CN102055652A - Dynamic bandwidth allocation method and device - Google Patents

Dynamic bandwidth allocation method and device Download PDF

Info

Publication number
CN102055652A
CN102055652A CN2009102096525A CN200910209652A CN102055652A CN 102055652 A CN102055652 A CN 102055652A CN 2009102096525 A CN2009102096525 A CN 2009102096525A CN 200910209652 A CN200910209652 A CN 200910209652A CN 102055652 A CN102055652 A CN 102055652A
Authority
CN
China
Prior art keywords
bandwidth
logical channel
token
byte number
current logical
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN2009102096525A
Other languages
Chinese (zh)
Inventor
罗强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
ZTE Corp
Original Assignee
ZTE Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by ZTE Corp filed Critical ZTE Corp
Priority to CN2009102096525A priority Critical patent/CN102055652A/en
Publication of CN102055652A publication Critical patent/CN102055652A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The invention discloses dynamic bandwidth allocation method and device. The method comprises the steps of: obtaining the number of bandwidth bytes requested by the current logic channel from an optical network unit; when the number of bandwidth bytes is greater than the current token number in an average token bucket, borrowing the token in a credit token bucket, wherein the current token number of the credit token bucket is a nonnegative number; and when the sum of the borrowed token number and the current token number in the average token bucket is greater than the number of bandwidth bytes, distributing bandwidth for the current logic channel. According to the technical scheme provided by the invention, the problems of lower bandwidth use ratio and longer data frame delay during distributing the bandwidth in relevant technologies are solved, and the effects of improving the bandwidth use ratio and reducing data frame delay are achieved.

