Method for allocating uplink time resources in a fibre-optic communication system, and corresponding uplink transmission method, user module and optical line terminal

Abstract

The invention relates to a technique for allocating time resources in a fibre-optic communication system comprising an optical line terminal connected to a plurality of user modules via a single optical fibre, the optical line terminal implementing a schedule for allocating time resources to the user modules updated periodically, allocating to a user module a time slot for the transmission of at least one frame from the user module. Such a technique is based on the implementation of the following steps: updating (100) the duration of the header fields, establishing (200) the allocation schedule taking into account the update of the duration of the header fields, and broadcasting (300) the established allocation schedule to all the user modules.

Description

Upstream time resource allocation method in a fiber optic communication system, upstream transmission method, corresponding user module, and optical line terminal

[0001] The invention falls within the field of optical fiber communications, and in particular within the field of passive optical networks or PON networks (for "Passive Optical Network" in English).

[0002] More specifically, the invention relates to the optimization of data traffic flow and in particular the management of a temporal resource allocation schedule in a PON network. State of the art

[0003] In recent years, increasing user expectations for traffic and quality of service have led to the development of new architectures and protocols for optical communication networks.

[0004] In particular, the PON architecture, to which the invention applies, is a passive point-to-multipoint architecture that provides high-speed optical access to a user population, for example, a remote residential population (typically distributed over several tens of kilometers). This architecture is characterized by the absence of active equipment along the branches of the optical fiber network connecting the optical exchange to the user modules.

[0005] Traditionally, a PON network comprises, as illustrated in the figure, an optical line terminal (OLT) located at the optical exchange (CO) connected to an optical coupler C by a single optical fiber OF, as well as a set of user modules (ONUs for Optical Network Units) designated ONU-1 to ONU-m (m typically being up to 64). The user modules ONU-1 to ONU-m are respectively connected to the coupler C by means of dedicated optical fibers OF-1 to OF-m.

[0006] The OLT transmits a downlink optical signal to the ONUs, and each ONU transmits an uplink optical signal to the OLT within its own specific time interval. To enable the OLT to identify the uplink optical signal specific to each ONU, a transmission schedule is pre-defined for each ONU, allocating it a time window for transmitting optical data in the form of a frame or an optical burst. In other words, this transmission schedule defines the allocation of time resources to the various ONUs in the network. The OLT is therefore configured to optically serve the ONUs via the passively shared OF fiber through the coupler / combiner C, according to a 1-to-m topology and a defined communication schedule.The process of allocating temporal resources in a PON aims in particular to ensure that each user receives a fair share of available traffic allocation while optimizing overall network performance.

[0007] A first known traffic flow allocation protocol consists of allocating a given amount of bandwidth (i.e., a specific quantity of traffic): this is called a fixed bandwidth allocation. However, this type of protocol cannot meet demand beyond the fixed allocation and will generate, on the one hand, underutilization when clients do not use their bandwidth, and on the other hand, a significant bandwidth limitation. Furthermore, this type of protocol is not suitable for IP (Internet Protocol) traffic.

[0008] Thus, a second known bandwidth allocation protocol involves dynamically adjusting the time allocation based on real-time traffic conditions and user requests: this is referred to as dynamic bandwidth allocation or DBA (for "Dynamic Bandwidth Allocation"). Furthermore, this protocol provides a better match between IP flows and bandwidth management, and also enables a high level of integration between heterogeneous QoS (Quality of Service) services.

[0009] In the case of dynamic bandwidth allocation for uplink communication, the resource allocation schedule includes transmit windows dynamically allocated to each of the ONUs. This allocation schedule is updated periodically by the OLT. Each transmit window constitutes a time slot of traffic allocated to one of the network's ONUs for the transmission of an uplink optical signal to the OLT. Each time slot is itself composed of several time intervals, primarily intended for the transmission of header data and payload data.

[0010] However, the time intervals for transmitting header data are identical for each of the ONUs, even though they have different transmission parameters. Thus, some ONUs do not require such a large time interval, while others would require a larger one, and the allocation schedule therefore appears suboptimal.

[0011] Therefore, there is a need to optimize the management of the communication schedule between an OLT and ONUs within a PON network. In particular, there is a need to optimize the time intervals allocated to the transmission of header data.

[0012] The present invention aims to resolve all or part of these drawbacks. Presentation of the invention

[0013] In a particular embodiment of the invention, a method for allocating time resources in a fiber optic communication system is proposed. This method includes an optical line terminal connected to a plurality of user modules via a single optical fiber. The optical line terminal implements a periodically updated time resource allocation schedule for said user modules, allocating to each user module a time slot for the transmission of at least one frame. This at least one frame comprises: a header field, including data relating to optoelectronic physical layer functions; and a payload field, including payload data.

