The present invention relates to an optical line termination and a data reception processing method in a PON (Passive Optical Network) system.
In a PON system, a plurality of termination devices (ONU: Optical Network Unit, ONT: Optical Network Termination) are connected to an OLT (Optical Line Termination) placed on a station side with an optical coupler and an optical fiber in a 1-to-multiple connection method. Each ONU transmits data in a time division manner, by which an access control is performed to prevent collision of data in the optical coupler. This control causes the OLT to receive data intermittently. The OLT regenerates data by extracting clock from the intermittently received data. Furthermore, a SERDES (Serializer/Deserializer) in the OLT parallelize data by extracting clock from the regenerated data.
A circuit that regenerates data (hereinafter, “data regenerating circuit”) by extracting clock from the received data outputs data with an indefinite rate that is different from a transmission line rate (a rate of reception data) in an interval where there is no reception data. When a PLL (Phase Locked Loop) or the like is used in the SERDES, if the circuit receives data with such an indefinite rate, the SERDES becomes unable to extract a proper clock. Therefore, once the data regenerating circuit is to output reception data again, it takes some time for the SERDES to become able to receive data in a proper manner by extracting the clock in properly. Therefore, when the ONU performs intermittent data transmission in the PON system, a standby period is set until the OLT becomes able to receive data.
As a technique for the OLT to shorten a synchronizing time, for example, Patent Literature 1 mentioned below discloses a technique of preventing a PLL from being out of synchronization by switching a reference of the PLL from input data to a reference clock.
Patent Literature 1: Japanese Patent Application Laid-open No. 2007-159145
However, in the conventional technique described above, although a standby period is set when the ONU performs intermittent data transmission, it is not possible to perform any data transmission in that period. Therefore, there has been a problem in that a bandwidth is wasted as the standby period becomes longer.
The present invention has been made in view of the above problems, and an object of the present invention is to provide an optical line termination and a data reception processing method by use of which it is possible to shorten a standby period until data can be received.
In order to solve the above problems and achieve an object, according to the present invention, included are: a deserializer that parallelizes burst data received from a termination device; a dummy-data generating unit that generates dummy data; and a data selecting unit that detects whether there is the burst data, selects either the dummy data or the burst data based on the detection result, and inputs selected data to the deserializer.
With the optical line termination, the PON system, or the data reception processing method according to the present invention, it is possible to shorten a standby period until data can be received.
Exemplary embodiments of an optical line termination and a data reception processing method according to the present invention will be explained below in detail with reference to the accompanying drawings. The present invention is not limited to the embodiments.
The data regenerating unit 1 is configured by a clock extracting unit 6 that extracts clock from the burst data and a data extracting unit 7 that regenerates data based on the extracted clock and the burst data. Although a configuration example is shown here in which a PLL is employed in the data regenerating unit 1, the configuration for regenerating the burst data is not limited to a particular method, but can be any other configuration such as a configuration without employing any PLL.
The data-regeneration selecting unit 10 is configured by the data selecting unit 3, the data regenerating unit 1, and the fixed-pattern generating unit 2 shown in
The data to be transmitted to the ONU (downlink data) is input to the MAC processing unit 11 via the frame buffer 13. The MAC processing unit 11 makes the transmission data into a PON frame and outputs it to the serializer 42 of the SERDES unit 4. The serializer 42 serializes the transmission data input as a parallel signal and transmits a serialized signal.
The reception data from the ONU (uplink data) is subjected to a clock extraction process and a data regeneration process by the data-regeneration selecting unit 10 and then input to the deserializer 41 of the SERDES unit 4. The deserializer 41 parallelizes the reception data input as the serial signal and outputs it to the MAC processing unit 11. The MAC processing unit 11 performs a process such as a conversion of the parallelized data into a predetermined data format and outputs the result of the process to a user device and the like via the frame buffer 13.
When the deserializer 41 of the SERDES unit 4 extracts the clock by using the PLL circuit (the PLL unit 43) as shown in
To deal with this problem, in a conventional case, when data is received after the state where a proper clock cannot be extracted in the interval where there is no reception data, there has been proposed a method in which the SERDES switches the reference of the PLL in order to shorten the time until the proper clock can be extracted (a synchronizing time). However, in this method, it is necessary to add an extra function to the SERDES.
On the other hand, in the present embodiment, as indicated below, it is configured that clock can be extracted continuously even in the interval where there is no reception data by causing a fixed pattern, which is a data string from which the clock can be extracted in the interval where there is no reception data, to be input to the deserializer 41 based on a transmission start time and a transmission period of each ONU stored in the uplink-slot managing unit 5. Therefore, without adding any extra function to the SERDES, it is possible to start a proper reception process rapidly when the data reception is resumed.
An operation of the present embodiment is explained next. First, an operation example of the uplink-slot managing unit 5 is explained. In the PON system, a data transmission timing of the ONU is controlled by the OLT, and in the present embodiment, the uplink-slot managing unit 5 executes this control.
The uplink-slot managing unit 5 stores the transmission start time and the transmission period of each ONU determined by the allocation-amount calculating unit 51 as allocation information and performs checking whether the data arrived at a right time from each ONU based on the allocation information.
