JP2012175525A - Dynamic band allocation method and passive optical network communication system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve data transmission efficiency and shorten a DBA calculation time.SOLUTION: The dynamic band allocation method comprises: an ONU grouping step in which ONUs among the ONUs connected to a multiplexing light transmission path in a linkup connection state are divided into a plurality of groups to divide the ONUs into the ones where an NSR-DBA is applied and the ones where an SR-DBA is applied; a fixed band allocation step in which a fixed band is allocated to the ONUs belonging to a group where the NSR-DBA is applied; and a best effort band allocation step in which a best effort band is allocated based on SR-DBA calculation to the ONUs belonging to a group where the SR-DBA is applied.

Description

  The present invention provides communication between a station-side terminal device (OLT: Optical Line Terminal) installed on the central office side of a telecommunications carrier and a terminal device (ONU: Optical Network Unit) installed on each of a plurality of subscribers. The present invention relates to bandwidth control in an optical network communication system configured by connecting via a passive optical network (PON).

  A communication network that connects a central office owned by a telecommunications carrier and a subscriber (sometimes referred to as a user) through an optical transmission line is called an optical access network. An EPON (Ethernet PON) system is widely used as one form of constructing an optical access network at a low cost. As an EPON system, a GE-PON (Gigabit Ethernet PON) system having a transmission rate (data rate) of 1 Gbit / s compliant with IEEE 802.3ah.std is widely used in Japan. In addition, the use of a 10G-EPON (10Gigabit Ethernet PON) system with a transmission speed (data rate) of 10 Gbit / s compliant with IEEE 802.3av.std is also expected.

  Here, Ethernet is a communication medium generally used in a LAN (Local Area Network) environment or the like, and “Ethernet” and Japanese “Ethernet” are registered trademarks. Hereinafter, Ethernet and Ethernet are used regardless of being a registered trademark.

  A PON connects a distribution multiplexer (sometimes called a star coupler) as an optical multiplexer / demultiplexer in the middle of an optical fiber transmission line, and connects one optical fiber transmission line to a plurality of branched optical fiber transmission lines. A network in which a plurality of terminal devices are connected in a star shape with the optical multiplexer / demultiplexer as the center (for example, Non-Patent Document 1). By adopting PON for the network connecting the central office and terminal equipment, the optical fiber transmission line between the central office and the optical multiplexer / demultiplexer can be shared by multiple terminal equipment, which can reduce equipment costs. It becomes.

  In an optical network communication system configured by connecting with a PON, a communication signal in a direction from a plurality of ONUs to an OLT (hereinafter also referred to as an upstream signal) is a star coupler provided in the middle of the optical fiber transmission line. ) Are combined and sent to the OLT. Therefore, it is necessary to prevent the transmission signals of different ONUs from colliding with each other so that the upstream signals do not collide with each other on the optical fiber transmission line. For this purpose, a technique for controlling the transmission timing and transmission amount of the ONU uplink signal is a method called time division multiple access (TDMA).

  In communication in the direction from ONU to OLT (uplink direction) and in the direction from OLT to ONU (downlink direction), signals are mounted and transmitted in a frame format which is a basic transmission unit. Therefore, when a signal is transmitted or received, this signal is mounted in a frame format, and this frame is transmitted or received.

  In TDMA, the time axis is divided into a plurality of short intervals, and this interval is assigned to each ONU as a communication time band for transmitting an uplink signal toward the OLT. That is, the OLT is instructed to transmit an uplink signal to each ONU. Assigning a communication time band to each ONU in this way is called bandwidth assignment.

  One of the bandwidth allocation methods is known as dynamic bandwidth allocation (DBA). DBA is a processing method in which an OLT receives a transmission bandwidth request from each ONU and dynamically allocates a communication bandwidth in consideration of this request.

  In the OLT, by receiving a transmission bandwidth request from each ONU, the traffic size requested by each ONU is grasped. Then, the transmission timing determined by the DBA is instructed from the OLT to each ONU. For example, in an optical network communication system configured by connecting with GE-PON, efficient and less wasteful bandwidth allocation is realized by DBA.

  Report frame (REPORT frame) and gate frame (GATE frame) standardized by IEEE 802.3av / ah.std for notification of traffic size from each ONU to OLT and transmission timing from OLT to each ONU Is used. That is, the traffic size is notified from each ONU to the OLT by the REPORT frame, and the OLT grasps the traffic size of all connected ONUs. As a result, in the OLT, each allocated bandwidth for each ONU is determined based on the traffic size notified from each ONU. The assigned bandwidth is notified by the OLT instructing the transmission timing to each ONU by the GATE frame.

  Bandwidth allocation in OLT is performed at regular time intervals called grant periods. That is, an upstream signal communication time band for each ONU is determined in each grant period, and this band is notified to each ONU as an allocated band in each grant period. Each ONU recognizes the allocated bandwidth based on the transmitted GATE frame, reads the uplink signal transmission time, and starts transmission of the uplink signal when the transmission time is reached. In this way, communication from ONU to OLT is performed.

  In this way, an optical network communication system configured using an optical network connected by PON, which employs a method of assigning a short section on the time axis to each ONU, TDM- Sometimes called a PON optical network communication system.

Koji Ochiai, et al., “Development of Gigabit Ethernet-PON (GE-PON) System”, NTT Technical Journal, March 2005, pp. 75-80

One index that determines the performance of an optical network communication system is data transmission efficiency, which means the amount of data that can be transmitted per unit time. One factor that determines the data transmission efficiency is T _RTT (Round Trip Time), which is the time from when the OLT sends a GATE frame to the ONU until the OLT receives the REPORT frame from the ONU. That is, _TRTT includes the uplink data transmission time. _TRTT also includes a delay time required for transmitting a signal transmitted from OLT to ONU or from ONU to OLT.

  The grant period is the time required for the OLT to receive traffic size from all ONUs (report reception time), the time required for DBA calculation in the OLT (DBA calculation time), and the time that the optical signal travels between OLT and ONU. It is set to a time longer than the sum of the required time (RTT time).

  In an optical network communication system using GE-PON, the number of ONUs to be connected, strictly speaking, the number of ONUs in the link-up connection state increases as the report reception time increases, and depending on the number of ONUs Since the amount of calculation of DBA increases, DBA calculation time also increases. Therefore, since the non-data transfer time (report reception time, DBA calculation time) within the grant period increases, the data transmission efficiency deteriorates.

  Accordingly, an object of the present invention is to provide a DBA capable of improving data transmission efficiency and shortening DBA calculation time. It is another object of the present invention to provide a passive optical network communication system suitable for realizing this DBA.

  The inventor of this application does not allocate the best effort bandwidth according to the traffic size for all ONUs that are linked up with the OLT and can communicate with each other. Has arrived at the idea that the above-mentioned problems can be solved by allocating a fixed bandwidth. That is, if the ONUs that are linked up and communicable can be divided into groups of ONUs that allocate the best effort band and groups of ONUs that allocate the fixed band, a REPORT frame for the ONUs that belong to the group that allocates the fixed band Since there is no need to perform bandwidth allocation calculation for ONUs belonging to this group, the data transmission efficiency can be improved and the DBA calculation time can be shortened.

  Here, referring to the REPORT frame that receives the bandwidth to be allocated from each ONU, the technology that is determined according to the declared queue length is called status report-dynamic bandwidth allocation (SR-DBA: Status Reporting-DBA). . In addition, a technique for determining the bandwidth without referring to the REPORT frame is called non-status reporting-dynamic bandwidth allocation (NSR-DBA).

  According to the gist of the present invention based on the above philosophy, the following DBA method is provided.

  The DBA method of the present invention comprises an optical transmission line, an optical multiplexer / demultiplexer coupled to one end of the optical transmission line, and a plurality of demultiplexing optical transmission lines formed by demultiplexing by the optical multiplexer / demultiplexer. A DBA method in the communication of a passive optical network communication system, in which an OLT is provided at the other end of the optical transmission line of the passive optical network, and one ONU is provided in each of the demultiplexed optical transmission lines. The first to third steps are included.

  The first step is to divide the ONUs in the link-up connection state into multiple groups among the ONUs connected to the demultiplexed optical transmission line in order to distribute the ONUs that apply NSR-DBA and ONUs that apply SR-DBA. This is a grouping step.

  The second step is a step of assigning a fixed bandwidth to ONUs belonging to a group to which NSR-DBA is applied.

  The third step is a step of assigning a best effort band based on the SR-DBA calculation to ONUs belonging to a group to which SR-DBA is applied.

  In the second step, a fixed value is set as the fixed bandwidth, or the traffic size of the uplink data received from the ONU belonging to the group to which NSR-DBA is applied is monitored, and based on the traffic size from this ONU It is preferable to set the calculated bandwidth prediction value.

  In the DBA method of the present invention, it is preferable to exchange the group to which NSR-DBA is applied and the group to which SR-DBA is applied every grant period.

