
FREQUENTLY ASKED QUESTIONS 1:
 |
 |
 |
What
are the basic components of a CallHandler system? |
| |
The
CallHandler Audio Conference System consists of four basic elements.
- The
PC Chassis Running Microsoft Windows NT
- Telephony
PCI Card(s) that plug into the Computer Slots. Analogue, ISDN Basic Rate, or
ISDN Primary Rate.
- In
the case of the American T1 Trunk which has 24 incoming lines x 4 per card. (4
max. telephone sockets per card) The trunks are supplied by your Telco.
operator in the in the form of a socket on your wall - probably next to your
company telephone exchange. A single trunk connection looks a bit like a normal
RJ45 computer network socket. You will also need a LAN/WAN socket next to the
system so that you can control it - if needed.
- You
connect Primary Rate ISDN cables into this board (Think of it as a very, very
sophisticated modem)
- For
a 24 line system you need one card with one socket - In the UK these trunks
(E1) have 30 lines capacity.
- You
need to talk to your telecom. provider for and incoming trunk(s) that connects
to this
- The
CallHandler Operating Software that manages the whole system and provides
other capabilities
- The
Application Software (e.g.CallHandler Audio Conference) Consisting of: a) Software that runs
on the CallHandler box b) Software that runs on PC on a LAN/WAN to set-up
and control the conferences
If
you want to add more lines, or change from analogue to digital as your business
grows, simply change the telephony card and the incoming telephone line type.
|
 |
What
is Power/Preview Dialling in Call Center Applications? |
| |
- What
are its dialling principles?
- The
agent script passes the power/preview dialler a
telephone number to dial, a message is sent to the CallHandler, the CallHandler
dials the number, then connects the agent to the ringing trunk line. When the
call is answered an event is sent from the CallHandler to the agent which can
trigger a screen pop on the agent's terminal.
- What
is its design criteria i.e. what are its benefits & attributes?
- The
benefit is that the dialling process is automatic, so less prone to error and
can be linked to an Agents script. It is generally used for smaller Agent
systems where Predictive Dialling is not effective
- Where
does it sit relative to the agent desktop and PABX?
- a)
The CallHandler can replace a PABX,
or b) The CallHandler can interface to
a PABX via a link e.g. DPNSS or QSIG
-
Does it interface to any common "Contact Management Systems" e.g. Sales
Logix
-
No direct interface, interfacing should be very easy using the
PowerDial/PreviewDial COM object.
- What
components are used in its construction?
- The
CallHandler Server
Windows NT4.0 OS Platform, Aculab Telco Spec Telephony
Cards, Pika Telco Spec Headset Connectivity Cards, Industrial specification
Chassis, and PC Components Modular COM/ActiveX software components, Hand
Optimised C/C++ Software, VB Scripting
- PowerDial
COM Object
C++ Software COM Component Communicates via TCP/IP to
CallHandler Server
|
 |
How
is SMS messaging dealt with? |
| |
CallHandler
version 2.0 has integrated SMS messaging. This means that you can send an SMS
command via a LAN/WAN to the CallHandler and this will send an SMS message
to a mobile phone.
We can support several delivery
methods: -
via GSM modem / Mobile phone attached to
CallHandler. -
ISDN/Modem dial-up to an SMSC (SMS Message
Centre) -
via Internet to an SMSC Chelston
has developed the SMS messaging software from the ground up to be very reliable
and efficient. Costs roughly:
No or very little set-up fee
Anything between 3p to 7p per message for 1-10000 messages
1. Web Servers - hundreds try these UK based ones:-
www.clickatell.com
www.bulksms.com
www.csoft.co.uk
or type: "bulk sms uk" into Google
2. There are several of ways to do this, depending what the service provider
supports:-
a. Forward the email to the SMS service provider using their SMTP
mail interface
b. Retrieve the email, convert it to an HTTP packet, queue it then
post it to the SMS service provider's server
c. As above, but using TCP/IP sockets connection. |
 | What low level logging info is available from E1/T1 |
| | Here is the actual low level logging information This is a sample log trace from the Traffic log.
