CN1750577B - Method and device for realizing multipath interleaving and deinterleaving - Google Patents
Method and device for realizing multipath interleaving and deinterleaving Download PDFInfo
- Publication number
- CN1750577B CN1750577B CN 200410051503 CN200410051503A CN1750577B CN 1750577 B CN1750577 B CN 1750577B CN 200410051503 CN200410051503 CN 200410051503 CN 200410051503 A CN200410051503 A CN 200410051503A CN 1750577 B CN1750577 B CN 1750577B
- Authority
- CN
- China
- Prior art keywords
- random asccess
- resource
- asccess memory
- user
- deinterleaving
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Mobile Radio Communication Systems (AREA)
Abstract
A method and device for multiplex interlacing and deinterlacing, which contains centralized controlling multi-channel ADSL user, pre collecting current user information and state, dynamically distributing interlacing and deinterlacing resource, said device containing a central control unit, an interlacing/deinterlacing RAM controller and an interlacing/deinterlacing RAM. Said invention saves RAM resource and user average cost.
Description
Technical field
The present invention relates in a kind of realization communication system multichannel interweaves and conciliates interweaving method and device thereof, relate in particular to (the Asymmetric Digital Subscriber Line of ADSL (Asymmetric Digital Subscriber Line) in the communication field, the method and apparatus of the chnnel coding abbreviation ADSL), this method and apparatus combines with forward error correction coding, can be used for processing and correction to unexpected error.
Background technology
The ADSL system can transmit the system of telephone service and broadband data service simultaneously on a pair of twisted-pair feeder, because the operational environment of twisted-pair feeder is more abominable, be subject to the interference of instantaneous spiking, therefore introduce the interleaving/deinterleaving link and can strengthen the ability of correcting the ADSL channel error.Provided as shown in Figure 1 in the prior art ADSL system transceiver channel, the basic functional block diagram of interleaving/deinterleaving data processing section, the ADSL framing does not directly send to physical layer, but sends to physical layer and transmit through scrambler, forward error correction coding with after interweaving; And at receiving terminal successively through carrying out the frame of separating of ADSL after deinterleaving, decoding FEC and the descrambling.
Adopt the interleaving/deinterleaving technology can effectively improve the ability of correcting the channel burst mistake, but realize that interleaving/deinterleaving need need take sizable random asccess memory (being called for short RAM) resource, the raising that can bring cost to the data buffer memory.For example, in first generation ADSL standard, as T1.413 and ITU-TG992.1, stipulated each ADSL user provided simultaneously to interweave and quick two logic channels, the depth capacity of Interleave Channel support is 64.Agreement has stipulated that as forward error correction coding, its code word size is 255 to use Reed-Solomon coding (being called for short the RS coding), and in order to support maximum interleave depth, the RAM size that needs is the RAM resource of 255 * 64=16320 byte like this.
In the ADSL of second generation standard I TU-T G992.3 and G992.5, stipulated that each user can have 4 delay paths at most, each all supports interleave function.On the other hand, along with the progress of integrated circuit technique, the ADSL number of users that single chip can be supported is more and more.Fig. 2 has provided and has supported a plurality of users of second generation ADSL standard to realize the functional schematic that multichannel interweaves, only support a user from an initial cover ADSL chip, till now, had digit chip can support 16 users, can predicting in the future, this numeral can further improve.Only support that with a nest plate 16 ADSL users are example, for each user provides 4 delay paths, each passage is supported interleave depth 64, and the use code word size is 255 RS coding, and needing altogether provides 1,044 for interweaving, the RAM resource of 480 bytes.Can see that in order to strengthen the ability of correcting the channel burst mistake, use interleaving technology to take a large amount of resources, the unique user cost also improves thereupon.Existing implementation method can be used interleave function in order to guarantee high-end user, obtains reliable and stable data service, has in fact kept a large amount of RAM resources for interleave function.
Analyze the ADSL business features and can find that the ADSL service is very similar with traditional telephone service, the ADSL user of a large amount of low sides just uses in the part period, and the most of the time does not have online.On the other hand, realize that interleaving/deinterleaving need carry out buffer memory to data, thereby can produce certain time-delay.Because the ADSL diversity of operations, quite a few is the requirement low delay, simultaneously the not too responsive business of mistake is not needed to adopt interleaving/deinterleaving.Even high-end user is not always to use the interleaving/deinterleaving function yet.In this case, the RAM resource that is used for interleaving/deinterleaving that existing implementation method keeps usually is in idle state, not only improved the cost of unique user, and the RAM resource is not fully used.
