Dialogic DM/V960A-4T1PCIW
(Universal)
Technical Specifications
Dialogic DM/V960A-4T1PCIW
Global Call
The DMV960A-4T1-PCIW cards support Global Call software, a unified
call control programming interface and protocol engine
that makes it easier to provide worldwide application
portability and can shorten development time by using
the same API for almost any network protocol.
Global Call software provides a common signaling
interface for network-enabled applications, regardless
of the signaling protocol needed to connect to the local
telephone network. Global Call is the recommended
API for unified call control for Springware and DM3
architectures. The signaling interface provided by
Global Call facilitates the exchange of call control
messages between the telephone network and virtually
any network-enabled application. Global Call lets
developers create an application that can work with
signaling systems worldwide, regardless of the network
to which they are connected.
Global Call is ideal for high-density, network-enabled
solutions for voice, data, and video, where the
supported hardware and signaling technology can vary
widely. Rather than requiring the application to handle
the low-level details, Global Call software offers a
consistent, high-level interface to the user, handling
each country’s unique protocol requirements in a way
that is transparent to the application.
Functional Description
The DM/V960A-4E1-PCIW combined media
boards are based on the DM3 architecture. The
architecture consists of a set of core specifications and
firmware modules that are implemented on boards with
various processors, including:
■ RISC processor for centralized control
■ DSP(s) for media processing
■ TDM bus interface (H.100/H.110)
■ Four digital telephony network interfaces
■ PCI and CompactPCI bus interfaces
These boards support up to 96 (T-1) channels of voice processing via a bank of DSPs and
up to four T-1 digital trunk interface (DTI)
circuits. The DTI circuits contain signaling services
(ISDN, channel associated signaling [CAS], and CCS),
plus any alarm handling and line maintenance services
required by the installed networks. Each DTI includes
software switchable clock circuits that can be set to:
■ loop mode — transmit clocking is slaved to the
external network
■ independent mode — transmit clocking is derived
from an onboard oscillator
■ expansion or system mode — transmit clocking is
slaved to the TDM; receive clocking is always slaved
to the trunk interface
The control processor is a general-purpose Dialogic i960®
RISC microprocessor, responsible for the initialization,
configuration, and control of the various elements that
make up these specific boards. It controls the TDM bus
interface, as well as the signaling protocols for the DTIs
installed on the platform.
The DMV960A-4T1-PCIW cards support various DSP configurations for
voice processing and call progress analysis
capabilities. These features are provided by a
daughterboard using Motorola* DSPs. The DSPs
process the digitized voice data using downloaded
resource firmware. Each DSP can perform the following
signal analysis and operations:
For incoming data
■ automatic gain control (AGC), which compensates
for variations in the level of the incoming audio signal
■ adaptive differential pulse code modulation
(ADPCM), pulse code modulation (PCM), LinearWAV,
Global System for Mobile Communications (GSM),
G.726, and TrueSpeech* algorithms that compress
digitized voice and save disk storage space
■ tone detection of DTMF, MF, or application-defined
single or dual tones
■ silence detection to determine whether the line is
quiet and the caller is not responding
For outbound data
■ expands stored, compressed audio data for
playback
■ adjusts the volume and pitch of playback upon
application or user request
■ generates tones — DTMF, MF, or any application defined
general-purpose tone
■ performs outbound dialing
■ monitors call progress functions, including
— line busy
— operator intercept
— ring
— no answer
— answered; the DSP detects whether the answering
party is a person, answering machine, a fax
machine, or modem
While recording speech, the DSP can use different digitizing
rates from 8.5 Kb/s to 176 Kb/s, selectable by
the application for the best speech quality and most
efficient storage. The digitizing rate is selected on a
channel-by-channel basis, and can be changed each
time a record or play function is initiated.
