Method for reporting the time distribution of a succession of events

• US 20020188420A1
• Filed: 05/03/2002
• Published: 12/12/2002
• Est. Priority Date: 05/16/2001
• Status: Active Grant

• Current Assignee: Alcatel-Lucent SA (Nokia Corporation)
• Original Assignee: Nortel Networks Limited (Nortel Networks Corporation)
• Inventors: Isnard, Olivier, Pison, Laurent, Calmel, Jean-Marie
• Granted Patent: US 6,745,141 B2
• Time in Patent Office: Days
• US Class Current: 702/178
• CPC Class Codes: H04L 43/024 by adaptive sampling
• CPC Class Codes: H04L 43/045 for graphical visualisation of monitoring data
• CPC Class Codes: H04L 43/18 Protocol analysers

Abstract

To report the time distribution of a succession of specified events such as occurrences of particular data elements in telecommunications equipment, the method comprises an initialization of a sampling quantity consisting of a positive integer and several successive iterations of a scheme comprising the following steps: detection of a number of events equal to the sampling quantity and metering of a sampling time taken to detect said number of events; quantization of the sampling time metered in the previous step so as to output coding data representing a quantized value of the metered sampling time; and updating of the sampling quantity as a function of parameters including the quantized value of this sampling time.

Claims

1. A method for reporting a time distribution of a succession of specified events, comprising an initialization of a sampling quantity consisting of a positive integer and a plurality of successive iterations of a scheme comprising the steps of:

detecting a number of events equal to the sampling quantity and metering a sampling time taken to detect said number of events;

quantizing the sampling time metered in the preceding step to output coding data representing a quantized value of said sampling time; and

updating the sampling quantity as a function of parameters including said quantized value of the sampling time.

2. A method according to claim 1, wherein the step of updating the sampling quantity comprises calculating an estimated event rate as a function of parameters including a preceding sampling quantity and the quantized value of the sampling time, whereby the updated sampling quantity is deduced from the estimated event rate.

3. A method according to claim 2, wherein the estimated event rate is calculated in the form (1−

x).λ

+x.K/t, where x is a weighting coefficient lying between 0 and 1, λ

is the estimated event rate in force before the current iteration, K is the preceding sampling quantity and t is the quantized value of the sampling time.

4. A method according to claim 2, wherein the event rate estimated in an i-th iteration of the scheme (i≧

1) is calculated in the form

5. A method according to claim 2, wherein the updated sampling quantity is an integer substantially proportional to the estimated event rate.

6. A method according to claim 5, wherein the updated sampling quantity is equal to ┌

Z.λ

.n/S┐

, where Z is a predefined total duration of observation, λ

is the estimated event rate, n is a number of bits used to form coding data representing a quantized time value, S is a predefined data size expressed as a number of bits, and ┌

y┐

represents the integer equal to or immediately greater than a number y.

7. A method according to claim 1, wherein the step of quantizing the sampling time comprises producing coding data including at least one index of n bits, n being a predefined integer greater than 3, and wherein the producing of said coding data comprises the steps of:

determining a prediction time interval;

if the sampling time is below the prediction interval, outputting an index having a first value and replacing the prediction interval by a new time interval situated immediately below the preceding interval;

if the sampling time is above the prediction interval, outputting an index having a second value distinct from the first value and replacing the prediction interval by a new time interval situated immediately above the preceding interval; and

if the sampling time is within the prediction interval, outputting an index having a value distinct from the first and second values and representative of the position of the sampling time in the prediction interval.

8. A method according to claim 7, wherein the step of updating the sampling quantity comprises calculating an estimated event rate as a function of parameters including the preceding sampling quantity and the quantized value of the sampling time, whereby the updated sampling quantity is deduced from the estimated event rate, and wherein the prediction time interval initially determined is positioned with respect to a prediction value equal to a ratio of the sampling quantity to the estimated event rate.

9. A method according to claim 1, wherein a shift parameter is initialized to zero and, in each iteration of said scheme, the shift parameter is updated as a function of a difference between the metered sampling time and the quantized value of said sampling time, and wherein the shift parameter is involved in counting down the sampling time in at least some of the iterations.

10. A method according to claim 1, wherein the quantizing of the sampling time consists in outputting, as coding data, a number expressed in floating point representing a quantized value equal to the metered sampling time.

17. Telecommunication equipment according to claim 1, wherein the means for quantizing the sampling time comprise means for producing coding data including at least one index of n bits, n being a predefined integer greater than 3, and wherein the means for producing said coding data comprises:

means for determining a prediction time interval;

means for outputting an index having a first value if the sampling time is below the prediction interval, and for replacing the prediction interval by a new time interval situated immediately below the preceding interval;

means for outputting an index having a second value distinct from the first value if the sampling time is above the prediction interval, and for replacing the prediction interval by a new time interval situated immediately above the preceding interval; and

means for outputting an index having a value distinct from the first and second values if the sampling time is within the prediction interval, said index being representative of the position of the sampling time in the prediction interval.

11. Telecommunication equipment, comprising means for transmitting and/or receiving traffic including successive data elements, means for detecting occurrences of data elements of a particular type in said traffic and coding means for outputting coding data reporting a time distribution of the occurrences detected, wherein the coding means comprise means for initializing a sampling quantity consisting of a positive integer and iterative processing means including:

means for metering a sampling time taken in one iteration to detect a number of events equal to the sampling quantity;

means for quantizing the sampling time metered in said iteration to output coding data representing a quantized value of said sampling time; and

means for updating the sampling quantity for a next iteration as a function of parameters including said quantized value of the sampling time.