Summary of the invention
The purpose of this invention is to provide a kind of realization multichannel interweaves and conciliates interweaving method and device, in order to overcome in existing ADSL technology and the standard, the diversity of the ADSL type of service that does not fully take into account, and the characteristics that when using ADSL, have nothing in common with each other of different ADSL users, when conciliating interleave function at realizing interweaving, the shortcoming that causes a large amount of interleaving/deinterleaving RAM resources to fail to make full use of, adopt method of the present invention, when realizing that the multichannel de-interleaving interweaves, distinguish different users earlier, then can be according to current state reasonable disposition resource, dynamic assignment interleaving/deinterleaving RAM resource.The present invention can significantly reduce realization interleaving/deinterleaving function and take the RAM resource, effectively reduces cost, and can reduce power consumption to a certain extent simultaneously.
Technical scheme of the present invention is as follows:
Multichannel de-interleaving interweaving method in a kind of realization communication system, this method may further comprise the steps: for multi-channel A SL user centralized control, carry out that dynamic assignment interweaves and the resource of deinterleaving; Whether collection active user's information and state use the interleaving/deinterleaving function according to current online user earlier, and the situation of interleave depth, distinguish the attribute and the demand of different user in advance, dynamically are each user resource allocation.
Wherein, described method also comprises:
A) central control unit is all ADSL subscriber's line configuration preference levels of being controlled, and finishes initial configuration;
B) for newly setting up ADSL user's dynamic assignment interleaving/deinterleaving random asccess memory resource of link;
C) together realize interleave function by interleaving/deinterleaving control unit and the random asccess memory memory space that is distributed;
D) discharge the shared interleaving/deinterleaving random asccess memory resource of ADSL user of cancelling link.
Described method, wherein, described step a) comprises the following steps: again
A1) described central control unit is all ADSL subscriber's line configuration preference levels of being controlled;
A2) whole interleaving/deinterleaving random asccess memory resources are divided into the basic block that size is 2 * N byte, wherein N is the maximum length of the code word of RS error correcting code, is all basic random asccess memory Resource Block numberings;
A3) described central control unit disposes the maximum interleaving/deinterleaving resource that can take according to the service scenario of different user application for each user;
A4) when initial condition, all users also do not set up the ADSL link, set up idle random asccess memory resource table earlier, set up the formation that needs force to discharge the random asccess memory resource again, and it is empty that this pressure of initial condition discharges formation.
Described method, wherein, described system random asccess memory resource can be configured as the summation of all user's requests, when expanding with convenient random asccess memory resource in system and according to customer requirements, can provide upgrade service easily.
Described method, wherein, described step b) also comprises
B1) when setting up link, obtain the size of the interleaving/deinterleaving random asccess memory that the user need distribute, with itself and the idle random asccess memory resource of system relatively, if the idle random asccess memory resource of system is bigger, then according to the user need distribute; Whether the resource of judge distributing simultaneously surpasses the maximum interleaving/deinterleaving resource that this user can take, if surpass then this user put into the formation that needs to force to discharge resource; When resource is nervous, then system will force this user that resource is discharged;
If system's idling-resource is less and the resource of customer requirements surpasses the service of its pre-first to file, then system will the principle according to first in first out discharges the resource that it takies more one by one the formation from forcing to discharge, the idle random asccess memory resource of update system tabular can satisfy till the demand of user to resource of newly setting up link up to system's idling-resource simultaneously;
B2) after the Dynamic Resource Allocation for Multimedia success, needing to distribute to the up-to-date piece label of setting up the interleaving/deinterleaving random asccess memory resource of link removes in the tabular of idle random asccess memory Resource Block, upgrade available random asccess memory piece tabulation, calculate residue random asccess memory resource;
B3),, set up the mapping table of interleaving/deinterleaving random asccess memory rank addresses and actual random asccess memory rank addresses according to the random asccess memory resource table that described central control unit distributes corresponding to the interleaving/deinterleaving control unit that Adds User.
Described method, wherein, described step c) also comprises
Continuous data flow R
I, jAt time t
I, j=i*N+j enters interleaver, R
I, jBe j byte of i RS code word, i belongs to the integer in [0, D-1] scope, and j is the integer that belongs in [0, N-1] scope; Convolutional encoding postpones d to each byte
I, j=j* (D-1), D represents interleave depth, and constant N represents the quantity of RS code word joint; So R
I, jOutput time be:
T
i，j＝t
i，j+d
i，j＝i*N+j+j*(D-1)＝i*N+j*D
For fear of conflict, D and N should satisfy relatively prime condition.