[0014] the process comprising the following steps, implemented by the optical line terminal: updating the duration of at least one header field; establishing the allocation schedule, taking into account the update of the duration of said at least one header field; disseminating the allocation schedule to all said user modules.

[0015] Thus, the present invention is based on a new approach consisting of optimizing the timing of allocation of time resources allocated to the different user modules of the communication system by an operation of updating the duration of the header fields associated with these user modules for the transmission of uplink optical frames.

[0016] According to one particular implementation, the duration update step consists of reducing the duration of at least one header field to make at least one time resource available and allocating that time resource to at least one load field. Thus, the allocation schedule optimization relies on a temporal optimization of the header fields, allocating fewer resources to header fields and more resources to guard fields of uplink optical frames. In other words, this approach allows more time to be dedicated to traffic allocation and less to optical frame header patterns.

[0017] According to a particular implementation, the update of the duration of said at least one header field is performed for a given user module of said plurality.

[0018] According to one implementation variant, the update of the duration of said at least one header field is performed for a pair of user modules succeeding one another in the allocation schedule.

[0019] In a particularly advantageous implementation, the process includes at least two iterations of the duration update step, in which, at each iteration, the duration of a candidate header field is reduced as long as a quality criterion remains above a predetermined threshold. This allows for a gradual and precise adjustment of the header field duration.

[0020] According to another feature of the method, said at least one header field comprising a guard time block and a preamble block carrying a signal enabling a user module to detect the power of an optical signal and / or to adjust the time phase of an optical signal, said duration update step consists of adjusting the duration of said guard time block and / or the duration of said preamble block. Thus, the allocation of time resources (also called speaking time) can relate to one or more time parameters of the calendar.

[0021] More specifically, since the preamble block consists of a repetition of a binary pattern, the duration update step consists of modifying the number of repetitions of said binary pattern.

[0022] According to a particular implementation, the adjustment of the duration of said guard time block and / or the duration of said preamble block takes into account at least one of the criteria belonging to the group including: information representative of the optical power received by the optical line terminal, from the user module for which the update is considered; information representative of the fiber distance between the optical line terminal and the user module for which the update is considered; information representative of the minimum power received and the maximum power received from the user modules already connected; information representative of the minimum distance and the maximum distance of the user modules already connected; information representative of the type of user module for which the update is considered.

[0023] According to a first implementation method, the process is implemented in different ways depending on whether a user module is in a hooking phase with the optical line terminal or in a traffic flow phase from at least one user module to the optical line terminal. The update step may include optimizing the duration of the preamble block prior to the hooking phase. The update step may also include iteratively optimizing the duration of the guard time block and / or the duration of the preamble block prior to the traffic flow phase.

[0024] According to a second implementation method, the process is implemented for an emergency shutdown phase of a user module, the update step including an iterative optimization of the duration of the preamble time, prior to said emergency shutdown phase.

[0025] In another particular embodiment of the invention, a method for transmitting frames by a user module is proposed, in a fiber optic communication system comprising an optical line terminal connected to a plurality of user modules via a single optical fiber, the optical line terminal implementing a periodically updated time resource allocation schedule for said user modules, allocating to a user module a time slot for the transmission of at least one frame from the user module, said at least one frame comprising: a header field, including data relating to optoelectronic physical layer functions; a payload field, including payload data,

[0026] the process comprising the following steps, implemented by said user module: receiving a time resource allocation calendar update signal; extracting information relating to a change in the duration of the header field of frames emitted by said user module, within said calendar from the update signal; constructing a header according to the extracted information; transmitting a frame including the constructed header.

[0027] Thus, on the user module side, receiving the update signal means that a time resource allocation must be implemented for the uplink frame to be transmitted. Specifically, the update signal includes data relating to the duration of the header fields associated with the user modules, from which each user module is configured to construct its own optical frame, taking into account the new time resource allocation specifically imposed upon it.

[0028] In another embodiment of the invention, an optical line terminal for a fiber optic communication system is proposed, comprising a plurality of user modules, said optical line terminal implementing a periodically updated time resource allocation schedule for said user modules, allocating to a user module a time slot for the transmission of at least one frame from the user module, said at least one frame comprising: a header field, including data relating to optoelectronic physical layer functions; a payload field, including payload data,

[0029] the optical line terminal comprising: means for updating the duration of at least one header field; means for establishing the allocation schedule, taking into account the update of the duration of said at least one header field; means for disseminating the allocation schedule to all said user modules.

[0030] In another embodiment of the invention, a user module for a fiber optic communication system is proposed, comprising an optical line terminal and a plurality of user modules, said communication system implementing a time resource allocation schedule allocating said user module a time slot for the transmission of at least one frame to the optical line terminal, said at least one frame comprising: a header field, including data relating to opto-electronic physical layer functions; a load field, including useful data;

[0031] the user module comprising: means for receiving a time resource allocation calendar update signal; means for extracting information relating to a change in the duration of the header field of frames emitted by said user module from the update signal; means for constructing a header according to the extracted information; means for transmitting a frame including the constructed header.

