WO2005096123A2 - Procede de synchronisation de donnees, notamment distribuees, prenant en compte les imprecisions et derives des horloges. - Google Patents
Procede de synchronisation de donnees, notamment distribuees, prenant en compte les imprecisions et derives des horloges. Download PDFInfo
- Publication number
- WO2005096123A2 WO2005096123A2 PCT/FR2005/000379 FR2005000379W WO2005096123A2 WO 2005096123 A2 WO2005096123 A2 WO 2005096123A2 FR 2005000379 W FR2005000379 W FR 2005000379W WO 2005096123 A2 WO2005096123 A2 WO 2005096123A2
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- time
- clock
- values
- common
- interval
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04J—MULTIPLEX COMMUNICATION
- H04J3/00—Time-division multiplex systems
- H04J3/02—Details
- H04J3/06—Synchronising arrangements
- H04J3/0635—Clock or time synchronisation in a network
- H04J3/0638—Clock or time synchronisation among nodes; Internode synchronisation
- H04J3/0658—Clock or time synchronisation among packet nodes
- H04J3/0661—Clock or time synchronisation among packet nodes using timestamps
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/04—Generating or distributing clock signals or signals derived directly therefrom
- G06F1/12—Synchronisation of different clock signals provided by a plurality of clock generators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/04—Generating or distributing clock signals or signals derived directly therefrom
- G06F1/14—Time supervision arrangements, e.g. real time clock
Definitions
- the invention relates to a method for determining a correspondence between the times indicated by the internal clocks of a plurality of machines connected to a network, such as sensors or computers, as well as to a method for synchronizing the data originating from said machines. .
- a network such as sensors or computers
- synchronizing data means ordering them chronologically according to their acquisition time; in a more limited sense, it also means “dating” these same data with respect to a single time reference.
- Synchronize clocks means forcing them to indicate the same time at a given time.
- a data synchronization process is necessary, for example, in the field of validation of driving assistance devices.
- test motor cars are equipped with various sensors connected in a network and making it possible to acquire information on the vehicle at the same time (average speed, acceleration ... ), on the interaction with other vehicles (relative distance and speed %) and on the driver's behavior (reaction time, gaze direction ).
- the data acquired by these various sensors and those recorded by the driving assistance devices must be synchronized in order to be able to be used by an ergonomist, and the precision of the synchronization must be better than a millisecond, and preferably better than 100 ⁇ s, for data acquisition rates up to 1 kHz.
- the most obvious solution to the problem of synchronizing data from a plurality of devices is to synchronize all clocks. Each device assigns a “date” to the data it records and, since the clocks are synchronized with each other, all the dates can be directly compared.
- Clock synchronization can be performed in hardware or software: a general introduction to these techniques is provided in the article by JA Fonseca and P. Fonseca "An overview of clock synchronization solutions", 4 th IFAC International Symposium, Sicica, Argentina , 2000. Hardware solutions are based on the use of an external clock signal, common to all devices. In this way one can obtain a very high precision, for example by the use of an atomic clock, such as that of the GPS system. On the other hand, these solutions have the drawback of requiring dedicated hardware: it is not possible to connect devices that are not designed to use an external clock signal to the network.
- Software solutions include the use of software clocks (variables stored in memory) controlled by the "hardware" clock of each device connected to the network. These software clocks are synchronized, so as to indicate approximately the same time at all times.
- software clocks variables stored in memory
- These software clocks are synchronized, so as to indicate approximately the same time at all times.
- two classes can still be distinguished: either a single synchronization is carried out at the start of the network usage session, and it is assumed that the clocks drifting can be neglected; either resynchronizations are also carried out during use; this is the technique adopted, for example, by the NTP protocol (Network Time Protocol).
- the clocks of the latter exhibit drifts which can reach 300 ⁇ s / s, more than one second per hour. Replacing these clocks with better quality components amounts to a material, expensive and inflexible solution.
- the second class of solutions presents a difficulty linked to the monotony of clocks. If, for example, a clock has moved ahead of the reference time, it will be brought back during resynchronization. If a first data item was recorded immediately before the resynchronization instant and a second data item immediately after, the date of the second recording is earlier than that of the first one and the temporal order of the data is reversed.