  Further, as the DBA method of the present invention, the following DBA method of the first embodiment or the DBA method of the second embodiment is preferable.

  The DBA method of the first embodiment includes a grouping step (step S1), an overhead bandwidth calculation step (step S2), a remaining bandwidth calculation step (step S3), a fixed bandwidth allocation step (step S4), and a re-remaining bandwidth calculation step ( Step S5) and a best effort bandwidth allocation step (Step S6).

  The grouping step (step S1) is a step of dividing the ONUs in the link-up connection state among a plurality of ONUs into N groups (N is an integer of 2 or more).

  The overhead bandwidth calculation step (step S2) is a step of calculating an overhead bandwidth necessary for receiving the REPORT frame and the data frame transmitted from the ONU.

  The remaining bandwidth calculating step (step S3) is a step of calculating an allocated bandwidth that is a bandwidth obtained by subtracting the overhead bandwidth from the all uplink signal transmission bandwidth.

  The fixed bandwidth allocation step (step S4) is a step of allocating a fixed bandwidth to ONUs belonging to a group to which NSR-DBA is applied among N groups.

  The remaining bandwidth calculation step (step S5) is a step of subtracting the fixed bandwidth allocated in the fixed bandwidth allocation step from the allocated bandwidth and calculating the remaining bandwidth again.

  In the best effort bandwidth allocation step (step S6), the ONU belonging to the group to which SR-DBA is applied among the N groups for the remaining bandwidth calculated in the re-remaining bandwidth calculation step is the best based on the SR-DBA calculation. This is a step of assigning an effort band.

  The DBA method of the second embodiment includes a grouping step (step S1), an overhead bandwidth calculation step (step S2), a remaining bandwidth calculation step (step S3), a data flow bandwidth allocation calculation step (step S7), and a fixed bandwidth allocation step. (Step S4 ′), a remaining bandwidth calculation step (Step S5 ′), and a best effort bandwidth allocation step (Step S6).

  The grouping step (step S1), the overhead bandwidth calculating step (step S2), the remaining bandwidth calculating step (step S3), and the best effort bandwidth allocation step (step S6) are steps included in the DBA method of the first embodiment described above. Is the same.

  In the data flow bandwidth allocation calculation step (step S7), the ONU belonging to the group to which the NSR-DBA is applied in the time zone in which data is received from the ONU belonging to the group to which the NSR-DBA is applied among the N groups. This is a step of monitoring the traffic size of the uplink data received from, and calculating a predicted bandwidth value based on the traffic size of each ONU.

  The fixed band allocation step (step S4 ′) is a step of allocating the band predicted value calculated in the data flow band allocation calculation step as a fixed band to ONUs belonging to the group to which NSR-DBA is applied among N groups. is there.

  The re-remaining bandwidth calculation step (step S5 ′) is a step of subtracting the fixed bandwidth allocated as the fixed bandwidth from the bandwidth prediction value in the fixed bandwidth allocation step from the allocated bandwidth and calculating the remaining bandwidth again.

  In the DBA method of the present invention, it is preferable to exchange the group to which NSR-DBA is applied and the group to which SR-DBA is applied every grant period.

  In the DBA method of the present invention, among the plurality of ONUs, when the ONUs in the link-up connection state are divided into the group to which NSR-DBA is applied and the group to which SR-DBA is applied, Divided into N groups of N groups, and for each grant period, SR-DBA is applied sequentially to ONUs belonging to Group 1 to Group N, and ONUs belonging to groups where SR-DBA is not applied It is preferable to apply NSR-DBA.

  A passive optical network communication system that implements the DBA method according to the first embodiment of the present invention may be configured as follows.

  An optical transmission line, an optical multiplexer / demultiplexer coupled to one end of the optical transmission line, and a passive optical network including a plurality of demultiplexing optical transmission lines formed by being demultiplexed by the optical multiplexer / demultiplexer. In this passive optical network communication system, an OLT is provided at the other end of the transmission line, and one ONU is provided for each of the demultiplexed optical transmission lines.

  The OLT includes an uplink band allocation unit that calculates an uplink transmission band for each ONU, a control signal read generation unit that receives an uplink control signal frame from an uplink signal receiving unit and reads information described in the frame, and ONUs in a plurality of groups. Grouping means for grouping, fixed bandwidth allocation means for allocating fixed bandwidth to ONUs belonging to groups to which NSR-DBA is applied, and SR-DBA calculation for ONUs belonging to groups to which SR-DBA is applied And a best-effort bandwidth allocating means for allocating the best-effort bandwidth based on the station-side control section.

  ONU temporarily stores uplink data frames transmitted from user equipment, monitors the uplink data frame buffer unit that outputs the stored uplink data frames, and monitors the uplink data frame buffer unit, and installs it in the uplink data frame in the transmission standby state And a subscriber-side control unit that obtains a queue length indicating the amount of the received data.

  A passive optical network communication system that implements the DBA method according to the second embodiment of the present invention is preferably configured as follows.

  The OLT of the passive optical network communication system that implements the DBA method of the first embodiment described above further includes a traffic monitoring unit that monitors the traffic size of the uplink data frame received from the ONU. The configuration is the same as that of the first passive optical network communication system except that the OLT further includes a traffic monitoring unit.

  According to the dynamic bandwidth allocation method of the present invention, a plurality of ONUs are grouped in the first step, and a fixed bandwidth is allocated to ONUs belonging to the group to which NSR-DBA is applied in the second step. Fixed bandwidth allocation is performed, and in the third step, best effort bandwidth allocation is performed in which the best effort bandwidth based on the SR-DBA calculation is assigned to ONUs belonging to the group to which SR-DBA is applied.

  Therefore, it is not necessary to receive a REPORT frame for an ONU belonging to a group to which a fixed bandwidth is allocated, and it is not necessary to perform a bandwidth allocation calculation for an ONU belonging to this group. Calculation time can be shortened.

  Compared to setting a fixed value as the fixed bandwidth in the second step, the traffic size of uplink data received from ONUs belonging to the group to which NSR-DBA is applied is monitored, and the calculation is based on this traffic size. If the estimated band predicted value is set, it is possible to allocate the band more effectively than setting a predetermined value as a fixed band. Moreover, even when the predicted bandwidth value is set as a fixed bandwidth, it is not necessary to refer to the REPORT frame, so that the data transmission efficiency can be improved and the DBA calculation time can be shortened.

  If the NSR-DBA application group and SR-DBA application group are replaced at each grant cycle, finer and more appropriate bandwidth allocation can be performed according to the temporal change in the communication status of each ONU. Is done.

  Also, divide multiple ONUs into 1st group to Nth group, and apply SR-DBA to ONUs belonging to 1st group to Nth group for each grant period, and to the group where SR-DBA is not applied By applying NSR-DBA to the ONU to which it belongs, the larger N is, the more the number of ONUs that belong to the group to which SR-DBA is applied can be reduced. DBA calculation time can be shortened.

It is a figure with which it uses for the rough description about the uplink communication and downlink communication of the optical network communication system of a TDM-PON method. (A) is a diagram for explaining a signal flow in uplink communication, and (B) is a diagram for explaining a signal flow in downlink communication. 1 is a schematic block configuration diagram of an optical network communication system of a TDM-PON method. FIG. It is a figure where it uses for description of the mode of transmission / reception of the REPORT frame and GATE frame exchanged between OLT and ONU in a DBA method. FIG. 6 is a diagram for explaining a state of transmission / reception of frames exchanged between an OLT and an ONU in the DBA method according to the first embodiment of the present invention. 1 is a schematic block configuration diagram of an optical network communication system for explanation of a DBA method according to a first embodiment of the present invention. 3 is a flowchart for explaining the DBA method according to the first embodiment of the present invention. It is a detailed flowchart about step S1 which is a grouping step. It is a detailed flowchart about step S2 which is an overhead band calculation step, and step S3 which is a remaining band calculation step. 5 is a detailed flowchart of step S4 as a fixed bandwidth allocation step and step S5 as a remaining bandwidth calculation step. It is a detailed flowchart about step S6 which is the best effort bandwidth allocation step. It is a figure which shows the structure of the REPORT frame and data frame which ONU transmits. FIG. 6 is a diagram showing results of calculating data transfer efficiency in the conventional DBA method and the DBA method of the first embodiment of the present invention. FIG. 7 is a schematic block configuration diagram of an optical network communication system for explaining a DBA method according to a second embodiment of the present invention. It is a figure with which it uses for description of the mode of transmission / reception of the frame exchanged between OLT and ONU in the DBA method of 2nd Embodiment of this invention. 7 is a flowchart for explaining the DBA method according to the second embodiment of the present invention. It is a detailed flowchart about step S7 which is a data flow band allocation calculation step.