It is assumed that the reference time of each ONU is delayed compared to the reference time of the OLT by a delay time from the OLT to each ONU. Therefore, when the ONU transmitted data at a time T, the data is transmitted at a time T+A (where A is the delay time from the OLT to each ONU) in the reference time of the OLT. The OLT receives the data transmitted from the ONU at a time (T+a+B=T+RTT) when a delay time B is further elapsed from T+A.
As described above, in the OLT, it is possible to obtain the time to start receiving data and the time to end receiving data transmitted from each ONU based on the transmission start time, the transmission period, and the RTT. In the present embodiment, the uplink-slot managing unit 5 outputs the burst head instruction signal at the time to start receiving the data transmitted from each ONU (a burst start position) and the burst end instruction signal at the time to end receiving the data (a burst end position) to the data selecting unit 3.
Dummy data is data with a sufficiently long hamming distance between a burst head identifier and a burst end identifier. The dummy data is transmitted according to clock and makes a transition to 0 or 1 according to a phase of the clock. Therefore, the dummy data can supply required phase information to a phase synchronizing circuit.
The fixed pattern generated by the fixed-pattern generating unit 2 is configured to be generated as data having a sufficient data change for the SERDES unit 4 to extract the clock. Therefore, the SERDES unit 4 becomes in a stable state where it can extract a proper clock even when there is no transmission data from the ONU. Although the fixed pattern can be any type of data, for example, it is possible to use a PRBS7 generating polynomial or a pseudo-random pattern based on PRBS31. Furthermore, although it is configured to generate fixed data as the fixed pattern in the present embodiment, it is not limited to the fixed data, but any data can be used as long as the data is dummy data having comparable rates of 0's and 1's in a predetermined period. The data may be different for each time. That is, the fixed-pattern generating unit 2 is a sort of dummy-data generating unit. The dummy-data generating unit is a unit that generates dummy data (including fixed data) according to a predetermined operation rule.
Furthermore, when the burst head instruction signal becomes High, the data selecting unit 3 determines that the burst head instruction is received, and selects data output from the data regenerating unit 1 as data to be output to the SERDES unit 4 (the burst data #1′ in
In this manner, because the SERDES unit 4 is in a stable state even when there is no transmission data from the ONU, the SERDES unit 4 can receive data in a stable state at a later time when the transmission data is received from the ONU, which can parallelize proper data rapidly. If the burst end instruction signal is further input from the uplink-slot managing unit 5 (that is, the burst end instruction signal becomes High), the data selecting unit 3 selects the fixed data from the fixed-pattern generating unit 2, outputs the selected data to the SERDES unit 4, and becomes in a standby state waiting for the burst head instruction.
Thereafter, the data selecting unit 3 determines whether the burst end instruction is received or not (Step S13). When it is determined that the burst end instruction is received (YES at Step S13), the data selecting unit 3 selects data output from the fixed-pattern generating unit 2, outputs the selected data to the SERDES unit 4 (Step S14), and the process returns to Step S11. On the other hand, when it is determined that the burst head instruction is not received at Step S11 (NO at Step S11), the process proceeds to Step S14. When it is determined that the burst end instruction is not received at Step S13 (NO at Step S13), the process returns to Step S12.
As described above, in the present embodiment, it is configured that the burst head instruction and the burst end instruction are output based on information stored in the uplink-slot managing unit 5 that performs an access control of the PON, and the data selecting unit 3 selects either the fixed pattern or the burst data as data to be output to the SERDES unit 4 based on these signals. Therefore, the SERDES unit 4 can maintain a stable state even when there is no transmission data from the ONU, with a simple structure, which makes it possible to shorten the standby period until data can be received from the ONU. Furthermore, depending on a configuration of a phase synchronizing circuit, it is possible to further shorten the standby period by matching a phase of clock for transmitting dummy data and a phase of a reference clock.
A difference between the present embodiment and the first embodiment is that the burst end instruction signal is output from the block-synchronization detecting unit 8. Operations of the present embodiment other than this are same as those of the first embodiment.
In the first embodiment, the uplink-slot managing unit 5 outputs the burst start instruction signal and the burst end instruction signal based on the control information of the transmission timing of the ONU; however, there is an error between the instruction of the transmission timing to the ONU and a timing at which the OLT actually receives the burst data transmitted by the ONU. This error is generally considered to be smaller than the OLT synchronization period, but is considerably larger than a period of the burst end identifier. In the example shown in
On the other hand, as for the end of the burst data, when the error is larger than a period of a termination code, an error between an actual end timing of the burst data and an instruction of the end by the burst end instruction signal becomes large, and there is a possibility that the data selecting unit 3 switches to the fixed pattern (selects the fixed pattern as the output to the SERDES unit 4) while data is output from the data extracting unit 7. Conversely, it can be considered that a time from the end of the burst data to the switch to the fixed pattern is increased.