- An object of the present invention is a method for establishing a correspondence between the local times of two or more machines whose clocks are not synchronized. Another object of the present invention is such a method, with improved precision compared to the prior art. Yet another object of the present invention is such a method, with known precision and modeled by an interval.
- the present invention also relates to a method for synchronizing data recorded by two or more machines, or generated by two or more devices, which does not require synchronization of the clocks of said devices or machines, and which is based on establishing a correspondence between the local times of two or more machines whose clocks are not synchronized.
- At least one of the aforementioned objects is achieved using a method for establishing an interval correspondence between the time indicated by a first monotonic clock and the time indicated by a second clock, also monotonic, characterized in that it puts implementing, over at least one time range, a first time reference common to the first and the second clock and monotonic on said range, and in that said method comprises the following steps: a) determination of a first delimited time interval by a first pair of time values of the first clock and belonging to a first time range over which said said exists first time reference common to the first and to the second clock; b) determining a second time interval delimited by a second pair of values of the time of the first clock and belonging to a second time range over which there is said first time reference common to the first and to the second clock; c) determination, using the common time reference, of a third time interval, delimited by a first pair of time values of the second clock, and containing the first time interval; d) determination, using the common time reference, of a fourth time interval, de
- step c) comprises the following operations: d) determination of a seventh and an eighth time interval, delimited by a fourth and fifth pair of time values of the second clock respectively and belonging to the first time range, such that said first time interval is between said seventh and eighth time intervals; c2) determining a first, second and third value of the first common time reference, included in said first, seventh and eighth time intervals respectively; c3) calculation by interpolation of said third time interval, using said first, seventh and eighth time intervals and said first, second and third values of the first common time reference; and step d) comprises the following operations: d1) determination of a ninth and a tenth time interval, delimited by a sixth and seventh pair of time values of the second clock respectively and belonging to the second time range, such that said second time interval is between said ninth and tenth intervals; d2) determining a fourth, fifth and sixth value of the first common time reference, included in said second, ninth and tenth time intervals respectively; d3) calculation by inter
- operation c3) is carried out by replacing said first, second and third values of the common time reference by time intervals whose width is equal to the discretization of the first common time reference over the first time range, and l operation d3) is performed by replacing said fourth, fifth and sixth values of the common time reference with time intervals whose width is equal to the discretization of the first common time reference over the second time range.
- a first reading of the first clock is repeatedly recorded, followed by a reading of the first common time reference and then by a second reading of the first clock.
- step c) comprises the following operations: d) determination of a seventh and an eighth value of the time of the second clock belonging to the first time range, such that said first time interval is between said seventh and eighth values; c2) determining a first, second and third interval of values of the first common time reference, comprising said first time interval and said seventh and eighth time values of the second clock respectively; c3) calculation by interpolation of said third time interval, using said first time value interval of the first clock, said seventh and eighth time values of the second clock and said first, second and third value intervals of the first common time reference; and step d) comprises the following operations: d1) determining a ninth and a tenth time value of the second clock belonging to the second time range, such that said second time interval is between said ninth and tenth values; d2) determining a fourth, fifth and sixth value interval of the first common time reference, comprising said second time interval and said ninth and tenth time values of the second clock; d3) calculation by interpolation of said fourth
- operations c3) and d3) are carried out by replacing said seventh, eighth, ninth and tenth values of the time of the second clock by time intervals whose width is equal to the discretization of the time of the second clock.
- a first reading of the first common time reference is repeatedly recorded, followed by a reading of the first clock and then by a second reading of the first reference common time, and independently a first reading of the first common time reference is also recorded several times, followed by a reading of the of the second clock (and then by a second reading of the first common time reference, and operations d), c2), c3), d1), d2) and d3) are performed using these records.
- the calculations by interpolation or extrapolation mentioned above are more precisely linear interpolations or extrapolations, or linear by intervals.
- the first time reference common to the first and to the second clock exists over at least two distinct time ranges and may include a break in monotony from one time range to another, a second common monotonic time reference is used in order to to resolve the ambiguities which result from the non-monotony of the first common time reference.
- the invention also relates to a method for synchronizing the data recorded and dated by a first machine, having a first clock, with respect to the local time of a second machine, having a second clock, characterized in that the dating takes place by associating with each data recorded by the first machine a fifth time interval, delimited by a third pair of values of the time of the first clock, and in that the synchronization is effected by determining, by a method as described above, a sixth time interval, delimited by a third pair of values of the time of the second clock and containing said fifth time interval.