  Embodiments of the present invention will be described below with reference to the drawings. Note that the block diagram shows a configuration example according to the present invention, and merely schematically shows the arrangement relationship of each component to the extent that the present invention can be understood, and the present invention is limited to the illustrated example. It is not a thing. Each block configuration diagram showing a schematic configuration of the optical network communication system shown below is not a characteristic part of the present invention, as long as it is a system of the same type as the optical network communication system of the embodiment of the present invention. Well-known components are not shown. Moreover, in each block block diagram, the same component is shown with the same number, and the duplicate description thereof may be omitted.

  First, prior to the description of the DBA method of the present invention, the configuration and operation of an optical network communication system of a typical TDM-PON method in which the DBA method is executed, and the outline of the DBA will be described.

<Optical network communication system of TDM-PON method>
With reference to FIG. 1 (A) and FIG. 1 (B), the configuration of a general optical network communication system of the TDM-PON method will be described, and communication in the direction from a plurality of ONUs to OLT (uplink communication) and The outline of communication (downlink communication) from OLT to multiple ONUs will be described.

  An optical network communication system of the TDM-PON method is configured by connecting an OLT 10 and three ONUs 14-1 to 14-3 via a PON including a star coupler 12 that is an optical multiplexer / demultiplexer. In FIG. 1 (A) and FIG. 1 (B), ONUs 14-1 to 14-3 are shown as ONU-1, ONU-2, and ONU-3, respectively. The numerical values of 1, 2, and 3 when indicated as ONU-1, ONU-2, and ONU-3 are sometimes referred to as ONU numbers.

  In the optical network communication system shown in FIGS. 1 (A) and 1 (B), three ONUs are connected, but the number of ONUs to be connected is not limited to three and two or more ONUs. May be configured.

  Also, not all ONUs connected via PON are always set in a state where they can communicate with OLT. That is, there are ONUs that are turned off and are not operating in the ONUs connected via the PON. Of all the connected ONUs, ONUs that are set in a communicable state may be referred to as ONUs in a link-up connection state. Therefore, in the following description, an ONU that is in a link-up connection state may be simply described as an ONU that is connected via a PON, regardless of the ONU that is in the link-up connection state.

  FIG. 1 (A) shows the flow of upstream signals from three ONUs 14-1 to 14-3 to the OLT 10. The upstream signal is multiplexed by the star coupler 12 from signals transmitted from the ONUs 14-1 to 14-3, multiplexed by this, and transmitted to the OLT 10. For this reason, when uplink signals are sent randomly from each ONU, there is a possibility of causing a collision on the optical transmission line. One technique for avoiding this is called TDMA. In TDMA, uplink signals are multiplexed by controlling the transmission timing from each ONU.

  FIG. 1B shows the flow of downstream signals in the direction from the OLT 10 to the three ONUs 14-1 to 14-3. In downlink communication, a downlink signal multiplexed from the OLT 10 by the TDM method is demultiplexed by the star coupler 12 and transmitted to each of the three ONUs 14-1 to 14-3. Therefore, the same downstream signal is sent to all three ONUs 14-1 to 14-3. Each ONU extracts only the signal sent to itself and discards the other signals.

  With reference to FIG. 2, a schematic configuration and operation of a general optical network communication system of the TDM-PON method will be described.

  Since the DBA processing of the present invention relates to the uplink communication of TDM-PON, the description of the configuration of the optical network communication system described below will focus on the components related to the uplink communication. Therefore, in the block configuration diagram of the optical network communication system of the TDM-PON method shown in FIG. 2, some of the functional blocks required for downlink communication are omitted.

  The optical network communication system of the TDM-PON method shown in FIG. 2 is an OLT 20 and ONUs installed on the subscriber side such as ONU 40-1, ONU 40-2, ONU 40-3 (hereinafter referred to as ONU 40-1). , ONU 40-2, ONU 40-3, etc. may be collectively referred to as ONU 40.) is an optical network communication system configured by being connected by a PON including a star coupler 50.

  The OLT 20 includes an interface (I / F) converter 22, a downlink signal transmitter 23, a station-side controller 24, an uplink signal receiver 25, and an optical multiplexer / demultiplexer 26. The interface conversion unit 22 performs communication protocol processing with the upper network 21. The downlink signal transmission unit 23 converts the downlink signal (electric signal) obtained by time-multiplexing the downlink data signal transmitted from the interface conversion unit 22 and the downlink control signal transmitted from the station side control unit 24 from an electric signal to an optical signal. Conversion (hereinafter sometimes referred to as electro-optical conversion) generates a downstream signal in the form of an optical signal.

  The upstream signal receiving unit 25 converts the upstream signal (optical signal) transmitted from the ONU 40 from an optical signal to an electrical signal (hereinafter also referred to as optical-electrical conversion), and converts the upstream signal in the form of an electrical signal. Generate. Then, the uplink signal is separated into an uplink control signal and an uplink data signal. The uplink control signal is transferred to the station-side control unit 24, and the uplink data signal is transferred to the interface conversion unit 22.

  The station-side control unit 24 includes a grouping unit 24-1, a fixed band allocation unit 24-2, a best effort band allocation unit 24-3, a control signal read generation unit 27, and an uplink band allocation unit 28. The control signal reading / generating unit 27 receives an uplink control signal from the uplink signal receiving unit 25 and reads information described in the signal. In addition, the control signal reading generation unit 27 generates a downlink control signal and passes it to the downlink signal transmission unit 23. The upstream bandwidth allocation means 28 refers to the queue length notification sent from each ONU 40 and calculates the upstream transmission bandwidth of each ONU 40. The queue length indicates the amount of data (traffic size) loaded in the data frame, and is also called a bandwidth request amount.

  The grouping means 24-1 executes a grouping step for dividing the ONU 40 into a plurality of groups. The fixed bandwidth allocation unit 24-2 executes a step of allocating a fixed bandwidth to the ONUs belonging to the group to which NSR-DBA is applied. The best effort bandwidth allocation unit 24-3 executes a step of allocating the best effort bandwidth based on the SR-DBA calculation to the ONUs belonging to the group to which the SR-DBA is applied.

  The optical multiplexer / demultiplexer 26 combines and demultiplexes the upstream signal and downstream signal in the form of an optical signal. In general, the center wavelengths of the upstream and downstream signal carriers are different. Therefore, the optical multiplexer / demultiplexer 26 is realized using an optical wavelength filter or the like.

  On the other hand, each of the ONUs 40 includes an interface (I / F) converter 42, a downlink signal receiver 43, a subscriber side controller 44, an uplink signal transmitter 45, an uplink data frame buffer 49, and an optical multiplexer / demultiplexer 46. Consists of. In FIG. 2, these components are shown as, for example, an interface converter 42-1, a downlink signal receiver 43-1, etc., so that these components can be identified as comprising ONU-1. . However, since all the ONU configurations are the same, in the following description, such identification numbers are omitted unless they are particularly required to be identified, and are simply referred to as an interface conversion unit 42, a downlink signal reception unit 43, and the like.

  The interface conversion unit 42 performs communication protocol processing with the user device 41. The upstream signal transmission unit 45 performs electro-optical conversion on the upstream signal (electric signal) obtained by time-multiplexing the upstream data signal read from the upstream data frame buffer unit 49 and the upstream control signal transmitted from the subscriber-side control unit 44. Thus, an upstream signal in the form of an optical signal is generated.

  The downlink signal receiving unit 43 performs optical-electrical conversion on the downlink signal (optical signal) transmitted from the OLT 20 to generate a downlink signal in the form of an electric signal. Then, the downlink signal is separated into a downlink control signal and a downlink data signal. The downlink data signal is transferred to the interface conversion unit 42, and the downlink control signal is transferred to the subscriber side control unit 44.

  The subscriber-side control unit 44 includes a control signal reading / generating unit 47 and an uplink transmission time control unit 48. The control signal reading / generating unit 47 receives the downlink control signal from the downlink signal receiving unit 43 and reads information described in the signal. Further, the control signal reading generation unit 47 generates an uplink control signal and passes it to the uplink signal transmission unit 45. The uplink transmission time control means 48 instructs the timing for transmitting the uplink signal. Further, the subscriber-side control unit 44 monitors the uplink data frame buffer unit 49, and acquires the traffic size (queue length) of the uplink data signal in the transmission standby state.

  The optical multiplexer / demultiplexer 46 combines and demultiplexes the upstream signal and downstream signal in the form of an optical signal. In general, the center wavelengths of the upstream and downstream signal carriers are different. Therefore, the optical multiplexer / demultiplexer 46 is realized using an optical wavelength filter or the like. The upstream data frame buffer unit 49 temporarily stores the upstream data signal transmitted from the user device 41. Then, the stored uplink data signal is output in accordance with the read instruction from the uplink signal transmission unit 45.

  The star coupler 50 distributes the downstream signal transmitted from the OLT 20 to each ONU 40. Further, the star coupler 50 combines the upstream signals transmitted from the respective ONUs 40 and transmits them to the OLT 20.

