GL Communicationsï¿½ Digital Central Office Switch Simulator (DCOSS) platform converts a Pentium PC into a digital central office switch simulator, including T1, E1, SS7, and POTS interfaces. It can also test for fax signaling, DTMF/MF tones and line noise/energy. It can also measure voice quality via the PSQM, PAMS, and PESQ+ voice quality standards, which is significant for VoIP testing.
The DCOSS system we had ran on Windows NT 4.0 and utilizes NMS
hardware. Using the GUI, the DCOSS may be initialized for a variety of
protocols such as Robbed-Bit R1 (wink) (along with additional CAS
protocols), MFC-R2, PRI-ISDN (includes T1 ISDN-NFAS), SS5, or SS7 (ISUP or
TUP). Also, analog telephones and/or BRI ISDN (National and International)
telephones may be used for placing or receiving calls. (The DCOSS in a
typical test scenario can be seen in Figure 4.)
Some of the more important functions include the ability to send/record PCM
voice files, send/detect DTMF/MF digits, send/detect single or dual
frequency tones, send/receive faxes, and send/receive modem traffic. The
nice thing about DCOSS is that these functions can be performed
simultaneously. In addition, these functions can be executed either on an
individual timeslot or all the timeslots on all the trunks.
DCOSS features both manual and bulk call generation. Manual call
generation allows the user to place calls on any timeslot/trunk manually
simply by clicking the mouse on the timeslot or trunk. Calls may be made
to/from a standard phone, fax machine, or modem without physical
connections. Once the manual calls are established, the user may talk over
the phones, send/capture PCM voice files, send/ detect DTMF/MF digits,
send/detect frequency tones, send or receive faxes (requires optional DCOSS
Fax licenses), and send/receive modem traffic (V.90).
Bulk call generation allows the user to automatically generate a
multitude of calls (for single, many, or all specified timeslots). The user
may load multiple trunks with a variety of calls such as send/record PCM
voice file, send DTMF/MF Digits, send single/dual frequency tones,
send/receive fax in both ï¿½Transmitï¿½ and ï¿½Receiveï¿½ directions,
send/receive modem traffic. In addition, the DCOSS features a bulk call
scheduler to allow the user to set up a series of actions or events, which
can be executed immediately or at user-defined start time. If the turnkey
tests are not adequate, users can write scripts using the DCOSS Bulk Call
Script Editor for more complex and flexible testing.
Other features include:
- Call Status Screen ï¿½ provides Real Time Status of all calls on a
particular trunk. This screen also displays the current function being
performed on any given timeslot (send file, record file, send tones,
send fax, receive fax, send modem, receive modem).
- Phone Handset status ï¿½ Phone Handset status displays the Real Time
status of the associated analog or BRI-ISDN phones. Information such as
On Hook/Off Hook/Monitor mode as well as the current timeslot and trunk
are displayed. The phone initiating the call will display the phone
- Bulk Call Status ï¿½ The Bulk Call Status screen displays to the user
Start / Stop Bulk Call activity from the Bulk Calling functionality.
- Captured SS7 Events ï¿½ The Captured SS7 Events lists chronologically
all the SS7 messages generated by the trunk and timeslot. Each SS7 event
displays timeslot/trunk, timestamp, SS7 event, OPC, DPC, and CIC.
- Trunk Status ï¿½ All digital trunks are displayed on a single screen
showing alarm status, current protocol and protocol configuration, as
well as additional information concerning the configuration of the
- Timeslot Status ï¿½ A single screen showing all timeslots of all
trunks and the current status of each timeslot.
- DID Routing, switch routing, and dedicated switching.
- Fax Generation/Reception (optional).
The DCOSS comes as a turnkey solution, so no software installation was
required. We connected the various physical connections pretty easily. We
soon discovered that the DCOSS has more features than you can shake a stick
at! Although we did test the traditional T1/E1 simulation and bulk call
generation, we focused most of our testing on checking out the PSQM, PAMS,
and PESQ voice quality measurement benchmarks, which are essential tools to
test VoIP applications.
Voice files may be sent or recorded using the DCOSS PCM Voice File
Generation/Reception Function, which can continuously transmit multiple PCM
voice files or WAV voice files. Both A-Law and m-Law PCM Voice Files or WAV
Voice Files may be transmitted or received. Fortunately, the DCOSS comes
with several reference PCM files to use for testing voice quality across the
system under test, so we didnï¿½t need to record our own reference files.