- the first common time reference is provided by the clock of a synchronous bus which at least temporarily connects said first and second machines.
- such a method of synchronizing data is broken down into a first sub-method of recording clock readings as described above, performed locally by each machine for which it is desired to be able to synchronize the data, and a second synchronization process proper, carried out by a single machine to which all the data has been transferred after the end of the data recording session.
- FIG. 1A and 1B illustrate how the determination can be made a correspondence between the times indicated by two separate and non-synchronized clocks
- Figures 2, 3A - 3L, 4 and 5, as well as tables Tab.A, Tab.B and Tab.C, illustrate by a concrete example a method constituting an embodiment of the invention
- Figure 6 shows a flow diagram of this process.
- real time is meant physical time, which cannot be known precisely and of which the clocks give only an approximation; in this document, “t” indicates a real time value.
- clock is meant a digital clock, constituted by an oscillator coupled to a counter.
- a "synchronous bus” is a bus on which a clock signal is broadcast.
- local time of a machine is meant the time measured by the clock of this machine.
- a local time value is indicated by "T”.
- date of a data recorded by a machine is meant either the time read on the clock of the machine at the time of the recording of the data, or the interval between two readings of this clock, carried out before and after said registration. Any ambiguity will be removed by the context and by the notation: thus, the date indicated by "[TJ" is in fact the interval between T and T. With T ⁇ T. In reality, any date can be considered as an interval: even if we know with certainty what was the value indicated by the clock of the machine at the time of the recording of a data, there remains an indeterminacy ⁇ due to the discretization of said clock.
- offset is meant the difference between the time indicated by two clocks at a given instant, or between the time of a clock and the real time.
- drift between two clocks, or a clock and real time, we mean the derivative of the offset from local time of one of the clocks or real time. Unless otherwise indicated in this document, it is considered that, for each pair of clocks, the drift is constant over time and that, consequently, the offset is a linear function with respect to time (assumption of linear drifts). In this case, the drift can be calculated by knowing two offsets determined on different dates. If the offsets and / or dates are intervals, the drift is also one:
- Exhibitors 1 and 2 refer to the first and second dates on which an offset is determined.
- synchronization by intervals of the data of machine B with respect to machine A is meant the determination, for each data recorded by machine B with a date [T B ], of an interval [T AB ] of local time of machine A as T AB precedes
- FIGS. 1A and 1 B illustrate how one can proceed to determine the correspondence between the times T A and T B indicated by two clocks A and B.
- the discretization of the clocks has been neglected ( ⁇ -> 0) and therefore the lines T A (t) and T B (t), which actually have a staircase shape, have been approached by straight lines.
- A, B and C such as computers, each having an internal clock H A , H B and H c which measures a local time 7 A (t), T B (t) and 7 " c (t) respectively, as well as a synchronous BUS bus, such as a FireWire ® bus (IEEE 1394) having its own clock H BU s having an accuracy better than 3 ⁇ s in the most case unfavorable according to the IEEE1394a standard.
- the machines can connect and disconnect from the BUS bus independently of each other, whenever a reconfiguration of the system occurs (for example, each time a machine connects or disconnects), an interrupt signal is transmitted on the bus.
- reference machine One of the machines connected to the network is named "reference machine” and its name is known by the other machines; in the figure it is machine B, characterized by a thick outline.
- the reference machine cannot be considered as a “master” machine in a centralized system: any machine on the network can be chosen to temporarily assume this function. Indeed, a new reference machine is chosen randomly at each interruption, which makes it possible to ensure the functioning of the system even in the event of disconnection of the old reference machine.
- Each machine is also equipped with a bus interface l A , l B and l c , comprising a clock, HI A , HI B and Hl c respectively, controlled by H BU s-
- the clock of the bus H BU s occurred by one of the I interfaces (called “Cycle Master” in the IEEE FireWire standard).
- the Hl clocks are slaved on H BU s.
- the bus clock H B us has a limited capacity counter, which is reset to zero every 128 s, but the interface clocks have a software extension with a capacity large enough so that overruns do not occur during the duration of the system.