  The uplink signal is composed of an uplink data signal and an uplink control signal. The upstream data signal is transmitted from the user device 41 to the upper network 21. That is, the uplink data signal is a signal carrying information requested by the user. On the other hand, the uplink control signal is a signal that the subscriber-side control unit 44 transmits to the station-side control unit 24. The uplink control signal describes queue length information.

  The downlink signal is composed of a downlink data signal and a downlink control signal. The downlink data signal is transmitted from the upper network 21 to the user equipment 41. That is, the downlink data signal is a signal carrying information requested by the user. On the other hand, the downlink control signal is a signal transmitted from the station side control unit 24 to the subscriber side control unit 44. In the downlink control signal, information on the uplink transmission band and the uplink transmission timing is described.

  The uplink control signal and downlink control signal may carry other information for controlling the network. In order to perform stable communication, it is necessary to periodically transmit and receive these control signals between the OLT 20 and the ONU 40.

  The processing in the upstream bandwidth allocation unit 28, the control signal read generation unit 27 of the OLT 20, the control signal read generation unit 47 of the ONU 40, the upstream transmission time control unit 48, and the like are, for example, a central processing unit CPU (not shown) In other words, each function is performed as a function unit that is expressed when a program stored in advance in a storage device (not shown) is executed.

<Outline of DBA>
As described above, the TDMA method is used for uplink communication in a TDM-PON optical network communication system. TDMA is a technique in which the OLT 20 manages the transmission timing of each ONU 40 and controls so that optical signals transmitted by different ONUs do not collide with each other on the time axis. The OLT 20 granting each ONU 40 permission to transmit an uplink signal for a certain period is referred to as “allocating a band”.

  In addition, as described above, a technique for dynamically changing the allocated bandwidth according to the situation is called DBA. In particular, SR-DBA is a method for determining the bandwidth to be allocated according to the declared queue length with reference to the REPORT frame that receives each ONU 40. In SR-DBA, the bandwidth allocated for each grant period is updated. The unit indicating the “band” here is time (for example, second), and indicates the length of time permitted for transmission within the grant period. A method for determining the bandwidth without referring to the REPORT frame is called NSR-DBA.

  The state of uplink communication using SR-DBA will be described with reference to FIG. FIG. 3 shows a time axis indicating the passage of time in the horizontal direction, and shows time charts for OLT 20, ONU-1, and ONU-2 in order from the top. In FIG. 3, the REPORT frame is surrounded by a rectangle and indicated as “R”, and the GATE frame is surrounded by a rectangle and indicated as “G”. Similarly, the data frame is surrounded by a rectangle and indicated as “data”.

  Here, for the sake of simplicity, it is assumed that the ONUs connected to the OLT 20 are two units, ONU-1 and ONU-2. Further, in FIG. 3, the grant period is shown for the tth and (t + 1) th. t is a parameter introduced for generalization and indicates an arbitrary integer value of 1 or more. Therefore, FIG. 3 shows a series of operations until the uplink data signal generated from the user equipment 41 is transmitted to the OLT 20 in the t-th grant period. In the following description, communication between ONU-1 and OLT 20 will be described, but the same applies to ONUs other than ONU-1.

  When an upstream data signal is generated, the upstream data signal is transmitted from the user equipment 41-1 to ONU-1 (ONU 40-1) in the ONU 40 and temporarily stored in the upstream data frame buffer unit 49. The band required to transmit this uplink data signal is represented as RBW (t) (not shown in FIG. 3). The value given by RBW (t) may be called queue length or requested bandwidth.

  After that, the required bandwidth is indicated using a notation such as RBW (t) indicating the identification number and grant period number, but when there is no need to identify the ONU in particular, or when it is not necessary to specify the grant period number The notation of the parameter “t” in parentheses may be omitted.

  Next, in order to notify the OLT 20 of the queue length for notifying the requested bandwidth, an uplink control signal in which the requested bandwidth in the (t + 2) th grant period is written is transmitted toward the OLT 20. This uplink control signal is transmitted so as to reach the OLT 20 at the beginning of the start of the (t + 1) th grant period. The OLT 20 reads all bandwidth request information of each ONU (here, ONU-1 and ONU-2) from the uplink control signal.

  Subsequently, the bandwidth allocated to each ONU is calculated. This calculation is executed by the upstream bandwidth allocating means 28 included in the OLT 20 in the (t + 1) th grant period. The allocated bandwidth RBW (t + 2) for each ONU in the (t + 2) th grant period is determined with reference to the queue length information RBW (t) obtained by this calculation.

  Following this, the allocated bandwidth RBW (t + 2) is notified to each ONU. This notification is executed in the (t + 1) th grant period. The OLT 20 embeds the allocated bandwidth RBW (t + 2) in the downlink control signal and transmits it to each ONU. Each ONU receives this downlink control signal and reads the allocated band RBW (t + 2).

  Subsequently, an uplink data signal is transmitted from each ONU to the OLT 20. This transmission is performed in the (t + 2) th grant period. Each ONU transmits the uplink data signal stored in the uplink data frame buffer unit 49 toward the OLT 20 according to the allocated band RBW (t + 2) transmitted to each ONU. The OLT 20 receives the upstream data signal sent from each ONU.

  As described above, the data signal generated in the t-th grant cycle is transmitted toward the OLT 20 in the (t + 2) -th grant cycle even in the earliest case. That is, the uplink data signal is transmitted toward the OLT 20 with a delay of two grant periods from the generation of the uplink data signal.

  The time from when the uplink data signal is generated until this uplink data signal is transmitted toward the OLT 20 may be referred to as an uplink transmission delay time. In consideration of the existence of transmission services such as voice communication, where the uplink transmission delay time becomes longer, the quality of service is reduced, and the uplink transmission delay time is preferably as short as possible. In order to shorten the uplink transmission delay time, it is effective to set the grant period short.

On the other hand, the minimum value T _{cyc_min} of the grant period T _cyc is given by the following equation (1).
T _{cyc_min} = T _ROH + T _calc + T _RTT (1)
Here, T _ROH is the time required for OLT 20 to receive the uplink control signal transmitted from each ONU (report reception time), and T _calc is the time required for OLT 20 to calculate the bandwidth (DBA calculation time) ), T _RTT is the time (RTT time) required for the optical signal to reciprocate between OLT 20 and ONU. In order to carry out SR-DBA reliably, T _calc must be set to a value equal to or greater than the minimum value T _{cyc_min} calculated by the above equation (1).

<DBA method of the first embodiment>
In the DBA method according to the first embodiment of the present invention, first, a link-up connection is performed in order to distribute an ONU that applies an NSR-DBA that does not refer to a REPORT frame and an ONU that applies an SR-DBA that refers to a REPORT frame. All ONUs in state are grouped into N groups. Here, a case will be described in which ONUs 1 to ONU-128 connected in link-up are ONU-1 to ONU-128, and these ONU-1 to ONU-128 are divided into two groups (N = 2).

  Table 1 divides ONU-1 to ONU-128 into two groups, and shows which ONU is applied with SR-DBA and which ONU is applied with NSR-DBA for each grant period.

  Table 1 shows ONUs to which SR-DBA is applied and ONUs to which NSR-DBA is applied for the t-th, (t + 1) -th, (t + 2) -th, etc. grant periods. . In Table 1, ONU-1 to ONU-64 are indicated as “ONU 1 to 64”, and ONU-65 to ONU-128 are indicated as “ONU 65 to 128”. As shown in Table 1, ONUs belonging to a group to which NSR-DBA is applied and ONUs belonging to a group to which SR-DBA is applied are switched every grant period.

  By the way, when ONU-1 to ONU-128 are divided into three groups, SR-DBA is applied to the ONUs belonging to each group at a rate of once every three grant cycles, and the other two times are NSR- DBA is applied. In general, when dividing into N groups, SR-DBA is applied to the ONUs belonging to each group at a rate of once every N grant periods, and NSR-DBA is applied the other (N-1) times. The

  With reference to FIG. 4, the exchange of signals (transmission / reception of frames) exchanged between the OLT and the ONU in the DBA method of the first embodiment of the present invention will be described. Fig. 4 shows the time axis indicating the passage of time to the right in the horizontal direction, and time charts for OLT, ONU-1 to ONU-64 (group), and ONU-65 to ONU-128 (group) in order from the top. Show. In FIG. 4, the grant period is shown for the t-th and (t + 1) -th.

  In Fig. 4, thin line arrows indicate transmission of GATE frames transmitted from OLT to ONU, and hatched wide arrows indicate transmission of REPORT frames transmitted from ONU to OLT. The white wide arrow indicates the transmission of the data frame indicated by applying NSR-DBA, and the shaded wide arrow indicates the application of SR-DBA sent from ONU to OLT. Shows the transmission of data frames. In FIG. 4, ONU-1 to ONU-64 are notified of the transmission bandwidth of the data frame determined by applying the REPORT frame transmission permission and NSR-DBA by the GATE frame from OLT. On the other hand, a state is shown in which the ONU-65 to ONU-128 are notified of the transmission band of the data frame determined by applying SR-DBA by the GATE frame from the OLT. In this case, REPORT frame transmission permission is not made.