Described method, wherein, the process of described realization convolutional interleave is specific as follows: storage array is divided into the capable D row of N, carries out write operation by row earlier when interweaving, carry out read operation by row then;
At moment t
I, j=i*N+j, the ram address that interweaves that the write operation elder generation calculating that interweaves need write, row address is [(i * N)/D+j] modN, column address is the (modD of i * N), utilize described step b3) mapping table of interweave random asccess memory rank addresses and the actual random asccess memory rank addresses that obtain, draw the rank addresses of actual random asccess memory correspondence;
The rank addresses of the described read operation that interweaves, be respectively (i * N+j)/D and (modD of i * N+j), table look-up obtain actual ram address after, data are read from this address;
The described read-write process that interweaves all is to be the cyclic behaviour in cycle with N * D.
Described method, wherein, described step d) also comprises
D1) as ADSL user the cancellation set up link the time, the random asccess memory Resource Block label of distributing to this user is added in system's idle random asccess memory resource tabular again, upgrade available random asccess memory piece tabulation, calculate residue random asccess memory resource, use in order to divide timing next time;
D2) if this user in the formation of forcing the release resource, then deletes it from this formation.
A kind ofly realize that multichannel interweaves and the device of deinterleaving, it comprises with the lower part: a central control unit, an interleaving/deinterleaving ram controller and an interleaving/deinterleaving random asccess memory;
Described central control unit is the key control unit in the ADSL system, it is a programmable microprocessor unit, when individual system was handled a plurality of ADSL user, described central control unit was used for the control and the random asccess memory parameter configuration of multichannel interleaving/deinterleaving;
Described interleaving/deinterleaving ram controller is the unit that concrete control data stream buffer memory is realized the interleaving/deinterleaving function, it writes random asccess memory according to certain rules with the input data of each interleaving/deinterleaving path, simultaneously according to rule from the address of random asccess memory sense data as the output of interleaving/deinterleaving unit; Described interleaving/deinterleaving ram controller has comprised each user's interleaving/deinterleaving random asccess memory table, and described central control unit can write the numbering of the basic random-access memory unit that distributes for certain user in this table; The correspondence table that also comprises a random asccess memory basic resource blocks and actual address in the controller to described random asccess memory piece label, obtains the address of corresponding actual random asccess memory with easily;
Described random asccess memory is to realize the unit of storage in the interleaving/deinterleaving process, according to the read-write control signal and the address of interleaving/deinterleaving controller, data is write buffer memory in the address of appointment, perhaps data in buffer in the assigned address is read.
Described device, wherein, the size of described/deinterleaving random asccess memory can dispose and expand according to user's different demands.
A kind of realization multichannel provided by the present invention interweaves and conciliates interweaving method and device, compared with prior art, distinguished user's priority in the system of the present invention in advance, can guarantee that like this high-end user can enjoy the service quality of its application, low end subscriber is under the condition that obtains the basic service assurance, can be in the professional non-crowded time period of interleaving/deinterleaving, also can obtain the same service quality of high-end user, can utilize the RAM resource more fully, the method of dynamic assignment and management, can obviously reduce and realize the needed RAM resource of multichannel interleaving/deinterleaving, save each user's average unit cost; Owing to reduced the use of RAM resource, also reduced the power consumption of system to a certain extent simultaneously.
Description of drawings
Fig. 1 is the partial function block diagram of prior art ADSL master data transceiver channel;
Fig. 2 is the functional schematic that the prior art multichannel interweaves;
Fig. 3 is the device schematic diagram that the realization multichannel of the embodiment of the invention interweaves;
Fig. 4 is the process flow figure that the newly-built upright link of user is arranged of the embodiment of the invention;
Fig. 5 is the processing method that the online user leaves system in the embodiment of the invention method and apparatus.
Embodiment
Below will be described in detail each preferred embodiment of the present invention.
Present embodiment is to come technical scheme of the present invention is described with situation about interweaving, and the situation of deinterleaving is basic identical with the principle that interweaves, and repeats no more.
As shown in Figure 3, the device that multichannel interweaves in the realization communication system of the present invention consists of the following components, a central control unit, interweave a ram controller and a random asccess memory; What use when realizing the function of multichannel deinterleaving is deinterleaving ram controller and deinterleaving random asccess memory.
Described central control unit is the key control unit in the ADSL system, and it is the microprocessor unit that can programme.When individual system was handled a plurality of ADSL user, described central control unit need be finished user's parameter configuration, control of information and State Control and initialization procedure or the like.In the present invention, main pay close attention to be based on control and the random asccess memory parameter configuration function that its realizes that multichannel de-interleaving interweaves.
