Method and apparatus for testing telephone sound quality

Description

RELATED APPLICATIONS

This application incorporates and claims the benefit of U.S. Provisional Application 60/369,540 filed Apr. 1, 2002.

FIELD OF THE INVENTION

This invention relates generally to telephones and more particularly to a method and apparatus for testing telephone sets to determine their sound quality.

BACKGROUND OF THE INVENTION

Telephone sets, i.e., base unit and handset, whether corded, cordless, or speakerphone, typically employ Automatic Gain Control (AGC) or other nonlinear circuits to suppress telephone line noise, static, and/or acoustic feedback. Such circuits are intended to eliminate or reduce noise that would otherwise be heard by a user at the handset speaker during a telephone communication. However, in some designs, such circuits can operate excessively with the result that the user is likely to hear a “choppy” sound quality. That is, the user may find that a far end talker with a quiet voice is difficult or impossible to understand because too many of the talker's syllables are suppressed or cutoff, resulting in the “choppy” sound.

Additionally potential sources of poor telephone sound quality include:

• • Signal overload with resulting distortion
  • Signal underload with resulting noise
  • Insufficient digital quantization resolution and/or range
  • Low digital sampling rate
  • Excessive low pass and/or dynamic filtering
  • Excessive cordless telephone RF carrier dropout

Currently used standard acoustic tests frequently produce results that do not adequately correlate to the human perception of a telephone's sound quality. For instance, the measured audio frequency response of a telephone unit A can fail to meet specified limits and yet be ranked much higher by a typical user than a telephone unit B whose measured audio frequency response complies with the specified limits.

The present invention is directed to an improved method and apparatus intended to determine the subjective sound quality of a telephone unit.

SUMMARY OF THE INVENTION

More particularly, the present invention is directed to a system which repeatedly applies a sequence of pulse tones of different frequency and amplitude to a telephone set under test (i.e., “test telephone”) to determine its sound quality.

In accordance with a preferred embodiment, each pulse tone sequence (PTS) is comprised of three tones (sine waves) which are electronically injected into the test telephone to simulate a talker's voice. More particularly, the three tones are preferably selected to have frequencies in the middle of the human voice range (e.g., 800 Hz, 1000 Hz, 1200 Hz) that most users can readily hear and distinguish.

Additionally, in accordance with the preferred embodiment, the multiple tones in each sequence have amplitudes selected to simulate a moderately quiet talker that most listeners can readily hear in a normal room environment.; In the preferred three tone sequence, the first and last tones are preferably selected to have amplitudes noticeably lower than the middle tone to simulate the lower amplitude syllables in human speech that are often lost, i.e., chopped off, in telephone communications.

BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 is a circuit diagram of an exemplary arrangement for testing a telephone set in accordance with the present invention;

FIG. 2A depicts an oscilloscope display produced by a reference corded telephone excited by a pulse tone sequence in accordance with the present invention and FIG. 2B comprises a table describing test results associated with the display of FIG. 2A.

FIG. 3A depicts an oscilloscope display produced by an exemplary cordless test telephone excited by the same pulse tone sequence as used in FIG. 2A and FIG. 3B comprises a table describing test results associated with the display of FIG. 3A;

DETAILED DESCRIPTION

Attention is initially directed to FIG. 1 which depicts a preferred arrangement for testing a telephone set 10 in accordance with the present invention. The telephone set 10 is comprised of a base unit 12 which communicates with a cordless handset 14 via an RF link 16.

In order to perform the test to be described hereinafter, a signal generator 20 and battery feed circuit 22 are connected to electrically inject a pulse tone sequence into base unit 12 via a simulated non-loaded cable 23 for transmission to handset 14. The handset 14 typically includes a transmit, or microphone, portion 26 and a receive, or speaker, portion 28. As illustrated, the transmit portion 26 is preferably sealed with a sound attenuating cover 30.

The receive portion 28 is acoustically coupled to an artificial ear 32 which responds to applied acoustic energy to drive amplifier 34. The output of amplifier 34 is applied to an oscilloscope 36 for displaying the characteristics of the acoustic energy produced by the handset receive portion 28.

In order to test the sound quality of the telephone set 10 in accordance with the invention, a sequence of multiple pulse tones produced by signal generator 20 is repeatedly applied to base unit 12. The pulse tone sequence (PTS) is selected to simulate a moderately quiet talker in a normal environment. In accordance with a preferred embodiment of the invention, the selected pulse tone sequence is comprised of three distinct successively generated sine waves of different frequencies within the human voice range. The following specification describes a preferred embodiment of the present invention.

Preferred Test Requirements

The telephone set shall be excited by a selected three tone sequence to determine whether it clearly reproduces all three tones of the sequence. The telephone set shall be considered as passing the test if either the Objective or the Subjective PTS Test requirement is met:

• • (a) Objective PTS Test—The first and third tones, at 10% of the amplitude of the middle tone, shall be evident at the oscilloscope 36. Each of the three tones shall be 200 milliseconds wide.
  • (b) Subjective PTS Test—All three tones shall be clearly heard without audible suppression and without audible rounding off of any of the tones.
    Standard Method of Measurement
  • (a) The PTS test shall be performed using the test circuitry shown in FIG. 1.
  • (b) The signal generator 20 shall produce the pulse tone sequence (i.e., “test signal”) repeatedly. Each cycle of the test signal shall include a triple-tone sequence, followed by 400 milliseconds of silence, for a total cycles duration of 1000 milliseconds (1 second). The sine wave frequencies, amplitudes, and durations shall be generated in the sequence listed in the following Table 1.

