FR2658969A1 - SYSTEM CONSTITUTED IN A NETWORK SUCH AS A CELLULAR RADIOTELEPHONE SYSTEM, FOR MEASURING THE TRANSMISSION DELAY BETWEEN NODES OF THE NETWORK AND SYNCHRONIZING THEM BETWEEN THEM. - Google Patents

Abstract

A networked system, such as a cellular radiotelephone system, is provided which allows network nodes to be synchronized to a pilot node. The pilot node (201) starts a timer (209) at the same time as it sends (203) a predetermined signal (token) to a target node. Upon receipt of the token, the target (221) starts a timer (229) and waits for the next available time when it can return the token to the pilot. As this is done, the target stops its timer, which produces (225) a token "hold time". Upon receipt of the token, the pilot stops its timer, which produces (205) an "elapsed time" for the token. The delay time of the channel interconnecting a target and the pilot can then be calculated (205; 225) as the half difference between the elapsed time and the guard time. Based on this, the target can then be synchronized to the pilot.

Description

The present invention relates to a system for

  synchronize the nodes of a network, for example a radio system

  cellular telephone containing several nodes.

  Networked systems, such as systems

  cellular radiotelephones, for example, are well known.

  Thus, these systems typically have a large number of nodes which are dispersed over a large geographical area. It is understood that it is necessary to be able to control the temporal relationships existing between the various constituents of such a system. Typically, this means that it is necessary to order all the clocks of the respective nodes relative to a "pilot clock" valid for the whole system, the clocks not necessarily being associated with those which can intervene in the transport of

  data inside the network.

  Therefore, the time relationship must be synchronized.

  real existing between all the clocks of the system, not only in a relative direction, but also in an absolute direction Each node of the system must therefore be controlled so that its clock has the same phase and the same frequency, or almost, as the pilot clock In addition, each clock of a node must also be synchronized so that it has the same absolute reading as the pilot clock. It is an object of the invention to provide a system which can be used to synchronize adjacent nodes in a

  network made up of a system, for example a radiotelephone system

  cell nick therefore it is produced a technique

improved timing.

  According to the invention, the pilot node starts a timer at the same time as it sends a predetermined signal called "token" to a target node. Upon receipt of the token, the target node starts a timer and stores the token. The target node waits then the next available time when it can return the token to the pilot node When it later returns the token to the pilot node, the target node stops its timer, so that a "guard time" of the token is produced Upon receipt of the token , the pilot node stops its timer, so that a "time el Lé" is created for the token The delay time relative to the channel interconnecting a target node and a pilot node can therefore be calculated Lé in the form of the half -difference between elapsed time and guard time On the basis of this time delay information, the target node can also be then synchronized

on the pilot node.

  The following description, intended as an illustration of

  the invention aims to give a better understanding of its characteristics and advantages; it is based on the appended drawings, among which: FIG. 1 is a block diagram showing a communication system comprising several nodes; Figure 2 is a block diagram showing a

  first embodiment of an improved synchronization system

station according to the invention;

  FIG. 3 is a first flow chart according to the invention.

tion;

  Figure 4 is a second flowchart according to the invention.

  tion; Figure 5 is a third flowchart according to the invention; and Figure 6 is a fourth flowchart according to the invention. FIG. 1 shows a block diagram showing a networked system, for example a cellular radiotelephone communication system, comprising several nodes. This system comprises a hierarchical set of nodes,

  like for example 111, 113, 115, 117 and 119, which are interconnected

  nected in a tree through channels such as for example 101, 103, 105 and 107 Each node of the network can for example be interconnected using a channel of type T 1 or CEPT, or else by another means any digital transmission With a

  similar communication system, we will understand that it is necessary

  keep synchronization throughout the network As we will see, it is also possible to use other types of channels There is virtually no Limit on the number of nodes

which can be synchronized.