Description

Distribution method of dynamic bandwidth and device
Technical field
The present invention relates to the communications field, in particular to a kind of Dynamic Bandwidth Allocation (Dynamic Bandwidth Assignment abbreviates DBA as) method and device.
Background technology
Gigabit passive optical network (Gigabit-capable Passive Optical Network, abbreviate GPON as) be a kind of gigabit optical access network, by optical line terminal (Optical LineTerminal, abbreviate OLT as), Optical Distribution Network (Optical Distribution Network, abbreviate ODN as) and optical network unit (Optical Network Unit abbreviates ONU as) composition.GPON is the tree structure of one-to-many on physical connection, connect by passive optical distribution network ODN, transfer of data adopts descending (from OLT to ONU) broadcasting, up (from ONU to OLT) time division multiplexing (Time Division Multiplexing, abbreviate TDM as) mode, in order to produce conflict, OLT distributes an ascending time slot for each ONU, ONU can only transmit data in the time slot that is authorized to, OLT is that the process of ONU arrangement time slot is exactly the process of the up realization Dynamic Bandwidth Allocation of GPON (DBA).Wherein, be dynamic change according to the active state of ONU, the bandwidth that flow is reported and configuration parameter distributes.
Typical DBA method has two kinds: the one, and SR (Status reporting) pattern, OLT requires ONU to submit to buffer memory to take report.The 2nd, TM (Traffic monitoring) pattern, OLT monitors the IDLE frame and the bandwidth mapping of up gigabit passive optical network Transmission Convergence (GTC) frame.
At first, at typical DBA method, below primary variables is described:
C is called up GTC frame capacity
RF is called fixed-bandwidth, distributes to the static state of this logical channel (explaining with TCONT), bandwidth independently, and this bandwidth can not change along with the loading condition at whole PON networking.Because bandwidth fixed, the delay of packet is very little, is used for the exigent business of real-time more, for example, and speech business.
RA is called the assurance bandwidth, when TCONT has unsatisfied traffic requirement, and the bandwidth that expectation distributes.
RM is called maximum bandwidth, the bandwidth of the maximum that TCONT can obtain.
RNA is called non-assured bandwidth, and OLT dynamically is assigned to a kind of additional bandwidth of authorizing TCONT, converts according to the ratio of fixed-bandwidth and assurance bandwidth.
RBE is called the bandwidth of doing one's best, and OLT dynamically is assigned to a kind of additional bandwidth of authorizing TCONT, and proportionally conversion non-guaranteed bandwidth.
χ AB is called the additional bandwidth indication, is used for illustrating the additional bandwidth type.
Secondly, at typical DBA method, below the DBA main processes of calculation is described:
Each flow descriptor comprises three bandwidth parameters, RF, and RA and RM, an additional bandwidth distributes selects indication parameter: χ AB.Additional bandwidth distribute to select indication parameter mark non-assured bandwidth to share (NA), does one's best shared (BE) or does not have sharing mode (None).
Wherein, RF and RA need meet the following conditions:
R M i ≥ R F i + R A i
χ AB i = NA , onlyif R M i > R F i + R A i > 0
χ AB i = BE , onlyif R M i > R F i + R A i ≥ 0 - - - ( 1 )
In addition, the summation of fixed-bandwidth and assurance bandwidth need meet the following conditions:
Σ i ( R F i + R A i ) ≤ C - - - ( 2 )
The dynamic flow load (Offered traffic load) of each TCONT is represented with RL (t), RL (t) is defined as the Mean Speed of the logical cache consumption of this TCONT, time is represented system constants (equal 1 GTC frame at least, in fact adopt 8 GTC frames usually)
R L ( t ) = B ( t ) + A ( t , t + Δ ) Δ - - - ( 3 )
The bandwidth Ri (t) 〉=0 that dynamically is assigned to TCONT is by guaranteeing that bandwidth and additional bandwidth form.Additional bandwidth can be the non-assurance bandwidth or the bandwidth of doing one's best.
If χ is iAB=NA
R i ( t ) = R G i ( t ) + R NA i ( t ) - - - ( 4 )
If χ is iAB=BE.
R i ( t ) = R G i ( t ) + R BE i ( t ) - - - ( 5 )
Wherein, guarantee that bandwidth is determined by equation:
R G i ( t ) = min { R F i + R A i ; max { R F i ; R L i ( t ) } } - - - ( 6 )
Wherein, non-ly guarantee that the summation of bandwidth can be expressed as following equation:
S NA ( t ) = C - Σ i min { R F i + R A i ; max { R F i ; R L i ( t ) } } - - - ( 7 )
For the fairness doctrine, participate in sharing any TCONTi and the j of non-assured bandwidth, all should satisfy this two conditions (χ AB=NA), unsaturation (RG (t)+RNA (t)<min{RM, RL (t) }), distribute according to the ratio of following equation simultaneously:
R NA i ( t ) R F i + R A i = R NA j ( t ) R F j + R A j - - - ( 8 )
Other TCONT need satisfy following identity: R (t)=min{RM; RL (t) }.
Non-ly guarantee that the summation of bandwidth can be expressed as following equation:
S BE ( t ) = C - Σ i ∈ { χ AB i = NA } min { R M i ; maz { R F i ; R L i ( t ) } }
- Σ i ∈ { χ AB i ≠ NA } min { R F i + R A i ; max { R F i ; R L i ( t ) } } - - - ( 9 )
For the fairness doctrine, participate in sharing any TCONTi and the j of non-assured bandwidth, all should satisfy this two conditions (χ AB=BE), unsaturation (RG (t)+RBE (t)<min{RM; RL (t) }), distribute according to the ratio of following equation simultaneously:
R BE i ( t ) R M i - ( R F i + R A i ) = R BE j ( t ) R M j - ( R F j + R A j ) - - - ( 10 )
Other TCONT need satisfy following identity: R (t)=min{RM; RL (t) }.
This shows the R in formula (4) and the formula (5) i(t) be the bandwidth byte number that distributes for current TCONT, and guarantee bandwidth in formula (4) and the formula (5)
Figure B2009102096525D0000052
Be to determine through the calculating of formula (6), in the formula (4)
Figure B2009102096525D0000053
Determine by formula (8), in the formula (5)
Figure B2009102096525D0000054
Determine by formula (10).
In the correlation technique, carrying out OLT medium access control (Media AccessControl, abbreviate MAC as) during chip design, generally adopt the strategy of timer-triggered scheduler that the bandwidth request of each ONU is distributed, the advantage of above-mentioned allocation strategy is that the bandwidth of request is distributed according to timeslice.For example, the bandwidth attribute of a TCONT is the fixed-bandwidth type, and then it all equates at the up byte number of permission that each timeslice obtains.
