FR3109040A1 - Method of counting events from an external network - Google Patents

Abstract

Method for counting events of an external network The method for counting events produced by an external source and distributed according to a cyclic pattern reproduced periodically and in which two successive events are separated by a time interval, the or each interval being associated with a rank in the pattern, comprising the steps of: a/ determining (112), for each rank of intervals, a reference duration of the time interval between two successive events; b/ detecting (122) events of the external source; c/ for each detected event: c1/ calculating (122) the duration of the interval between the detected event and the preceding detected event; c2/ comparing (124, 128) the duration of the calculated time interval with the reference duration of the interval of the same rank; - if the duration of the calculated time interval is equal to within a tolerance of the reference duration of the interval of the same rank, increment (126) by a unit value an event counting counter; - otherwise, search for an integer such that satisfies a condition on the difference between the duration of the calculated time interval and a sum of the reference durations of the same rank for the next time intervals repairing the successive events, and increment (130) the event counting counter by the value. Figure for the abstract: Figure 3

Description

Method for counting events from an external network

The present invention relates to a method for counting events produced by an external source and distributed according to a cyclic pattern reproduced periodically and in which two successive events are separated by a time interval , the or each interval being associated with a rank in the pattern.

It is known, in particular from document Devialet US 10134442, to synchronize audio systems from a time base generated by a local clock of each audio system. These local clocks are synchronized from a common time base at 50 or 60 Hz resulting from the detection of events consisting of the instants of passage through the zero value of the mains voltage supplied by the power supply network.

The detection of the instants of passage of the sector by a zero value, or any other predetermined value is not completely reliable. Some events may be duplicated or, conversely, some others may not be detected.

For example, when using mains voltage zero crossings, high amplitude harmonics can create double or triple crossings near the zero crossing.

Conversely, events may be missing due to saturation of the queues of the software in charge of the processing or any other problem occurring on the actual detection of the passage through a predetermined value.

In the context of the audio synchronization of several sound reproduction devices, the consequence of an error in counting an event is the lag of a sound reproduction device in relation to the others. In the case of mains use, in Europe, this offset is at least 10 ms, i.e. more than 1000 times the acceptable tolerance for the synchronization of sound reproduction equipment.

It is therefore imperative to make the counting of events robust.

The object of the invention is therefore to propose a method and equipment allowing reliable counting of events produced by an external source, even in the event of an error in the detection of certain events, or in the event of multiple detections thereof.

To this end, the subject of the invention is a process for counting events of the aforementioned type, comprising the steps of:

a/ determine, for each rank of intervals, a reference duration the time interval between two successive events;

b/ detect events from the external source;

c/ for each event detected:

c1/ calculate the duration of the interval between the detected event and the previous detected event;

c2/ compare duration from the calculated time interval to the reference duration of the interval of the same rank ;

- if the duration of the calculated time interval is equal to within a tolerance at the reference duration of the interval of the same rank , increment a counter by a unit value event counting;

- otherwise, search for an integer such as satisfies a condition on the difference between the duration of the calculated time interval and a sum of the reference times of the same rank for the next time intervals repairing the successive events, and increment the counter value event count ;

According to particular modes of implementation, the method comprises one or more of the following characteristics:

- the or each reference duration ( is calculated as a function of a predetermined number of previously calculated durations, time intervals of the same rank ;

- if the previous durations calculated for a predetermined number of successive time intervals of the same rank are equal to within a tolerance , the or each reference duration is calculated as a function of a predetermined number of last calculated past durations time intervals of the same order ; And

- otherwise, the or each reference duration is equal to a reference duration previously calculated;

- if the duration of the time interval is not equal to the tolerance near the reference duration of the interval of the same rank, a new determination of the reference duration is performed, from durations of the time interval calculated after the event for which the duration of the time interval is not equal to the tolerance near the reference duration of the interval of the same rank ;

- tolerance is calculated as a function of the reference duration ;

- the condition (124) on the difference between the duration of the calculated time interval and a sum of the reference times satisfied with the number of is the fact that this difference is less than a predetermined threshold;

- the condition (124) on the difference between the duration of the calculated time interval and a sum of the reference times satisfied with the number of is the fact that minimize this difference.

The invention also relates to an event counting device comprising means for implementing the event counting method as defined above.

It also relates to sound reproduction equipment comprising sound reproduction means and an event counting device as defined above.

The invention will be better understood on reading the following description, given solely by way of example and made with reference to the drawings in which:

Figure 1 is a schematic view of a sound reproduction assembly comprising two synchronized equipment;

Figure 2 is a schematic representation of the synchronization signal; And

FIG. 3 is a flowchart of a method implemented by the method according to the invention.

