Adaptive system for measuring the broadband signal-to-noise ratio of a transmission channel

• US 5,166,954 A
• Filed: 03/05/1991
• Issued: 11/24/1992
• Est. Priority Date: 03/05/1991
• Status: Expired due to Term

• Current Assignee: American Telephone & Telegraph Company (AT&T, Inc.)
• Original Assignee: AT&T, Inc.
• Inventors: Levonas, Robin, Grizmala, Fredrick
• Primary Examiner(s): Kuntz, Curtis
• Assistant Examiner(s): TSE, YOUNG TOI
• Application Number:
  US07/664,908
• Time in Patent Office: 630 Days
• Field of Search: 375/10, 375/14, 375/27, 375/103, 370/13, 370/17, 328/162, 379/6, 455/226, 364/572, 364/481, 371/22
• US Class Current: 375/227
• CPC Class Codes: H04B 14/068 with adaptive feedback

Abstract

A facility is provided for measuring the signal-to-noise ratio of a transmission channel using the results obtained by transmitting over the channel a pseudorandom noise signal, in which the value of the signal-to-noise ratio may be used to determine whether the channel contains various types of pulse code modulation equipment.

Claims

1. An arrangement for measuring signal-to-noise ratio of a communications channel comprising

means for generating a pseudorandom noise signal constrained to a predetermined bandwidth and transmitting said constrained signal to said channel,

means for sampling at a predetermined rate said constrained signal after it has propagated through said channel,

means for generating a filtered response, y_i, with respect to each said sample of said constrained signal to determine the level of energy in each such sample and for calculating an error value, e_i, between each such filtered response, y_i, and an associated receive sample of said constrained signal, and

means for calculating the signal-to-noise ratio of said channel as a function of each said filtered response, y_i, and each said error signal, e_i.

2. The arrangement set forth in claim 1 wherein said means for generating said constrained signal includes means for transmitting a signal having a predetermined frequency and phase to said channel, in which said transmission occurs for a predetermined duration commencing prior to the generation of said constrained signal, and in which the phase of said predetermined frequency signal is reversed after a predetermined segment of said duration has elapsed.

3. The arrangement set forth in claim 2 wherein said means for sampling includes means for detecting said phase reversal to determine the end of said duration and thus synchronize said sampling means with the transmission of said constrained pseudorandom signal over said channel.

4. The arrangement set forth in claim 2 wherein said means for sampling includes

first and second filtering means,

means for sampling said predetermined frequency signal and for passing said sample to said first and second filtering means to generate a first sample and a second weighted sample of said sample of said predetermined frequency signal, said first filtering means filtering out of said sample frequencies other than said predetermined frequency and said second filtering means filtering out of said sample said predetermined frequency and then generating a weighted version of the filtered result,

means for calculating a first power summation involving the value of each said first filtered sample and a second power summation involving the value of each said second weighted sample, and

means for generating a signal-to-noise ratio of said channel with respect to said predetermined frequency signal as a function of said first and second power functions.

5. The arrangement set forth in claim 1 wherein said means for sampling includes means for storing in a respective one of a first table of sequential locations of a memory said sample of said constrained signal as said sample is generated, and wherein said arrangement further comprises

means, responsive to the generation of each said sample of said constrained signal, for generating a reference sample of said constrained signal, such that said reference sample is correlated in time with an associated sample of the propagated constrained signal and for storing in a respective one of a second table of sequential locations of said memory said reference sample as it is generated.

6. The arrangement set forth in claim 5 wherein said means for generating said filtered response is an adaptive finite impulse response filter having a predetermined number of tap coefficients.

7. The arrangement set forth in claim 6 wherein said channel has an impulse response and wherein said arrangement further comprises

means for changing the values of said coefficients as a function of the value of each said error, e_i, so that said adaptive finite impulse response filter may converge on the impulse response of said channel.

8. The arrangement set forth in claim 7 wherein said means for generating said filtered response further comprises

means, operative when said adaptive finite impulse response filter has filtered a predetermined number of said samples of said constrained signal, for generating an offset value as a function of the position within said adaptive finite impulse response filter of that one of said tap coefficients having the largest magnitude,

means for refiltering the samples stored in said second table and, responsive to each succeeding filtered response, y_i, recalculating said error value, e_i, using a respective one of said reference samples stored in said second table, in which the first of said receive samples that is used is stored at a location of said first table that is offset by the value of said offset value,

means, operative just prior to said refiltering, for initializing the values of said tap coefficients to predetermined values and, responsive to each new value of said error, e_i, for changing the values of said tap coefficients so that said adaptive finite impulse response filter may reconverge on said impulse response, and

means, responsive to the refiltering of said predetermined number of samples contained in said second table and using the last values of said tap coefficients that were calculated as a result of such refiltering, for causing said adaptive finite impulse response filter to refilter those samples using just said last values of said tap coefficients, in which said signal-to-noise ratio of said channel with respect to said constrained signal is then calculated as a function of each filtered response, y_i, generated as a result of using the last values of said tap coefficients.

