of its unique features and
capabilities.
- Physical Structure - Multiple Devices
- Basic Rate Frame Structure
- Bearer Services, Teleservices and
Supplementary Services
- Circuit Mode, Packet Mode and Frame
Mode Services
11
Voice/Speech Compression
Voice compression is one of the prime
reasons for integrating a PBX network with a Wide Area Network. Voice
compression has been available on TDMs for many years. Over time, more and more
compression techniques have been developed to improve the efficiency of the
bandwidth used. Key to the use of voice compression are the effects it has upon
the quality of the voice being carried both in terms of distortion and, in some
cases, the additional delay added. Furthermore, when we try to support voice
band data such as modem traffic, fax traffic or possibly in-band signalling, a
number of issues arise. An appreciation of these factors is key to the
successful implementation of voice compression.
- What is voice compression and why do
it?
- A look at various different types of
speech coder - Waveform, Source and Hybrid coders
- ADPCM - G.726 (16, 24, 32 &
40kbit/s)
- Embedded ADPCM (E-ADPCM) - G.727 (16,
24, 32 & 40kbit/s)
- LD-CELP - G.728 (16kbit/s)
- CS-ACELP - G.729 (8kbit/s)
- ACELP/MP-MLQ - G.723.1 (5.3k and
6.3kbit/s)
- 7kHz ADPCM - G.722
- Other compression systems
- Support of voice band data on voice
compression circuits
12
Speech Impairments
Essential to the successful support of
voice in any network whether it be a traditional TDM (Time Division Multiplex)
environment or a packet based network such as Voice over IP, is the minimisation
of speech impairments. This section looks at some of the key potential
impairments including delay and distortion.
- End-to-end delay including
packetisation delay, transmitter delay and packet jitter delay
- What delays can we tolerate? - ITU-T
G.114
- Distortion - What can we tolerate? -
ITU-T G.113
13
Echo and Echo Control
Today, the use of such technologies as
ATM and Frame Relay, as well as techniques such as voice compression, are
resulting in more and more delay being imposed onto voice connections across our
networks. The combination of signal reflections and delay create the effect of
echo. In practice, we can tolerate a certain amount of echo, although only very
little, and steps should be taken to eliminate it. The objective of this section
is to look at various causes of echo and to discuss how it can be removed. Both
echo suppressors and echo cancellers are discussed, although the primary focus
is on echo cancellers.
- Delay and what we can tolerate
- Causes of echo
- Echo Suppressors
- Echo Cancellers
- G.165 vs. G.168 echo cancellers
14
Voice and Data Integration
This section investigates a number of
potential ways that voice and data may be integrated and some of the key aspects
that need to be considered in each case.
- The transport of voice across a data
network when supporting analogue interfaces and digital interfaces with CAS and
CCS signalling types
- The interworking of DS-1 and E1
- Using the "data network" to
support voice switching
15
Voice over Frame Relay (VoFR)
This section looks at how voice is
actually carried across Frame Relay as defined by the Frame Relay Forum FRF.11
Implementation Agreement.
- Introduction to Frame Relay
- How voice samples are carried in
frames
- The use of Virtual Circuits for voice,
data and voice/data
- Multiplexing of multiple channels
within a frame
- The support of signalling: DTMF, CAS
and CCS
- The support of fax traffic
- FRF.12 - Frame Relay Fragmentation
16
Voice and Telephony Over ATM (VTOA)
In this section we look at methods used
to carry voice over ATM. The ATM Forum has produced a number of standards for
VTOA which we shall look at. In addition we will also look at the specification
for a new AAL, AAL2 for the support of voice trunking including compressed
speech support. We shall look at the issues that surround some of these
techniques along with many of the benefits that can be derived from them.
- Introduction to ATM
- Transport of a complete E1/DS-1
digital voice stream - Circuit Emulation
- Support of a single voice channel in a
single Virtual Circuit
- Support of multiple voice channels in
a single Virtual Circuit
- Silence Suppression (Speech Activity
Detection, Voice Activity Detection)
- af-vtoa-0078.000: Circuit Emulation
Service Interoperability Specification V2
- af-vtoa-0083.000: Voice and Telephony
Over ATM to the Desktop Specification
- af-vtoa-0085.000: Specifications of
(DBCES) Dynamic Bandwidth Utilisation - In 64kbit/s Timeslot Trunking Over ATM -
Using CES
- af-vtoa-0089.000: Voice and Telephony
Over ATM - ATM Trunking using AAL1 for Narrowband Services Version 1
- ITU-T I.363.2: Voice Trunking on ATM
including compressed speech support over AAL2
17
Voice over IP (VoIP)
Today the topic of supporting voice
over, traditionally, data based networks is one of the hottest around. There is
a huge move towards the support of voice over IP, both across Local Area
Networks (LANs) and across Wide Area Networks (WANs). This section looks at how
voice is supported over IP. It also discusses the idea of running voice over the
Internet.
- Introduction to Voice over IP
- A look as H.323 - Terminal equipment,
Gatekeepers and Gateways
- A look at H.225.0 and H.245
- How speech, DTMF, signalling etc.. is
carried in IP packets
- What are the issues of supporting
voice over IP?
|