DSP-processed speech is transmitted by the control
processor to the host for disk storage. When playing
back a stored file, the processor retrieves voice
information from the host CPU and passes it to the
DSP, which converts the file into digitized voice. The
DSP sends the digitized voice responses to the caller
via the network interface or TDM bus. In addition,
cache prompts now let you store 4 MB to 8 MB of
onboard cache for the storing and playback of voice
files directly on the board, eliminating the need to send
voice files to and from the host/server.
Shared RAM on these boards enables communication
between the host system and the i960 control
processor. A bank of global memory is also provided to
facilitate communications between the control
processor and the various DSPs. In addition to
providing a data pathway between processors, the
global memory can also serve as a repository for data
that is to be shared among processors, or which may
not be storable within local memory associated with
the processor.
Downloadable Firmware
The hardware for the DM/V960A-4E1-PCIW combined media board consists of a baseboard with
a RISC processor and two or four DS-1 digital network
interfaces. (Different assemblies are used for T-1 and
E-1.) An array of DSPs resides on a low-profile
daughterboard. Telephony signaling protocols and
voice processing features are downloaded as firmware
to the board on power up and reside on the various
onboard processors. This downloadable firmware
approach enables easy feature upgrade and
expansion. Individual firmware components, such as a
network interface protocol, or a voice recording
function, are referred to as resources.
Network Interface
The T-1 versions of The DMV960A-4T1-PCIW cards support all T-1
robbed-bit signaling protocols and are fully compatible
with all interface devices that use, or can be set to use,
1.544 MHz clocking and µ-law PCM. The E-1 versions
of these boards support all CEPT CAS protocols and
are fully compatible with interface devices that use, or
can be set to use, 2.048 MHz clocking and A-law PCM
(ITU-T Recommendation G.703/704/711). The boards
also support the clear channel feature, thus providing
up to 124 bearer channels.
The board also support ISDN PRI access for
T-1. The T-1 protocol implementations comply
with the North American standard ISDN PRI and the
INS-1500 standard used in Japan. In North America
and Japan, the ISDN Primary Rate includes 23
voice/data channels (B channels) and one signaling
channel (D channel).
The key ISDN PRI features include
■ Non-Facility Associated Signaling (NFAS) lets a
single D-channel control up to 20 PRI trunks,
providing significant savings in ISDN service
subscription costs available on NI-2, 4ESS, Lucent
5ESS*, DMS100, and DMS250
■ D-channel backup (on NI-2 only) lets another
D-channel takeover should the main D-channel fail
■ Facility, notify, and optional Information Elements
(IEs) let applications work with network-specific
supplementary services
■ Direct Dialing In (DDI), also known as Dialed Number
Identification Service (DNIS), lets an application route
incoming calls by automatically identifying the
number the caller dialed
■ Call-by-call service selection lets an application
select the most efficient bearer channel service, such
as an toll-free line or a WATS line, on a call-by-call
basis
■ User-to-user information lets an application send
proprietary messages to remote systems during call
establishment
■ LAP-D Layer 2 access lets developers build a
customized Layer 3 protocol
■ The ability to dynamically set protocol timers through
a configuration file
■ A maskable Layer 2 Control lets the application
toggle between bringing Layer 2 up and down as
desired
Voice processing features, downloaded to the onboard
DSPs at power up, let these combined media boards
play and record voice messages to and from callers
through the digital network interface. Messages can be
stored using G.711 µ-law or A-law PCM, at a rate of 64
Kb/s, as is used by the public switched telephone
network (PSTN). To reduce storage requirements and
help developers implement unified messaging
applications that meet VPIM standards, voice coding
algorithms can compress recordings as low as 8.5
Kb/s using low-bit rate coders such as TrueSpeech,
GSM, and G.726. Sampling rates and coding methods
are selectable on a channel-by-channel basis.
Applications can dynamically switch sampling rate and
coding method to optimize data storage or voice
quality as needed.
AGC is provided to automatically adjust the signal level
of incoming calls for recording at normal levels,
compensating for adverse line conditions, distance,
and other factors. Playback volume can also be
dynamically adjusted over a 40 dB range using DTMF
input or directly from the application.