  The SR-DBA calculation in the t-th grant period (grant t) is the allocated bandwidth for ONU-65 to ONU-128 in the next (t + 1) -th grant period (grant t + 1). Therefore, in the t-th grant period, the OLT collects a REPORT frame for a transmission bandwidth request from the ONU-65 to ONU-128. The allocated bandwidth allocated to each of ONU-65 to ONU-128 is instructed to each ONU from the OLT by the GATE frame. In addition, for ONU-1 to ONU-64, in order to perform SR-DBA calculation at the (t + 1) -th grant period, the uplink data is transmitted in the OLT during the REPORT frame transmission instruction and the NSR-DBA data reception time. Instructed to be able to arrive.

  In ONU-65 to ONU-128, bandwidth allocation calculation is performed by NSR-DBA in the (t + 1) th grant period, so transmission of a REPORT frame is not required. Therefore, the REPORT frame transmission instruction is not performed, and the uplink data arrives at the OLT during the SR-DBA data reception time.

  Thereafter, similar band allocation is performed while alternately switching the group to which NSR-DBA is applied and the group to which SR-DBA is applied for each grant period.

  Here, the DBA method will be specifically described on the assumption that the DBA method of the first embodiment of the present invention is applied to the GE-PON system. The DBA method of the first embodiment of the present invention can be similarly applied to EPON systems other than the GE-PON system as described below.

  The DBA method according to the first embodiment of the present invention can be applied to a conventional GE-PON system. In other words, when applying the DBA method of the first embodiment of the present invention, the conventional GE-PON system can be used as it is.

  A DBA method according to the first embodiment of the present invention will be described with reference to FIG. FIG. 5 is a diagram showing the same system as the optical network communication system of the TDM-PON method shown in FIG. 2, but shows only the components that are particularly necessary for explaining this DBA method, and shows other configurations. The element is omitted. That is, FIG. 5 emphasizes the station-side control unit 24 including the ONU upstream bandwidth allocation unit 28 and the control signal read generation unit 27, and the ONU upstream data frame buffer unit 49 and the subscriber-side control unit 44, respectively. The other components are not shown.

  In the DBA method of the first embodiment of the present invention, when applying NSR-DBA, a predetermined band is set as a fixed band. This fixed bandwidth is determined based on a contract or the like exchanged between the communication carrier and the user, or by taking into consideration the actual state of uplink communication actually performed by the user.

  Uplink data generated from the user equipment 41 connected to the ONU is generated as a data frame for transmission and is temporarily stored in the uplink data frame buffer unit 49. At this time, the stored traffic size (queue length) indicating the amount of data stored in the upstream data frame buffer unit 49 is notified from the upstream data frame buffer unit 49 to the subscriber side control unit 44.

  The REPORT frame transmitted from ONU to OLT has stored traffic size information. The OLT receives the REPORT frame transmitted from each ONU, extracts the accumulated traffic size information, and executes DBA based on this information. The allocated bandwidth for uplink transmission determined by the DBA is notified from the OLT to each ONU by the GATE frame.

  With reference to FIG. 6, the DBA method of the first embodiment of the present invention will be schematically described.

  Step S1 is a grouping step, and ONU grouping processing is performed. This step corresponds to the first step described above, which is a step of dividing ONUs in a link-up connection state into a plurality of groups among the plurality of ONUs.

  The number of groups is determined in advance. The number of groupings (the parameter N described above) to be set is a design item that should be determined by comprehensively considering the number of ONUs, the average traffic size of the ONUs, and the like. Here, assuming that 128 ONUs are connected to the OLT, these ONUs are assumed to be divided into two groups as shown in Table 1. However, in this case, the ONUs are divided into three or more groups. Even if it exists, it is the same.

  Step S2 is an overhead bandwidth calculation step, which is a step of calculating an overhead bandwidth necessary for receiving a REPORT frame and a data frame transmitted from the ONU. The overhead band is composed of Laser on time, Laser off time, sync time, and guard time. “Laser on time” and “Laser off time” indicate the time taken to turn on / off the laser element. The sync time indicates the time until the clock extractor provided in the OLT locks the clock. The guard time indicates the guard time of the uplink signal, and no signal is transmitted during this time period. The maximum values of parameters such as Laser on time, Laser off time, sync time, and guard time are defined in IEEE 802.3ah.std.

  In this embodiment, since the ONU is divided into two groups, the transmission time of the REPORT frame can be halved, and the overhead bandwidth for REPORT frame transmission can be halved. Therefore, the shortened time band can be allocated to the data frame transmission time band.

  Step S3 is a remaining bandwidth calculation step, which is a step of calculating an allocated bandwidth that is a bandwidth obtained by subtracting the overhead bandwidth from the all uplink signal transmission bandwidth. The band left after the subtraction is a band allocated to each ONU for data frame transmission.

  Step S4 is a fixed bandwidth allocation step, which is a step of allocating a fixed bandwidth to an ONU belonging to a group to which NSR-DBA is applied among the two groups. This allocation is performed regardless of whether the ONU is requesting bandwidth.

  Step S5 is a re-remaining bandwidth calculation step, in which the fixed bandwidth allocated in the fixed bandwidth allocation step is subtracted from the allocated bandwidth, and the remaining bandwidth is calculated again.

  Step S6 is a best effort bandwidth allocation step, in which the best effort bandwidth based on the SR-DBA calculation is allocated to ONUs belonging to the group to which SR-DBA is applied among the two groups. Since the SR-DBA calculation in this step is not performed for all ONUs but for half of the ONUs, the number of calculations can be reduced to half.

  Hereinafter, step S1 to step S6 will be described in detail. Table 2 collectively shows parameters used in the flowchart referred to in this detailed description. The DBA method according to the first embodiment of the present invention is characterized in that new steps S1 and S4 that are not included in the conventional similar method are included.

  With reference to FIG. 7, step S1, which is a grouping step, will be described in detail. As shown in FIG. 7, it is determined whether the grant period number t is an odd number or an even number.

  In FIG. 7, N / G_num means the number of groups divided by G_num, which is 2 here. N means the total number of ONUs, here 128. Therefore, in the flowchart, the instruction n = 1 to (N / G_num) means that the parameter n is changed from 1 to 64 and the type of ONU assigned to the group n1 is determined. That is, when ONU-1 to ONU-64 are assigned to group n1 and ONU-65 to ONU-128 are assigned to group n2, the display of group n1 indicates that the ONU included in this group is ONU-1 Indicates 64 ONU-64 ONUs. Similarly, the display of group n2 indicates that the ONUs included in this group are 64 ONUs of ONU-65 to ONU-128.

  According to the flowchart shown in FIG. 7, when the grant period number t is an even number, ONU-1 to ONU-64 are assigned to the group n1, and ONU-65 to ONU-128 are assigned to the group n2. On the other hand, when the number of grants is an odd number, ONU-65 to ONU-128 are assigned to group n1, and ONU-1 to ONU-64 are assigned to group n2.

  Referring to FIG. 8, step S2 that is an overhead bandwidth calculation step and step S3 that is a remaining bandwidth calculation step will be described in detail. Step S2 includes steps S2-1 and S2-2. As shown in FIG. 8, in the flow shown in step S2-1, the report overhead (ROH) for the ONU for transmitting the REPORT frame is integrated (ROH_sum). Here, since all ONUs are divided into two groups in all steps, REPORT frames are transmitted only from half of all ONUs. Therefore, the loop in this step S2-1 covers all of the ONUs 1 to 64 belonging to the group n1, and the last ONU is ONU-64. In the flowchart shown in FIG. 8, “For n = 1 to n1_num” in the flow of step S2-1 means this.

  In the flowchart shown in FIG. 8, ROH_sum = ROH_10G and ROH_sum = ROH_1G are shown in the flow of step S2-1 because the report overhead ROH_10G with a data rate of 10 Gbit / s or the data rate of 1 Gbit / s It shows that the report overhead ROH_1G, which is s, is sequentially added to ROH_sum.

  A signal transmitted intermittently at a time when ONU is allowed to transmit is called a burst signal. In the flow indicated as step S2-2, the burst overhead (BOH) of the data frame carrying the burst signal is accumulated (BOH_sum). Since the burst overhead is a band accompanying data frame transmission, the number of loops is N, which is the number of all ONUs in the link-up connection state. In the flowchart shown in FIG. 8, “For n = 1 to N” in the flow of step S2-2 means this.

  Since both report overhead (ROH) and burst overhead (BOH) differ depending on the transmission rate (data rate) of the PON system, the data rate is the same in both steps S2-1 and S2-2. Have made a selection. Here, 10 Gbit / s and 1 Gbit / s are selected.

  Step S3, which is a remaining bandwidth calculation step, is a step of subtracting the sum of ROH_sum and BOH_sum (ROH_sum + BOH_sum) from the bandwidth (DBA_period) corresponding to the grant period.