But, there is a bigger defective in above-mentioned implementation: if TCONT data carried by data frame length is a change at random, although statistics in the long period, for example, in 1 second time, the bandwidth of this TCONT can be stabilized in one and comparatively determine at flow value, but, in less timeslice, for example, in 10 milliseconds, will find that the flow jitter of this TCONT may be very violent.The timer-triggered scheduler strategy is to distribute identical byte number in identical timeslice in essence, but because the flow jitter of TCONT, this scheduling strategy is not well positioned to meet user's actual need, because user's actual flow is in the fluctuation of short time, and the strategy of Dynamic Bandwidth Allocation is to equate according to the byte number that distributes on each less timeslice, this just causes allocated bandwidth efficient not enough, and bandwidth availability ratio reduces, and causes Frame to produce bigger time delay simultaneously.
ITUT G984.3 agreement recommends to adopt SR (Status reporting) pattern, and this is the mode that a kind of active bandwidth reports.
But in practical application, under SR (Status reporting) pattern, ONU reports bigger impact flow sometimes, thereby can make the flow of other TCONT in the GPON system supervene vibration.Therefore, be necessary the peak value of upstream bandwidth is limited.
And in the GPON system, the time delay of upstream bandwidth is a very important index.The mechanism of allocated bandwidth should guarantee: long-time statistical, and keep the TCONT flow in the scope that network management is formulated, short-term fluctuation is within the peak-peak that can control.The upstream bandwidth distribution has only the byte number of following substantially in the ONU current cache to change, and can keep stable delay character.
The problem lower at bandwidth availability ratio in the correlation technique, that data frame delay is bigger does not propose effective solution at present as yet.
Summary of the invention
At in the correlation technique when carrying out allocated bandwidth, the problem that bandwidth availability ratio is lower, data frame delay is bigger and propose the present invention, for this reason, main purpose of the present invention is to provide a kind of Dynamic Bandwidth Allocation (Dynamic Bandwidth Assignment, abbreviate DBA as) method, one of to address the above problem at least.
According to an aspect of the present invention, provide a kind of distribution method of dynamic bandwidth.
Distribution method of dynamic bandwidth according to the present invention comprises: obtain the bandwidth byte number that comes from the current logical channel request of optical network unit; When the bandwidth byte number greater than average token bucket in during current token number, use the token in the credit token bucket, wherein, the current token number in the credit token bucket is a nonnegative number; Under the situation of current token number sum in token number of using and average token bucket, be that current logical channel is distributed bandwidth greater than the bandwidth byte number.
According to a further aspect in the invention, provide a kind of dynamic bandwidth allocation apparatus.
Dynamic bandwidth allocation apparatus according to the present invention comprises: acquiring unit is used to obtain the bandwidth byte number that comes from the current logical channel request of optical network unit; Use the unit, be used for using the token in the credit token bucket under the situation of bandwidth byte number greater than the average current token number of token bucket, wherein, the current token number in the credit token bucket is a nonnegative number; The allocated bandwidth unit is used under the situation of current token number sum greater than the bandwidth byte number of token number of using and average token bucket, is that current logical channel is distributed bandwidth.
By the present invention, bandwidth byte number (DBRu) in current logical channel request reports under the bigger situation, allow to use bandwidth and give this logical channel (TCONT), allow it obtain bandwidth response, can control simultaneously and still meet the flow that the user is distributed in network management under the long-term flow, solved in the correlation technique when carrying out allocated bandwidth, the problem that bandwidth availability ratio is lower, data frame delay is bigger, and then reached the effect that improves bandwidth availability ratio, reduces data frame delay.
Description of drawings
Accompanying drawing described herein is used to provide further understanding of the present invention, constitutes the application's a part, and illustrative examples of the present invention and explanation thereof are used to explain the present invention, do not constitute improper qualification of the present invention.In the accompanying drawings:
Fig. 1 is the flow chart according to the distribution method of dynamic bandwidth of the embodiment of the invention;
Fig. 2 is the flow chart of distribution method of dynamic bandwidth according to the preferred embodiment of the invention;
Fig. 3 is the structured flowchart according to the dynamic bandwidth allocation apparatus of the embodiment of the invention;
Fig. 4 is the structured flowchart of dynamic bandwidth allocation apparatus according to the preferred embodiment of the invention;
Fig. 5 is the structured flowchart of the dynamic bandwidth allocation apparatus of example according to the present invention.
Embodiment
Functional overview
Core thinking of the present invention is as follows: at first, reduce the random data packet delay of ONU to OLT, and reduce the time delay of the discharge model of square wave type, accelerate response speed, improve the QOS ability.Secondly, in order to improve the allocative efficiency of upstream bandwidth, make it surpass 90%.Under the guidance of above-mentioned thinking, main policies allows to use bandwidth and gives this TCONT for to report under the bigger situation at DBRu, allows it obtain bandwidth response at once, can control simultaneously still to meet the flow that the user is distributed in network management under the long-term flow.
Hereinafter will describe the present invention with reference to the accompanying drawings and in conjunction with the embodiments in detail.Need to prove that under the situation of not conflicting, embodiment and the feature among the embodiment among the application can make up mutually.