9. The arrangement set forth in claim 8 wherein said means for generating said filtered response further comprises means for determining channel shaping imposed by said channel on said constrained signal, in which said channel shaping is determined as a function of the last values of said tap coefficients.

10. The arrangement set forth in claim 8 wherein said means for generating said filtered response further comprises means for determining group delay imposed by said channel on said constrained signal.

11. A method for measuring signal-to-noise ratio of a communications channel comprising the steps of

generating a pseudorandom noise signal constrained to a predetermined bandwidth and transmitting said constrained signal to said channel,

sampling at a predetermined rate said constrained signal after it has propagated through said channel,

generating a filtered response, y_i, with respect to each said sample of said constrained signal to determine the level of energy in each such sample and for calculating an error value, e_i, between each such filtered response, y_i, and an associated receive sample of said constrained signal, and

calculating the signal-to-noise ratio of said channel as a function of each said filtered response, y_i, and each said error signal, e_i.

12. The method set forth in claim 11 wherein said step of generating said constrained signal includes the step of transmitting a signal having a predetermined frequency and phase to said channel, in which said transmission occurs for a predetermined duration commencing prior to the generation of said constrained signal, and in which the phase of said predetermined frequency signal is reversed after a predetermined segment of said duration has elapsed.

13. The method set forth in claim 12 wherein said step of sampling includes the step of detecting said phase reversal to determine the end of said duration and thus synchronize said step of sampling with the step of transmitting said constrained pseudorandom signal over said channel.

14. The method set forth in claim 12 wherein said step of sampling includes the step of

providing first and second filters,

sampling said predetermined frequency signal and passing said sample to said first and second filters to generate a first sample and a second weighted sample of said sample of said predetermined frequency signal, in which said first filter is used to filter out of said sample frequencies other than said predetermined frequency and in which said second filter is used to filter out of said sample said predetermined frequency and then generating a weighted version of the filtered result,

calculating a first power summation involving the value of each said first filtered sample and a second power summation involving the value of each said second weighted sample, and

generating a signal-to-noise ratio of said channel with respect to said predetermined frequency signal as a function of said first and second power functions.

15. The method set forth in claim 11 wherein said step of sampling includes the step of storing in a respective one of a first table of sequential locations of a memory said sample of said constrained signal as said sample is generated, and wherein said method further comprises the step of

responding to the generation of each said sample of said constrained signal by generating a reference sample of said constrained signal that is correlated in time with an associated sample of the propagated constrained signal and for storing in a respective one of a second table of sequential locations of said memory said reference sample as it is generated.

16. The method set forth in claim 15 wherein said step of generating said filtered response is an adaptive finite impulse response filter having a predetermined number of tap coefficients.

17. The method set forth in claim 16 wherein said channel has an impulse response and wherein said method further comprises the step of

changing the values of said coefficients as a function of the value of each said error, e_i, so that said adaptive finite impulse response filter may converge on the impulse response of said channel.

18. The method set forth in claim 17 wherein said step of generating said filtered response further comprises the steps of

responding to said adaptive finite impulse response filter having filtered a predetermined number of said samples of said constrained signal by generating an offset value as a function of the position within said adaptive finite impulse response filter of that one of said tap coefficients having the largest magnitude,

refiltering the samples stored in said second table and responding to each succeeding filtered response, y_i, by recalculating said error value, e_i, using a respective one of said receive samples stored in said first table, in which the first of said receive samples that is used is stored at a location of said first table that is offset by the value of said offset value,

prior to said refiltering, initializing the values of said tap coefficients to predetermined values and, responding to each new value of said error, e_i, by changing the values of said tap coefficients so that said adaptive finite impulse response filter may reconverge on said impulse response, and

responding to the refiltering of said predetermined number of samples contained in said second table by causing said adaptive finite impulse response filter using just the last values of said tap coefficients that were calculated as a result of such refiltering, to refilter those samples, in which said signal-to-noise ratio of said channel with respect to said constrained signal may then be calculated as a function of each filtered response, y_i, generated as a result of using the last values of said tap coefficients.

19. The method set forth in claim 18 wherein said step of generating said filtered response further comprises the step of determining channel shaping imposed by said channel on said constrained signal, in which said channel shaping is determined as a function of said last values of said tap coefficients.

20. The method set forth in claim 18 wherein said step of generating said filtered response further comprises the step of determining group delay imposed by said channel on said constrained signal.