  With reference to FIG. 9, step S4 that is a fixed bandwidth allocation step and step S5 that is a remaining bandwidth calculation step will be described in detail. As shown in FIG. 9, the ONU to which fixed band allocation is performed is determined by the result of step S1, which is the above-described grouping step. Here, the ONUs to which fixed bandwidth allocation is performed in the first grant period are the n2 groups of ONU-65 to ONU-128. Therefore, the loop in this flow covers all 65 to 128 of ONUs belonging to the group n2, and the first ONU is ONU-65 and the last ONU is ONU-128.

  Alloc [n] = Fix [n] means that the fixed allocated bandwidth (Fix [n]) is sequentially allocated to the allocated bandwidth (Alloc [n]) (step S4). RBW_ = Fix [n] means that the remaining bandwidth is updated by subtracting the fixed allocation bandwidth (Fix [n]) from the remaining allocation bandwidth (RBW). That is, after allocating a fixed band to the ONU to be allocated with the fixed band, the band is subtracted from the remaining band, and the remaining band is updated (step S5).

  With reference to FIG. 10, step S6 which is the best effort bandwidth allocation step will be described in detail. Step S6 includes steps S6-1 to S6-4.

  As shown in FIG. 10, in the flow shown in Step S6-1, the number of ONUs requesting bandwidth, that is, the number of ONUs subject to best effort calculation N_BE, and the ONU accumulated uplink traffic size (minimum guaranteed bandwidth) R_min The sum of [n1 [n]] is calculated. ONU accumulated uplink traffic size (maximum) R_max [n1 [n]] is determined whether it is positive or negative. If it is positive, it is added as the number of ONUs subject to best effort calculation (N_BE ++), and the sum of ONU accumulated uplink traffic size is calculated. Is executed. In this flowchart, Sum_Rmin + = Rmin [n1 [n]] is shown in the flowchart for calculating the sum of ONU accumulated uplink traffic sizes. If it is negative, it is not added as the number of ONUs subject to best effort calculation.

  Next, step S6-2 is executed to determine whether there is a remaining bandwidth (RBW> 0) and whether there is an ONU requesting bandwidth (N_BE> 0). If both are true, a best effort calculation is performed in step S6-3. If not true, the calculation is terminated.

  In step S6-3, the best effort calculation is performed for the ONU subject to the best effort calculation. Here, the parameter R_min [n1 [n]] is used as a proportional coefficient. Therefore, by setting R_min [n1 [n]] to a value reflecting the contract service level of each user, bandwidth allocation corresponding to the service level is realized.

  In the flowchart shown in FIG. 10, since the ONU that is subject to the best effort calculation is the ONU that belongs to the group n1, the loop in this flow is the last ONU across all 1 to 64 of the ONUs that belong to the group n1. Is ONU-64. In Step S6-3 in the flow shown in FIG. 10, “For n = 1 to n1_num” means that this loop is a loop that sequentially performs processing from ONU-1 to ONU-64. Yes. Also, in this loop, BE [n] = RBW × R_min [n1 [n]] / sum_Rmin is shown in the best effort band BE [n] and the remaining allocated band (RBW) is a proportional coefficient R_min This shows that the calculation of sequentially multiplying values obtained by multiplying [n1 [n]] and dividing by the total value Sum_Rmin of the minimum guaranteed bandwidth is executed.

  Following step S6-3, step S6-4 is executed. In step S6-4, the best effort band (BE [n]) calculated in step S6-3 is allocated to the target ONU. This allocation is performed for the ONU requesting the bandwidth.

  If the best effort bandwidth (BE [n]) is equal to or greater than the requested bandwidth (R_max [n1 [n]]), that is, BE [n1 [n]] ≧ R_max [n1 [n]], the target ONU Since all of the requested bandwidth (Rmax [n1 [n]]) can be allocated to the requested bandwidth, the requested bandwidth (Rmax [n1 [n]]) is allocated to the allocated bandwidth (Alloc [n1 [n]]). Next, the values of RBW, sum_Rmin, N_BE_n, and Rmax [n1 [n]] are updated. In the flow shown in FIG. 10, this series of operations is “Alloc [n1 [n]] = Rmax [n1 [n]]”, “RBW = Rmax [n1 [n]]”, “sum_Rmin = _Rmin [n1 [n] ]] ”,“ N_BE_n ”, and“ Rmax [n1 [n]] = 0 ”.

  On the other hand, when the requested bandwidth (R_max [n1 [n]]) is larger than the best effort bandwidth (BE [n1 [n]]), that is, BE [n1 [n]] <R_max [n1 [n]] In this case, since the entire requested bandwidth cannot be allocated, the best effort bandwidth (BE [n1 [n]]) is allocated to the allocated bandwidth (Alloc [n1 [n]]). Also, N_BE_n and sum_Rmin are not updated because the number of ONUs requesting bandwidth has not decreased. In the flow shown in FIG. 10 for this series of operations, "Alloc [n1 [n]] = BE [n1 [n]]", "RBW- = BE [n1 [n]]", "sum_Rmin- = BE [n1 [n]] ”.

  After step S6-4 ends, the flow continues as long as there is a remaining bandwidth and there is an ONU requesting bandwidth. In the conventionally known DBA method, the processing loop shown in step S6-1, step S6-3, and step S6-4 is not for ONU-1 to ONU-64, but to ONU-1 to ONU-128. Is executed against. In contrast, in the DBA method according to the first embodiment of the present invention, the number of ONUs included in the processing loop is halved, so the time required for the DBA processing is also halved.

  In order to verify the effect of improving the data transfer efficiency by the DBA method of the first embodiment of the present invention, a comparison was made with the data transfer efficiency in the case of the conventional DBA method. Table 3 lists the parameters that represent the data rate, grant period, overhead, and REPORT frame length used to make this comparison.

  FIG. 11 shows the configuration of the REPORT frame and data frame transmitted by ONU used for numerical calculation. In FIG. 11, the horizontal axis is the time axis, and each band “Guard Time”, “Laser on time”, “sync Time”, “Report”, “Laser off time” and data constituting the REPORT frame are shown. “Guard Time”, “Laser on time”, “sync Time”, “Data”, and “Laser off time” constituting the frame are shown side by side on the time axis.

  In FIG. 11, a block described as “Data” is a data transfer band, and other blocks are non-data transfer bands.

  FIG. 12 shows the results of calculating the data transfer efficiency in the conventional DBA method and the DBA method of the first embodiment of the present invention. The horizontal axis in FIG. 12 indicates the number of ONUs that are linked up, and the vertical axis indicates the data transfer efficiency as a percentage.

  The data transfer efficiency by the conventional DBA method is shown by a solid line 1. In addition, the data transfer efficiency according to the DBA method of the first embodiment of the present invention is indicated by broken lines 2, 4, 8, and 16. The broken line 2 is divided into 2 groups for DBA (N = 2), the broken line 4 is divided into 4 groups for DBA (N = 4), and the broken line 8 is divided into 8 groups for DBA (N = 8), the broken line 16 indicates the data transfer efficiency when DBA is performed in 16 groups (N = 16).

  When the number of ONUs linked up is 128, the data transfer efficiency between the conventional DBA method and the DBA method of the first embodiment of the present invention is 17% different as shown in FIG. is there. That is, according to the DBA method of the first embodiment of the present invention, the data transfer efficiency is improved by 17%. Also, in the DBA method of the first embodiment of the present invention, it is shown that the data transfer efficiency improves as the number of groupings N is increased to 2, 4, 8, and 16.

  The data transfer efficiency shown in FIG. 12 assumes a data rate of 10 Gbit / s assuming 10G-EPON, but similar results can be obtained even when the data rate is 1 Gbit / s.

<DBA method of the second embodiment>
In the above-described DBA method according to the first embodiment of the present invention, a predetermined value is used as the fixed band. On the other hand, in the DBA method of the second embodiment of the present invention, the traffic size of the uplink data received from all ONUs is monitored as a fixed bandwidth, and the bandwidth prediction calculated based on the traffic size of each ONU Use the value. As described above, even if the bandwidth prediction value is used as the fixed bandwidth, it is not necessary to refer to the REPORT frame in the NSR-DBA calculation process. Therefore, as in the case of applying the DBA method of the first embodiment, the REPORT The number of frame transmissions can be reduced, improving data transmission efficiency and reducing DBA calculation time.

  A DBA method according to the second embodiment of the present invention will be described with reference to FIG. FIG. 13 is a diagram showing the same system as the optical network communication system of the TDM-PON method shown in FIG. 2 as in FIG. 5, but only the components that are particularly necessary for explaining this DBA method. The other components are omitted. The difference between the optical network communication system of the TDM-PON method shown in FIG. 5 and the optical network communication system of the TDM-PON method shown in FIG. 13 is that the traffic size of the uplink data frame received from the ONU is further changed to the OLT. The traffic monitoring unit 30 for monitoring is provided. The configuration is the same as that of the optical network communication system of the TDM-PON method shown in FIG. 5 except that the OLT further includes a traffic monitoring unit 30, and thus the description of the overlapping components is omitted.

