Networking / Beginners

Understanding dedicated call quality issues

If you have call quality issues that exist only on your dedicated circuit, you can track down their sources much faster than you can call quality issues that you notice in your carrier's switched network. The calls running through your carrier's switched network may have 20 or 30 dedicated spans and pieces of hardware associated with them, each one requiring tests and validation. If the quality issue is truly confined to your dedicated circuit, on the other hand, your problem is confined to a finite space.

Quality issues such as static, low volume, clipping (where you hear continuous split-second gaps in your conversation), and possibly even echo, leave a trail of clues and evidence for you to follow as you troubleshoot. All these issues can be the result of poorly provisioned hardware; maybe one section of the circuit is set up for loopstart signaling and the rest of the circuit is set up for E&M Wink signaling. A piece of hardware wearing out and failing could also cause quality issues. Confirming what's what with defective hardware is not difficult because all network switches in North America have computer files that can be activated to collect performance data on the circuit.

Files saved by computer software to monitor call performance are unoriginally called performance monitors (PMs), and they collect information on the quantity and type of errors on a dedicated circuit. Your phone system may have a similar feature, allowing you to collect the errors experienced by your hardware. If your hardware isn't as advanced, you can request the information from your carrier by calling to open a trouble ticket. PM software categorizes errors in several categories. If there was no error on a circuit in a particular category, the "0" error count is listed. The quantity of each type of error guides you to other areas of inquiry when you see them. Here are some common PM errors:

  • Erred seconds: You find erred seconds on a circuit with a minor problem. An erred second indicates that for the duration of one second, the communication of the circuit was distorted. The overhead information may have been lost for that specific second because some piece of hardware experienced a low-level, intermittent electrical short. You might experience minor static or quality issues that are more of an annoyance than a concern.
  • Severely erred seconds: You can find severely erred seconds on a circuit with large issues. The severely erred seconds indicate a larger issue whereby information was not only missing, but replaced with aberrant data. These errors often cause static or line noise so severe that you can't hear the person you are calling. If your circuit has severely erred seconds but you aren't experiencing any regular quality issues, you might instead notice that your calls disconnect prematurely or that your entire circuit drops unexpectedly.
  • Framing slip: Framing slips signal a configuration problem, generally a timing issue where your hardware is attempting to correct a lag behind the master clock of your carrier. If you aren't clocking off your carrier or if your hardware is set up for an outpulse signal and start that doesn't match your carrier, you can expect frame slips.
    Remember The bad news about frame slips is that they may be small, minor, and go unnoticed by you as they accumulate, until the point where they drop your circuit.
  • Unavailable seconds: Unavailable seconds are just that, seconds of time when the network thinks your circuit is unavailable. Unplugging your hardware generates unavailable seconds in your carrier's network as easily as a fiber cut or an accidental removal of your cross-connect by your local loop provider.

TipOne of the simplest things you can do to see whether a problem on your dedicated circuit can resolve itself is to reboot the whole system. Sometimes all you need is PFM (pure friggin' magic) to get everything to right itself. If you decide to reboot your system, make sure that your carrier pulls the PMs before you reboot your hardware; this way, you can prevent confusion. Even if you have a clean circuit, as soon as you power down your channel bank and reboot it, you create a handful of errors on the circuit. Even if you have a legitimate issue on your circuit, rebooting has the potential to mask the true issue and enlarge the number of errors your carrier finds.

Remember Without controlled testing, your carrier won't know which piece of hardware is generating the errors, or when the errors were created. Errors that exist in the PM files don't have time stamps next to them, only total error counts for a given period of time. If the PMs show your circuit received 176 erred seconds and 56 severely erred seconds in the past 24 hours, you have no way of identifying whether all the errors occurred in one group as your circuit took a hit, or if they were generated periodically over the past 24 hours. The only way to validate the frequency of errors is to purge the PM files and check it every 15 minutes, in order to determine the rate at which the errors are generated.

[Previous] [Contents] [Next]