^-- Date/Time Trace Was Recorded ^-- CallHandler Unique Resource Identifier
01/07/2002 04:40:16 | 005 | ResHandle:0x86200800
^-- Start Date/Time Of Call ^-- Call Duration In Seconds
StartDate:01/07/2002 StartTime:00:01:10 Dur:43
^- DDI Number (If Set) ^- CLI Number (If Set) ^- Real Destination Number ^-Real Originating Number
DDI: CLI: DestAddr:96789354112 OrigAddr:
^-- The Reason The Call Ended ^-- The Raw Clearing Cause Code Taken From The Protocol
Cause:NORMAL:0x0 Raw:0x0 |
 |
What
signalling protocol over that E1? Is it MFC-R2? Is it "Digital E and M" with
DTMF?. Maybe "Primary-Rate" (PRI) with signalling over the D-Channel
(Q.931). |
| |
We
can support just about any signalling protocol by loading firmware drivers to
support the protocol. For example, we support PRI Q931, T1 PRI, MFC-R2
(Standard in South Africa, Egypt and Israel) also Basic Rate Euro ISDN CTR3 for
connection to COs. We support both 'User End' and 'Network End' connections, so
the CallHandler system can also be a CO. Also, we support Q.SIG and DPNSS
for direct connection to PBXs. Your chassis should behave as if it is a PBX
towards the "Central-Office" exchange. This is the 'User End' configuration,
this is no problem. We fully implement all features of the protocols we
support, this includes Call Clearing reporting, Direct Dial Inward (DDI) and
Caller Line Identification (CLI) signalling |
 |
Can
CallHandler support more than a single E1? |
| |
We
can support up to 80 E1 connections, or 2400 lines from a single chassis
depending on the application. Multiple PC Chassis can be clustered to gether to form very large systems. *** Compact PCI (cPCI) Solutions ***
2400 Lines in a Compact PCI System Dual Processor
(Limited by H110 Bus Capacity)
*** Passive Backplane PCI Solutions ***
2040 Lines Call Switching Only
(Limited by Chassis Size, 17 slots)
1080 Lines Audio Record/Playback on Dual 1GHz PC
(Limited by Host Processor / PCI Bandwidth) |
 |
How
many mailboxes? |
| |
There
are a number of things that limit the number of mailboxes you can have:
i)
Disk Space Every message left uses up disk space at the rate of 3.7MB per
minute (uncompressed). So, 5000 mailboxes with say, 5 messages each with an
average 2 minute length will use up 190GB, for example. The maximum size we can
support in a single chassis is 109GB, or 2900 mailboxes. We can make better use
of the storage space available by using audio compression 'on-the-fly' as the
messages are recorded. We can support several rates of compression, 4:1 or 2:1,
although audio quality degrades with compression. To support a larger number of
mailboxes we would use an external RAID array. One possible RAID system would
take disk capacity up to 3.4TB, or 91000 mailboxes. ii)
Number of DDI Numbers Available If you have 5000 mailboxes and you want to give
every user a separate telephone number, you will need 5000 DDI numbers from
your telecoms provider. You need to ask your telecoms provider whether they can
supply this many numbers. Assuming they can, whenever someone wishes to leave a
message, they will dial a phone number say: '01189016812', the telecoms company
will switch the call through to the CallHandler and pass either the whole
number or the last few digits of the number to the CallHandler, in the example
'6812', before the call is answered. The CallHandler will use this number to
locate the correct mailbox. We can support either a part of the number or the
whole number being forwarded as a DDI, so can support any number of DDI
numbers, and hence mailboxes. |
 |
Aculab
Text to Speech - what do I need? |
| |
This
uses a separate PC to run the TTS Engine, and links to the CallHandler via
100BaseT LAN link. This configuration can support 120 concurrent TTS channels.
|
 |
Lernout
& Hauspie Text to Speech - what do I need? |
|
This
uses the DSP resources (opposed to a separate host PC). We can support up to 26
concurrent channels per card - 8 Cards per PC |
 |
Microsoft
SAPI Support - what can it do? |
|
This
will allow any SAPI compatible TTS engine to work with the CallHandler e.g.
Lucent TTS, Microsoft (L&H) TTS, L&H, Philips etc. This will support
any number of TTS channels, distributed over one or more TTS servers, and
linked to the CallHandler via a high speed LAN. |
 |
What
custom controls does CallHandler have? |
| |
There
are several custom controls. Each Device and Device Manager in the CallHandler
has two controls: Configuration Control and Status Control. The Configuration
Control is used to configure the device prior to it being "Run" by the
CallHandler. The Status Control is used to monitor the status of the device and
to alter device settings while it is being "Run". The Status and Configuration
Controls run from within the IE5 browser and form part of the CallHandler Maintenance Centre. |
 |
What
does the Developer Kit include regarding Text to Speech |
| |
Developers
Kit including a)
API Documentation b)
Example Code c)
Hooks for Aculab and Lernout & Hauspie Text to Speech Systems |
 |
Do
I need special equipment to run the Conference Demo Disc? |
| |
No,
the demo software is Install Shield based and contains all the necessary
modules and simulators. |
 |
Can
we use the existing E1 Trunk which has 15 lines currently used for the office
exchange? |
| |
Yes,
providing the E1 is looped through the CallHandler. The CallHandler
would strip out its 15 lines, and then pass the remaining ones to the local
exchange. We would supply a system with additional physical ports to facilitate
this. |
 |
Why
do you still use NT when there are later versions of MS Windows
available |
| |
CallHandler
uses Windows NT 4.0 Service Pack 6 because this is a tried a tested platform.
