2.2.  Learn Mode Requirements and Limitations

The following requirements apply to learn mode:

• The tone to be learned must be either a continuous tone or a repeating tone.
• To learn a continuous tone, the tone must occur continuously for a minimum duration of one second without any other tone being present. (If another tone is present, you may be able to exclude it from being learned by restricting the lm_LearnTone( ) learning parameters specified in tn_rangep.) The duration of the tone is configurable in the LM_PARM structure.
• To learn a tone having a cadence, several repetitions of the tone must occur without any other tone being present. The cadence must contain at least five tone-on/tone-off transitions, although 10 transitions will characterize the tone with more accuracy. (The default is 10.)

The following limitations apply to learn mode:

• Learn mode requires that the channel be completely dedicated to its operations while learn mode is active. Simultaneous use of any other signal or tone detection on the channel is not compatible, including DTMF, MF, R2 MF, Socotel, user-defined tone detection, PerfectCall call progress analysis, and basic call progress analysis. However, you can perform these operations on another channel and route the learn mode channel's receive time slot to the other channel's time slot.
• Since learn mode can characterize either a tone that has several cadence repetitions or a tone that is continuous, it is invalid to use learn mode to characterize the following:
  • A sequence of different tones, such as a SIT tri-tone sequence or a PBX warble tone (see call progress analysis to detect SIT tones).
  • Short duration, single instance tones that do not repeat, such as a "bong" tone (unless the duration is long enough to qualify as a continuous tone).

 In these cases, to characterize a tone that occurs in a single instance or is one in a sequence, you must simulate the repetitions by recording the tone and then playing the recorded tone repeatedly. (Of course, the simulation will be more robust if you concatenate multiple recordings of the tone).

• Learn mode can accurately characterize almost all tones having a complex cadence. In the laboratory, it is possible to create a situation where the initial definition of a complex cadence overlaps with another similar tone description. In this unusual case, adjusting the tone description can correct the problem. Note that such a conflict is extremely rare. [One unlikely scenario of such a conflict would be the following (this is hypothetical and has never been encountered): If your environment includes 2 PBXs that each come from a different manufacturer, the PBXs produce a signal for different purposes, the signal has the same frequency or frequencies, the signal from one PBX has a complex cadence, and the signal from the second PBX has a simple or complex cadence that matches in part the cadence of the signal from the first PBX, a tone description conflict may occur when characterizing the tones with learn mode.] [Jigar removed previous s3.1.5 from new doc: how to use GTD learn mode; this is outdated info]

Click here to contact Telecom Support Resources