Method embodiment
According to embodiments of the invention, provide a kind of distribution method of dynamic bandwidth.
Fig. 1 is the flow chart according to the distribution method of dynamic bandwidth of the embodiment of the invention.As shown in Figure 1, this method comprises following processing (step S101-step S105):
Before execution in step S101, need be according to the bandwidth byte number of current logical channel request, be current logical channel dynamic-configuration dispatching cycle, wherein, dispatching cycle and bandwidth byte number are inversely proportional to.
Preferably, can be configured by following formula this dispatching cycle:
t i=MT;
Wherein, t iBe dispatching cycle, T is default minimum dispatching cycle, and M be the natural number according to the bandwidth byte number setting of current logical channel request, and M and bandwidth byte number are inversely proportional to.
In specific implementation process, can adopt strategy dispatching cycle of variation.For example, timeslice is configured to 1ms.Be exactly every interval 1ms, dispatch once.If isochronous surface is too little, cause expense excessive, be exactly to distribute fixing byte number in addition in each timeslice, cause up packet, not necessarily fill up the bandwidth of being distributed, cause bandwidth availability ratio to descend.In order to improve allocated bandwidth efficient, can control time delay effectively again simultaneously, therefore the size of selecting the control time sheet is arranged.For example bandwidth attribute is divided into 3 classes, greater than 5Mbps, between 2~5Mbps, less than 2Mbps, for these three kinds of discharge patterns, respectively according to 1: 2: 4 multiple, control scheduling time, if be T the minimum dispatching cycle that allows, be T the dispatching cycle greater than 5Mbps so, be 2T the dispatching cycle between 2~5Mbps, less than the dispatching cycle of 2Mbps be 4T.
At bandwidth byte number according to current logical channel request, be current logical channel dynamic-configuration before dispatching cycle, also need to carry out following processing:
(1) obtain the attribute configuration information that comes from current logical channel, wherein, attribute configuration information is included as the default flow attribution information of each logical channel;
(2) not under the situation of last logical channel in the attribute configuration information in definite current logical channel, do not distribute bandwidth for current logical channel according to flow attribution information corresponding priorities.
Step S101: obtain the bandwidth byte number that comes from the current logical channel request of optical network unit;
Step S103: when the bandwidth byte number greater than average token bucket in during current token number, use the token in the credit token bucket, wherein, the current token number in the credit token bucket is a nonnegative number;
According to scheme provided by the invention, in bandwidth allocation, use to allow the bandwidth token bucket, and the credit of token is used mechanism.Wherein, R FOr R A, or R NA, or R BESpeed add token, if the bandwidth that TCONT obtains then can be used in the credit token bucket, if less than the token number in the current token bucket greater than the token number in the current token bucket, then this TCONT obtains the permission of up data, and the DBA distributor will distribute bandwidth to give this TCONT.
B C i ( t ) = B C i ( t - 1 ) + R 1 i ( t ) - - - ( 11 )
Figure B2009102096525D0000102
Value be R F, R A, R NA, R BEIn one of any.Wherein, R NA, R BECalculate acquisition according to formula (8) and formula (10).
The mechanism of using of token bucket.Allow token bucket to be negative number, system adopts single to allow to use the token number that maximum can be used in time of token and time, the treatment mechanism of using as token.This mechanism is by two parameter control t, and the multiple n of average discharge.Work as token bucket
Figure B2009102096525D0000103
Be negative number, and token bucket The token number less than after can sending byte number, stop to send.Up to the token number that allows the bandwidth token bucket is positive number.
B p i ( t ) = B p i ( t - 1 ) + n R 1 ( t ) i - - - ( 12 )
Figure B2009102096525D0000106
Enable control by the control of square-wave generator, the feature of this square-wave generator that formula (13) is expressed.In time, square wave is a high level at t/k, the token number in the token bucket bucket with
Figure B2009102096525D0000107
Speed increase, wherein n is a constant that can dispose, and is a positive integer.
f ( t ) = 0 1 ( t / k ) - - - ( 13 )
Introduce to allow the bandwidth token bucket, and the credit of token using mechanism, is current logical channel distribution bandwidth according to constraints formula (14) and formula (15).
Figure B2009102096525D0000111
(
Figure B2009102096525D0000112
For negative,
Figure B2009102096525D0000113
For just) (14)
(
Figure B2009102096525D0000115
With
Figure B2009102096525D0000116
Just be) (15)
Wherein,
Figure B2009102096525D0000117
Be formula (11) and formula (12) result of calculation.
Step S105: under the situation of current token number sum in token number of using and average token bucket, be that current logical channel is distributed bandwidth greater than the bandwidth byte number.
Preferably, distribute bandwidth according to one of following condition (being above-mentioned formula (14) or formula (15)) for current logical channel among the step S105;
(1) exists For negative,
Figure B2009102096525D0000119
Be timing,
(2) exist
Figure B2009102096525D00001111
With
Figure B2009102096525D00001112
Be timing,
Figure B2009102096525D00001113
Wherein,
Figure B2009102096525D00001114
Be the current token number in the average token bucket,
Figure B2009102096525D00001115
Be the current token number in the credit token bucket,
Figure B2009102096525D00001116
For guaranteeing bandwidth, R i(t) be the bandwidth of distributing for current logical channel.
Preferably, be after current logical channel is distributed bandwidth, also comprise following processing: if there is the remaining bandwidth byte number in the bandwidth byte number in dispatching cycle, then the remaining bandwidth byte number is distributed to the next logical channel of current logical channel, until the whole bandwidth byte numbers that distribute in dispatching cycle.