The described ram controller that interweaves is the unit that concrete control data stream buffer memory is realized interleave function.It writes random asccess memory according to certain rules with the input data of each path that interweaves, simultaneously also according to rule from the address of random asccess memory sense data as the output of interleave unit.The method that the ram controller that interweaves calculates read/write address has detailed explanation in the implementation method of back.The ram controller that interweaves has comprised each user's the random asccess memory table that interweaves, and central control unit can write the numbering of the basic random-access memory unit that distributes for certain user in this table.The correspondence table that also comprises a random asccess memory basic resource blocks and actual address in the controller can obtain the address of corresponding actual random asccess memory easily from random asccess memory piece label.
Described random asccess memory is to realize the unit of storage in the interleaving process.According to the read-write control signal and the address of the described ram controller that interweaves, data are write buffer memory in the address of appointment, perhaps data in buffer in the assigned address is read.According to user's different demands, the size of random asccess memory can dispose and expand.
Multichannel de-interleaving interweaving method is as follows in the realization communication system of the present invention:
Core algorithm is: for multi-channel A SL user centralized control, dynamic assignment interweaves and the resource of deinterleaving.Collect active user's information and state earlier, whether use the interleaving/deinterleaving function according to current online user, and the situation of interleave depth, dynamically be each user resource allocation.By attribute and the demand of distinguishing different user in advance, can dispose interleaving/deinterleaving random asccess memory resource according to demand.
Owing to the process that interweaves with deinterleaving is symmetrical, its resource dynamic is distributed and the method for management is identical.In the following description, will only itemize the process that interweaves, the process of deinterleaving is adopted and is used the same method, no longer repeat specification.
The step of described multichannel interweaving method of the present invention is:
The first step, described central control unit are all ADSL subscriber's line configuration preference levels of being controlled, and finish initial configuration.The priority of different user can be determined according to each ADSL user's different attribute (such as rate and the miscellaneous service information opened of application etc.) by the upper layer network unit, and send to described central control unit.
This step can comprise the following steps: again
A1) described central control unit is all ADSL subscriber's line configuration preference levels of being controlled;
A2) the random asccess memory resource that will all interweave is divided into the basic block that size is 2 * N byte, and wherein N is the maximum length of the code word of RS error correcting code, for all basic random asccess memory Resource Block numberings, in order to using;
A3) described central control unit is according to the service scenario of different user application, the maximum that can take for each user's configuration random asccess memory resource that interweaves.System's random asccess memory resource can be configured as the summation of all user's requests, and the random asccess memory resource of system can expand, like this user improve service quality require the time, upgrade service can be provided easily;
A4) when initial condition, all users also do not set up the ADSL link, set up idle random asccess memory resource table earlier, comprise all basic random asccess memory Resource Block numberings this moment in the idle random asccess memory resource table, the size of table has been represented the size of idle random asccess memory simultaneously; Set up needs again and force to discharge the formation of resource, this formation of initial condition is empty.
In second step, be ADSL user's dynamic assignment of newly setting up link random asccess memory resource that interweaves; Concrete this step may further comprise the steps again:
B1) when setting up link, can obtain the random asccess memory size that interweaves that the user need distribute, it and the idle random asccess memory resource of system are compared.If system's idling-resource is bigger, then according to the user need distribute; Whether the resource of judge distributing simultaneously surpasses maximum that this user can the take resource that interweaves, if surpass then this user put into the formation that needs to force to discharge resource.When resource is nervous, then system will force this user that resource is discharged.
If system's idling-resource is less and the resource of customer requirements surpasses the service of its pre-first to file, then system will the principle according to first in first out discharges the resource that it takies more one by one the formation from forcing to discharge, simultaneously the idle random asccess memory resource of update system tabular.Can satisfy till the demand of user of newly setting up link up to system's idling-resource resource.
B2) after the Dynamic Resource Allocation for Multimedia success, needing to distribute to the up-to-date piece label of setting up the random asccess memory resource that interweaves of link removes in the tabular of idle random asccess memory Resource Block, upgrade available random asccess memory piece tabulation, calculate residue random asccess memory resource.
B3),, set up the mapping table of interweave random asccess memory rank addresses and actual random asccess memory rank addresses according to the random asccess memory resource table that central control unit distributes corresponding to the control unit that interweaves that Adds User.
In the 3rd step, together realize interleave function by the random asccess memory memory space that interweaves control unit and distributed.
