Monitoring system for a communications network

• US 7,136,772 B2
• Filed: 11/08/2002
• Issued: 11/14/2006
• Est. Priority Date: 11/08/2002
• Status: Active Grant

• Current Assignee: Avago Technologies International Sales Pte Limited (Broadcom, Inc.)
• Original Assignee: Avago Technologies Fiber IP Singapore PTE Limited (Broadcom, Inc.)
• Inventors: Elsheimer, Robert, Duchi, Ceceli
• Primary Examiner(s): Raymond, Edward
• Application Number:
  US10/291,259
• Publication Number:
  US 20040091032A1
• Time in Patent Office: 1,467 Days
• Field of Search: 702/118, 702/188, 702/117, 709/224, 714/25, 714/26
• US Class Current: 702/118
• CPC Class Codes: H04B 3/46 Monitoring; Testing

Abstract

A monitoring system for a communications network includes a set of components for transmitting and receiving data signals over the communications network. Each component includes a built-in self-test system for performing tests within the component and outputting corresponding test results. Each self-test system is configured to non-invasively determine waveform characteristics of the data signals. The system includes a set of processors. Each processor is coupled to a subset of the set of components, and is configured to evaluate the test results output by the components coupled to the processor. Each processor is configured to detect and predict faults in the communications network based on the evaluation of the test results.

Claims

1. A monitoring system for a communications network, the monitoring system comprising:

a set of components for transmitting and receiving data signals over the communications network, each component including a built-in self test system for performing tests within the component and outputting corresponding test results, each self-test system configured to non-invasively determine waveform characteristics of the data signals; and

a set of processors, each processor coupled to a subset of the set of components, and configured to evaluate the test results output by the components coupled to the processor, each processor configured to associate identification information with each test result and detect and predict faults in the communications network based on the evaluation of the test results.

2. The monitoring system of claim 1, wherein the self-test system of each component further comprises:

a stimuli generator for generating start signals and stop signals based on the data signals; and

a time interval analyzer for determining the waveform characteristics of the data signals based on the generated start and stop signals.

3. The monitoring system of claim 2, wherein the waveform characteristics include at least one of jitter, pulse width, rise and fall times, and inter-channel skew.

4. The monitoring system of claim 1, wherein the self-test system of each component further comprises:

a signal amplitude measurement circuit for measuring an amplitude of the data signals;

a power supply measurement circuit for measuring power generated by a power supply of the component; and

a temperature sensor for measuring temperature within the component.

5. The monitoring system of claim 4, wherein the self-test system of each component further comprises:

a first analog-to-digital converter coupled to the signal amplitude measurement circuit for generating digital signal amplitude data;

a second analog-to-digital converter coupled to the power supply measurement circuit for generating digital power supply data; and

a third analog-to-digital converter coupled to the temperature sensor for generating temperature data.

6. The monitoring system of claim 1, wherein the self-test system of each component further comprises:

an optical power detector for measuring optical power of the data signals; and

an analog-to-digital converter coupled to the optical power detector for generating digital optical power data.

7. The monitoring system of claim 1, wherein each processor is configured to associate a time stamp and a component identifier with each set of test results output by the components coupled to the processor.

8. A method for non-invasively detecting and predicting faults in a communications network, the method comprising:

providing a set of components for transmitting and receiving data signals over the communications network;

non-invasively performing parametric tests within each component and outputting corresponding test results;

associating identification information with each test result;

generating a first fault indication indicating an existing fault if a test result is outside of predetermined specification limits;

identifying whether a failure trend exists based on the test results; and

generating a second fault indication indicating a potential future fault if a failure trend exists.

9. The method of claim 8, and further comprising:

identifying a test result that is outside of predetermined specification limits;

calculating a time value for the identified test result indicating an amount of time that the identified test result has been outside of the specification limits; and

comparing the calculated time value with a threshold time value.

10. The method of claim 9, wherein the first fault indication is generated only if the calculated time value exceeds the threshold time value.

11. The method of claim 8, and further comprising:

identifying a mathematical function of time that provides a best fit representation of the test results;

determining whether the test results are degrading over time, thereby identifying whether a failure trend exists; and

calculating a time to failure based on the mathematical function and the predetermined specification limits.

12. The method of claim 11, and further comprising:

comparing the calculated time to failure to a threshold time; and

wherein the second fault indication is generated only if the calculated time to failure is less than the threshold time.

13. The method of claim 11, and further comprising:

calculating a correlation coefficient that is indicative of the accuracy of the mathematical function to predict future values of the test results; and

comparing the calculated correlation coefficient to a minimum threshold value.

14. The method of claim 13, wherein the step of identifying whether a failure trend exists is performed only if the calculated correlation coefficient exceeds the minimum threshold value.

15. The method of claim 8, wherein the identification information includes a test name, a time stamp, and a component identifier that identifies the component that output the test result.

16. The method of claim 8, and further comprising:

sorting the test results based on the identification information associated with each test result.

17. A built-in self-test system for a component that transmits or receives data signals over a communications network, the system comprising:

a stimuli generator for generating stimuli signals based on the data signals, the stimuli generator including a set of flip-flops for generating stimuli signals based on rising and falling edges of the data signals, and including a set of comparators for generating stimuli signals based on a magnitude of the data signals;

a time interval analyzer for non-invasively determining waveform characteristics of the data signals based on the generated stimuli signals;

a set of monitoring circuits for monitoring operating characteristics of the component; and

a controller for controlling the time interval analyzer and the set of monitoring circuits to continually generate test results.

18. The built-in self-test system of claim 17, wherein the waveform characteristics include at least one of jitter, pulse width, rise and fall times, and inter-channel skew.

19. The built-in self-test system of claim 18, wherein the set of monitoring circuits include a temperature sensing circuit and a power supply monitoring circuit.