12. Telecommunication equipment according to claim 11, wherein the means for updating the sampling quantity comprise means for calculating an estimated event rate as a function of parameters including a preceding sampling quantity and the quantized value of the sampling time, whereby the updated sampling quantity is deduced from the estimated event rate.

13. Telecommunication equipment according to claim 12, wherein the estimated event rate is calculated in the form (1−

x).λ

+x.K/t, where x is a weighting coefficient lying between 0 and 1, λ

is the estimated event rate in force before said iteration, K is the preceding sampling quantity and t is the quantized value of the sampling time.

14. Telecommunication equipment according to claim 12, wherein the event rate estimated in an i-th iteration (i≧

1) is calculated in the form

15. Telecommunication equipment according to claim 12, wherein the updated sampling quantity is an integer substantially proportional to the estimated event rate.

16. Telecommunication equipment according to claim 15, wherein the updated sampling quantity is equal to ┌

Z.λ

.n/S┐

, where Z is a predefined total duration of observation, λ

is the estimated event rate, n is a number of bits used to form coding data representing a quantized time value, S is a predefined data size expressed as a number of bits, and ┌

y┐

represents the integer equal to or immediately greater than a number y.

18. Telecommunication equipment according to claim 17, wherein the means for updating the sampling quantity comprise means for calculating an estimated event rate as a function of parameters including the preceding sampling quantity and the quantized value of the sampling time, whereby the updated sampling quantity is deduced from the estimated event rate, and wherein the prediction time interval initially determined is positioned with respect to a prediction value equal to a ratio of the sampling quantity to the estimated event rate.

19. Telecommunication equipment according to claim 11, wherein a shift parameter is initialized to zero and, in each iteration, the shift parameter is updated as a function of a difference between the metered sampling time and the quantized value of said sampling time, and wherein the shift parameter is involved in counting down the sampling time in at least some of the iterations.

20. Telecommunication equipment according to claim 11, wherein the means for quantizing the sampling time comprise means for outputting, as coding data, a number expressed in floating point and representing a quantized value equal to the metered sampling time.

21. A computer program product, loadable into a memory associated with a processor, and comprising code portions for implementing, in an execution of said program by the processor, a method for reporting a time distribution of a succession of specified events, comprising an initialization of a sampling quantity consisting of a positive integer and a plurality of successive iterations of a scheme comprising the steps of:

detecting a number of events equal to the sampling quantity and metering a sampling time taken to detect said number of events;

quantizing the sampling time metered in the preceding step to output coding data representing a quantized value of said sampling time; and

updating the sampling quantity as a function of parameters including said quantized value of the sampling time.

22. A computer program product according to claim 21, wherein the step of updating the sampling quantity comprises calculating an estimated event rate as a function of parameters including a preceding sampling quantity and the quantized value of the sampling time, whereby the updated sampling quantity is deduced from the estimated event rate.

23. A computer program product according to claim 22, wherein the estimated event rate is calculated in the form (1−

x).λ

+x.K/t, where x is a weighting coefficient lying between 0 and 1, λ

is the estimated event rate in force before the current iteration, K is the preceding sampling quantity and t is the quantized value of the sampling time.

24. A computer program product according to claim 22, wherein the event rate estimated in an i-th iteration of the scheme (i≧

1) is calculated in the form

25. A computer program product according to claim 22, wherein the updated sampling quantity is an integer substantially proportional to the estimated event rate.

26. A computer program product according to claim 25, wherein the updated sampling quantity is equal to ┌

Z.λ

.n/S┐

, where Z is a predefined total duration of observation, λ

is the estimated event rate, n is a number of bits used to form coding data representing a quantized time value, S is a predefined data size expressed as a number of bits, and ┌

y┐

represents the integer equal to or immediately greater than a number y.

27. A computer program product according to claim 21, wherein the step of quantizing the sampling time comprises producing coding data including at least one index of n bits, n being a predefined integer greater than 3, and wherein the producing of said coding data comprises the steps of:

determining a prediction time interval;

if the sampling time is below the prediction interval, outputting an index having a first value and replacing the prediction interval by a new time interval situated immediately below the preceding interval;

if the sampling time is above the prediction interval, outputting an index having a second value distinct from the first value and replacing the prediction interval by a new time interval situated immediately above the preceding interval; and

if the sampling time is within the prediction interval, outputting an index having a value distinct from the first and second values and representative of the position of the sampling time in the prediction interval.

28. A computer program product according to claim 27, wherein the step of updating the sampling quantity comprises calculating an estimated event rate as a function of parameters including the preceding sampling quantity and the quantized value of the sampling time, whereby the updated sampling quantity is deduced from the estimated event rate, and wherein the prediction time interval initially determined is positioned with respect to a prediction value equal to a ratio of the sampling quantity to the estimated event rate.

29. A computer program product according to claim 21, wherein a shift parameter is initialized to zero and, in each iteration of said scheme, the shift parameter is updated as a function of a difference between the metered sampling time and the quantized value of said sampling time, and wherein the shift parameter is involved in counting down the sampling time in at least some of the iterations.

30. A computer program product according to claim 21, wherein the quantizing of the sampling time consists in outputting, as coding data, a number expressed in floating point representing a quantized value equal to the metered sampling time.