  The traffic monitor unit 30 monitors the traffic size of the upstream data frame transmitted from each ONU to the OLT. The monitoring result is sent to the station side control unit 24. In the station-side control unit 24, a fixed band determined by reflecting this monitoring result is allocated to the ONU to which the band is allocated by the NSR-DBA.

  With reference to FIG. 14, the exchange of signals (transmission / reception of frames) exchanged between the OLT and the ONU in the DBA method of the second embodiment of the present invention will be described. In FIG. 14, the time axis indicating the passage of time is shown to the right in the horizontal direction, and time charts for OLT, ONU-1 to ONU-64 (group), ONU-65 to ONU-128 (group) are shown in order from the top. Show. Further, in FIG. 14, the grant period is shown for the tth, (t + 1) th, and (t + 2) th.

  In FIG. 14, thin line arrows indicate transmission of GATE frames transmitted from OLT to ONU, and hatched wide arrows indicate transmission of REPORT frames transmitted from ONU to OLT. The white wide arrow indicates the transmission of the data frame indicated by applying NSR-DBA, and the shaded wide arrow indicates the application of SR-DBA sent from ONU to OLT. Shows the transmission of data frames.

  In implementing DBA according to the second embodiment of the present invention, first, bandwidth allocation is performed according to either SR-DBA or NSR-DBA bandwidth allocation method in order to transmit an uplink data frame.

  In the example shown in FIG. 14, in the t-th grant period (grant t), a predetermined fixed band is allocated to ONU-65 to ONU-128, and the next (t + 1) -th grant period (grant) DBA is performed in which a predicted bandwidth calculated based on the traffic size from t + 1) is assigned as a fixed bandwidth.

  Band allocation can also be performed by a method other than the example shown in FIG. For example, SR-DBA can be used as a bandwidth allocation method in the t-th grant period. In this case, it is the same as the method of executing the conventional DBA, and in the t-th grant period, bandwidth allocation by SR-DBA is performed for all ONUs that are linked up. Then, DBA is performed in which the predicted bandwidth value calculated based on the traffic size from the next (t + 1) th grant period is assigned as a fixed bandwidth.

  The monitoring time zone of the traffic size by the traffic monitoring unit 30 is set to the transmission time zone of the data frame instructed by applying the NSR-DBA of OLT. Based on the traffic size of each ONU ascertained based on the monitoring result, a bandwidth prediction value is calculated.

  This bandwidth calculation calculation processing is completed before the SR-DBA calculation executed in the (t + 1) -th grant period (grant t + 1) is started, and the calculation result is SR-DBA. Delivered to calculation.

  In the SR-DBA calculation, this predicted bandwidth value is used as an allocated bandwidth for NSR-DBA calculation (corresponding to the fixed allocated bandwidth Fix [n] in the DBA method of the first embodiment of the present invention). The SR-DBA calculation result is notified to each ONU by the GATE frame as the allocated bandwidth of each ONU. The allocated bandwidth notified here is the allocated bandwidth in the (t + 2) th grant period (grant t + 2) of each ONU.

  Therefore, the traffic monitoring result in the t-th grant period (grant t) is reflected as the allocated bandwidth in the (t + 2) -th grant period (grant t + 2), which is two grants away. The REPORT frame transmission instruction is issued only to the ONU that is the target of bandwidth allocation by the SR-DBA in the (t + 1) th grant period (grant t + 1).

  The DBA method according to the first embodiment of the present invention described above is that the same bandwidth allocation is performed while alternately switching the group to which NSR-DBA is applied and the group to which SR-DBA is applied for each grant cycle. It is the same.

  Here, the DBA method will be specifically described assuming that the DBA method of the second embodiment of the present invention is applied to the GE-PON system. The DBA method of the second embodiment of the present invention can be similarly applied to EPON systems other than the GE-PON system as described below.

  Referring to FIG. 15, the DBA method according to the second embodiment of the present invention will be schematically described. The DBA method according to the second embodiment of the present invention includes steps S1 to S3, step S7, step S4 ′, step S5 ′, and step S6. Of these steps, steps S1 to S3 and step S6 are included. Since this is the same as the steps included in the DBA method of the first embodiment described above, description thereof is omitted.

  In the DBA method according to the second embodiment of the present invention, steps S1 to S3 are executed before step S7 which is a data flow bandwidth allocation calculation step. After step S7, steps S4 ′, S5 ′, and S6 of the same type as the DBA method of the first embodiment described above are executed.

  In the data flow bandwidth allocation calculation step (step S7), the ONU belonging to the group to which the NSR-DBA is applied in the time zone in which data is received from the ONU belonging to the group to which the NSR-DBA is applied among the N groups. This is a step of monitoring the traffic size of the uplink data received from, and calculating the bandwidth prediction value based on the traffic size of this ONU.

  The fixed band allocation step (step S4 ′) is a step of allocating the band predicted value calculated in the data flow band allocation calculation step as a fixed band to ONUs belonging to the group to which NSR-DBA is applied among N groups. is there.

  The re-remaining bandwidth calculation step (step S5 ′) is a step of subtracting the fixed bandwidth allocated as the fixed bandwidth from the bandwidth prediction value in the fixed bandwidth allocation step from the allocated bandwidth and calculating the remaining bandwidth again.

  With reference to FIG. 16, step S7 which is a data flow band allocation calculation step will be described in detail. Step S7 includes steps S7-1 to S7-3. Table 4 collectively shows the parameters used in the flowchart referred to in this detailed description.

  As shown in FIG. 16, in the flow shown as step S7-1, the traffic monitoring unit 30 monitors the traffic size (traffic [n2 [n]]) indicating the upstream data flow rate of all ONUs belonging to the group n2. Based on this monitoring result, the integrated value (traffic_sum) of the traffic size of the data amount is calculated.

  In the flow of step S7-1 in the flowchart shown in FIG. 16, “acquisition of traffic [n2 [n]] of each ONU” indicates that the traffic size indicating the upstream data flow rate of the ONU is acquired. I mean. In addition, “traffic_sum + = traffic [n2 [n]]” in the same flow indicates that traffic [n2 [n]] is sequentially integrated with traffic [n2 [n]] in this loop. It means that.

Subsequent to step S7-1, step S7-2 is executed. Step S7-2 is a step of obtaining a band that can be used by the NSR-DBA and is used as a fixed band among all the allocated bands. Bands other than this band are used for best effort band allocation by SR-DBA. This fixed bandwidth is determined in advance. That is, the total bandwidth allocated as fixed bandwidth in the DBA method of the second embodiment is determined in advance, but within this bandwidth, the data flow bandwidth allocation for each ONU bandwidth allocated by the NSR-DBA Bandwidth allocation based on calculation is performed. In the flow of step S7-1 in the flowchart shown in FIG. 16, the parameter indicated as RBW _NSR means a band that can be used in NSR-DBA.

  Subsequent to step S7-2, step S7-3 is executed. Step S7-3 is a step of calculating the allocated bandwidth predicted from the traffic size acquired by the traffic monitoring unit 30. In the example of the data flow bandwidth allocation calculation step shown in FIG. 16, the bandwidth is distributed in proportion to the traffic size. The data flow allocation band NSR-BN [n] obtained from the calculation result is delivered as the fixed allocation band Fix [n].

In the loop of step S7-3, NSR-BW [n2 [n]] = RBW _NSR × traffic [n2 [n]] / traffic_sum indicates that NSR-BW [n] This shows that a calculation is performed in which the bandwidth RBW _NSR usable by the DBA is multiplied by the upstream data flow rate traffic [n2 [n]] of each ONU and the value divided by the traffic integrated value traffic_sum is sequentially integrated.

  Steps S7-1 to S7-3 are executed for the ONU that is the target of bandwidth allocation by NSR-DBA. This is reflected in the expression For n = 1 to n2_num in the flow of steps S7-1 and S7-3 in the flowchart shown in FIG. In the loop shown here, Forn = 1 to n2_num because ONUs belonging to group n2 are targeted for bandwidth allocation by NSR-DBA. Depending on the grant period, ONUs belonging to group n1 are subject to bandwidth allocation by NSR-DBA. In this case, For n = 1 to n1_num.

  Steps S4 ′ and S5 ′, which are fixed bandwidth allocation steps, differ only in that the predicted bandwidth calculated in the data flow bandwidth allocation calculation step is used for the fixed bandwidth. It is the same.