[0032] In another embodiment of the invention, a computer program product is proposed comprising program code instructions for implementing a process as described above, in any of its embodiments, when executed by a processor.

[0033] In another embodiment of the invention, a computer-readable recording medium is proposed on which a computer program is stored, comprising program code instructions for executing the steps of the method according to the invention as described above. Such a recording medium can be any entity or device capable of storing the program. For example, the medium can include a storage means, such as a ROM, for example a CD-ROM or a microelectronic circuit ROM, or a magnetic recording means, for example a USB flash drive or a hard drive. Alternatively, such a recording medium can be a transmissible medium, such as an electrical or optical signal, which can be transmitted via an electrical or optical cable, by radio, or by other means, so that the computer program it contains is executable remotely.The program according to the invention can, in particular, be downloaded onto a network, for example, the Internet. Alternatively, the storage medium can be an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of the method that is the subject of the aforementioned invention. Presentation of the figures

[0034] Other aims, characteristics and advantages of the technique will become apparent upon reading the following description, given for illustrative purposes only and not as a limitation, and which refers to the attached figures, in which:

[0035] , already described in relation to earlier art, schematically represents a traditional PON communication network;

[0036] represents an example of a time window made available to a user module for the transmission of an uplink optical frame to an optical line terminal;

[0037] is a flowchart of a particular embodiment of the process according to the invention;

[0038] presents an event diagram showing in a generic way the sequence of the process between a user module and an optical line terminal according to a particular embodiment of the invention;

[0039] presents an event diagram illustrating the step of updating the duration of a header field, according to an example of an embodiment of the invention;

[0040] represents, in the form of a flowchart, the phase of progressive adjustment of the duration of the header field according to an example of an embodiment of the invention;

[0041] presents the simplified structure of a device implementing the process according to a particular embodiment of the invention Detailed description of the invention

[0042] In all figures in this document, identical elements and steps are designated by the same numerical reference.

[0043] In the following description, we consider an example of an implementation of the invention in a PON (Passive Optical Access Network) type fiber optic communication system, such as the one illustrated in relation to the figure. This could be a PON based on a point-to-multipoint transmission mode (downstream TDM (Time-Division Multiplexing) and upstream TDMA (Time-Division Multiple Access), for example). Such an optical network comprises an optical line terminal (OLT) and a set of user modules (ONU1-ONU-m), typically 64 ONUs. The OLT is configured to optically serve all the remote user modules ONU1 to ONU-m using the OF optical fiber via the optical coupler C.

[0044] In this configuration, in order for the OLT to identify the uplink optical signal specific to each ONU, a time allocation schedule is periodically imposed on the ONUs, allocating each ONU a time slot for the transmission of an optical frame from that ONU to the OLT. This is referred to as uplink time allocation (or bandwidth allocation). The schedule is thus divided into a plurality of time slots (commonly called "talk time"), each allocated to one of the ONUs in the optical network, resulting in a succession of 64 time slots in the implementation example presented here.

[0045] An example of a time slot T ONU2 The method made available to an ONU2 user module for transmitting an uplink optical frame to the OLT is illustrated in the figure. Each successive time slot in the calendar is considered to be structured in the same way. The time slot T ONU2includes a first time interval (T plo ) dedicated to the transmission of an optical frame header field and a second time interval (T u ) dedicated to the transmission of a load field of the optical frame. The header field contains data intended for optoelectronic physical layer functions. The load field contains payload data. Thus, the second time slot constitutes a dedicated traffic time slot, which is allocated to the user module for the transmission of its payload data into the network.

[0046] The header field includes a guard time block GT and a preamble block PR. A first portion of the time interval T plo (reference T) g) is made available to the user module ONU2 to allow the guard time, which is a pause time to prevent the overlap of signals sent by two successive user modules, typically between the user module ONU1 (previous ONU) and the user module ONU2 (current ONU). The user module ONU2 has a second portion of the time interval T plo (reference T) p ) for the transmission of the preamble block to the OLT, this preamble block carries an optical signal sized to allow power detection of this signal and / or adjustment of its temporal phase. The preamble block is composed of a binary pattern (referenced P in the figure) which is repeated "n" times. For example, if the preamble pattern is of the type AAAAAAAAAAAAAAA in hexadecimal, this corresponds to 8 bytes in binary with an alternation of 1s and 0s which is repeated 255 times, or 4080 bits.

[0047] The ONU2 user module also has a third portion of the time interval T plo (reference T) d ) dedicated to the transmission of a delimitation block (DE). This delimitation block serves to temporally mark the start of useful data with a numerical information.

[0048] The load field includes, more specifically, a BH header block (for "Burst header") used for the identification of the optical frame in particular, a DB state block (for "Dynamic Bandwidth Report upstream") used to report the state of the frame queues destined for the OLT (on which the dynamic bandwidth allocation function is based) and a PL payload block containing the useful information to be transmitted by the ONU.