- the time measured by the clock of the interface HI A , HI B and Hic ("interface time") of machine A, B and C is indicated by NT A (t), NT B (t) and NT c (t) respectively.
- One of the machines is chosen to "give the network time”: this means that the interface clocks of all the other machines are synchronized with its own.
- a new machine is chosen for this task and the software extension of the interface clocks is reset to zero: after the interruption, all clocks therefore indicate the same time as the clock on the bus H B us-
- the machine which gives the network time is not necessarily the reference machine. If a machine is disconnected from the network, it is its own reference machine and its network time is the time of its own interface clock.
- time NT of the network constitutes a first temporal reference common to all the machines connected to the bus BUS, but not monotonic.
- the time of the reference machine constitutes a second time reference common to all the machines connected to the BUS bus and, moreover, monotonous.
- This second time reference is nevertheless very imprecise because, in order to know it, a machine other than the reference one must make a request via the bus, which requires a relatively long time which cannot be determined a priori.
- each machine records, independently of the others, data and their date in the form of an interval, for example sensor acquisitions.
- each machine builds a so-called “parking meter” file, which consists of a series of lines.
- the lines are filled at approximately regular intervals, for example of about one second, and they include five fields: TL1 is a first reading of the local time of the machine; TR is a reading of the network time; TL2 is a second reading of the local time of the machine; NMR is the name of the reference machine; and TMR is the local time of the reference machine.
- the first three readings must be taken in order, so that TR is between TL1 and TL2.
- the machine fills an "exception line" by assigning the fields TL1, TR, TL2 and TMR the value 0 and the field NMR the conventional name "BUS_RESET".
- Filling time stamp files is a low-priority, low-cost process that does not significantly interfere with recording and dating events.
- the time stamp files are transferred to the same computer and a proper resynchronization sub-procedure can be started.
- Tables Tab.A, Tab.B and Tab.C show part of the content of the time stamp files of machines A, B and C respectively, hereinafter called FHA, FHB and FHC respectively. Only the two lines which precede and follow each interruption are shown; moreover, for the sake of clarity and space saving, the TL2 field does not contain the value of the second reading of the local time of the machine, but the difference between the second and the first reading, which is clearly equivalent.
- the times indicated are in microseconds ( ⁇ s), but this is not essential for understanding the process.
- the interrupts subdivide each file into sections ⁇ 1 - ⁇ , ⁇ 1 - ⁇ 9 and ⁇ 1 - ⁇ 9, separated by exception lines. It is important to observe that there is an interruption at each break in monotony, so we can consider that the network time is in fact “locally” monotonous on each section.
- the NMR and TMR fields are used to determine a temporal correspondence between the sections of the different files. For example, we see that sections ⁇ 1 and ⁇ 2 were recorded simultaneously, while machines A and C were both connected to the BUS bus, with machine A as the reference machine. In this way, the "history" of the system can be reconstructed, as illustrated in FIGS. 3A - 3L. Initially (fig.
- FIG. 3F the three machines are disconnected from the bus and fill sections ⁇ .2, ⁇ 3 and ⁇ 5 with the respective time stamp files (machine A, which was already disconnected, does not receive the interrupt signal caused by the disconnection of B and C).
- machines A and B connect again to the bus, with A as the reference machine, and thus pass to sections ⁇ 3 and ⁇ 4, while machine C remains disconnected and, not receiving an interrupt signal, continues to complete section ⁇ 5.
- Figure 3H the three machines are connected to the bus (sections ⁇ 4, ⁇ 5 and ⁇ 6), with B as the reference machine. When the latter (B) disconnects (figure 31, sections ⁇ 5, ⁇ 6 and ⁇ 7), A becomes the reference machine.
- TMR time stamp file
- NT first time reference
- FIG. 4 is a graphic representation of the information contained in the FHA and FHB files (the discretization of the clocks is neglected: the steps corresponding to the discrete increments of the counters are not distinguished).
- the ordinate axis carries the local times of machines A and B and the times of the networks to which said machines are connected, that of the abscissae the real time.
- the lines TMA and TMB respectively represent the local times of A and B.
- the lines TRA (dotted line) and TRB (circles) represent the times of the network of A and B respectively.
- the common areas P1, P2, P3 and P4 are highlighted by shaded areas.