The operating process that this step is concrete is as follows: continuous data flow R
I, jAt time t
I, j=i*N+j enters interleaver, R
I, jBe j byte of i RS code word, i belongs to the integer in [0, D-1] scope, and j is the integer that belongs in [0, N-1] scope.Convolutional encoding postpones d to each byte
I, j=j* (D-1), D represents interleave depth, and constant N represents the quantity of RS code word joint.So R
I, jOutput time be:
T
i，j＝t
i，j+d
i，j＝i*N+j+j*(D-1)＝i*N+j*D
For fear of conflict, D and N should satisfy relatively prime condition.
The process of specific implementation convolutional interleave is as follows: storage array is divided into the capable D row of N, carries out write operation by row earlier when interweaving, and carries out read operation by row then.
At moment t
I, j=i*N+j, the ram address that interweaves that the write operation elder generation calculating that interweaves need write, row address is [(i * N)/D+j] modN, column address is the (modD of i * N), utilize front second step b3) mapping table of step interweave random asccess memory rank addresses and the actual random asccess memory rank addresses that obtain, can draw the rank addresses of actual random asccess memory correspondence.
Equally, the rank addresses of the read operation that interweaves is respectively (i * N+j)/D and the (modD of i * N+j).Table look-up obtain actual ram address after, data are read from this address.
The read-write process that interweaves all is to be the cyclic behaviour in cycle with N * D.
In the 4th step, discharge the shared random asccess memory resource that interweaves of ADSL user of cancelling link; This step specifically may further comprise the steps again:
D1), the random asccess memory Resource Block label of distributing to this user can be added to again in system's idle random asccess memory resource tabular during as ADSL user at the link that cancellation has been set up.Upgrade available random asccess memory piece tabulation, calculate residue random asccess memory resource, use in order to divide timing next time.
D2) if this user in the formation of forcing the release resource, also needs it is deleted from this formation.
Focusing on 16 ADSL users with a system below is example, supposes wherein to have 2 high-end users, and 4 interleave channels are supported in application, and maximum support interleave depth is 64, and the maximum length of RS code word is 255.Other is 14 general users, has only applied for supporting 1 interleave channel, and maximum interleave depth is 16.For different users, be the random asccess memory that interweaves of maximum length * support interleave channel number * maximum support interleave depth of number of users * RS code word as long as byte number is provided, just can satisfy the service quality of user applies.Can calculate according to this formula, 2 high-end users need the random asccess memory that interweaves of 130,560 bytes, 14 low end subscribers need 57, the random asccess memory that interweaves of 120 bytes, this multichannel interlacing system only need to dispose the random asccess memory resource and be 187,680 bytes.
Hardware unit of the present invention part comprises central control unit, ram controller and random asccess memory interweave as shown in Figure 3.The specific embodiment process is as follows:
The first step is finished initial configuration by described central control unit.
A1) described central control unit is all ADSL subscriber's line configuration corresponding priorities of being controlled.
A2) the random asccess memory resource that will all interweave is divided into the basic block that size is 2 * 255 bytes, the 255th, the maximum length of the code word of RS error correcting code.Then the basic random asccess memory Resource Block in this example is numbered 1,2...368 is in order to using.
A3) when initial condition, set up idle random asccess memory resource table.Comprise [1,368] all random asccess memory Resource Block numberings this moment in the idle random asccess memory resource table, the size of table is 368.
A4) set up the formation of forcing to discharge resource, this formation at present is empty.
Second step, the dynamic assignment random asccess memory that interweaves.
After a period of time, can enter following state by supposing the system in operate as normal: new link is attempted to set up by a domestic consumer, and the resource that interweaves of its application is D=32, and needing the random asccess memory Resource Block is 16.And 365,366,367,368 4 random asccess memory Resource Block numberings are only arranged in the idle random asccess memory resource table this moment, the size of table is 4.Then take following steps to carry out the dynamic assignment memory immediately that interweaves:
B1) as shown in Figure 4, described central control unit comparison user need distribute interweave random asccess memory size and system's free time random asccess memory resource after, enter the less branch of idling-resource.
B2) described central control unit comparison user need distribute random asccess memory size and system the maximum resource (domestic consumer's maximum interleave depth is 16, and correspondence can take 8 basic blocks) that interweaves for its assurance.Find to surpass its maximum guarantee resource after, then its application is revised as interleave depth is 16 to central control unit, needs 8 basic blocks of application.
B3) as shown in Figure 4, user more after treatment need distribute the idle random asccess memory resource of interweave random asccess memory and system after, it is less still to obtain system's idling-resource, need enter this moment to force the resource dispose procedure.
B4) force the process of release random asccess memory resource that different selections can be arranged.Here provide a simple processing method.Because the system resource anxiety must exist some or several users has taken system resource more.So it is not empty that the pressure in the described central control unit discharges the Subscriber Queue of resource, described central control unit can be read the user who enters this formation the earliest, adopt a kind of method or process, force it to discharge the resource that takies more, such as forcing it to enter the process that the online user leaves system.