[0049] Figure 1 represents a flowchart of a particular embodiment of the allocation process according to the invention. This flowchart includes the main steps for implementing the process, which are applied each time the schedule needs to be updated. This update can be implemented on an ad hoc basis (for example, upon the occurrence of a specific event such as the detection of a malfunction or the introduction of new equipment) or periodically. Some of these steps are performed by the OLT, and others are performed by a ONU in the network. To facilitate understanding of the present invention, these steps are also represented in Figure 1 as an event diagram illustrating the flow of the process between the OLT and a given ONU in the network.

[0050] As discussed in more detail below, the process relies on a novel approach that involves updating the duration of the header fields associated with user modules to establish an optimized resource allocation schedule tailored to each user module, characterized by increased bandwidth allocation for transmitting payload data. This optimization of the allocation schedule is assumed to be performed before each uplink optical frame transmission to the OLT.

[0051] At step 100 (labeled "MAJ_CE"), the OLT updates the header field associated with the OLT user module. This update is based on obtaining a reduced duration of the time interval (T). plo) dedicated to transmitting the header field associated with the ONU (this is then referred to, somewhat inaccurately, as reducing the header field duration). This reduced header field duration can be achieved through a progressive header field tuning phase (symbolized by loop I on the ) in which several header field duration values ​​are tested sequentially and incrementally, as long as a quality criterion remains above a predetermined threshold. This allows for precise adjustment of the header field duration, and therefore the associated bandwidth. A detailed example of implementing this tuning phase is illustrated later in relation to the .

[0052] The resource(s) made available by a reduction in the duration of the header field associated with the UN is / are intended to be reallocated to the load field.

[0053] The header field setting can be adjusted over one or more time windows dedicated to the header field, and in particular over the time interval dedicated to the guard time T g and / or on the time interval dedicated to the preamble T p Thus, the OLT is configured to update the header field duration by adjusting the GT guard time block duration and / or the PR preamble block duration.

[0054] The adjustment of the duration of the guard time block and / or the duration of the preamble block takes into account at least one of the following criteria (without being exhaustive): information representative of the optical power received by the optical line terminal, from the user module for which the update is considered; information representative of the fiber distance between the optical line terminal and the user module for which the update is considered; information representative of the minimum power received and the maximum power received from the user modules already connected; information representative of the minimum and maximum distance of the user modules already connected; information representative of the type of user module for which the update is considered.

[0055] According to one particular implementation, the update step consists of modifying the number of repetitions of the binary pattern in order to reduce the duration of the PR preamble block.

[0056] Update step 100 is performed by the OLT for each of the 64 ONUs in the optical network. The updated data relating to the duration of the header fields obtained at the end of this step is stored, for example, in a local OLT table.

[0057] At step 200 (denoted "ETA_CA"), the OLT establishes an optimized time resource allocation schedule, taking into account the updated duration of the header field associated with each ONU in the network. More precisely, for a given ONU and a fixed associated time slot, the OLT reduces the duration of the header field (i.e., reduces the duration of the time interval T). plo ) to the benefit of the charge field (i.e., increasing the duration of the time interval T) uThus, the resource made available by reducing the duration of the header field associated with the UN is allocated to the load field associated with that UN, resulting in a header field time interval T plo smaller and by a charge field time interval T u larger than in the previous allocation schedule. This operation therefore amounts to allocating increased speaking time for payload data traffic by the relevant ONU. Such an approach thus allows for more bandwidth to be granted to the uplink transmission of payload data by network equipment.

[0058] Once the schedule is established for all ONUs in the network, at step 300 (labeled "DIF_CA"), the OLT broadcasts the optimized allocation schedule to all ONUs in the network. To do this, the OLT constructs an optical signal, called the update signal, and transmits it downlink to all ONUs via the network's dedicated communication channels (signal represented by the arrow Fd on the diagram). This update signal includes a header field, a payload field, and optionally a protection field. The payload field contains data relating to any changes in the duration of the header fields and the charge fields to be allocated to each ONU in the network.

[0059] Such optimization of the allocation schedule, as illustrated by steps 100, 200, and 300, can be implemented on an ad hoc basis (for example, upon the occurrence of a specific event such as the detection of an equipment malfunction or the introduction of new equipment) or periodically, for example, according to a periodicity predefined by the OLT, typically before each transmission of an optical frame or uplink optical burst, or of a predefined number of optical frames or optical bursts. It is worth recalling here that an optical burst is a grouping of several optical frames (or packets) transmitted by an ONU in the network.