  As shown in FIG. 1, the node 111 located at the top of the tree structure contains a source of reference time signals. The invention aims to produce a technique making it possible to synchronize other nodes on this source of time signals. context, the node that contains the source of reference clock signals is called the "pilot" node, while the adjacent node, which is synchronized on the pilot node, is called a "target" node So if node 111 is in the process of synchronizing the node 113, then, in this context, the node 111 is the pilot node and the node 113 is the target node Although, in FIG. 1, it is shown that the node 111 being at the top of the tree structure contains the source of reference time signals, any node of the network can contain this source of reference time signals For example, if node 117 of FIG. 1 should contain the source of time signals s reference, the synchronization sequence would start from node 117 to synchronize node 113, then pass from node 113 to node 111 to synchronize the latter, after which node 111 would synchronize

  the other nodes to which it is connected.

  Figure 2 is a block diagram showing a pilot node 201 coupled to a target node 221 via a channel 261 It is assumed that the channel 261 is an element of the T1 type

with 24 time slots.

  The system of the invention allows the node 201, called

  pilot node synchronizes node 221, called target node Syn-

  timing of Target 221 on pilot 201 can be set

  works with the help of an algorithm now described that we understand

  bine with a material means of time measurement specially designed for this purpose, which is present on the interface of each node with its telecommunication interconnection This synchronization system does not require that this interconnection be specially reserved for the synchronization process, but simply that a time slice taken in this interconnection operation in association

  with channel 261 be allocated to the synchronization process.

  Furthermore, it is only necessary to use the time slot for the duration of the synchronization operation, which is

typically a few seconds.

  Inside each node, there is an interface with the respective interconnection means, which will be called interconnection control device (ICC), a calculation means which will be called central processing unit (CPU), and a source

  local clock signals which will be designated by the reference CLK.

  Thus, the pilot node 201 contains an ICC 203, a CPU 205 and a CLK 207 Similarly, the target node 221 contains an ICC 223, a

CPU 225 and a CLK 227.

  When the network is fully synchronized, all

  CLKs in the network have a predetermined time relationship.

  According to the invention, the pilot first establishes a communication path 261 via a time slot, designated for this purpose, of the telecommunications interconnection with the target The pilot 201 then starts his timer 209 and simultaneously sends a signal predetermined, called "token", to target 221 The operation in which the pilot node 201 initially sends the token to

  target node 201 is represented by the arrow 263 in FIG. 2.

  Upon receipt of the token, the target node 221 starts its timer 229 and stores the token. The target node then waits for the next available time slot by which it can return the token to the pilot node 201 The operation in which the target node 221 keeps temporarily the token until it is able to return it to the pilot node 201 is

  represented by arrow 265 in Figure 2.

  When the appropriate time frame becomes available

  target node 221 stops its timer 229 and sends simul-

  the token at the pilot node 201 The duration for which the target node 221 retains the token before returning it to the pilot node is defined as "on-call time", or occupation time, and can be conveniently measured by the timer 229 The operation in which the target node 221 returns the token to the pilot node 201 is represented by The arrow 267 on the

figure 2.

  Upon receipt of the token, the pilot node 201 stops its timer 209 The time elapsing between the time when the pilot node 201 initially sends the token to the target node 221 and the time when it subsequently receives it from the target node 221 is called the "elapsed time", and can be conveniently measured by

timer 209.

  It will be understood that, as soon as one of the nodes 201 and 221 has become aware of the two durations, the guard time and the elapsed time, it can calculate the delay At relative to the channel 261 on the basis of the half difference between the elapsed time and the guard time In addition, as soon as the target node 221 becomes aware of the value of At, the target node can be synchronized with the pilot node 201 in the following manner The pilot node 201 informs the target node 221 that it will send a predetermined signal to the target at a predetermined time TO O Then, the pilot node 221 waits to receive this predetermined signal At the moment when it receives this signal, the target node 221 adjusts its clock 227 to the value

T + At.

  The operation described above provides a resolution

  lution better than 1 juice for absolute synchronization from one node to another, and it can be used to synchronize everything

a network of elements.

  The ICC device has three synchronization modes,

as shown below.