DTMF detection is provided to control record and play
functions using DTMF input. Local echo cancellation
techniques are used to improve DTMF cut-through and
talk-off/play-off suppression over a wide variety of
telephone line conditions.
The voice player and recorder resources are linked
with the DTMF detection resources using run-time
control (RTC) messages. This lets play or record
functions be initiated or terminated quickly using DTMF
input from the caller. The RTC function off-loads the
host application from involvement in every interaction,
thereby enabling voice processing applications to
scale to hundreds of ports per system.
Continuous speech processing (CSP) enables
software-based ASR and the ability to speak over
speech prompts. It processes the incoming voice
signal using DSP-based echo canceller (EC) and voice
activity detector (VAD) integrated on the board. The
incoming voice signal is then streamed to the host
system and the ASR engines only when voice energy is
detected. Features such as the pre-speech buffer and
the onboard VAD let the system attain higher accuracy
and efficiency.
The transaction record feature lets voice activity on two
channels be summed and stored in a single file, or in a
combination of files, devices, and/or memory. When it
is necessary to archive a verbal transaction or record a
live conversation, the silence compressed record
feature, when enabled, eliminates silence from
recorded data, thus saving disk storage space. Speed
and volume control are also provided to let the
application or user adjust the speed and volume during
playback. Silence compressed streaming to the host
improves performance by removing the silence when
data is sent to the host CPU. Streaming to the CT Bus
lets echo cancelled data be streamed into the
TDM bus.
Conferencing
The conferencing solution on the DMV960A-4T1-PCIW cards is implemented using onboard DSPs. The
conferencing resource sums incoming voice signals on
the board. Higher quality conferencing is attained
using sophisticated summing algorithms and echo
cancellation (EC) and tone clamping (TC) integrated on
the board. The advanced algorithm distinguishes
between noise and speech dynamically and prevents
noise build up. The incoming voice signal is then
streamed out to the CT Bus where it can be transmitted
to any network interface — either PSTN or IP telephony.
This enables very large size conferences (1000+) such
as analyst calls or broadcast calls where most of the
callers are in listen-only mode. In addition, bridging
(also known as cascade conferencing) lets you bridge
together conferences (up to 60 participants in each
conference) from different DSPs and boards,
consuming just one extra time slot per bridge. This
maximizes the flexibility of your conferencing solution
by letting you create high-density conferences where
any party has the capability to speak and be heard by
other participants.
The conferencing resource supports the active talker
feature that identifies which conferees are actively
talking at any given time and suppresses the
background noise from all the silent parties. It also lets
applications dynamically choose between the summing
mode — active talker vs. pure summation — based on
the conference size. Applications can set this
parameter during configuration time or change
dynamically during runtime. For a small number of
parties, pure summation might be preferred so all
conferees are heard; and as a conference size
increases, the active talker feature might be enabled
so conferees can hear the most active participants.
Tone Signaling
In addition to the DTMF signaling commonly used for
voice processing, the DM/V1200A-4E1 combined media board also contains
a robust set of features used for network tone signaling
and control. The global tone detection (GTD) and
global tone generation (GTG) features provide the
capability to detect and generate user-defined tones
for solving special application situations, such as
integration with PBX or dealing with unique tones.
Perfect Call call progress analysis accurately monitors
outbound calls, detects when calls are answered, and
distinguishes
■ line ringing with no answer
■ line busy
■ problem completing call (such as operator intercept)
■ call answered by a human or answering machine
■ call answered by a fax machine or modem
Perfect Call is Dialogicligently tolerant of the wide variation
in call progress signaling tones found in central offices
and PBXs around the globe and offers accurate
performance right out of the box. Patented DSP-based
algorithms are used to accurately discriminate human
speech from recorded human voice and from
network noise.