[0060] At step 400 (denoted "Rx_SM"), upon receiving the update signal, the ONU extracts data related to a modification of the header and load field durations specific to it from said update signal, and then modifies the optical frame intended for transmission over the network. To do this, at step 500 (denoted "CON_TO"), the ONU constructs an uplink optical frame in which the header and load field durations are sized according to the data extracted from the update signal and specifically imposed by the OLT in order to optimize the allocation of time resources for the benefit of the data traffic. Typically, the operation consists of allocating a reduced amount of time resources for the transmission of the header field (i.e., a reduced duration of the time interval T). plo) and an increased amount of temporal resources for the transmission of the charge field (i.e., an increased duration of the time interval T u ) in accordance with the data extracted from the update signal relating to the change in the allocation schedule.

[0061] Finally, at step 600 (noted "Tx_TO"), the UN proceeds to transmit the uplink optical frame thus constructed (materialized by the arrow Fm on the) by integrating the useful data into the load field of greater capacity thanks to the invention.

[0062] Steps 400 to 600 described above can be implemented by each of the 64 UNs in the network.

[0063] Thus, at the request of the OLT, each ONU is able to construct its own uplink frame, the allocation of which in temporal resources (and therefore in bandwidth) has been optimized in favor of useful data, from the modification data received from the OLT.

[0064] In the implementation example described above, the update of the header field duration by the OLT is specific to a particular user module of the network.

[0065] According to one embodiment, the OLT's update of the header field duration can be specific to a sequence of two user modules following one another in the calendar. Indeed, as described in more detail later, due to the possible sensitivity of pairs of ONUs following one another in the calendar to the guard time and / or preamble time intervals of the header field (as illustrated with the user modules whose time slots T) ONU1 and T ONU2 (are successive in the calendar), the OLT can take into account the UN pairs imposed by the calendar in order to optimize the duration of these time intervals as part of its update.

[0066] The following details of the implementation of the allocation process on the OLT side are presented through three specific examples of the implementation of the process: during a phase of attaching a new user module, during a phase of traffic flow, and / or during an emergency stop phase.

[0067] 1. Phase of attaching a new user module

[0068] According to a first embodiment, the method according to the invention can be implemented for the integration phase of a new ONU with the OLT. Indeed, it may be advantageous to implement the method of the invention when a new user module is introduced into the PON network and an integration phase is initiated, in order to optimize the allocation of bandwidth resources during this integration phase.

[0069] For this lock-on phase, the duration of the preamble block is considered the only temporal element to be modified in order to optimize the recognition of the transmitted signal, based on the received optical power level. Thus, at least one parameter of the pattern contained in the header field can be adjusted by the OLT, taking into account one or more user parameters.

[0070] Adjusting at least one pattern parameter means the ability to adjust the shape and / or length of the pattern itself and / or its repetition rate to increase its detectability by the OLT during the hooking phase. The user parameter is a parameter belonging to the group comprising: a parameter representing the type of user modules provisioned but not hooked onto the optical line terminal interface; a parameter representing the maximum and minimum optical fiber distances between the optical line terminal and the already connected user modules; a parameter representing the maximum and minimum power received from the already connected user modules; and a parameter representing the preamble duration of the already connected user modules, stored by the optical line terminal.

[0071] In addition, it is also possible to reduce the duration of the preamble block, depending on the parameters taken into account by the OLT, and as long as the minimum detectability threshold is guaranteed, in order to allow an allocation of the freed-up time to a load field for the transmission of useful data in addition to a certain guarantee of UN hooking by the OLT. 2. Traffic flow phase

[0072] According to a second embodiment, the method according to the invention can be implemented for a traffic flow phase between at least one user module and the optical line terminal.

[0073] Before the flow phase begins, the OLT can implement a schedule optimization phase by modifying the duration of the guard time block and / or the duration of the preamble block, for example, using incremental adjustment through successive iterations. Specifically, these durations can be reduced to allocate the freed bandwidth to the load fields for transmitting useful data.

[0074] Optimization of the preamble block

[0075] According to one implementation method, the bandwidth allocation process relies on optimizing the duration of the preamble block using the progressive tuning phase discussed above. A detailed example of this particular implementation is illustrated in the form of an event diagram.

[0076] In this example, the OLT iteratively tests different duration values ​​that could potentially be assigned to the preamble block associated with a given UN. The value retained at the end of the test constitutes the header field duration updated during step 100.

[0077] In step 110, the OLT modifies the header pattern (at least one of the parameters, such as pattern length and repetitions) to reduce the header field duration. This typically involves assigning the preamble block a reduced candidate duration T test (for example, compared to the time allocated in the previous schedule). In step 115, the OLT sends a test request (indicated by the "Req" arrow) to the relevant UN, containing data representing the proposed modification to the header pattern (for example, the candidate duration T). iassociated with iteration i). Upon receipt of the request, the UN establishes a test signal in accordance with the request in which the allocated preamble block duration corresponds to the candidate duration T i Then, at step 120, the ONU transmits the test frame to the OLT (represented by the "Sig" arrow) for signal analysis against a predefined communication quality criterion.