- each datum recorded by machine B is characterized by a date [T B ], which is an interval between two readings of the clock H B surrounding a value of the network time, [NT] .
- the objective is to determine [T AB ], the local time interval of the machine A corresponding to [T B ].
- the first operation consists in determining two dates [T A ⁇ 1 ] and [T A2 1 ] of machine A belonging to the same range and located before and after [T B 1 ]. In fact, these “dates” are intervals determined by the method described with reference to FIG. 1 B and corresponding to the times of the network [NT A 1 ] and [NT A2 1 ] respectively.
- the linear clock drift hypothesis makes it possible to calculate [T AB 1 ] by a linear interpolation:
- [T B ] ⁇ [T B 1 ] is a simplified notation for T B ⁇ T B and [T B ] ⁇ [T B 2 ] for T B > T B.
- Figure 5 is a graphic representation of equations 6 - 8. We observe that the width of the interval [T AB ], and therefore the imprecision of the synchronization, is roughly constant for
- FIG. 6 is illustrated a flowchart of the method of interval resynchronization of the data of the machine B with respect to the machine A described above. The process can be reproduced for the synchronization of several machines.
- the first step (E1) involves filling in the FHA and FHB time stamp files for machines A and B, as well as those of all the other machines in the system (C, in particular) and, in parallel, recording locally dated data . These operations are executed until the end of the data recording session.
- the second step (E2) involves determining, using the NMR and TMR fields of the FHA, FHB and FHC files, the temporal relationships between the different sections of these files, as well as "Common time ranges" of machines A and B, that is to say periods during which these two machines were connected to the same network. Then, step E3, two lines LB1 and LB2 of the file FHB are chosen, each belonging to a "common time range".
- step E4 two lines (LA1 and LA2) of the FHA file belonging to the same time range as the line LB1 of FHB and recorded respectively before and after the latter are determined. In the same way, the lines LA3 and LA4 which “surround” LB2 are determined.
- LA1 and LA3 are the last lines recorded before LB1 and LB2, and that LA2 and LA2 are the first lines recorded after LB1 and LB2 respectively.
- step E6 all the data from machine B (or only part of it) is resynchronized by interpolation or extrapolation, using equations 6.1, 6.2, 7 and 8.
- a method of resynchronizing the data according to the flow diagram of FIG. 6 has been experimentally tested by the inventors.
- two computers based on an Intel ® Pentium ® IV processor with a clock rate of 1, 8 and 2 GHz respectively and a RAM of 512 MB, equipped with Microsoft operating system ® Windows 2000 Professional ® and a FireWire ® MindReady ® interface with the Sednet 2 ® API, linked together and to a UNIBRAIN ® Fire-I ® digital camera by a FireWire ® network bus.
- determining an offset [off AB 1 ] requires the use of a couple of network time readings, [NT A - ⁇ 1 ] and [NT A2 1 ].
- any couple surrounding [T B ] can be chosen, and the result is always an interval [off A ⁇ 1 ] containing with certainty the "true" value off AB 1 V r a ⁇ of the offset (impossible to know).