TABLE 1
PTS Test Signal Frequencies, Amplitudes, and Durations

Sequence	1	2	3
Frequency	800 Hz	1000 Hz	1200 Hz
Amplitude	5 mV peak	50 mV peak	5 mV peak
Duration	200 milliseconds	200 milliseconds	200 milliseconds

• • (c) The measurement shall be made for 0 km of 26 AWG non-loaded cable as shown in FIG. 1.
  • (d) If the telephone handset has a Receive volume control, the control shall be set to the nominal gain level.
  • (e) The handset should be in an ambient noise environment no greater than 40 dB(A) Sound Pressure Level referenced to 20 micropascals, and the handset transmitter should be covered, as by cover 30.
  • (f) To evaluate the Objective Receive Sound Quality, an oscilloscope 36 or equivalent instrument is connected to the output of amplifier 34 and the instrument time base is adjusted to display two complete cycles of the PTS test signal. The display is adjusted so that the amplitude of the middle (1000 Hz) tone is near full scale. The relative amplitudes of the first (800 Hz) and third (1200 Hz) tones are observed and compared to the middle tone, as well as their pulse widths.
  • As an example, the display of a reference corded telephone which passes this test is shown in FIG. 2A. Note that the first and third tones are evident at 10% of the amplitude of the middle tone, and each tone is about 200 milliseconds wide. The display of a cordless telephone which fails this test is shown in FIG. 3A; the first and third tones are not evident, and the second tone is noticeably less than 200 milliseconds wide.
  • (g) To evaluate the Subjective Receive Sound Quality, a technician shall listen to the handset receiver for all three tones of the PTS test signal. A reference corded telephone should preferably be used for comparison purposes; e.g., the reference corded telephone of FIG. 2A which produces the subjective results summarized in the Table of FIG. 2B where all three tones are clearly heard by the testing technician. The performance of an exemplary cordless telephone which fails this subjective test is summarized in the Table of FIG. 3B where the first and third tones are not clearly heard, sounding suppressed or rounded off.
    Expanded Measurements

To help determine the level of “choppy” sound in a telephone, the PTS test may be performed at amplitudes above and below the standard (5, 50, 5 mV peak) level; the following Table 2 lists the recommended PTS levels. All other parameters (e.g., sequence, frequency, duration, and silence) and procedures are identical to those previously discussed.

TABLE 2
Expanded PTS Test Amplitudes

	PTS Level	800 Hz	1000 Hz	1200 HZ
	Very High	20 mV peak	200 mV peak	20 mV peak
	High	10 mV peak	100 mV peak	10 mV peak
	Standard	5 mV peak	50 mV peak	5 mV peak
	Low	2 mV peak	20 mV peak	2 mV peak
	Very Low	1 mV peak	10 mV peak	1 mV peak

If a telephone set fails to reproduce all three tones of the PTS test signal at the Very High Level, the unit has a very high level of “choppy” sound and thus has very poor Receive Sound Quality. If a telephone reproduces all three tones of the PTS test signals at he Very Low Level, the unit has a very low level of “choppy” sound and thus has very good Receive Sound Quality.

As an example, a reference corded telephone which passes all of the expanded measurements is summarized in the following Table 3;

TABLE 3
Expanded Objective/Subjective PTS Test-
Reference Corded Telephone

				Results
PTS Level	800 Hz	1000 Hz	1200 Hz	Pass/Fail
Very High	Yes	Yes	Yes	Pass
(20, 200, 20 mV peak)
High	Yes	Yes	Yes	Pass
(10, 100, 10 mV peak)
Standard	Yes	Yes	Yes	Pass
(5, 50, 5 mV peak)
Low	Yes	Yes	Yes	Pass
(2, 20, 2 mV peak)
Very Low	Yes	Yes	Yes	Pass
(1, 10, 1 mV peak)
Key:
Yes = tone clearly visible or audible
No = tone suppressed or rounded off

Note in Table 3 that all three tones are clearly reproduced at all PTS levels. A cordless telephone which fails most of the expanded measurements is summarized in Table 4.

TABLE 4
Expanded Objective/Subjective PTS Test-Cordless Telephone

				Results
PTS Level	800 Hz	1000 Hz	1200 Hz	Pass/Fail
Very High	Yes	Yes	Yes	Pass
(20, 200, 20 mV peak)
High	No	Yes	Yes	Fail
(10, 100, 10 mV peak)
Standard	No	Yes1	No	Fail
(5, 50, 5 mV peak)
Low	No	Yes1	No	Fail
(2, 20, 2 mV peak)
Very Low	No	No	No	Fail
(1, 10, 1 mV peak)
Key:
Yes = tone clearly visible or audible
No = tone suppressed or rounded off
Note:
1Objectively, the tone was rounded off (“No”), but Subjectively, the tone was clearly audible (“Yes”) for an overall “Yes”.

Note that at the High (10, 100, 10 mV peak) level, the first (800 Hz) tone is not clearly reproduced, so the result is a “Fail” for the High PTS level.

Additionally, the PTS test measurement may be repeated with longer loops (e.g., 2.7 and 4.6 km of 26 AWG non-loaded cable) to determine the level of “choppy” sound with a range of loop conditions.

From the foregoing, it should now be appreciated that a method and apparatus have been described for testing telephone sets to determine their sound quality. Although only a limited number of specific embodiments have been described, it is recognized that various modification and equivalents will occur to those skilled in the art falling within the spirit and intended scope of the invention.