In Figure 1 are illustrated two sound reproduction equipment 12, 14.

These two items of equipment each comprise means 20 for sound reproduction of an audio stream comprising, for example, an information processing unit 22 and an amplifier 24 connected to a loudspeaker 26.

The information processing unit 22 is connected to an audio signal source 23, such as a wifi network, a wired network, or any other means of transmitting information capable of providing an audio stream consisting of a succession of samples each associated with restitution time information.

Each piece of equipment 12, 14 is connected to a collective power supply network 30 for its power supply and to supply a common time base to each of the two pieces of equipment.

As known per se, the network provides an electric current with a frequency of 50 or 60 Hz depending on the geographical area, commonly called mains.

Each piece of equipment 12, 14 comprises, connected to the supply network, a module 32 for detecting events from the electrical network. This module implements a hysteresis comparator to filter any interference present on the power supply. The switching thresholds for a supply voltage of 230 V are taken equal for example to 11 V and 43 V for the positive alternations and -11 V and -43 V for the negative alternations. An event is detected when the mains voltage changes from a value below 11 V respectively -11 V to a value above 43 V respectively -43 V.

As illustrated in FIG. 2, the signal obtained at the output of the detection module 32 consists of events distributed according to a cyclic pattern reproduced almost periodically with a period slowly varying. The detection module 32 normally generates events per period, these events being distributed in the same way from one period to another but not necessarily in a uniform way. In the example shown in Figure 4, the number number of events per period is 4. Three periods are illustrated.

Advantageously, when the hysteresis comparator proves to be insufficient to infiltrate the parasites from the power supply network, a digital filter is implemented in a processor, for example the power supply processor of each piece of equipment.

The various alternation detection modules 32 are capable of supplying at output the same clock signal with a frequency equal to 200 Hz or 240 Hz depending on the mains frequency, the output signal forming a common time base for the two equipment, the frequency of which depends only on variations in mains frequency.

Each item of equipment 12, 14 comprises a main clock 34 which is specific to it. The main clocks of the various sound reproduction equipments are synchronized in frequency and in phase on the basis of the events supplied by the module 32 for detecting alternations by any suitable means.

The output of the main clock 34 is connected to the sound reproduction means 20 to provide them with a time base enabling them to synchronize the samples and their simultaneous restitution.

According to the invention, each piece of equipment also comprises a timestamping and counting module 40 connected to the output of the main clock 34 and of which a trigger input is connected to the output of the electronic event detection module 32 of the equipment considered.

This same timestamping and counting module 40 has an output connected to clock 34 for its time synchronization on network events.

These counting and time-stamping modules 40 are suitable for time-stamping each event from the module 32. At the output, the module 40 provides the clock 34 for each event with time-stamping information consisting of a pair consisting of the number of order of the event detected by the module 32 and of the timestamp instant of this event provided by the main clock 34.

The counting and timestamping modules 40 are suitable for numbering each event by an integer as illustrated in Figure 2, whether or not these events have been directly detected by the detection module 32.

The counting and time-stamping modules 40 comprise a computer, integrating software means suitable for implementing the flowchart described in FIG. 3 and means for temporary storage of the data necessary for its operation.

The processing carried out by the counting and timestamping module 40 will now be described with reference to FIG. 3 and are executed by the computer under the control of suitable software.

The module 40 comprises means for continuously recording the numbers of the events detected by the module 32 as well as the instant of detection of the event in the repository of the main clock 34.

Certain events may be missed by the detection means 32 and they are then missing from the memory of the counting and timestamping module 40.

The module 40 includes means for calculating the duration noted time intervals between two successive detected events.

Durations intervals are illustrated in Figure 2. For the rest of the description, we note THE subsequences such as obtained from calculated durations .

The module 40 determines in step 112 the average of the last durations for each time interval of the same rank . This noted average is defined by the following formula:

At step 114, the minimum of the averages calculated for the different ranks of the intervals is calculated. This noted minimum is expressed in the form:

.

A tolerance is then defined in step 116. It is expressed in the form:

With a number between 0 and 1 and in particular substantially equal to 0.1.

During a test carried out at step 118, the module 40 determines whether the last durations of the intervals for each row are of equal duration, to the tolerance close. For this purpose, the following test is carried out:

When this condition is not satisfied, the steps 112 and following are implemented again for the rank considered.

On the other hand, as soon as the condition of step 118 is satisfied, and as long as this condition remains true, the reference duration is stored in step 120 for each rank apart.

At the end of the detection of an event in step 122, a duration of the corresponding interval is calculated as the time difference between the timestamp instants of the last two detected events.