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Synchronisation In Digital Transmission Systems (AREA)
- Electric Clocks (AREA)
Abstract
Description
Claims
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2007501304A JP4829876B2 (ja) | 2004-03-03 | 2005-02-17 | クロックドリフトおよび不正確さを考慮した分散型データなどのデータ同期方法。 |
US10/598,442 US7889764B2 (en) | 2004-03-03 | 2005-02-17 | Method for the synchronization of data, such as distributed data, taking account of clock drifts and inaccuracies |
EP05729336.7A EP1728140B1 (fr) | 2004-03-03 | 2005-02-17 | Procede de synchronisation de donnees, notamment distribuees, prenant en compte les imprecisions et derives des horloges. |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
FR0402189A FR2867288B1 (fr) | 2004-03-03 | 2004-03-03 | Procede de synchronisation de donnees, notamment distribuees prenant en compte les imprecisions et derives des horloges |
FR0402189 | 2004-03-03 |
Publications (2)
Publication Number | Publication Date |
---|---|
WO2005096123A2 true WO2005096123A2 (fr) | 2005-10-13 |
WO2005096123A3 WO2005096123A3 (fr) | 2006-03-02 |
Family
ID=34855009
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/FR2005/000379 WO2005096123A2 (fr) | 2004-03-03 | 2005-02-17 | Procede de synchronisation de donnees, notamment distribuees, prenant en compte les imprecisions et derives des horloges. |
Country Status (5)
Country | Link |
---|---|
US (1) | US7889764B2 (fr) |
EP (1) | EP1728140B1 (fr) |
JP (1) | JP4829876B2 (fr) |
FR (1) | FR2867288B1 (fr) |
WO (1) | WO2005096123A2 (fr) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
NO20076533L (no) * | 2007-12-18 | 2009-06-19 | Tandberg Telecom As | Fremgangsmåte og system for klokkedriftskompensering |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2424136B1 (fr) | 2010-08-24 | 2013-10-02 | Siemens Aktiengesellschaft | Système et procédé pour obtenir une synchronisation temporelle dans un réseau de communication |
EP2567741A1 (fr) * | 2011-09-07 | 2013-03-13 | Research In Motion Limited | Appareil et procédé associé de fourniture d'une lecture de média synchronisée |
US11776053B2 (en) * | 2014-09-07 | 2023-10-03 | Codrut Radu Radulescu | Synchronized exchange system |
KR20180074684A (ko) * | 2015-10-23 | 2018-07-03 | 퀄컴 인코포레이티드 | 제어기와 센서들을 동기화하기 위한 장치 및 방법들 |
US20180224887A1 (en) * | 2017-02-06 | 2018-08-09 | Qualcomm Incorporated | Apparatus and method for asynchronous event timestamping |
CN112689963B (zh) * | 2020-03-30 | 2022-01-11 | 华为技术有限公司 | 一种数据处理方法及装置 |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0042924A2 (fr) * | 1980-06-30 | 1982-01-06 | International Business Machines Corporation | Appareil de transfert de données |
US20020150189A1 (en) * | 1997-10-10 | 2002-10-17 | Rambus Incorporated | Apparatus and method for generating a distributed clock signal using gear ratio techniques |
EP1380918A2 (fr) * | 2002-07-12 | 2004-01-14 | Ensequence, Inc. | Appareil et procédé associé permettant de synchroniser l'operation d'un compteur temporel éloigné avec une centrale de commande |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH07128463A (ja) * | 1993-11-08 | 1995-05-19 | Nippon Telegr & Teleph Corp <Ntt> | 通信装置の時刻管理方式 |
JP2000010933A (ja) * | 1998-06-25 | 2000-01-14 | Nippon Telegr & Teleph Corp <Ntt> | データ同期補正方法及び装置及びデータ同期補正プログラムを格納した記憶媒体 |
US6278694B1 (en) * | 1999-04-16 | 2001-08-21 | Concord Communications Inc. | Collecting and reporting monitoring data from remote network probes |
US7668099B2 (en) * | 2003-06-13 | 2010-02-23 | Apple Inc. | Synthesis of vertical blanking signal |
-
2004
- 2004-03-03 FR FR0402189A patent/FR2867288B1/fr not_active Expired - Lifetime
-
2005
- 2005-02-17 EP EP05729336.7A patent/EP1728140B1/fr not_active Not-in-force
- 2005-02-17 JP JP2007501304A patent/JP4829876B2/ja not_active Expired - Fee Related
- 2005-02-17 US US10/598,442 patent/US7889764B2/en not_active Expired - Fee Related
- 2005-02-17 WO PCT/FR2005/000379 patent/WO2005096123A2/fr active Application Filing
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0042924A2 (fr) * | 1980-06-30 | 1982-01-06 | International Business Machines Corporation | Appareil de transfert de données |
US20020150189A1 (en) * | 1997-10-10 | 2002-10-17 | Rambus Incorporated | Apparatus and method for generating a distributed clock signal using gear ratio techniques |
EP1380918A2 (fr) * | 2002-07-12 | 2004-01-14 | Ensequence, Inc. | Appareil et procédé associé permettant de synchroniser l'operation d'un compteur temporel éloigné avec une centrale de commande |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