[0078] If the test signal is received correctly by the OLT (level above a threshold), then the OLT compares, in step 130, the value of at least one reception quality indicator of the test signal with a threshold (such as a received power level measured by the OLT or a bit error rate estimated by it). If the test is successful, the OLT proceeds to step 131, stores the current tested value in a dedicated local table, and then returns to step 110 to test a new candidate duration value (duration T). i+1associated with iteration i+1), reduced compared to the value tested in the previous iteration (T i+1 > T i If the test is negative, the OLT proceeds to step 132 and updates the duration of the preamble block, taking into account not the current candidate duration tested (iteration i), but the candidate duration tested in the previous iteration (duration T). i-1 ).

[0079] If the test signal is not received by the OLT (insufficient power level in reception for example or presenting too high an error rate), the OLT goes to step 140 and proceeds to update the duration of the preamble block taking into account the candidate duration tested in the previous iteration (i-1).

[0080] The duration values ​​obtained during this gradual tuning phase, along with information related to the user modules, are collected for all ONUs and stored in a local OLT table. These values ​​can be compiled into charts to define optimal operating parameters for these ONUs. In particular, the optical line terminal can store the optimized durations of the header fields for each module connected to the OLT.

[0081] According to a particular implementation, the progressive adjustment phase is initiated by a step of determining an approximate initial duration, intended to be taken into account by the OLT during the first iteration of the adjustment phase.Typically, the header pattern or its repetition rate is predetermined by the OLT based on at least one of the following ONU parameters: a parameter representing the type of ONU involved in the tuning phase (for example, the user terminal may use a received signal strength indicator (or RSSI)); a parameter representing the distance between the OLT and the ONU involved in the tuning phase (for example, the OLT may use information transmitted during the ONU lock-on phase which allows measurement of the time of flight between the OLT and the ONU); a parameter representing the power received from the ONU involved; a parameter representing the optical budget to be allocated to the ONU involved; a parameter representing the duration of the preamble block last allocated to the ONU involved.

[0082] For example, the OLT determines an initial approximate duration of the preamble block taking into account: standard parameters relating to the preamble block (for example a preamble pattern of type AAAAAAAA in hexadecimal, corresponding to 2 bytes in binary with an alternation of 1 and 0 repeated 90 times, which corresponds to 16 x 90 = 1440 bits); information relating to the received power associated with the user module and the distance between the OLT and the ONU.

[0083] As an alternative or complement, the OLT can also rely on one or more charts to obtain an approximate initial duration for the preamble block before proceeding with the actual iterative, incremental tuning. These charts define ONU operating zones based on various data such as: pattern repetition rate, fiber distance between the OLT and the ONUs, received power at different pattern repetition rates, Ethernet packet error rate, optical budget allocated to the ONUs, etc. These charts can be initially integrated into the OLT before it is commissioned and can be updated progressively with information obtained through the process (this information can itself be cross-referenced with the aforementioned ONU parameters).

[0084] The progressive tuning phase allows the OLT to calculate a new ascending time allocation for all ONUs using the resources made available by optimizing the duration of the header fields. For example, with one user module connected to 128 user modules, if the optimization frees up 1440 – 288 = 1152 bits for each, this corresponds to 147,456 bits that can be reallocated for useful data traffic during each allocation period for all user modules.

[0085] We now present, in relation to this, an organizational chart illustrating the implementation of the progressive adjustment phase according to a particular implementation example in which two UNs are involved, referenced UN-1 and UN-2. This particular implementation example is based on an overall optimization of the time to be allocated to the network UNs.

[0086] At step 110', the OLT determines an initial candidate duration (Ti ) to allocate to the preamble block to each of the UNs UN-1 and UN-2 and transmits to each of them a test request including data representative of the candidate duration T i Upon receiving the request, each of UN-1 and UN-2 establishes a test signal in accordance with the received request, in which the allocated preamble block duration corresponds to the candidate duration T i (steps 120-1 and 120-2), and transmits it to the OLT.

[0087] If the test signal is received correctly by the OLT (level above a threshold), then the OLT compares the value of at least one signal reception quality indicator for each of the two received test signals with a threshold (steps 130-1 and 130-2). If the test is successful for both received test signals, the OLT proceeds to step 131', stores the current tested values ​​in a dedicated local table, and then returns to step 110' to test a new candidate duration value (duration T). i+1with T i+1 > T i If the test is negative for at least one of the two received test signals, the OLT updates the duration of the preamble block, taking into account the candidate duration tested in the previous iteration (duration T). i-1 ) (steps132-1 and 132-2).

[0088] If at least one of the two test signals is not received by the OLT, the OLT proceeds to update the duration of the preamble block by taking into account the candidate duration tested in the previous iteration (i-1) (steps140-1 and 140-2).