Suppose a n of domestic consumer owing to the whole random asccess memory basic blocks that taken between [1,32], it is forced to off-line in this process, discharges its whole resources.
B5) upgrade idle random asccess memory resource tabular, obtain a size and be 36 new tabular.Comprising [365,368] and [1,32].
B6) as shown in Figure 4, in comparison procedure next time, enter the more branch of system's idling-resource.
B7) described central control unit distributes 8 basic random asccess memory pieces according to the demand that interweaves of newly setting up link user D=16.And the quantity of random asccess memory piece and concrete label [365,368] and [1,4] are write the controller that interweaves in should user's random asccess memory resource table.
B8) upgrade idle random asccess memory resource tabular, obtain a size and be 28 new tabular [5,32], return.
In the 3rd step, together realize interleave function by the random asccess memory memory space that interweaves control unit and distributed.
In the present embodiment, can suppose the maximum length N=255 of forward error correction RS code word, interleave depth D=16.The implementation process of this step, the method for describing in the technical scheme according to the front earlier calculates the row address and the column address of the random asccess memory that needs operation.Here the row address of random asccess memory does not need to change, and it is constant that the column address of random asccess memory is removed lowest bit, and other bit has been represented 1～8 position of this user's random asccess memory resource table in the ram controller.From this user's random asccess memory resource, can find corresponding basic random asccess memory piece label, can find actual address from the correspondence table of label and actual address subsequently, this address is the address of the read-write random-access memory finally sent of ram controller.
The 4th step, suppose that this user leaves system, can be according to step shown in Figure 5, the random asccess memory resource label that discharges this CU earlier is [365,368] and [1,4], upgrade idle random asccess memory resource tabular then, obtain a size and be 36 new tabular [1,32] and [365,368].Judge this user then not in forcing release resource queue, then directly return.
In the present embodiment, if do not distinguish user type, all need the random asccess memory resource of 1,044,480 bytes according to the high-end user configuration; And current system only needs to dispose about 8% resource.Because the continuous progress of semiconductor technology, the scale and the integrated level of chip improve constantly, and single chip is certain to the business of supporting that simultaneously more user is different with Geng Duo in the future.Adopt the method for the invention, according to different user property dynamic assignment and the management random asccess memory resource that interweaves, will obviously reduce and realize the multichannel needed random asccess memory resource that interweaves, improve the random asccess memory utilization ratio, reduce cost and power consumption, guarantee the service quality of different user.
Claims (10)
1. realize multichannel de-interleaving interweaving method in the communication system for one kind, this method may further comprise the steps: for multi-channel A SL user centralized control, carry out that dynamic assignment interweaves and the resource of deinterleaving; Whether collection active user's information and state use the interleaving/deinterleaving function according to current online user earlier, and the situation of interleave depth, distinguish the attribute and the demand of different user in advance, dynamically are each user resource allocation.
2. method according to claim 1 is characterized in that, described method also comprises:
A) central control unit is all ADSL subscriber's line configuration preference levels of being controlled, and finishes initial configuration;
B) for newly setting up ADSL user's dynamic assignment interleaving/deinterleaving random asccess memory resource of link;
C) together realize the interleaving/deinterleaving function by interleaving/deinterleaving control unit and the random asccess memory memory space that is distributed;
D) discharge the shared interleaving/deinterleaving random asccess memory resource of ADSL user of cancelling link.
3. method according to claim 2 is characterized in that described step a) comprises the following steps: again
A1) described central control unit is all ADSL subscriber's line configuration preference levels of being controlled;
A2) whole interleaving/deinterleaving random asccess memory resources are divided into the basic block that size is 2 * N byte, wherein N is the maximum length of the code word of RS error correcting code, is all basic random asccess memory Resource Block numberings;
A3) described central control unit disposes the maximum interleaving/deinterleaving resource that can take according to the service scenario of different user application for each user;
A4) when initial condition, all users also do not set up the ADSL link, set up idle random asccess memory resource table earlier, set up the formation that needs force to discharge the random asccess memory resource again, and it is empty that this pressure of initial condition discharges formation.
4. method according to claim 3 is characterized in that, the random asccess memory resource distribution of described system becomes the summation of all user's requests, according to customer requirements, can provide upgrade service easily when expanding with convenient random asccess memory resource in system.