Preferably, after the whole bandwidth byte numbers that distribute in dispatching cycle, also comprise following processing: write down bandwidth allocation information, and bandwidth allocation information is sent to optical network unit by descending GTC frame.
Fig. 2 is the flow chart of distribution method of dynamic bandwidth according to the preferred embodiment of the invention.As shown in Figure 2, distribution method of dynamic bandwidth comprises following processing (step S201-step S219) according to the preferred embodiment of the invention:
Step S201:OLT disposes minimum DBA dispatching cycle, and all be this minimum integral multiple of dispatching cycle any dispatching cycle;
Step S203: read TCONT attribute configuration table, wherein, this attribute configuration table has disposed the flow attribution of an expection for each TCONT;
Step S205: judge that whether current TCONT is last of this TCONT attribute configuration table, if, then change execution in step S217 over to, if not, execution in step S207 then;
Step S207: read TCONT traffic policy table,, read TCONT traffic policy table, because different TCONT has different priority levels, so need distribute bandwidth successively according to priority according to the flow attribution of traffic policy list deciding TCONT;
Step S209: from the ONU data field, read DBRu, with the token number in the average token bucket
Figure B2009102096525D0000121
Compare, if
Figure B2009102096525D0000122
Greater than DBRu, then directly carry out allocated bandwidth and calculate, otherwise, change execution in step over to.
Step S211: use the token number from the credit token bucket, may occur in the Bc token bucket this moment
Figure B2009102096525D0000123
Less than 0, the token number of then carrying out next stage computing is
Figure B2009102096525D0000124
In the Bc token bucket
Figure B2009102096525D0000131
Greater than 0, the token number of then carrying out next stage computing is
Figure B2009102096525D0000132
Step S213: the token number that adopts step S211 to calculate, compare with the DBRu byte number, if the former is bigger, execution in step S213 calculates the bandwidth of actual allocated so, otherwise, distribute bandwidth then no longer for this TCONT, change step S203 over to.
Step S215: distribute R in the total bytes that from dispatching cycle, can distribute i(t).
Step S217: whether calculating also has remainder bytes to distribute, if also have byte to distribute in dispatching cycle, continues to change over to step S205 so, till all bytes all have been assigned with.
Step S219: the bandwidth that TCONT is distributed writes the BWMap table, and the control information formation bandwidth control table with relevant sends to ONU by descending GTC frame.
By the foregoing description, provide a kind of distribution method of dynamic bandwidth.Handle through this method, can improve the utilization ratio of upstream bandwidth, through data computation, can be under random data bag condition, raise the efficiency and surpass 90%, surpass present 80% situation.For the random data bag, adopt the mode of using to handle, can improve fixed-bandwidth, guarantee bandwidth, the time delay of the non-assured bandwidth and the bandwidth of doing one's best.As calculated, the time delay of system's maximum can be tapered to 8ms, even in the 4ms.
Device embodiment
According to embodiments of the invention, provide a kind of dynamic bandwidth allocation apparatus.
Fig. 3 is the structured flowchart according to the dynamic bandwidth allocation apparatus of the embodiment of the invention.Fig. 4 is the structured flowchart of dynamic bandwidth allocation apparatus according to the preferred embodiment of the invention.As shown in Figure 3, comprise: acquiring unit 30, use unit 32, allocated bandwidth unit 34, each unit is described below in conjunction with Fig. 4 according to the dynamic bandwidth allocation apparatus of the embodiment of the invention.
Acquiring unit 30 is used to obtain the bandwidth byte number that comes from the current logical channel request of optical network unit;
Use unit 32, be connected with acquiring unit 30, be used for using the token in the credit token bucket under the situation of bandwidth byte number greater than the average current token number of token bucket, wherein, the current token number in the credit token bucket is a nonnegative number;
Allocated bandwidth unit 34 and is used unit 32 and is connected, is used under the situation of current token number sum greater than the bandwidth byte number of the token number of using and average token bucket, and be that current logical channel is distributed bandwidth.
Preferably, said apparatus also comprises: dispatching cycle, dispensing unit 36, were used for the bandwidth byte number according to current logical channel request, were current logical channel dynamic-configuration dispatching cycle, and wherein, dispatching cycle and bandwidth byte number are inversely proportional to.
Preferably, above-mentioned acquiring unit 30 also is used to obtain the attribute configuration information that comes from current logical channel, and wherein, attribute configuration information is included as the default flow attribution information of each logical channel; Allocated bandwidth unit 34, also being used in definite current logical channel is not under the situation of last logical channel of attribute configuration information, does not distribute bandwidth for current logical channel according to flow attribution information corresponding priorities.
Preferably, above-mentioned allocated bandwidth unit 34, when also being used for bandwidth byte number in dispatching cycle and having the remaining bandwidth byte number, the remaining bandwidth byte number is distributed to the next logical channel of current logical channel, until the whole bandwidth byte numbers that distribute in dispatching cycle.
Preferably, said apparatus can also comprise: record cell 38 is used to write down bandwidth allocation information; Transmitting element 40 is used for bandwidth allocation information is sent to optical network unit by descending GTC frame.
Fig. 5 is the structured flowchart of the dynamic bandwidth allocation apparatus of example according to the present invention.As shown in Figure 5, the dynamic bandwidth allocation apparatus of example comprises according to the present invention: minimum dispatching cycle of control unit 100 (be equivalent to above-mentioned dispatching cycle dispensing unit), TCONT traffic policy table unit 101, credit token bucket unit 102, Mean Speed control token bucket unit 103, allocated bandwidth unit 104 (being equivalent to above-mentioned allocated bandwidth unit), BWMAP table dispensing unit 105 (being equivalent to above-mentioned record cell), BWMAP transmitting element 106 (being equivalent to above-mentioned transmitting element), use control unit 107, DBRu reports table unit 108.Below in conjunction with Fig. 5 each unit is described.