[0089] The same duration for the preamble block can be updated for both ONU-1 and ONU-2. This approach can of course be extended to a larger number of user modules, typically to all 64 ONUs of the PON network.

[0090] Thus, according to this global optimization method, it is possible to obtain the same duration of preamble block, which is suitable for all user modules.

[0091] According to one embodiment, the optimal duration of the guard preamble block can be specific to the pair of user modules whose time slots are consecutive in a calendar. Thus, if there are "m" user modules connected to the OLT, then there are "mx (m-1)" pairs of successive ONUs that could potentially be involved in updating the allocation calendar. In this case, the OLT constructs a table containing, for each pair of successive user modules, the optimized value of the duration of the preamble block associated with it.

[0092] Optimizing guard time

[0093] According to a second implementation method, the time resource allocation process relies on optimizing the duration of the guard time block. As illustrated in Figure 1, the guard time block lies between the effective transmission intervals of two successive UNs (T1). ONU1 and TONU2 Its role is to avoid the overlap of optical signals sent by two user modules whose time slots follow each other in the calendar.

[0094] As with determining the optimal duration of the preamble block, this optimization can also be performed using the incremental tuning phase, in any of the implementation examples described above: the optimal duration of the guard time block can be specific to each ONU in the network, specific to all ONUs in the network, or specific to successive pairs of ONUs. The detailed examples above for optimizing the preamble block, particularly those illustrated in Figures 5 and 6, can therefore be applied mutatis mutandis to optimizing the guard time block. The duration of the guard time block can thus be identical for all ONUs connected to the OLT, or adapted to each ONU connected to the OLT, or adapted to each successive pair of ONUs in the schedule. 3. Emergency shutdown phase

[0095] According to a third embodiment, the method according to the invention can be implemented for an emergency shutdown phase of a ONU, in order to ensure the correct transmission of a final signal by the ONU before its shutdown, while limiting the use of temporal resources by the signal header field. Thus, the method makes it possible to determine an optimal duration of the header field during a specific optimization phase.

[0096] More specifically, when a ONU needs to send a "dying gasp" message to the OLT to which it is connected, for example because its power supply is failing, it has very little time to send this message. Therefore, the allocation method according to the invention can reduce the duration of the preamble block. The specific optimization phase for the duration of the preamble block comprises the following steps: in a "normal" operating state (outside of the emergency shutdown phase), the OLT proposes a pattern and a repetition rate for that pattern; the OLT proposes to test a shorter pattern and repetition rate, if possible, to the user module; if the test is successful, this pattern and repetition rate are taken into account by the ONU when a "dying gasp" message needs to be sent.

[0097] Therefore, the optimization steps for the preamble block for an emergency stop phase are similar to the steps described above in relation to the optimization of the preamble block for the traffic flow phase.

[0098] Thus, applying the process to the preamble block used during an emergency shutdown phase ensures correct and efficient transmission of a final signal by a ONU before its shutdown, to the OLT.

[0099] Figure 700 presents a simplified structure of a device implementing the allocation method according to the invention (for example, the particular embodiment described above in relation to Figures 3 to 4). This device comprises a random access memory 730 (for example, RAM), a processing unit 710, equipped, for example, with a processor, and controlled by a computer program stored in a read-only memory 720 (for example, ROM or a hard drive). At initialization, the code instructions of the computer program are, for example, loaded into the random access memory 730 before being executed by the processor of the processing unit 710.Such a computer program, if executed by an intelligent entity internal to an optical line terminal, allows the execution of part of the steps of the algorithm described above (steps 100, 200, 300), or if executed by a user module, of the other part of the steps of the algorithm described above (steps 400, 500, 600).

[0100] In one embodiment, an intelligent entity external to the optical termination could be responsible for updating the header fields and establishing the process allocation schedule. More specifically, this external entity (for example, located in the central office) could initiate the allocation algorithm and send commands to execute the steps dedicated to the optical termination and the user module, respectively, according to the principle described above.

[0101] This illustrates only one particular way, among several possible ways, of implementing the various algorithms detailed above, in relation to the invention. Indeed, the technique of the invention can be implemented interchangeably:

[0102] - on a reprogrammable computing machine (a PC, a DSP processor, or a microcontroller) executing a program comprising a sequence of instructions, or

[0103] - on a dedicated computing machine (for example a set of logic gates such as an FPGA or an ASIC, or any other hardware module).

[0104] In the case where the invention is implemented on a reprogrammable computing machine, the corresponding program (i.e. the sequence of instructions) may be stored in a removable storage medium (such as, for example, a floppy disk, a CD-ROM or a DVD-ROM) or not, this storage medium being readable partially or totally by a computer or a processor.