  In the first mode, the ICC device will offer the possibility

  ability to produce a single bit pattern to be transmitted over a given interconnect time frame, to start a timer when that pattern is issued, and to stop the timer when the pattern is

  returned, on the corresponding entry time slot The reso-

  lution of the timer will be related to the clock used on the interconnection (1.544 M Hz for Tl, or 2.048 M Hz for CEPT) The

CPU can read this timer.

  In the second mode, the ICC device offers the possibility

  ability to start the clock when the single bit configuration is received on a given interconnection time slot and to stop this timer, or time counter, when this configuration is re-transmitted, on the corresponding transmission time, as a result of an internal loop mechanism La

  CPU can also read this timer.

  The third mode of the ICC device extends the second mode so that it includes delivering a signal to its

  CLK clock in competition with the recognition of the configuration

  single bit tion on the designated incoming time slot of the given interconnection The CLK will accept this signal and initialize

  its time chain to a value preset by the CPU.

  The interconnection interface present on the ICC plate should normally contain an elastic buffer capable of accommodating the time differences between the input and output data rates. To be more precise, it will be said that the elastic buffer is here intended to hold account for the difference in "phase" between incoming and outgoing data Due to the fact that the CLK is synchronized with the input interconnect, the input clock frequency and node clock frequencies can be considered as equal, although there may be a difference in phase (implementing the preferred embodiment will further establish phase equality and therefore give better synchronization resolution) While this may not not be true in the long term due to drift, it is true in the short term (for a duration measured in seconds) As a result, the time that any given data spends da ns the elastic buffer can be considered as constant during

  the interval required for the synchronization operation.

  The ICC device uses the CLK clock as its own source of clock signals with regard to the data which is transmitted via the interconnection. Thus, all the information which is transmitted on the interconnection is aligned in terms of blocks on the CLK time signal chain and are synchronized

with these.

  The information that is received on the Interconnection is not aligned in terms of blocks with the time chain of The CLK

  nor is it synchronized with it.

  The synchronization operation can be performed via the CPU / ICC assembly in the driver and the target according to the procedure

  which is defined below In this discussion, the actions between

  taken by the pilot and, or, the target indicate a cooperative effort implementing the CPUs of the two ends of the interconnection It is assumed that the ICC devices are directed by their respective CPUs The term "pilot" will be used to indicate the node which controls the synchronization operation, and the term "target" will be used to indicate the node which is in

being synchronized.

  The procedure includes the following operations.

  Operation 1 The pilot sends a message to the target, asking it to enter a mode by which it will buckle on a

  designated time slot of a designated interconnect.

  Operation 2 The cib The place The designated time frame

in loopback mode.

  Operation 3 Coinciding with the next main block lock clock signal (in seconds), the pilot's ICC device will transmit to the target ICC device the unique bit pattern. This bit pattern will be on the edge of Designated Interconnection Designated Time The pilot's ICC device will start counting equipment when the single bit configuration is issued. Operation 4 The target's ICC device will search for the

  designated time slot of designated interconnection is related

  both to the unique bit configuration When it has found this unique bit configuration, it will start its counting equipment When the unique bit configuration will have been returned to the pilot as a result of the fact that the target ICC device is in a loopback mode on the designated time slot, the counter will be stopped The count value represents the time spent in the target The target will send the measured value to the pilot

by the counter.

  Operation 5 When the single bit configuration is received in the pilot's ICC device after being returned by the target, the pilot's ICC device will stop sounding

  counting equipment The value measured by the counter represented

  feel the total duration of the round trip.

  Operation 6 It is possible to repeat several

  times operations 1 to 5 and represent them as histo-

  gram, if necessary, to ensure that the measured times are reproducible The pilot then calculates the one-way time, which is obtained by dividing by two the result of the difference between the value measured by the pilot's counter and the

  value measured by the target counter.

  Operation 7 The pilot then sends a message to the target asking it (1) to remain without its looping mode on the designated time slot of the designated interconnection, but now validating the synchronization signal for its CLK clock, and ( 2) load a specified value, as calculated

  in operation 6 above, in the prepositioning mechanism

  ment of its CLK clock, as part of its time chain.