Resource Technical Specifications
Audio Signal
Usable receive range –40 dBm0 to 0 dBm0 nominal, configurable by parameter**
Automatic gain control Application can enable/disable output level, configurable by parameter**
Silence detection –40 dBm nominal, software adjustable**
Transmit level (weighted average) –12.5 dBm nominal, configurable by parameter**
Transmit volume control 40 dB adjustment range, with application-definable increments and legal
limit cap
Frequency Response
24 Kb/s 300 Hz to 2600 Hz ±3 dB
32 Kb/s 300 Hz to 3400 Hz ±3 dB
64 Kb/s 300 Hz to 3400 Hz ±3 dB
Audio Digitizing
8.5 Kb/s TrueSpeech
13 Kb/s GSM (TIPHON, MSGSM)
16 Kb/s, 24 Kb/s, 32 Kb/s, and 40 Kb/s G.726
24 Kb/s OKI* ADPCM @ 6 kHz sampling
32 Kb/s OKI ADPCM @ 8 kHz sampling
32 Kb/s IMA ADPCM @ 8 kHz sampling
48 Kb/s G.711 PCM (µ-law for T-1 and A-law for E-1) @ 6 kHz sampling rate
64 Kb/s G.711 PCM (µ-law for T-1 and A-law for E-1) @ 8 kHz sampling rate
64 Kb/s Linear 8 kHz 8-bit WAV
128 Kb/s Linear 8 kHz 16-bit WAV
88 Kb/s Linear 11 kHz 8-bit WAV
176 Kb/s Linear 11 kHz 16-bit WAV
Digitization selection Selectable by application on function call-by-call basis
Pitch controlled
Playback speed control Available on the following 8 kHz coders: OKI ADPCM, G.711 PCM, Linear
Adjustment range: ±50%
Adjustable through application or programmable DTMF control
DTMF Tone Detection
DTMF digits 0 to 9, *, #, A, B, C, D per Bellcore LSSGR Sec. 6
Dynamic range (T-1) –36 dBm to +3 dBm per tone, configurable by parameter**
(E-1) –39 dBm to 0 dBm per tone, configurable by parameter**
Minimum tone duration 32 ms; can be increased with software configuration
Interdigit timing Detects like digits with a >45 ms interdigit delay
Detects different digits with a 0 ms interdigit delay
Acceptable twist and frequency (T-1) Meets Bellcore LSSGR Sec 6. Meets ITU-T Q.23 recommendations**
Noise tolerance Meets Bellcore LSSGR Sec 6 and EIA 464 requirements for Gaussian,
impulse, and power line noise tolerance
Cut-through (T-1) Local echo cancellation permits 100% detection with a >4.5 dB
return loss line.
Performance dependent on far-end handset’s match to local analog loop.
Talk off Detects less than 10 digits while monitoring Bellcore TR-TSY-000763 standard
speech tapes. (LSSGR requirements specify detecting no more than
470 total digits.) Detects 0 digits while monitoring MITEL speech tape
#CM 7291.
Resource Technical Specifications (cont.)
Global Tone Detection
Tone type Programmable for single or dual
Max. number of tones Application-dependent
Frequency range Programmable within 300 Hz to 3500 Hz
Max. frequency deviation Programmable in 5 Hz increments
Frequency resolution ±5 Hz. Separation of dual frequency tones is limited to 62.5 Hz at a
signal-to-noise ratio of 20 dB.