5. method according to claim 3 is characterized in that described step b) also comprises
B1) when setting up link, obtain the size of the interleaving/deinterleaving random asccess memory that the user need distribute, with itself and the idle random asccess memory resource of system relatively, if the idle random asccess memory resource of system is bigger, then according to the user need distribute; Whether the resource of judge distributing simultaneously surpasses the maximum interleaving/deinterleaving resource that this user can take, if surpass then this user put into the formation that needs to force to discharge resource; When resource is nervous, then system will force this user that resource is discharged;
If system's idling-resource is less and the resource of customer requirements surpasses the service of its pre-first to file, then system will the principle according to first in first out discharges the resource of forcing the user in the release formation to take one by one the formation from forcing to discharge, the idle random asccess memory resource of update system tabular can satisfy till the demand of user to resource of newly setting up link up to system's idling-resource simultaneously;
B2) after the Dynamic Resource Allocation for Multimedia success, needing to distribute to the up-to-date piece label of setting up the interleaving/deinterleaving random asccess memory resource of link removes in the tabular of idle random asccess memory Resource Block, upgrade available random asccess memory piece tabulation, calculate residue random asccess memory resource;
B3),, set up the mapping table of interleaving/deinterleaving random asccess memory rank addresses and actual random asccess memory rank addresses according to the random asccess memory resource table that described central control unit distributes corresponding to the interleaving/deinterleaving control unit that Adds User.
6. method according to claim 5 is characterized in that described step c) also comprises
Continuous data flow R
I, jAt time t
I, j=i*N+j enters interleaver, R
I, jBe j byte of i RS code word, i belongs to the integer in [0, D-1] scope, and j is the integer that belongs in [0, N-1] scope; Convolutional encoding postpones d to each byte
I, j=j* (D-1), D represents interleave depth, and constant N represents the quantity of RS code word joint; So R
I, jOutput time be:
T
i，j＝t
i，j+d
i，j＝i*N+j+j*(D-1)＝i*N+j*D
For fear of conflict, D and N should satisfy relatively prime condition.
7. method according to claim 6 is characterized in that, the process of described realization convolutional interleave is specific as follows: storage array is divided into the capable D row of N, carries out write operation by row earlier when interweaving, carry out read operation by row then;
At moment t
I, j=i*N+j, the ram address that interweaves that the write operation elder generation calculating that interweaves need write, row address is [(i * N)/D+j] modN, column address is the (modD of i * N), utilize described step b3) mapping table of interweave random asccess memory rank addresses and the actual random asccess memory rank addresses that obtain, draw the rank addresses of actual random asccess memory correspondence;
The rank addresses of the described read operation that interweaves, be respectively (i * N+j)/D and (modD of i * N+j), table look-up obtain actual ram address after, data are read from this address;
The described read-write process that interweaves all is to be the cyclic behaviour in cycle with N * D.
8. method according to claim 2 is characterized in that described step d) also comprises
D1) as ADSL user the cancellation set up link the time, the random asccess memory Resource Block label of distributing to this user is added in system's idle random asccess memory resource tabular again, upgrade available random asccess memory piece tabulation, calculate residue random asccess memory resource, use in order to divide timing next time;
D2) if this user in the formation of forcing the release resource, then deletes it from this formation.
9. realize that multichannel interweaves and the device of deinterleaving for one kind, it comprises with the lower part: a central control unit, an interleaving/deinterleaving ram controller and an interleaving/deinterleaving random asccess memory;
Described central control unit is the key control unit in the ADSL system, it is a programmable microprocessor unit, when individual system was handled a plurality of ADSL user, described central control unit was used for the control and the random asccess memory parameter configuration of multichannel interleaving/deinterleaving; For multi-channel A SL user centralized control, carry out that dynamic assignment interweaves and the resource of deinterleaving; Whether collection active user's information and state use the interleaving/deinterleaving function according to current online user earlier, and the situation of interleave depth, distinguish the attribute and the demand of different user in advance, dynamically are each user resource allocation;
Described interleaving/deinterleaving ram controller is the unit that concrete control data stream buffer memory is realized the interleaving/deinterleaving function, it writes random asccess memory according to certain rules with the input data of each interleaving/deinterleaving path, simultaneously according to rule from the address of random asccess memory sense data as the output of interleaving/deinterleaving; Described interleaving/deinterleaving ram controller has comprised each user's interleaving/deinterleaving random asccess memory table, and described central control unit can write the numbering of the basic random-access memory unit that distributes for certain user in this table; The correspondence table that also comprises a random asccess memory basic resource blocks and actual address in the controller to described random asccess memory piece label, obtains the address of corresponding actual random asccess memory with easily;
Described random asccess memory is to realize the unit of storage in the interleaving/deinterleaving process, read-write control signal and address according to the interleaving/deinterleaving ram controller, data are write buffer memory in the address of appointment, perhaps data in buffer in the assigned address is read.