  Operation 8 The pilot's ICC device transmits the unique bit configuration to the target, coinciding with the next main block clock signal When this unique bit configuration is received by the target, the value noted in (2) of operation 7 above is loaded in the

  CLK clock time chain of the target When the confi-

  Single bit guration is returned to the pilot's ICC device due to the fact that the target's ICC device is in loopback mode, the synchronization operation is then complete At this time, the CLK clock synchronization time strings

  of the pilot and the target are synchronized better than 1 ps.

  Referring now to FIG. 3, which represents a first flow diagram according to the invention This flow diagram describes a

  method for calculating the delay relating to the inter-channel

  connecting the pilot node and the target node In this organization

  gram, we assume that the pilot node finally calculates the value

of this delay time.

  We start first with the pilot side The pilot node

  starts the process at step 301, then goes to step 303.

  There, the pilot sends a unique bit configuration, or token, to the target. Coincidentally with this step 303, the pilot starts a clock, or timer, in step 305. It will be noted that the timer cited here may be, for example for example, any suitable time signal source, hardware or software based,

  which is available in the pilot node.

  The token is then transmitted to the target via the channel. This transmission or sending operation is represented by the dashed line 321 The pilot then goes to step 307, o

  it waits to receive the token from the target.

  We now refer to the target node Initially, the target node waits for the pilot to receive the token. This is represented by the determination operation 331, where the target

  waits, until it has determined that the token has been received.

  After receiving the token, the target starts a

  clock, or timer, in step 333 Note that the minu-

  The term cited here can be, for example, any suitable source of time signals, based on hardware or software, which

is available in the target node.

  The target then goes to step 335, where it waits for the

  next available time slot of the inter-

  connection T 1, to send the token to the pilot When the section

  time has become available, the token is sent The operation

  ration consisting in returning the token to the pilot is represented by

dashed line 323.

  After having returned the token to the pilot, the target then goes to step 337, where it determines the duration for which it has kept the token. In other words, the "guard time" is the interval of time which begins upon receipt of the token by the

  target and ends when the target sends the Natural token -

  ment, this guard time can be obtained directly using the target's time signal source, which source was

  previously triggered in Step 333.

  After having produced the guard time, the target proceeds to step 339, where it sends the information giving the guard time to the pilot. The target can accomplish this task by coding, for example, the guard time in the form of '' one or more words

  binaries and sending them to the pilot via the Tl channel This operation

  the amount of on-call time sent to the pilot is represented by the

dashed line 325.

  We return to the pilot The pilot's next task is to wait to receive the token from the target This is

  determined by determination operation 307, where the pilot deter-

  mine whether or not it has received the token After receiving the token, the pilot goes to step 309, where he determines the time elapsed since the token was sent to the target In other words, this "elapsed time "is the time interval that begins with the token being sent by the

  pilot and ends when the pilot receives the Natural token-

  ment, this elapsed time can be directly obtained using the pilot's time signal source which was previously

triggered in step 305.

  The pilot then waits to have received the information relating to the on-call time from the target. This is presented by the determination step 311, where the pilot determines whether or not he has received the on-call time. received the guard time, the pilot goes to step 313, where he calculates the delay on the basis of the difference between the elapsed time and the guard time. After calculating the delay time, the flowchart

  performs a return (step 315) to the start step 301.

  Reference is now made to FIG. 4 which represents a second flowchart according to the invention. This flowchart describes a

  another method for calculating the delay due to the inter-channel

  connecting the pilot node and the target node In this organization

  gram, it is assumed that it is the target node which ultimately

  calculates the value of this delay time.

  We start with the pilot node. This starts the operation in step 401, then goes to step 403 There, the pilot sends the token to the target and starts a clock, or a

timer, in step 405.