NO20076533L (no) * | 2007-12-18 | 2009-06-19 | Tandberg Telecom As | Fremgangsmåte og system for klokkedriftskompensering |
WO2009078733A1 (fr) * | 2007-12-18 | 2009-06-25 | Tandberg Telecom As | Procédé et système de compensation de dérive d'horloge |
US8515086B2 (en) | 2007-12-18 | 2013-08-20 | Trygve Frederik Marton | Method and system for clock drift compensation |
Also Published As
Publication number | Publication date |
---|---|
JP2007526570A (ja) | 2007-09-13 |
EP1728140B1 (fr) | 2017-04-26 |
FR2867288A1 (fr) | 2005-09-09 |
FR2867288B1 (fr) | 2006-06-02 |
US7889764B2 (en) | 2011-02-15 |
US20070297457A1 (en) | 2007-12-27 |
EP1728140A2 (fr) | 2006-12-06 |
WO2005096123A3 (fr) | 2006-03-02 |
JP4829876B2 (ja) | 2011-12-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP1728140B1 (fr) | Procede de synchronisation de donnees, notamment distribuees, prenant en compte les imprecisions et derives des horloges. | |
WO2009043923A1 (fr) | Procede de synchronisation de flux video | |
EP2572319A1 (fr) | Procede et systeme pour fusionner des donnees issues de capteurs d'images et de capteurs de mouvement ou de position | |
EP3004912A1 (fr) | Procédé et dispositif de métrologie pour la calibration de la géométrie d'un réseau de balises acoustiques sous-marines | |
EP2572164A1 (fr) | Aide a la navigation a traitement de donnees d'affichage perfectionne | |
EP3793142B1 (fr) | Communication par bus can | |
EP2751637A1 (fr) | Procédé de synchronisation d'une grappe de serveurs et grappe de serveurs mettant en oeuvre ce procédé. | |
EP1118943B1 (fr) | Dispositif de régénération d'une horloge à partir d'au moins deux bits de synchronisation | |
EP3793132B1 (fr) | Communication par bus can | |
FR3093606A1 (fr) | Procédé et dispositif de correction de l’heure définie par une horloge interne d’une entité | |
EP4033680A1 (fr) | Procédé de synchronisation de domaines temporels d'un système sur puce | |
FR3073957B1 (fr) | Systeme et procede de datation d'un evenement detecte dans un vehicule automobile | |
FR2997206A1 (fr) | Unite de calcul d'un appareil de commande et son procede de gestion | |
EP1120714B1 (fr) | Dispositif de régénération d'une horloge | |
FR3065343A1 (fr) | Procede de traitement d’informations acquises par une pluralite de capteurs d’un vehicule automobile | |
EP1580893B1 (fr) | Dispositif et procédé de détection de phase d'un signal | |
EP1127305A1 (fr) | Procede et dispositif de synchronisation d'une horloge locale sur l'horloge d'un reseau de communication sans fil | |
FR2676152A1 (fr) | Procede, systeme et dispositif de synchronisation des horloges d'un reseau. | |
EP2791687B1 (fr) | Procede et systeme d'auto-calibration d'un dispositif de mesure de vitesse d'un vehicule circulant dans un espace tridimensionnel | |
JP7482291B2 (ja) | クロック同期 | |
WO2020104567A1 (fr) | Système et procédé de mise à jour d'horloge | |
WO2005064271A1 (fr) | Procede de capture du mouvement d'un solide, utilisant une mesure absolue associee a une mesure par double integration | |
EP3469986B1 (fr) | Procede de synchronisation d'un systeme de localisation magnetique | |
EP0077249B1 (fr) | Procédé de chronométrie, système de gestion centralisée mettant en oeuvre ce procédé, et son application à la détection de fuites dans un réseau de transport de fluides | |
EP0343062B1 (fr) | Procédé d'émission de données avec compression et de réception de données avec expansion sur ligne téléphonique numérisée |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A2 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NA NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SM SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW |
|
AL | Designated countries for regional patents |
Kind code of ref document: A2 Designated state(s): BW GH GM KE LS MW MZ NA SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LT LU MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2007501304 Country of ref document: JP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWW | Wipo information: withdrawn in national office |
Ref document number: DE |
|
REEP | Request for entry into the european phase |
Ref document number: 2005729336 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2005729336 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 2005729336 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 10598442 Country of ref document: US |
|
WWP | Wipo information: published in national office |
Ref document number: 10598442 Country of ref document: US |