The duration is compared in step 124 to the reference duration of the interval of the same rank to determine if these values are equal within a tolerance. The test performed is expressed in the form:

Or is the counter numbering the events.

If this condition is satisfied, the counter is incremented by one unit in step 126, this counter giving the number of the event. At the end of step 124, steps 112 and following are again implemented.

On the other hand, if the condition of step 124 is not satisfied, the module 40 determines at step 128 if there is a non-zero integer such as satisfies a condition on the difference between the duration of the calculated time interval and a sum of the reference times of the same rank for the next time intervals between successive events.

More specifically, in the embodiment described, the test carried out consists in determining whether the difference between the duration of the calculated time interval and a sum of the reference times is below the predetermined threshold for the number .

This test is expressed mathematically in the form:

.

If the condition is satisfied, the counter is incremented by units at step 130.

If this condition is not satisfied, or at the end of step 130 of incrementing the counter, the last durations used for the calculation of the averages in step 112 are reset to zero in step 134 then steps 112 and following are implemented again until the condition of step 118 is again satisfied .

It is understood that with such a method, even if certain events are not correctly detected, the following detected events are however correctly numbered since the counter at step 130 is incremented by a number corresponding to the number of events which should have been detected.

As known per se, the clock 34 of each piece of equipment provides time synchronization on the detected events taking into account their instant of detection and their sequence number.

As a variant, the test carried out at step 128 is replaced by the determination of a value which minimizes the function equal to the fractional part of with

, representing the sum of the estimators of the intervals of the next k expected events.

So, noting the whole part of And is the fractional part of , the value is then determined by the following formula:

Or .

In the example described with reference to FIG. 2, the number ranks of time intervals in the same period is equal to 4. As a variant, the module 32 is suitable for determining the positive and negative alternations of the voltage, that is to say determining the instants of passage through zero of the latter so that the number of ranks is equal to 2.

Each information processing unit 22 of the sound reproduction means of each piece of equipment is connected to the output of the main clock 34 and includes software means for synthesizing an audio clock from the main clock of the equipment. This audio clock is used for the restitution of the audio samples as known per se according to a rhythm defined by the temporal information of restitution contained in the audio signal.

The method described above is implemented continuously in equipment 12 and 14.

Simultaneously, the sound reproduction means 20 provide sound reproduction from samples associated with restitution time information received from the audio signal source 23. This restitution is performed on the basis of a time base produced by a clock audio specific to each equipment, each audio clock being synchronized on the sector.

Claims (9)

Method for counting events produced by an external source and distributed according to a cyclic pattern reproduced periodically and in which two successive events are separated by a time interval , the or each interval being associated with a rank in the pattern, comprising the steps of:
a/ determine (112), for each rank of intervals, a reference duration the time interval between two successive events;
b/ detecting (122) events from the external source; And
c/ for each event detected:
c1/ calculate (122) the duration of the interval between the detected event and the previous detected event;
c2/ compare (124, 128) the duration from the calculated time interval to the reference duration of the interval of the same rank ;
- if the duration of the calculated time interval is equal to within a tolerance at the reference duration of the interval of the same rank , incrementing (126) a counter by a unit value event counting;
- otherwise, search for an integer such as satisfies a condition on the difference between the duration of the calculated time interval and a sum of the reference times of the same rank for the next time intervals repairing the successive events, and incrementing (130) the counter value event count . Counting method according to Claim 1, in which the or each reference duration ( is calculated (112) as a function of a predetermined number of previously calculated durations, time intervals of the same rank . A method according to claim 2, wherein:
- if the previous durations calculated for a predetermined number of successive time intervals of the same rank are equal to within a tolerance , the or each reference duration is calculated (134, 112) as a function of a predetermined number of last calculated prior durations time intervals of the same order ; And
- otherwise, the or each reference duration is equal to a reference duration previously calculated (120). Method according to Claim 2 or 3, in which if the duration of the time interval is not equal to the tolerance near the reference duration of the interval of the same rank, a new determination of the reference duration is performed, from durations of the time interval calculated after the event for which the duration of the time interval is not equal to the tolerance near the reference duration of the interval of the same rank . A method according to any preceding claim, wherein the tolerance is calculated (116) as a function of the reference duration . Method according to any one of the preceding claims, in which the condition (124) on the difference between the duration of the calculated time interval and a sum of the reference times satisfied with the number of is the fact that this difference is less than a predetermined threshold. Method according to any one of Claims 1 to 5, in which the condition (124) on the difference between the duration of the calculated time interval and a sum of the reference times satisfied with the number of is the fact that minimize this difference. Device for counting events comprising means for implementing the method for counting events according to any one of the preceding claims. Sound reproduction equipment comprising sound reproduction means (20) and an event counting device according to claim 8.