  The token is then sent to the target, which is represented

  felt by the broken line 421 The pilot then goes to step 407, where he waits to receive the token from the target. If we now consider the target node, we see that it begins by waiting to receive the token from the pilot, in step 431 As soon as he received the token, the target does

  start a clock, or timer, in step 433.

  The target then goes to step 435, where it waits for the

  next available time slot of the T 1 inter installation

  connection, to return the token to the pilot When the desired time slot becomes available, the token is sent, which is

  represented by the broken line 423.

  After returning the token to the pilot, the target deter-

  then mines the guard time, in step 437 This guard time can naturally be obtained directly using the target time signal source, which source was previously triggered in step 433 After having produced the guard time, the target goes to step 461, where it waits to receive the information

  relating to the time elapsed, on the part of the pilot.

  The pilot's next task is to wait, until the token is received from the target. This is described by determination step 407, where the pilot determines whether he has or not received the token After receiving the token, the pilot goes to step 409, where he determines the elapsed time This elapsed time can naturally be obtained directly using the source of the pilot's time signals, which has

  previously triggered in step 405.

  After having produced the elapsed time (step 409), the pilot goes to step 451, where he sends the target the information relating to the elapsed time. The pilot can accomplish this task by coding for example the elapsed time one or more binary words and by sending them to the target via the channel T 1 This operation of sending the elapsed time to the target is

  represented by the broken line 453.

  We return to the target again. This then waits to receive the information relating to the elapsed time from the pilot. This is described by the determination step 461, where the target determines whether or not it has received the elapsed time. received the elapsed time, the target proceeds to step 463, where it calculates the delay time based on the difference between the elapsed time and the

on-call time.

  After the delay time has been calculated, the flowchart then returns (step 465) and returns to the start step 401. Reference is now made to FIG. 5, which represents a third flowchart according to the invention This flowchart describes a method in which the target node synchronizes itself with the pilot node, on the basis of the value (At) of the delay time of the interconnection channel In this flowchart, we will assume

  that the pilot node calculates the value of the delay time.

  We start with the pilot node The flowchart begins at step 501, then goes to step 503 There, the pilot node produces the value of the delay time At The pilot node can produce this delay time using the process described in Figure 3 by

example.

  After having produced the delay time (At), the pilot node goes to step 505, where it sends the target delay time information to the target The pilot node can accomplish this task by coding for example the value of time delay in the form of one or more binary words and by sending them to the target via the channel T 1 The operation in which the pilot node sends the value (At) of the delay time to the target node is represented by

the broken line 521.

  After sending the delay time value to the target, the pilot node goes to step 507, where it sends a message to the target node informing the latter of the existence of an instant to come (T 0) o the pilot node will send a synchronization signal to the target o The pilot node can transmit this message by any suitable method, for example by coding the message in the form of one or more binary words and sending these words to the target via the channel T1 The operation in which the pilot node sends the message indicating the instant (T) when the pilot node will send the synchronization signal to the target node is represented by the line in

broken line 523.

  The pilot node then goes to step 509, where L waits for equality "t = T" At the precise instant ot = T 0, the pilot node goes to Step 511, where it sends the synchronization signal to target node This synchronization signal can naturally be a single bit configuration which is identical to the token used to measure the delay time The operation in which the pilot node sends the synchronization token at the precise time ot =

  T O is represented by the broken line 525.

  We now return to the target node. This begins by waiting to receive the information relating to the delay time.

  on the part of the pilot, in step 531 After having received this information

  mation, the target node goes to step 533, where it receives the message indicating that the pilot will send a synchronization signal to the target at the precise time o t = T The target node then goes to

  step 535, where it receives the synchronization signal.

  Due to the delay of the channel, ie At, the target receives this synchronization signal, however, not at time t = T 0, but at a later time t = T + At The target node goes o then to step 537 o he synchronizes, or sets, his own clock

  based on the relation t = T + At.

o

  The synchronization operation being completed, the organizer

  gram returns (step 539) and returns to the start step

501.