Minimum control unit dispatching cycle 100: be used to dispose minimum dispatching cycle of DBA, this cycle is generally the integral multiple of GTC frame time, and wherein, the GTC frame time is 125us.
TCONT traffic policy table unit 101: be connected with allocated bandwidth unit 104, be used to dispose TCONT traffic policy table, wherein, this unit need be to RF, and RA and RM are configured;
Credit token bucket unit 102: be used for interpolation speed according to formula (12) control token.Can configuration parameter K control the shake of up TCONT flow.
Mean Speed control token bucket unit 103: be connected with allocated bandwidth unit 104, be used for generating rate according to formula (11) control token.
Allocated bandwidth unit 104: be used for carrying out the calculating of bandwidth authorizing according to formula (14),
Figure B2009102096525D0000151
Calculate R according to formula (6) i(t) according to formula (7), formula (8), formula (9) calculates, RF, the value of RA is the configuration of TCONT traffic policy table unit.The order of allocated bandwidth is fixed by the TCONT traffic policy voting of TCONT traffic policy table unit 101, at first from fixed-bandwidth (RF), secondly is to guarantee bandwidth (RA), is non-assurance bandwidth (RNA) once more, is the bandwidth (RBE) of doing one's best at last.For the fixed-bandwidth (RF) of low discharge, the bandwidth that need calculate according to maximum cycle is at first distributed.The allocated bandwidth unit is with the R of TCONT i(t) elder generation compares with the token number of average discharge token bucket, if bandwidth on demand less than token bucket, then allows to pass through.If this moment, bandwidth on demand surpassed the token number, then from the credit token bucket, use token, as long as the token number in the credit token bucket can be used not for negative.Satisfy bandwidth on demand if add the token number of using, then allowed the distribution of this TCONT participation bandwidth, if less than, do not participate in this so and take turns distribution.If epicycle is distributed bandwidth left is arranged still, the bandwidth of distributing for this TCONT is the token number in the average token bucket and the bandwidth of the token number sum correspondence in the credit token bucket so.
BWMAP shows dispensing unit 105: be connected with allocated bandwidth unit 104, be used for the bandwidth that unit 105 distributes is write the BWMap table, this table has stipulated that each descending GTC frame passes to the breath of taking a message of ONU, and other control informations.
BWMAP transmitting element 106: the Access clauses and subclauses that send the BWMap table from OLT send to ONU.The Access clauses and subclauses are parts of descending GTC frame structure, are used for transmitting bandwidth information and other control informations.The control bandwidth that is used at this unit arrives the mapping relations of GTC frame structure.
Use control unit 107: being used for controlling the speed of using in the unit interval according to formula (13), is the peak rate that is used for controlling TCONT in essence, avoids bringing spike to impact to system.
DBRu reports table unit 108: respectively with allocated bandwidth unit 104 with use control unit 107 and be connected, be used to write down the buffer status information that each TCONT of each ONU reports, wherein, above-mentioned information is the basis that DBA distributes, do not have DBRu accurately, just can't produce DBA result accurately.Wherein, the reporting of this information generally was controlled within minimum dispatching cycle, the chance that all TCONT must acquisition once report.The bandwidth that DBRu reports is also distributed by BWMAP table dispensing unit 105.
By the foregoing description, provide a kind of dynamic bandwidth allocation apparatus.Mutually combine by each unit, can improve the utilization ratio of upstream bandwidth, through data computation, can be under random data bag condition, raise the efficiency and surpass 90%, surpass present 80% situation.For the random data bag, adopt the mode of using to handle, improved fixed-bandwidth, guarantee bandwidth, the time delay of the non-assured bandwidth and the bandwidth of doing one's best.As calculated, the time delay of system's maximum can be tapered to 8ms, even in the 4ms.
In sum, pass through the above embodiment of the present invention, the Dynamic Bandwidth Allocation scheme that provides, average discharge by the up TCONT of token bucket control ONU, simultaneously at up burst flow, the mode that adopts token bucket to use can improve the up TCONT of ONU and obtain the distribution of bandwidth, thereby reach the time of delay that has shortened the random data bag.Apply for up specific discharge model, for example, the input rate of TCONT is the Burst mode, has accelerated the response time of OLT distribution bandwidth, has improved the QOS ability of system effectively.
Obviously, those skilled in the art should be understood that, above-mentioned each module of the present invention or each step can realize with the general calculation device, they can concentrate on the single calculation element, perhaps be distributed on the network that a plurality of calculation element forms, alternatively, they can be realized with the executable program code of calculation element, thereby, they can be stored in the storage device and carry out by calculation element, perhaps they are made into each integrated circuit modules respectively, perhaps a plurality of modules in them or step are made into the single integrated circuit module and realize.Like this, the present invention is not restricted to any specific hardware and software combination.
The above is the preferred embodiments of the present invention only, is not limited to the present invention, and for a person skilled in the art, the present invention can have various changes and variation.Within the spirit and principles in the present invention all, any modification of being done, be equal to replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (12)

