Method and apparatus for measuring jitter

• US 7,590,170 B2
• Filed: 09/29/2004
• Issued: 09/15/2009
• Est. Priority Date: 09/29/2004
• Status: Active Grant

• Current Assignee: Teradyne Incorporated
• Original Assignee: Teradyne Incorporated
• Inventors: Tek, Mehmet Nejat, Liu, Chibling
• Primary Examiner(s): Payne, David C
• Assistant Examiner(s): Wong, Linda
• Application Number:
  US10/954,032
• Publication Number:
  US 20060067390A1
• Time in Patent Office: 1,812 Days
• Field of Search: 702 67- 71, 702/189, 702/198, 714 47- 48, 714/55, 714 40- 42, 714701-707, 714752-755, 714761-762, 714/768, 714764-765, 714/792, 714/795, 714/798, 37524026-24027, 375242-245, 375/253, 375/254, 375/262, 375/265, 375/354, 375/373, 375/376, 375365-368, 375/362, 375224-228, 375/371
• US Class Current: 375/226
• CPC Class Codes: G01R 29/26 Measuring noise figure; Measuring signal-to-noise ratio
• CPC Class Codes: G01R 31/31709 Jitter measurements; Jitter generators measuring jitter, noise figure or signal-to-noise ratio per se G01R29/26; analysis of tester signals G01R31/31901
• CPC Class Codes: H04L 1/20 using signal quality detector

Abstract

A system and method for characterizing the jitter of a periodic signal. Samples of the signal are taken with a sampling device. A set of samples representing a particular value of the signal in multiple cycles of the periodic signal is collected. Those values are formed into a histogram. The histogram is matched to a probability distribution function. By identifying parameters that shape the probability distribution function to match the histogram of actual samples, characteristics of the jitter are determined. This technique may be employed as part of the calibration or verification of the jitter injection instrument such as might be used for testing semiconductor devices. Measurements may be made with a sampling device that is calibrated to NIST standards. In this way, the jitter measurements become NIST traceable.

Claims

1. A method of characterizing jitter in a signal, comprising:

obtaining a plurality of samples of the signal;

forming a histogram of the sample values;

fitting a probability distribution function to the histogram, wherein fitting comprises providing a plurality of probability distribution functions, each characterized by a set of parameters, and identifying one of the plurality of probability distribution functions that differs from the histogram by the smallest amount, wherein identifying comprises computing for each of the plurality of probability distribution functions an error value representative of the differences between said probability distribution function and the histogram, and identifying the probability distribution function having the smallest error value; and

determining characteristics of the jitter from parameters of the probability distribution function identified as fitting the histogram.

2. The method of characterizing jitter in a signal of claim 1 wherein the plurality of probability distribution functions are generated iteratively by varying parameters of the probability distribution function.

3. The method of characterizing jitter in a signal of claim 1 wherein computing an error value comprises computing an error value with a norm function.

4. The method of characterizing jitter in a signal of claim 1 wherein fitting a probability distribution function to the histogram comprises performing at least one of a least squares estimation, minimization and/or root finding of parameters of the probability distribution function.

5. The method of characterizing jitter in a signal of claim 1 wherein fitting a probability distribution function to the histogram comprises identifying values of a plurality of parameters of the probability distribution function.

6. The method of characterizing jitter in a signal of claim 5 wherein the probability distribution function comprises a combination of a probability distribution function of sinusoidal jitter and a probability distribution function of normal jitter and identifying values of a plurality of parameters comprises identifying values of the peak to peak deviation of the sinusoidal jitter and the standard deviation of the normal jitter.

7. The method of characterizing jitter in a signal of claim 1 wherein the signal is phase modulated according to a predetermined function and the probability distribution function comprises a combination of a probability distribution function of the predetermined function and a probability distribution function of random values.

8. The method of characterizing jitter in a signal of claim 1 wherein fitting a probability distribution function to a histogram comprises fitting a probability distribution function that represents the convolution of the probability distribution functions of two jitter components.