  Reference is now made to FIG. 6 which represents a fourth flowchart according to the invention. This flowchart describes a method in which the target node synchronizes itself on the pilot node, as a function of the value of the delay time (At) of the interconnection path In this flowchart, we will assume

  that the target node calculates the value of this delay time.

  We start with the target node The flowchart starts at step 601, then goes to step 603 There, the target node calculates and produces the value of the delay time, ie At The target node can produce this value of the time of delay using by

  example the process of figure 4.

  We now consider the pilot node At the appropriate time (step 607), the pilot node sends a message to the target node informing the latter of the existence of an upcoming time (T) o to which the pilot node will send a signal of synchronization at the target node The pilot node can transmit this message by any suitable method, for example by coding the message in the form of one or more binary words and by sending them to the target node via the channel T 1 The operation in which the pilot node sends the message indicating the instant (T) when the pilot node will send the O synchronization signal is represented by the line

interrupted 623.

  The pilot node then goes to step 609, where it waits for the relationship "t = To" to be verified. At the precise instant ot = To, the pilot node goes to step 611, where it sends the synchronization signal at the target node Naturally, the synchronization signal can be a single bit configuration which is

  identical to the token used to measure the delay time The ope-

  ration in which the pilot node sends the synchronization token

  sation at the precise instant o t = T is represented by the line at o

broken line 625.

  We return again to the target node At the appropriate time

  requested (step 633), the target node receives the message indicating that the pilot node will send a synchronization signal to the target node at the precise time o t = T The target node then goes to step o

  635, where it will receive the synchronization signal.

  Due to the delay time At of the channel, the target receives this synchronization signal, however, not at time t = To, but at a later time t = T + At The target node goes O next to step 637 , o it synchronizes, or else regulates, its own

  clock as a function of the relation t = T + At.

  The synchronization operation being completed, the organizer

  gram performs a return (step 639) and returns to the start step 601. Of course, those skilled in the art will be able to imagine,

  from the system whose description has just been given as

  merely illustrative and in no way limiting, various variants

  and modifications outside the scope of the invention.

Claims (10)