1. a distribution method of dynamic bandwidth is characterized in that, comprising:
Obtain the bandwidth byte number that comes from the current logical channel request of optical network unit;
When described bandwidth byte number greater than average token bucket in during current token number, use the token in the credit token bucket, wherein, the current token number in the described credit token bucket is a nonnegative number;
Under the situation of current token number sum in token number of using and described average token bucket, for described current logical channel is distributed bandwidth greater than described bandwidth byte number.
2. method according to claim 1 is characterized in that, is that described current logical channel is distributed bandwidth according to one of following condition:
Figure F2009102096525C0000011
For negative, Be timing,
Figure F2009102096525C0000013
Figure F2009102096525C0000014
With
Figure F2009102096525C0000015
Be timing,
Figure F2009102096525C0000016
Wherein, described
Figure F2009102096525C0000017
For the current token number in the described average token bucket, described
Figure F2009102096525C0000018
For the current token number in the described credit token bucket, described
Figure F2009102096525C0000019
For guaranteeing bandwidth, described R i(t) be the bandwidth of distributing for described current logical channel.
3. method according to claim 1 and 2 is characterized in that, before obtaining the bandwidth byte number that comes from the current logical channel request of optical network unit, described method also comprises:
According to the bandwidth byte number of described current logical channel request, be described current logical channel dynamic-configuration dispatching cycle, wherein, described dispatching cycle and described bandwidth byte number are inversely proportional to.
4. method according to claim 3 is characterized in that, be configured by following formula described dispatching cycle:
t i=MT, wherein, t iBe described dispatching cycle, T is default minimum dispatching cycle, and M be the natural number according to the bandwidth byte number setting of described current logical channel request, and M and described bandwidth byte number are inversely proportional to.
5. method according to claim 3 is characterized in that, according to the bandwidth byte number of described current logical channel request, is described current logical channel dynamic-configuration before dispatching cycle, and described method also comprises:
Obtain the attribute configuration information that comes from described current logical channel, wherein, described attribute configuration information is included as the default flow attribution information of each logical channel;
In definite described current logical channel is not under the situation of last logical channel in the described attribute configuration information, does not distribute bandwidth for described current logical channel according to described flow attribution information corresponding priorities.
6. method according to claim 3 is characterized in that, is being that described method also comprised after described current logical channel was distributed bandwidth:
If there is the remaining bandwidth byte number in the bandwidth byte number in described dispatching cycle, then described remaining bandwidth byte number is distributed to the next logical channel of described current logical channel, until the whole bandwidth byte numbers that distribute in described dispatching cycle.
7. method according to claim 6 is characterized in that, after the whole bandwidth byte numbers that distribute in described dispatching cycle, described method also comprises:
Write down bandwidth allocation information, and described bandwidth allocation information is sent to described optical network unit by descending gigabit passive optical network Transmission Convergence GTC frame.
8. a dynamic bandwidth allocation apparatus is characterized in that, described device comprises:
Acquiring unit is used to obtain the bandwidth byte number that comes from the current logical channel request of optical network unit;
Use the unit, be used for using the token in the credit token bucket under the situation of described bandwidth byte number greater than the average current token number of token bucket, wherein, the current token number in the described credit token bucket is a nonnegative number;
The allocated bandwidth unit is used under the situation of current token number sum greater than described bandwidth byte number of token number of using and described average token bucket, for described current logical channel is distributed bandwidth.
9. device according to claim 8 is characterized in that, described device also comprises:
Dispatching cycle, dispensing unit was used for the bandwidth byte number according to described current logical channel request, was described current logical channel dynamic-configuration dispatching cycle, and wherein, described dispatching cycle and described bandwidth byte number are inversely proportional to.
10. device according to claim 8 is characterized in that,
Described acquiring unit also is used to obtain the attribute configuration information that comes from described current logical channel, and wherein, described attribute configuration information is included as the default flow attribution information of each logical channel;
Described allocated bandwidth unit, also being used in definite described current logical channel is not under the situation of described last logical channel of attribute configuration information, does not distribute bandwidth for described current logical channel according to described flow attribution information corresponding priorities.
11. device according to claim 9 is characterized in that,
Described allocated bandwidth unit, when also being used for bandwidth byte number in described dispatching cycle and having the remaining bandwidth byte number, described remaining bandwidth byte number is distributed to the next logical channel of described current logical channel, until the whole bandwidth byte numbers that distribute in described dispatching cycle.
12. device according to claim 11 is characterized in that, described device also comprises:
Record cell is used to write down bandwidth allocation information;
Transmitting element is used for described bandwidth allocation information is sent to described optical network unit by descending gigabit passive optical network Transmission Convergence GTC frame.
CN2009102096525A 2009-11-03 2009-11-03 Dynamic bandwidth allocation method and device Pending CN102055652A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2009102096525A CN102055652A (en) 2009-11-03 2009-11-03 Dynamic bandwidth allocation method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2009102096525A CN102055652A (en) 2009-11-03 2009-11-03 Dynamic bandwidth allocation method and device