Timing Programmable cadence qualifier, in 10 ms increments
Dynamic range (T-1) Default set at –36 dBm to +3 dBm per tone
Global Tone Generation
Tone type Generate single or dual tones
Frequency range Programmable within 200 Hz to 4000 Hz
Frequency resolution 1 Hz
Duration 10 ms increments
Amplitude (T-1) –43 dBm0 to –3 dBm0 per tone nominal
MF Signaling (T-1) R1
MF digits 0 to 9, KP, ST, ST1, ST2, ST3 per Bellcore LSSGR Sec 6, TR-NWT-000506
and CCITT Q.321
Transmit level Complies with Bellcore LSSGR Sec 6, TR-NWT-000506
Signaling mechanism Complies with Bellcore LSSGR Sec 6, TR-NWT-000506
Dynamic range for detection –25 dBm to +3 dBm per tone
Acceptable twist 6 dB
Acceptable freq. variation Less than ±1 Hz
MF Signaling (E-1) R2
MF digits All 15 forward and backward signal tones per ITU-T Q.441
Transmit level –8 dBm0 per tone, nominal, per ITU-T Q.454; programmable
Signaling mechanism Supports the R2 compelled signaling cycle and non-compelled pulse
requirements per ITU-T Q.457 and Q.442
Dynamic range for detection –35 dBm to –5 dBm per tone
Acceptable twist 7 dB
Acceptable freq. variation Less than ±1 Hz
Call Progress Analysis
Busy tone detection Default setting designed to detect 74 out of 76 unique busy/congestion
tones used in 97 countries as specified by ITU-T Rec. E., Suppl. #2.
Default uses both frequency and cadence detection. Application can
select frequency only for faster detection in specific environments.
Ring back detection Default setting designed to detect 83 out of 87 unique ring back tones
used in 96 countries as specified by ITU-T Rec. E., Suppl. #2. Uses both
frequency and cadence detection.
Positive voice detection accuracy >98% based on tests on a call database
Positive voice detection speed Detects voice in as little as 1/10th of a second
Positive answering machine detection Standard
Fax/modem detection Preprogrammed
Intercept detection Detects entire sequence of the North American tri-tone
Other intercept tone sequences can be programmed
Dial tone detection before dialing Application enable/disable
Supports up to three different user-definable dial tones
Programmable dial tone drop out debouncing (when not part of regulatory
approval)
Tone Dialing
DTMF digits 0 to 9, *, #, A, B, C, D per Bellcore LSSGR Sec 6, TR-NWT-000506,
ITU-T Q.23
Frequency variation Less than ±1 Hz
Rate 10 digits/s, configurable by parameter**
Level (T-1) –4.0 dBm per tone, nominal, configurable by parameter**
Conferencing
Max. parties per conference...
Up to 60
Bridging/cascade conferencing. Lets you bridge together conferences from different DSPs and boards,
consuming just one extra time slot per bridge.
Echo cancellation 16 ms.
Tone clamping Enable/disable at board level.
Summing modes Automatically configures to active talker or pure summation based on
number of parties in a conference. Application can specify minimum number
of parties before active talker mode is used.
Automatic gain control Normalizes the parties’ power levels to a unified target. Key features
include speech/noise discrimination, tolerance to impulsive noise, faster
convergence, and increased steady-state stability.
Tone detection/generation Generates tariff tones and warning tones. Detects DTMF from each party
and can clamp those tones so that other members of the conference do
not hear them.
Active talker notification Can notify the application of which party is talking so the application can
process that information and act accordingly.
Number of active talkers Dynamically selectable.
Modes Duplex, monitor, coach, pupil.
Facsimile
Fax compatibility T.30, T.4, T.6, V.17, V29, V27ter, V.21.
Speed 14.4 Kbps with automatic fallback send and receive concurrently on all
channels.
TIFF Single page
Multipage
Compression MH (ITU T.4, 1D)
MR (ITU T.4 2D)
MMR (ITU T.6)
Onboard, on-the-fly
ECM Supported
ASCII to TIFF On-board,on-the-fly
Page headers Generated on board, on-the-fly
Width A4
Polling Normal and turnaround
Resolution Standard (100 dpi x 200 dpi)
Fine (200 dpi x 200 dpi)
Superfine (200 dpi x 400 dpi)
JPEG/JBIG Color fax and gray scale fax pass-through feature
Protocols
T-1 CAS E&M (wink start, immediate start), loop start, ground start;
feature group A, B, and D
T-1 ISDN NI-2, 4ESS, 5ESS*, DMS100, DMS250, INS1500