For our tests we performed a loop back configuration connecting a
crossover T1 cable from trunk 0 to trunk 1. We then configured the bulk call
generator to call out on trunk 0 and receive the calls on trunk 1. After
assigning a specific voice file to play from trunk 0, we specified the file
to receive on trunk 1 as well as selected all 24 channels. A few minor
configuration settings later, we clicked Begin Bulk Calling and the system
started making calls and playing the specified voice file across the T1
channels we selected. In addition, the main GUI features a scrolling test
call history and up to eight analog handset devices, which can be mapped to
a specific channel for listening in or for making manual outbound test
We especially liked the Measurement History, which shows several bits of
important voice quality information, including the trunk number/channel and
PSQM, PAMS, PESQ scores in a tabular format. Each record displayed is also
time/date stamped making diagnosis much easier. Another measurement of call
quality can be performed from the DCOSS Call Quality Test screen. From this
screen the user may define the criteria that determines if a call will yield
a Pass or Fail. If a call fails, a record of that failed call will be sent
to the error screen with an explanation of where the call failed.
Although it isnï¿½t fair to make a single feature comparison of the DCOSS
versus Empirixï¿½ competing Hammer system, we should point out that there
was one important feature that the DCOSS has that the Hammer product does
not. The DCOSS system is able to perform all the various speech quality
tests (PSQM, PAMS, PESQ, etc.) using just a single set of test calls,
whereas the Hammer system requires that you run separate test calls for each
type of speech quality test performed. In fact, this was one of our few ï¿½Room
for Improvementï¿½ items for the Empirix product reviewed in the November
2001 issue (www.tmcnet.com/it/1101/1101labs1.htm).
The way the DCOSS is able to perform multiple speech quality tests all at
once is that it actually stores both the reference voice file (original
file) and the corresponded degraded voice file received on the terminating
side. It then performs the various speech quality tests, comparing the
reference sound file and the terminating degraded sound file. In fact, the
DCOSS has an interesting ï¿½manualï¿½ voice quality test. You can take any
two voice files and perform manual voice quality tests without having to
perform any telephony calls. As an interesting verification test that the
PAMS standard does indeed work, we manually chose the same voice file for
both the reference and degraded voice file and the resultant PAMS score was
a perfect 5. Additionally, DCOSS has an ï¿½auto voice quality monitorï¿½
mode, which will ï¿½watchï¿½ a specified directory for voice files put
there. It will then automatically apply voice quality tests on the files
against the corresponding reference file.
The DCOSS features a useful pass/fail criteria setting. Thus, a user can
enter the desired threshold values for PAMS LE/LQ, PSQM/PSQM+, PAMS
Clipping, PAMS Latency, PESQ MOS, and PESQ Latency measurements. Then the
user could select ï¿½View Failed Measurements Onlyï¿½ in order to view only
Failed Calls and screen out calls that were successful. From this point, the
user can double-click on a failed call and open it in CoolEdit (graphical
oscilloscope-like sound editor) to visually compare the reference and
degraded voice files.
One of the features that we liked was that DCOSS can freeze the real-time
Call Records screen. This is useful for pausing the real-time flood of
statistics when making several test calls. Another useful screen was the
Call Statistics screen, which shows the progress of calls in real-time, as
well as any detected errors. The Call Statistics screen displays statistics
per timeslot, per trunk, per entire system as well as hourly statistics and
may be exported to an ASCII file. One final feature of note is the Remote
Access capability, which allowed us to remotely control the DCOSS using the
supplied DCOSS client. The DCOSS Remote Access capability can operate over
any IP connection, including a LAN, WAN, or the Internet.
ROOM FOR IMPROVEMENT
The DCOSS has an almost endless supply of ï¿½undockableï¿½ windows, which
can be both a good thing and a bad thing. We found it to a bit confusing
having several windows open and then trying to remember which icon at the
bottom of the screen pertained to the window we wanted to switch to. We
found ourselves resorting to several trial-and-error mouse clicks to find
the window we wanted. Thus, we would instead prefer some sort of hierarchal
tree-like structure within a single window to traverse the various
configuration screens. Of course, in some cases, having undockable windows
can be convenient, so we wouldnï¿½t want to lose this capability. One minor
improvement would be to see the system run on Windows 2000; besides, the
Windows 2000 drivers for the NMS hardware have been out for a while now.
The DCOSS is a great platform for simulating and testing telecom networks
and products, including VoIP gateways, softswitches, and PBXs. With its bulk
call generator, E1/T1 plus SS7 support, voice quality measurement tools, and
fairly easy to use interface, the DCOSS is a comprehensive telecom testing
solution perfect for service providers, and carriers, as well as testing and
product development labs.
To The March 2002 Table Of Contents ]