9. The method of characterizing jitter in a signal of claim 1 wherein obtaining a plurality of samples of the signal comprises under sampling or equivalent time sampling the signal.

10. The method of characterizing jitter in a signal of claim 1 wherein each of the sample values has a magnitude and a time associated with it and forming a histogram of sample values comprises forming a histogram of the times associated with samples having values in a predetermined range.

11. A method of verifying the performance of a programmable jitter injection device, comprising:

programming the jitter injection device to generate a signal having a programmed amount of jitter;

characterizing jitter in the signal according to the method of claim 1 to obtain characteristics of the signal; and

determining, from the characteristics of the signal, whether the generated signal contains the programmed amount of jitter.

12. The method of verifying the performance of a programmable jitter injection device of claim 11 wherein:

programming the jitter injection device to generate a signal having a desired amount of jitter comprises specifying a phase modulation function; and

fitting a probability distribution function to the histogram comprises fitting to the histogram a probability distribution function with a probability distribution matching the phase modulation function.

13. The method of verifying the performance of a programmable jitter injection device of claim 11 wherein the signal has a frequency above 10 GHz.

14. A method of verifying the performance of a programmable jitter injection device, comprising:

a) programming the jitter injection device to generate a jitter modulated signal having a programmed amount of jitter, the jitter modulated signal being modulated according to a modulating function;

b) forming a histogram of times of occurrence of a value of the jitter modulated signal;

c) fitting a probability distribution function to the histogram, the probability distribution function having a component proportional to a probability distribution function of the modulating function and a component proportional to a probability distribution function of a random function, wherein fining comprises providing a plurality of probability distribution functions, each characterized by a set of parameters, and identifying one of the plurality of probability distribution functions that differs from the histogram by the smallest amount, wherein identifying comprises computing for each of the plurality of probability distribution functions an error value representative of the differences between said probability distribution function and the histogram, and identifying the probability distribution function having the smallest error value; and

d) characterizing jitter from the component proportional to a probability distribution function of the modulating function of the probability distribution function identified as fining the histogram.

15. The method of verifying the performance of a programmable jitter injection device of claim 14 wherein the modulating function is a sinusoid and the random function is a Gaussian function.

16. The method of verifying the performance of a programmable jitter injection device of claim 14 wherein forming a histogram of times of occurrence of a value comprises:

sampling the jitter modulated signal;

selecting samples having values in a band around the value;

dividing the selected samples into bins.

17. The method of verifying the performance of a programmable jitter injection device of claim 14, additionally comprising:

changing the programmed characteristics of the jitter modulated signal and repeating steps b) to d).

18. A test system configured for providing a signal with a programmed amount of jitter in a signal, the test system comprising:

a reference clock;

a programmable jitter injection module comprising;

a phase modulator having a modulation input a signal input, and a modulated output, the modulated output being the signal input phase modulated by an amount proportional to the modulation input;

a synthesizer circuit having an output generated from the reference clock, the output of the synthesizer circuit utilized as the signal input;

a sampling device receiving the modulated output and providing as an output a plurality of samples of the modulated output;

a computer processor receiving the plurality of samples of the modulated output, the computer processor having a program associated therewith, the program controlling the computer to analyze the samples by forming a histogram of sample values and fitting to the histogram a probability distribution function having at least one component representative of the probability distribution function of the modulation input, wherein fitting comprises providing a plurality of probability distribution functions, each characterized by a set of parameters, and identifying one of the plurality of probability distribution functions that differs from the histogram by the smallest amount, wherein identifying comprises computing for each of the plurality of probability distribution functions an error value representative of the differences between said probability distribution function and the histogram, and identifying the probability distribution function having the smallest error value.

19. The test system of claim 18 wherein the program controlling the computer processor to analyze the samples comprises controlling the computer to fit to the histogram to a probability distribution function having at least two components comprising the probability distribution function of the modulation input and the probability distribution function of a source of random jitter.

20. The test system of claim 18 wherein the sampling device collects between 15,000 and 45,000 samples.