Publications (1)

Publication Number Publication Date
CN102055652A true CN102055652A (en) 2011-05-11

Family

ID=43959603

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2009102096525A Pending CN102055652A (en) 2009-11-03 2009-11-03 Dynamic bandwidth allocation method and device

Country Status (1)

Country Link
CN (1) CN102055652A (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103107959A (en) * 2011-11-14 2013-05-15 联想(北京)有限公司 Electronic equipment and allocation method of data bandwidth thereof
CN103248580A (en) * 2013-04-08 2013-08-14 汉柏科技有限公司 High-precision token bucket speed limit method based on major-minor buckets
CN103533660A (en) * 2012-07-02 2014-01-22 中兴通讯股份有限公司 Method and apparatus downlink LCH scheduling
CN106464835A (en) * 2015-04-30 2017-02-22 华为技术有限公司 Uplink bandwidth allocation method, device and system
CN106470431A (en) * 2015-08-14 2017-03-01 中国联合网络通信集团有限公司 A kind of method and device obtaining user's perception speed
CN110868359A (en) * 2019-11-15 2020-03-06 中国人民解放军国防科技大学 Network congestion control method
CN115348490A (en) * 2022-10-18 2022-11-15 武汉长光科技有限公司 Method for dynamically scheduling service wavelength channel and related device

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103107959A (en) * 2011-11-14 2013-05-15 联想(北京)有限公司 Electronic equipment and allocation method of data bandwidth thereof
CN103533660B (en) * 2012-07-02 2019-02-01 中兴通讯股份有限公司 Downlink LCH dispatching method and device
CN103533660A (en) * 2012-07-02 2014-01-22 中兴通讯股份有限公司 Method and apparatus downlink LCH scheduling
CN103248580B (en) * 2013-04-08 2016-08-24 汉柏科技有限公司 A kind of method of high accuracy limiting speed of token bucket based on major-minor bucket
CN103248580A (en) * 2013-04-08 2013-08-14 汉柏科技有限公司 High-precision token bucket speed limit method based on major-minor buckets
CN106464835A (en) * 2015-04-30 2017-02-22 华为技术有限公司 Uplink bandwidth allocation method, device and system
CN106464835B (en) * 2015-04-30 2019-08-13 华为技术有限公司 Distribution method, the device and system of upstream bandwidth
US10771397B2 (en) 2015-04-30 2020-09-08 Huawei Technologies Co., Ltd. Upstream bandwidth allocation method, apparatus, and system
CN106470431A (en) * 2015-08-14 2017-03-01 中国联合网络通信集团有限公司 A kind of method and device obtaining user's perception speed
CN106470431B (en) * 2015-08-14 2019-08-02 中国联合网络通信集团有限公司 A kind of method and device for obtaining user and perceiving rate
CN110868359A (en) * 2019-11-15 2020-03-06 中国人民解放军国防科技大学 Network congestion control method
CN110868359B (en) * 2019-11-15 2023-03-24 中国人民解放军国防科技大学 Network congestion control method
CN115348490A (en) * 2022-10-18 2022-11-15 武汉长光科技有限公司 Method for dynamically scheduling service wavelength channel and related device
CN115348490B (en) * 2022-10-18 2023-03-24 武汉长光科技有限公司 Method for dynamically scheduling service wavelength channel and related device

Similar Documents

Publication Publication Date Title
US7539211B2 (en) Dynamic bandwidth allocation method considering multiple services in ethernet passive optical network system
CN102055652A (en) Dynamic bandwidth allocation method and device
EP2467978B1 (en) Method for soft bandwidth limiting in dynamic bandwidth allocation
JP4331187B2 (en) Dynamic bandwidth allocation apparatus and method in Ethernet passive optical communication network, and EPON master apparatus using the same
KR100698571B1 (en) Dynamic bandwidth distribution method in passive optical network
CN101997761B (en) Bandwidth allocation method and optical line terminal (OLT)
US8040918B2 (en) Dynamic bandwidth allocation in a passive optical access network
CN101753421B (en) Method and device for dynamic bandwidth allocation in gigabit passive optical network
KR100801370B1 (en) Band allocation control apparatus, band allocation control method, and band allocation control program
US9819599B2 (en) Method and system for improving bandwidth allocation efficiency
US8326149B2 (en) Dynamic bandwidth allocation in a passive optical network in which different optical network units transmit at different rates
CN101924706A (en) Gigabit passive optical network (PON) bandwidth management method based on ONU port
KR20090119856A (en) A dynamic bandwidth allocation device for a passive optical network system and the method implemented
CN100546227C (en) The method that the Ethernet passive optical network multi service dynamic bandwidth distributes
CN107770875A (en) Aeronautical Ad hoc networks hybrid MAC protocols
Rodriguez-Perez et al. Improving energy efficiency in upstream EPON channels by packet coalescing
CN109348315B (en) Dynamic bandwidth allocation method for adjusting bandwidth allocation parameters based on time division multiplexing
US11683102B1 (en) Bandwidth allocation method and associated optical line terminal
CN102684975B (en) Dynamic bandwidth source allocation method and dynamic bandwidth source allocation device
JP2004289780A (en) Optical subscriber's line terminal station device, optical subscriber's line termination apparatus and band assignment method to be used thereby
CN106878201B (en) A kind of distribution method of dynamic bandwidth and device
CN104954285A (en) Dynamic power control method and device for OTN (optical transport network) system
Yang et al. Dynamic bandwidth allocation (DBA) algorithm for passive optical networks
CN115484516B (en) Bandwidth allocation method and device in passive optical network
CN104935529A (en) OFDMA-PON dynamic bandwidth allocation method

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C12 Rejection of a patent application after its publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20110511