Methods and systems for detecting metallic faults affecting communications links
View Patent ↗One exemplary aspect is to detect the presence of one or more types of faults affecting DSL communications. Upon their detection, a message can be generated that recommends corrective action. Furthermore, another exemplary aspect is to estimate the rate impact of the detected fault in some cases. One exemplary aspect of certain embodiments is to detect the presence of one or more faults that affect DSL communications, and upon their detection, generating, for example, a message that recommends corrective action. In this context, a fault can be one or more metallic faults, malfunctioning modems, or something else that affects the measured attenuation but is not normally attributable to normal DSL transmission lines.
1. A system comprising:
circuitry configured to detect a fault in a communication line, the circuitry configured to compare a measured attenuation curve with a best-fit attenuation curve, wherein the measured attenuation curve is reported by a modem, and the best-fit attenuation curve is determined iteratively by comparing to the measured attenuation curve is pre-stored models of attenuation curves that could occur for loops that do not contain metallic faults; and
a message generation module adapted to generate a confidence metric which quantifies an extent to which it is known that the fault is present on the communication line,
wherein after the fault is corrected the circuitry is configured to compare another measured attenuation curve to another best-fit attenuation curve to detect a presence of another fault.
2. A method of detecting the presence of a fault in a communication line comprising:
comparing with circuitry a measured attenuation curve with a best-fit attenuation curve, wherein the measured attenuation curve is reported by a modem, and the best-fit attenuation curve is determined iteratively by comparing to the measured attenuation curve pre-stored models of attenuation curves that could occur for loops that do not contain metallic faults; and
generating a confidence metric which quantifies an extent to which it is known that the fault is presented on the communication line,
wherein after the fault is corrected the circuitry is configured to compare another measured attenuation curve to another best-fit attenuation curve to detect a presence of another fault.
3. A non-transitory computer-readable information storage media having stored thereon instructions that when executed by a processor causes a comparison between a measured attenuation curve with a best-fit attenuation curve, wherein the measured attenuation curve is reported by a modem, and the best-fit attenuation curve is determined iteratively by comparing to the measured attenuation curve pre-stored models of possible attenuation curves that could occur for loops that do not contain metallic faults; and
generates a confidence metric which quantifies an extent to which it is known that the fault is present on a communication line,
wherein after the fault is corrected the circuitry is configured to compare another measured attenuation curve to another best-fit attenuation curve to detect a presence of another fault.
4. A system for detecting the presence of a fault comprising:
means for comparing a measured attenuation curve with a best-fit attenuation curve, wherein the measured attenuation curve is reported by a modem, and the best-fit attenuation curve is determined by iterative means that compare to the measured attenuation curve pre-stored models of attenuation curves that could occur for loops that do not contain metallic faults; and
means for generating a confidence metric which quantities an extent to Which it is known that the fault is present on a communication line, wherein after the fault is corrected the means for comparing also compares another measured attenuation curve to another best-fit attenuation curve to detect a presence of another fault.
5. The system of claim 1 , wherein a user is notified of the detection of a metallic fault or malfunctioning modem.
6. The system of claim 1 , wherein additional links are performed to identify additional metallic faults or malfunctioning modems.
7. The system of claim 1 , wherein the detection of a metallic fault or malfunctioning modem is reported to a management entity via a management interface.
8. A method for determining one or more faults in a communications channel comprising:
measuring an attenuation curve for the communications channel;
finding a best-fit attenuation curve, including determining the best-fit attenuation curve by iteratively comparing to the attenuation curve pre-stored attenuation curves for loops that do not contain metallic faults;
comparing the measured attenuation curve and the best-fit attenuation curve over one or more subchannels;
determining, based on the comparison, whether one or more detectable faults are present; and
generating a confidence metric which quantifies an extent to which it is known that the fault is present on a communication line, and repeating_the measuring, finding, comparing and determining steps once a first fault is corrected.
9. A system for determining one or more faults in a communications channel comprising:
means for measuring an attenuation curve for the communications channel;
means for finding a best-fit attenuation curve, including determining the best-fit attenuation curve by iteratively comparing to the attenuation curve pre-stored attenuation curves for loops that do not contain metallic faults;
means for comparing the measured attenuation curve and the best-fit attenuation curve over one or more subchannels;
means for determining, based on the comparison, whether one or more detectable faults are present; and
means for generating a confidence metric which quantifies an extent to which it is known that the fault is present on a communication line, and once a first fault is corrected identifying another fault with the means for measuring, means for finding, means for comparing and means for determining.
10. A non-transitory computer-readable information storage media having stored thereon instructions that if executed by a processor cause to be performed a method for determining one or more faults in a communications channel comprising:
measuring an attenuation curve for the communications channel;
finding a best-fit attenuation curve, including determining the best-fit attenuation curve by iteratively comparing to the attenuation curve pre-stored attenuation curves for loops that do not contain metallic faults;
comparing the measured attenuation curve and the best-fit attenuation curve over one or more subchannels;
determining, based on the comparison, whether one or more detectable faults are present; and
generating a confidence metric which quantifies an extent to which it is known that the fault is present on a communication line, and repeating the measuring, finding, comparing and determining steps once a first fault is corrected.
11. A system comprising:
circuitry configured to determine one or more faults in a communications channel, the circuitry configured to implement
an attenuation curve measurement module that is capable of measuring an attenuation curve for the communications channel;
a best-fit attenuation module capable of finding a best-fit attenuation curve, including determining the best-fit attenuation curve by iteratively comparing to the attenuation curve pre-stored attenuation curves for loops that do not contain metallic faults;
a comparison module capable of comparing the measured attenuation curve and the best-fit attenuation curve over one or more subchannels;
a fault detection module capable of determining, based on the comparison, whether one or more detectable faults are present; and
a message generation module adapted to generate a confidence metric which quantifies an extent to which it is known that the fault is present on a communication line, wherein said circuitry is configured to identify another fault after a first fault is corrected.
12. The method of claim 8 , further comprising applying a correction to the measured attenuation curve.
13. The method of claim 8 , further comprising reporting whether the first fault has been detected.
14. The method of claim 8 , further comprising reporting, that a determination regarding a presence of the first fault cannot be made.
15. The method of claim 8 , further comprising determining a rate impact.
16. The method of claim 8 , further comprising determining a rate impact and reporting the same.
17. The method of claim 8 , further comprising repeating the measuring, finding, comparing and determining steps.
18. The method of claim 8 , further comprising querying an attenuation curve library.
19. The method of claim 8 , further comprising applying a correction to the measured attenuation curve, wherein the correction is one or more of:
multiplying the measured attenuation by a constant scale factor,
multiplying each tone of the measured attenuation by a scale factor that changes for each tone,
changing tone-dependent scaled factors as a function of the loop topology
using pre-determined calibration curves, wherein attenuation across a range of known loops is measured, the measured attenuation curves per-tone are divided by the theoretical models in the attenuation curve library, the resulting corrections stored in a calibration table, and fitting corrections to a parametric model, wherein the model is stored for later use in dynamically generating the corrections.
20. The method of claim 8 , further comprising selecting one or more subchannels.
21. The method of claim 8 , wherein rate impact estimates can be determined using a link assumption that includes actual link parameters used to obtain the measured attenuation curve.
22. The method of claim 8 , wherein the comparing of the measured attenuation curve and the best-fit attenuation curve to determine whether a fault is present is based on determining whether the following inequality is true:
MeasuredAttenuation/ModeledAttenuation<AttenuationErrorThreshold,
where AttenuationErrorThreshold is a ratio that allows for modeling and measurement errors to occur without a tolerable rate of false fault detections and missed fault detections.