To deal with the problem, in the present embodiment, the block-synchronization detecting unit 8 stores a data pattern of the burst end identifier, and compares the stored pattern with burst data output from the data regenerating unit 1 (a pattern check). When the burst data and the stored pattern match each other, the block-synchronization detecting unit 8 determines that a code indicating the end of the burst data (a burst end identifier) is detected, and outputs the burst end instruction signal indicating the burst end.
The data selecting unit 3 then selects data to be output to the SERDES unit 4 based on the burst end instruction signal in the same manner as the first embodiment. Therefore, in the present embodiment, it is possible to reduce the error in switching to the fixed pattern. The uplink-slot managing unit 5a according to the present embodiment does not need to output the burst end instruction signal, but operations thereof other than this are same as those of the uplink-slot managing unit 5 according to the first embodiment. In addition, operations of the present embodiment other than those explained above are same as those of the first embodiment.
As described above, in the present embodiment, it is configured that the block-synchronization detecting unit 8 outputs the burst end instruction signal indicating the burst end when the termination code is detected based on the burst data. Therefore, it is possible to obtain the same effect as the first embodiment, and at the same time, it is possible to prevent an inappropriate switch of the data selecting unit 3 due to the error in the arrival timing of data.
In the present embodiment, the data selecting unit 3a performs a switching based on the burst end instruction signal output from the block-synchronization detecting unit 8 in the same manner as the second embodiment; however, the block-synchronization detecting unit 8 is set to an internal function of the data selecting unit 3a. With this configuration, the data selecting unit 3a detects the burst end by using data output from the data regenerating unit 1, so that it is possible to switch to the fixed pattern at a timing after an end of the actual burst data. That is, an error between the actual burst end and the burst end instruction is eliminated by performing a data selection while performing burst end detection, thus preventing an inappropriate switching to the fixed pattern.
An operation of the data selecting unit 3a including the block-synchronization detecting unit 8 therein is same as that of the data selecting unit 3 according to the second embodiment. In addition, operations of the present embodiment other than those explained above are same as those of the second embodiment.
In the above explanations, regarding the head of data, the data selecting unit 3a is configured to recognize the burst head position based on the burst head instruction signal output from the uplink-slot managing unit 5a in the same manner as the second embodiment. However, as shown in
Thereafter, the data selecting unit 3 compares the stored data pattern of the burst end identifier with the burst data output from the data regenerating unit 1 (burst end pattern check: Step S23). When the stored data pattern of the burst end identifier and the burst data output from the data regenerating unit 1 match each other (YES at Step S23), the data selecting unit 3a selects data output from the fixed-pattern generating unit 2, outputs the selected data to the SERDES unit 4 (Step S24), and returns to Step S21. On the other hand, when it is determined that the data do not match each other at Step S21 (NO at Step S21), the process proceeds to Step S24. When it is determined that the data do not match each other at Step S23 (NO at Step S23), the process returns to Step S22.
As described above, in the present embodiment, it is configured that the block-synchronization detecting unit 8 same as that of the second embodiment is included in the data selecting unit 3a. Therefore, the data selection can be performed while performing the burst end detection, and as a result, it is possible to perform the operation of the second embodiment more rapidly and accurately.
The SERDES unit 4a is configured by the fixed-pattern generating unit 2 that is same as that of the first embodiment, the data selecting unit 3a, a clock extracting unit 6a, a data extracting unit 7a, and a parallelizing unit 9. The data regenerating unit 1 according to the first embodiment includes the clock extracting unit 6 and the data extracting unit 7; however, the deserializer 41 of the SERDES unit 4 according to the first embodiment also performs the clock extraction and the data extraction, and the parallelization process is performed thereafter. In the present embodiment, as shown in
In the data selecting unit 3a, the input burst data is not data after the clock extraction and the data extraction (output from the data regenerating unit 1) but data before the clock extraction and the data extraction, and the data selecting unit 3a outputs the selected data (selects either the fixed pattern or the burst data) to the clock extracting unit 6a; however, functions and operations other than those explained above are same as those of the first embodiment, and the data selecting unit 3a selects data to be output based on the burst end instruction signal and the burst head instruction signal output from the uplink-slot managing unit 5.
Furthermore, the clock extracting unit 6a and the data extracting unit 7a are same as the clock extracting unit 6 and the data extracting unit 7 according to the first embodiment except that a data input source is the data selecting unit 3a and an output destination is the parallelizing unit 9 and to which data of a fixed pattern selected by the data selecting unit 3a is input. The parallelizing unit 9 is a functional unit that performs a parallelizing process from among processes of the SERDES unit 4 according to the first embodiment. Operations of the present embodiment other than those explained above are same as those of the first embodiment.
As described above, in the present embodiment, it is configured that the data regenerating unit 1 and the SERDES unit 4 according to the first embodiment are integrated to a single element, and processing units that perform the clock extraction and the data extraction are set to be common. Therefore, the process can be performed more efficiently than that of the first embodiment, and as a result, it is possible to obtain the same effect as the first embodiment and to shorten the standby period required for data regeneration.
As described above, the optical line termination and the data reception processing method according to the present invention are useful for a PON system, and are particularly suitable in a case of using a PLL for clock extraction.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/JP2009/063761 | 8/3/2009 | WO | 00 | 2/3/2012 |