We tend to software and hardware that has a track record, and has been fully
tested in our laboratory before making a general release. We can use Windows 2000 if required. |
 |
What
version of SQL do you support? |
| |
Latest Version |
 |
What
is the format of the audio data? |
| |
The
format of the audio data conforms to the ITU-T Recommendation G.711 Pulse Code
Modulation (PCM) of Voice Frequencies. This is a companded audio format for
improving noise performance of audio channels. The G.711 format covers two
incompatible companding formats: A-Law and mu-Law. These
format apply to specific countries depending on the telecommunications system
the country supports. Generally, Europe operates using the A-Law format, and
North America operates using the mu-Law format. This only applies when the
CallHandler system is installed in a particular country, it must support
the companding format for that country. But if a call is made from a
CallHandler in Europe to North America, the CallHandler would use A-Law
companding, and the international telecom provider will convert A-Law to mu-Law
and visa-versa between countries. |
 |
Do
you have any references on Audio Formats? |
| |
[1]
Microsoft Multimedia Standards Update - New Multimedia Data Types and Data
Techniques Rev3.0 April 15, 1994 - Available as part of the Multimedia
Registration Kit (MRK) from Microsoft, the MRK is distributed as an MSDOS
self-extracting archive, which is available from:
ftp://ftp.microsoft.com/softlib/mslfiles/mdrk.exe
filename: RIFFNEW.DOC [2]
ITU-T Recommendation G.711 [11/88] - Pulse code modulation (PCM) of voice
frequencies. Available from the International Telecommunications Union (ITU) at
http://www.itu.ch [3]
MSDN Library - Resource Interchange File Format Services - Multimedia: Platform
SDK |
 |
Do
your Telephony Cards work in Egypt? |
| |
Our
Digital Telecom cards have been tested in Egypt. We understand that Egypt uses
a standard MFC-R2 but there can be certain variations depending on the Egyptian
exchange the card is run against, and this may vary between the Egyptian
telecom providers. |
 | Can you give me more information on video over Voice Over IP (VoIP), this is a real benefit in
the call centre or strategic customer service environment. i.e. picture and
voice point to point. |
| |
Both the H323 and SIP standards support the streaming of video.
H323 and SIP are basically both protocols for establishing streaming
connections between 2 (or more parties). Both H323 and SIP use the
RTP (Realtime Transport Protocol) and RTCP (Realtime Transport Control
Protocol) for streaming. RTP can stream just about any format of audio
and video. So the protocol is pretty much in-place to support any sort
of multimedia. Comparing audio and video streaming, a typically compressed
audio stream uses 8kbits/sec, whereas a typically compressed video stream
uses typically 64-128kbits/sec. Compressing and streaming video around
and enterprise is a much larger problem than audio, you need much bigger
network bandwidth, and much bigger codec cards, and hence is a lot more
expensive. |
 |
What about security and Voice Over IP (VoIP), will or could the connectivity of telephone
networks to IP present some sort of security breach to the corporate
|
| | There are several issues with security:
i) At present with most VOIP systems it is not possible to encrypt the
audio channels, so anyone with a VOIP network monitor can listen in on
a VOIP conversation.
ii) H323 is not very good at connecting through firewalls. The H323
handshake is done via one IP port, then the actual stream connection is
made on another port, which is agreed upon dynamically via the handshake
process. To make connections through a firewall it is essential to know
which port connections to allow and which to deny, and H323's dynamic
port allocation makes this pretty much impossible. There are a couple of
products that get around this problem, but they're not very mature yet.
This is not an issue with SIP as the handshake and stream connections
are made on fixed pre-defined ports. |
| |
Can you support for D-channel for E1 line? |
| | We support D-channel protocol on E1. There are
2 D-channels on E1, one for framing and the other for protocol
signalling. An example of a D1 protocol is Q931 |
 |
PLEASE SUBMIT YOUR OWN
QUESTIONS |