  1 cellular radiotelephone system, comprising at least one pilot node coupled to a target node via a telecommunications channel such as a Tl or CEPT line, the system being designed to measure the delay time of said telecommunications line characterized in that: said pilot node (111; 201) comprises sending means (203) for sending a first signal to said target node, and means (205) responding to said sending means by starting a first timer (209); said target node (113; 221) comprises receiving means (223) for receiving said first signal, means (225) which responds to said receiving means by starting a second timer (229), means (223) serving sending a second signal to said pilot node, means (225) which responds to said first sending means by   producing a guard time, on the basis of said second minute   terie, and a second sending means (223) for sending said guard time to said pilot node; and said pilot node further comprises reception means (203) for receiving said second signal, first production means (205) which responds to said reception means by producing an elapsed time, based on said first timer, a second production means for producing said hold time, and means (205) for calculating said delay time based on the difference between said elapsed time and said on-call time.   2 Cellular radiotelephone system according to resale   dication 1, characterized in that the second means for sending said target node comprises means for sending a third signal to said pilot node on the basis of said guard time, and the second means for producing said pilot node comprises means for   receiving said third signal from said target node.   3 Cellular radiotelephone system, comprising at least one pilot node coupled to a target node via a telecommunications channel such as a T 1 or CEPT line, the system being designed to measure the delay time of said line telecommunications, characterized in that: said pilot node (111; 201) comprises: first sending means (203) for sending a first signal to said target node, and means (205) which responds to said first sending means by starting a first timer (209); said target node (113; 221) comprises: receiving means (223) for receiving said first signal, means (225) which responds to said receiving means by starting a second timer (229), means (223) serving to send a second signal to said pilot node, and   a means (225) which responds to said sending means by producing it   health guard time based on said second timer; said pilot node further comprises: receiving means (203) for receiving said second signal, means (205) which responds to said receiving means by   producing an elapsed time based on said first minute   * terie, and a second sending means (203) for sending said elapsed time to said target node; and said target node further comprises: second producing means (225) for producing said elapsed time, and means (225) for calculating said delay time based on the difference between said elapsed time and said on-call time.   4 Cellular radiotelephone system as claimed   cation 3, characterized in that the second means for sending the pilot node comprises means for sending a third signal to said target node on the basis of said elapsed time, and the second means for producing a target comprises means for   receiving said third signal from the pilot node.   5 Target node intended for use in a cellular radiotelephone system, said target node being intended to be coupled to a pilot node via a channel and comprising a means making it possible to measure the delay time of said channel, characterized in what it includes: first receiving means (223) for receiving a first signal sent from said pilot node, means (223) for sending a second signal to said pilot node, first producing means (225) generating a guard time based on the difference between the time when said target node has received said first signal and the time when said target node has received said second signal, second receiving means (223) for receiving   an elapsed time on the part of the pilot node on the basis of the difference   between the time when said pilot node sent said first signal and the time when said pilot node received said second signal, and means (225) for calculating said delay time based on the difference between said time elapsed and said on-call time.   6 Target node intended for use in a cellular radiotelephone system, said target node being intended to be coupled to a pilot node via a channel and comprising a means which makes it possible to measure the delay time of said channel, characterized in that it comprises: a first reception means (223) serving to receive a first signal sent by said pilot node, means (223) serving to send a second signal to said pilot node, OS means (225) serving producing a guard time based on the difference between the instant when said target node received said first signal and the instant when said target node sent said second signal, and means (223) for sending said delay time keep audit pilot node.   7 Cellular radiotelephone system comprising at least one target node, said target node being intended to be coupled to a pilot node via a channel and comprising a means which makes it possible to measure the delay time of said channel, characterized in that that it comprises: a first reception means (223) serving to receive a first signal sent by said pilot node, means (223) serving to send a second signal to said pilot node, a first production means (225) serving producing a guard time based on the difference between the time when said target node received said first signal and the time when said target node sent said second signal, a second receiving means (223) for receiving a elapsed time from said pilot node based on the difference between the time when said pilot node sent said first signal and the time when said pilot node received said second signal, and means (225) for calculating said delay time based on the difference between said elapsed time and said on-call time.   8 Cellular radiotelephone system comprising at least one target node, said target node being intended to be coupled with a pilot node via a channel and comprising a means which makes it possible to measure the delay time of said channel, characterized in that that it comprises: a first reception means (223) serving to receive a first signal sent by said pilot node, means (223) serving to send a second signal to said pilot node, means (225) serving to produce a time guard based on the difference between the time when said target node received said first signal and the time when said target node sent said second signal, and means (223) for sending said guard time audit pilot node.   9 Cellular radiotelephone system, which comprises at least one target node having a clock and coupled to a pilot node via a channel, said system being designed to synchronize the clock of the target node with said pilot node, characterized in that that: said pilot node (111; 201) comprises: means (205) serving to produce the delay time of said channel, first sending means (203) serving to send said delay time to said target node, a second sending means (203) for sending to said audit   target node a message that it will send a predefined signal   completed at a predetermined time, and third sending means (203) for sending said predetermined signal; and said target node (113; 221) comprises: first reception means (223) for receiving said delay time, second reception means (223) for receiving said message, third reception means (223) for serving receiving said predetermined signal, and means (225) for setting said clock (227) to   the basis of said predetermined instant and said delay time.   10 cellular radiotelephone system which comprises at least one target node having a clock and coupled to a pilot node via a channel, said system being designed to synchronize the clock of the target node on said characterized in that: said node pilot (111; 201) comprises: a first sending means (203) serving as target node a message indicating that it will send a finished one at a predetermined time, and a second sending means (203) serving as predetermined signal; and said target node (113; 221) comprises: pilot node, sending signal to said prede-   sending said means (225) for generating the delay time of said channel, first receiving means (223) for receiving said message, second receiving means (223) for receiving said first predetermined signal, and a means (225) for adjusting said clock (227)   based on said predetermined instant and said delay time.