Evaluation of signal-processor performance

• US 5,841,667 A
• Filed: 08/26/1996
• Issued: 11/24/1998
• Est. Priority Date: 02/25/1994
• Status: Expired due to Fees

• Current Assignee: Tele-Ip Limited
• Original Assignee: Martin Communications Pty., Ltd.
• Inventors: Martin, Andrew Louis
• Primary Examiner(s): Voeltz, Emanuel Todd
• Assistant Examiner(s): Smithers, Matthew
• Application Number:
  US08/696,833
• Time in Patent Office: 820 Days
• Field of Search: 364/554, 364/582, 364/551.01, 395/183.14
• US Class Current: 702/182
• CPC Class Codes: G06F 11/321 Display for diagnostics, e.g. diagnostic result display, self-test user interface
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3447 Performance evaluation by modeling
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/88 Monitoring involving counting
• CPC Class Codes: H04L 1/24 Testing correct operation

Abstract

The present invention provides a method and apparatus for evaluating the performance of a signal-processing device during a test period when an input signal is applied to the signal-processing device and an output signal is generated by it. A select profile of the input (or output) signal quality is derived at intervals in which the output (or input) signal lies within some predetermined range. The derived profile may be normalized with respect to the quality of the input (or output) signal throughout the test period. The profile is indicative of device performance of the device being tested. The method may be implemented in either hardware or software.

Claims

1. A method of evaluating the performance of a signal-processing device during a test period when an input signal is applied to the device and an output signal is generated thereby, the method comprising the steps of deriving for one of said input signal and said output signal a select profile of signal quality variation by increment of signal quality for time intervals in which the signal quality of the other of said input signal and said output signal falls within a predetermined range of values, and graphically representing said select profile as an indicator of device performance.

2. A method according to claim 1 including the step of normalising said select profile, increment by increment of overall quality of said one of said input signal and said output signal without regard to said predetermined range, to generate a normalised select profile as an indicator of device performance.

3. A method according to claim 2 wherein the select profile is a probability profile representing the likelihood that a measurement of the signal quality of said one of said input signal and said output signal will fall within a given increment of signal quality when the signal quality of said other of said input signal and said output signal falls within said predetermined range.

4. A method according to claim 1 wherein the select profile is a probability profile representing the likelihood that a measurement of the signal quality of said one of said input signal and said output signal will fall within a given increment of signal quality when the signal quality of said other of said input signal and said output signal falls within said predetermined range.

5. A method according to claim 1 for evaluating the relative performance of the device with respect to a similar reference device, including the steps of deriving the select profiles for each device, differencing or otherwise contrasting said profiles, and graphically representing the difference there between.

6. A method of evaluating the performance of a signal processing device during a test period, comprising the steps of:

applying an input signal to the device and generating an output signal therefrom,

sampling a quality of said input and said output signal at regular intervals throughout the test period,

generating a first overall profile of signal quality of one of said input signal and said output signal by accumulating counts of signal quality for each of a plurality of signal quality increments,

generating a first select profile of the signal quality of said one of said input signal and said output signal by accumulating counts of signal quality for each of said plurality of signal quality increments for intervals in which the signal quality of the other of said input signal and said output signal falls within a predetermined range, and

using said first select profile as an indicator of device performance.

7. A method according to claim 6 including the steps of:

generating an overall probability-profile from said first overall profile representing the likelihood that a measurement of the signal quality of said one of said input signal and said output signal will fall within a given increment of signal quality,

generating a select probability-profile from said first select profile representing the likelihood that a measurement of the signal quality of said one of said input signal and said output signal will fall within a given increment of signal quality for intervals in which the signal quality of said other of said input signal and said output signal falls within a predetermined range, and

using said select probability-profile as an indicator of device performance.

8. A method according to claim 7 including the steps of:

generating a normalised select profile by adjusting said first select profile or said select probability-profile, increment by increment, according to the respective counts or probabilities of said first overall profile or said probability-profile, respectively, and

using said normalised select profile as an indicator of device performance, or as the basis for an indicator of device performance.

9. A method according to claim 8 wherein the adjustment of the first select profile is effected by division of counts, and wherein adjustment of the select probability-profile is effected by division of probabilities, increment by increment of signal quality.

10. A method according to claim 8 including the steps of:

generating a test normalised select probability-profile for a test device,

generating a reference normalised select probability-profile for a similar reference device, and

comparing or contrasting said test and reference profiles to highlight the difference there between.

11. A method according to claim 6 comprising the steps of:

generating a family of test profiles for a plurality of said ranges of signal quality,

deriving a gradient-profile representing the gradient between the members of said family, and

using said gradient-profile as an indicator of device performance.

12. A method according to claim 11 including the steps of:

generating said family of profiles for a plurality of said ranges of signal quality for the test device,

generating said family of profiles for said plurality of said ranges of signal quality for a reference device, and

comparing or contrasting said test and reference families of profiles to highlight the difference there between.

13. Circuit means for use in evaluating the performance of a signal processing device which is adapted to receive an incoming signal and to generate an outgoing signal derived therefrom, the circuit means comprising:

a first input for receiving one of said device signals as a first input signal,

a second input for receiving the other one of said device signals as a second input signal,

first gate means connected to said first input for sampling the quality of the first input signal at regular time intervals during an evaluation period,

first accumulator means connected to said first gate means and adapted to accumulate sample counts according to increment of signal quality,

second gate means connected to said second input for sampling the quality of the second input signal at said time intervals,

and discriminator means connected to said second input and adapted to determine, for each sample of second input signal quality, whether or not the signal quality lies within a predetermined range, said discriminator being connected to prevent said first accumulator means from storing sample counts during time intervals when the quality of said second input signal lies outside said range,

the profile of counts in said accumulator at the end of the evaluation period providing an indicator of device performance.

14. Circuit means according to claim 13 including:

a second accumulator connected to receive all sample counts of first input signal quality according to increment of signal quality, or alternatively connected to receive sample counts only during intervals when said second input signal quality lies outside said range, and

normalisation circuit means adapted to adjust said profile of counts in accordance with the profile of total sample counts of first input signal quality.

15. Circuit means according to claim 13 including means for computing a probability profile from said count profile.

16. Circuit means according to claim 13, including a microprocessor with connected memory and visual output device adapted to store said profile in said memory together with a similar reference profile and to display the normalised and reference profiles on said visual output device in a contrasting manner.