10, 20: OLT
12, 50: Star coupler
14-1: ONU-1
14-2: ONU-2
14-3: ONU-3
21: Upper network
22, 42: Interface converter
23: Downlink signal transmitter
24: Station side controller
24-1: Grouping method
24-2: Fixed bandwidth allocation method
24-3: Best effort bandwidth allocation method
25: Upstream signal receiver
26, 46: Optical multiplexer / demultiplexer
27: Control signal reading and generating means
28: Up band allocation means
30: Traffic monitor
40, 40-1, 40-2, 40-3: ONU
41: User equipment
43: Downlink signal receiver
44: Subscriber side controller
45: Upstream signal transmitter
47: Control signal reading and generating means
48: Up transmission time control means
49: Uplink data frame buffer

Claims (9)

An optical transmission path, an optical multiplexer / demultiplexer coupled to one end of the optical transmission path, and a passive optical network comprising a plurality of demultiplexing optical transmission paths formed by demultiplexing by the optical multiplexer / demultiplexer A dynamic bandwidth allocating method in communication of a passive optical network communication system comprising a station-side terminating device at the other end of the transmission line and a subscriber-side terminal device provided at each of the demultiplexed optical transmission lines Because
Executed by the station-side terminal device;
A grouping step of dividing the plurality of subscriber-side terminal devices into a plurality of groups;
Non-status report-fixed bandwidth allocation step for allocating a fixed bandwidth to a subscriber-side terminal device belonging to a group to which dynamic bandwidth allocation is applied;
And a status report-a best effort bandwidth allocation step for allocating a best effort bandwidth based on a dynamic bandwidth allocation calculation for a subscriber-side terminal device belonging to a group to which the dynamic bandwidth allocation is applied. Dynamic bandwidth allocation method.   The dynamic band allocation method according to claim 1, wherein a predetermined value is set as the fixed band.   As the fixed band, the traffic size of the uplink data received from the subscriber side terminal device belonging to the group to which the non-status report-dynamic band allocation is applied is monitored, and based on the traffic size of each subscriber side terminal device The dynamic band allocation method according to claim 1, wherein the calculated band prediction value is set. An optical transmission path, an optical multiplexer / demultiplexer coupled to one end of the optical transmission path, and a passive optical network comprising a plurality of demultiplexing optical transmission paths formed by demultiplexing by the optical multiplexer / demultiplexer A dynamic bandwidth allocating method in communication of a passive optical network communication system comprising a station-side terminating device at the other end of the transmission line and a subscriber-side terminal device provided at each of the demultiplexed optical transmission lines Because
A grouping step of dividing the plurality of terminal devices on the subscriber side into N groups (N is an integer of 2 or more);
An overhead bandwidth calculating step for calculating an overhead bandwidth required for receiving a report frame and a data frame transmitted from the subscriber side terminal device;
A remaining bandwidth calculating step of calculating an allocated bandwidth that is a bandwidth obtained by subtracting the overhead bandwidth from a total uplink signal transmission bandwidth;
A fixed bandwidth allocation step of allocating a fixed bandwidth to a subscriber-side terminal device belonging to a group to which non-status report-dynamic bandwidth allocation is applied among the N groups;
Re-remaining bandwidth calculation step of subtracting the fixed bandwidth allocated in the fixed bandwidth allocation step from the allocated bandwidth and calculating the remaining bandwidth again;
For the remaining bandwidth calculated in the re-remaining bandwidth calculation step, among the N groups, the status report-dynamic bandwidth is sent to the subscriber side terminal device belonging to the group to which status report-dynamic bandwidth allocation is applied And a best effort bandwidth allocating step for allocating the best effort bandwidth based on the allocation calculation. An optical transmission path, an optical multiplexer / demultiplexer coupled to one end of the optical transmission path, and a passive optical network comprising a plurality of demultiplexing optical transmission paths formed by demultiplexing by the optical multiplexer / demultiplexer A dynamic bandwidth allocating method in communication of a passive optical network communication system comprising a station-side terminating device at the other end of the transmission line and a subscriber-side terminal device provided at each of the demultiplexed optical transmission lines Because
A grouping step of dividing the plurality of terminal devices on the subscriber side into N groups (N is an integer of 2 or more);
An overhead bandwidth calculating step for calculating an overhead bandwidth required for receiving a report frame and a data frame transmitted from the subscriber side terminal device;
A remaining bandwidth calculating step of calculating an allocated bandwidth that is a bandwidth obtained by subtracting the overhead bandwidth from a total uplink signal transmission bandwidth;
Among the N groups, the non-status report-the group to which the dynamic band allocation is applied in the time zone in which data is received from the subscriber side terminal device belonging to the group to which the dynamic band allocation is applied. A data flow rate band allocation calculating step of monitoring a traffic size of uplink data received from the subscriber side terminal device belonging to and calculating a predicted bandwidth value based on the traffic size of each subscriber side terminal device;
Of the N groups, a fixed band that allocates the band predicted value calculated in the data flow band allocation calculation step as a fixed band to a subscriber-side terminal device belonging to a group to which non-status report-dynamic band allocation is applied An assignment step;
Re-remaining bandwidth calculation step of subtracting the fixed bandwidth allocated as the fixed bandwidth in the fixed bandwidth allocation step from the allocated bandwidth and calculating the remaining bandwidth again;
For the remaining bandwidth calculated in the re-remaining bandwidth calculation step, among the N groups, the status report-dynamic bandwidth allocation to the subscriber side terminal device belonging to the group to which the status report-dynamic bandwidth allocation is applied. And a best effort bandwidth allocating step for allocating the best effort bandwidth based on the calculation.   The group which applies non-status report-dynamic bandwidth allocation and the group which applies status report-dynamic bandwidth allocation for every grant period are replaced. Dynamic bandwidth allocation method. In the grouping step, the plurality of subscriber side terminal devices are divided into N groups of a first group to an Nth group,
For each grant period, subscribers belonging to groups that sequentially apply status report-dynamic bandwidth allocation to subscriber-side terminal devices belonging to Group 1 to Group N, and status report-dynamic bandwidth allocation is not applied. 6. The dynamic band allocation method according to claim 1, wherein non-status report-dynamic band allocation is applied to the terminal device on the side. An optical transmission path, an optical multiplexer / demultiplexer coupled to one end of the optical transmission path, and a passive optical network comprising a plurality of demultiplexing optical transmission paths formed by demultiplexing by the optical multiplexer / demultiplexer A passive optical network communication system comprising a station-side terminating device at the other end of the transmission line and a subscriber-side terminal device in each of the demultiplexed optical transmission lines, and implementing the dynamic bandwidth allocation method Because
The station side termination device is:
An uplink band allocating unit that calculates an uplink transmission band for each subscriber-side terminal device; a control signal reading and generating unit that receives an uplink control signal frame from an uplink signal receiving unit and reads information described in the uplink control signal frame; Grouping means for grouping each subscriber-side terminal device into a plurality of groups, and fixed bandwidth allocation for allocating a fixed bandwidth to a subscriber-side terminal device belonging to a group to which non-status report-dynamic bandwidth allocation is applied And a best effort bandwidth allocating means for allocating a best effort bandwidth based on a status report-dynamic bandwidth allocation calculation for a subscriber side terminal device belonging to a group to which dynamic bandwidth allocation is applied. With a control unit,
The subscriber side terminal device is:
An upstream data frame buffer that temporarily stores upstream data frames transmitted from the user equipment and outputs the stored upstream data frames;
A passive optical network communication system comprising: a subscriber-side control unit that monitors the upstream data frame buffer unit and obtains a queue length indicating an amount of data mounted in an upstream data frame in a transmission standby state. . An optical transmission path, an optical multiplexer / demultiplexer coupled to one end of the optical transmission path, and a passive optical network comprising a plurality of demultiplexing optical transmission paths formed by demultiplexing by the optical multiplexer / demultiplexer A passive optical network communication system comprising a station-side terminating device at the other end of the transmission line and a subscriber-side terminal device in each of the demultiplexed optical transmission lines, and implementing the dynamic bandwidth allocation method Because
The station side termination device is:
An uplink band allocating unit that calculates an uplink transmission band for each subscriber-side terminal device; a control signal reading and generating unit that receives an uplink control signal frame from an uplink signal receiving unit and reads information described in the uplink control signal frame; Grouping means for grouping each subscriber-side terminal device into a plurality of groups, and fixed bandwidth allocation for allocating a fixed bandwidth to a subscriber-side terminal device belonging to a group to which non-status report-dynamic bandwidth allocation is applied And a best effort bandwidth allocating means for allocating a best effort bandwidth based on a status report-dynamic bandwidth allocation calculation for a subscriber side terminal device belonging to a group to which dynamic bandwidth allocation is applied. A control unit;
A traffic monitoring unit that monitors the traffic size of the uplink data frame received from the subscriber side terminal device;
The subscriber side terminal device is:
An upstream data frame buffer that temporarily stores upstream data frames transmitted from the user equipment and outputs the stored upstream data frames;
A passive optical network communication system comprising: a subscriber-side control unit that monitors the upstream data frame buffer unit and obtains a queue length indicating an amount of data mounted in an upstream data frame in a transmission standby state. .