Claims

Method for allocating time resources in an optical fiber communication system comprising an optical line terminal (OLT) connected to a plurality of user modules via a single optical fiber, the optical line terminal implementing a periodically updated time resource allocation schedule to said user modules, allocating to a user module (ONU) a time slot for the transmission of at least one frame from the user module, said at least one frame comprising: a header field, including data relating to opto-electronic physical layer functions;a load field, comprising useful data, characterized in that it comprises the following steps, implemented by the optical line terminal: updating (100) the duration of at least one header field, said update consisting of reducing the duration of said at least one header field so as to make at least one time resource available and allocating said at least one time resource to at least one load field; establishing (200) the allocation schedule, taking into account the update of the duration of said at least one header field; disseminating (300) the allocation schedule to all said user modules. A method according to claim 1, wherein the update of the duration of said at least one header field is performed for a given user module of said plurality. A method according to claim 1, wherein the update of the duration of said at least one header field is performed for a pair of user modules succeeding one another in the allocation schedule. A method according to any one of claims 1 to 3, comprising at least two iterations of the duration update step, wherein, at each iteration, the duration of a candidate header field is reduced as long as a quality criterion remains above a predetermined threshold. A method according to any one of claims 1 to 4, wherein said at least one header field comprises: a guard time block; a preamble block carrying a signal enabling a user module to detect the power of an optical signal and / or to adjust the time phase of an optical signal; and wherein said duration update step adjusts the duration of said guard time block and / or the duration of said preamble block. Method according to claim 5, wherein, said preamble block being made up of a repetition of a binary pattern, said duration update step consists of modifying the number of repetitions of said binary pattern. A method according to claim 5 or claim 6, wherein the adjustment of the duration of said guard time block and / or the duration of said preamble block takes into account at least one of the criteria belonging to the group comprising: information representative of the optical power received by the optical line terminal, from the user module for which the update is considered; information representative of the fiber distance between the optical line terminal and the user module for which the update is considered; information representative of the minimum power received and the maximum power received from the user modules already connected; information representative of the minimum distance and the maximum distance of the user modules already connected; information representative of the type of user module for which the update is considered. A method according to any one of claims 5 to 7, wherein it is implemented in distinct ways depending on whether a user module is in a phase of snapping with the optical line terminal or in a phase of flowing traffic from at least one user module to the optical line terminal. A method according to any one of claims 5 to 7, wherein it is implemented for an emergency shutdown phase of a user module, the update step comprising an iterative optimization of the duration of the preamble time, prior to said emergency shutdown phase. Method of transmitting frames by a user module (ONU), in an optical fiber communication system comprising an optical line terminal (OLT) connected to a plurality of user modules via a single optical fiber, the optical line terminal implementing a periodically updated time resource allocation schedule to said user modules, allocating to a user module a time slot for the transmission of at least one frame from the user module, said at least one frame comprising: a header field, comprising data relating to opto-electronic physical layer functions; a payload field, comprising payload data, characterized in that it comprises the following steps, implemented by said user module: reception (400) of a time resource allocation schedule update signal;extraction of information relating to a reduction in the duration of the header field of frames emitted by said user module, within said schedule from the update signal, so as to make at least one time resource available and to allocate said at least one time resource to at least one load field; construction (500) of a header according to the extracted information; transmission (600) of a frame including the constructed header. Product computer program downloadable from a communication network and / or stored on a computer-readable medium and / or executable by a microprocessor, characterized in that it includes program code instructions for the execution of a bandwidth allocation method according to any one of claims 1 to 10 when said program is implemented by a computer. Product computer program downloadable from a communication network and / or stored on a computer-readable medium and / or executable by a microprocessor, characterized in that it includes program code instructions for the execution of a frame transmission method according to claim 11 when said program is implemented by a computer. Optical line terminal (OLT) for optical fiber communication system comprising a plurality of user modules, said optical line terminal implementing a periodically updated time resource allocation schedule to said user modules, allocating to a user module (ONU) a time slot for the transmission of at least one frame from the user module, said at least one frame comprising: a header field, including data relating to opto-electronic physical layer functions; a payload field, including payload data, characterized in that it comprises: means for updating the duration of at least one header field consisting of reducing the duration of said at least one header field so as to make at least one time resource available and allocating said at least one time resource to at least one payload field;means of establishing the allocation schedule, taking into account the updating of the duration of said at least one header field; means of disseminating the allocation schedule to all said user modules. user module (ONU) for an optical fiber communication system comprising an optical line terminal (OLT) and a plurality of user modules, said communication system implementing a time resource allocation schedule allocating said user module a time slot for the transmission of at least one frame to the optical line terminal, said at least one frame comprising: a header field, including data relating to opto-electronic physical layer functions; a payload field, including payload data; characterized in that it comprises: means for receiving a time resource allocation schedule update signal;means of extracting information relating to a reduction in the duration of the header field of frames emitted by said user module from the update signal, so as to make at least one time resource available and to allocate said at least one time resource to at least one load field; means of constructing a header according to the extracted information; means of transmitting a frame including the constructed header.

Images