10. device according to claim 9 is characterized in that, the size of described interleaving/deinterleaving random asccess memory is according to user's different demand configurations and expansion.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200410051503 CN1750577B (en) | 2004-09-17 | 2004-09-17 | Method and device for realizing multipath interleaving and deinterleaving |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200410051503 CN1750577B (en) | 2004-09-17 | 2004-09-17 | Method and device for realizing multipath interleaving and deinterleaving |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1750577A CN1750577A (en) | 2006-03-22 |
| CN1750577B true CN1750577B (en) | 2010-06-16 |
Family
ID=36605828
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200410051503 Expired - Fee Related CN1750577B (en) | 2004-09-17 | 2004-09-17 | Method and device for realizing multipath interleaving and deinterleaving |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1750577B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101335732B (en) * | 2007-06-29 | 2011-11-02 | 华为技术有限公司 | Symbol interleaving mode selection method, apparatus and transmitter implementing symbol interleaving |
| CN101771498B (en) * | 2008-12-30 | 2013-04-03 | 易视芯科技(北京)有限公司 | System and method for expanding bit interleaving memories |
| CN110086574B (en) * | 2019-04-29 | 2020-09-29 | 京信通信系统（中国）有限公司 | Message processing method and device, computer equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1211869A (en) * | 1998-08-04 | 1999-03-24 | 国家科学技术委员会高技术研究发展中心 | Method for realization of data interweaving and deinterweaving using dynamic RAM |
| US6151690A (en) * | 1997-01-31 | 2000-11-21 | Alcatel | Interleaving and de-interleaving method for digital data, interleaving and de-interleaving devices, and communication system |
| US6707822B1 (en) * | 2000-01-07 | 2004-03-16 | Agere Systems Inc. | Multi-session asymmetric digital subscriber line buffering and scheduling apparatus and method |
-
2004
- 2004-09-17 CN CN 200410051503 patent/CN1750577B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6151690A (en) * | 1997-01-31 | 2000-11-21 | Alcatel | Interleaving and de-interleaving method for digital data, interleaving and de-interleaving devices, and communication system |
| CN1211869A (en) * | 1998-08-04 | 1999-03-24 | 国家科学技术委员会高技术研究发展中心 | Method for realization of data interweaving and deinterweaving using dynamic RAM |
| US6707822B1 (en) * | 2000-01-07 | 2004-03-16 | Agere Systems Inc. | Multi-session asymmetric digital subscriber line buffering and scheduling apparatus and method |
Non-Patent Citations (1)
| Title |
|---|
| 全文. |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1750577A (en) | 2006-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US5751741A (en) | Rate-adapted communication system and method for efficient buffer utilization thereof | |
| US8351537B2 (en) | Seamless change of depth of a general convolutional interleaver during transmission without loss of data | |
| US5737337A (en) | Method and apparatus for interleaving data in an asymmetric digital subscriber line (ADSL) transmitter | |
| US6754221B1 (en) | System and method for selecting a compression algorithm according to an available bandwidth | |
| US7657818B2 (en) | Dynamic minimum-memory interleaving | |
| US20040071138A1 (en) | Cell multiplexing apparatus handling multiple items of information | |
| US20050152382A1 (en) | Local communication system | |
| CN101394220B (en) | Time slot uniform distribution method oriented to MF-TDMA system | |
| EP1832028B1 (en) | Resource sharing in a dsl transceiver | |
| CN101282305B (en) | Bandwidth control method for distributed system as well as service plate | |
| CN101924627A (en) | Carrier management for a network | |
| JPH11234309A (en) | Asynchronous transfer mode cell transmission scheduling method | |
| US6606306B1 (en) | System and method for supporting a plurality of media conferences | |
| TW365730B (en) | Asynchronous transfer mode and media access protocol interoperability apparatus and method | |
| US20030021338A1 (en) | Device for transmitting/receiving digital data capable of variable-rate processing, in particular on a vdsl environment | |
| CN100459584C (en) | Multiple service supporting method and network equipment based on XDSL multiple-time-delay access | |
| CN1750577B (en) | Method and device for realizing multipath interleaving and deinterleaving | |
| US6707822B1 (en) | Multi-session asymmetric digital subscriber line buffering and scheduling apparatus and method | |
| CN101247348B (en) | Load sharing method and device | |
| JP2001119733A (en) | Asymmetrical digital subscriber line(adsl) down high- speed cell bus interface protocol | |
| US6377541B1 (en) | Method and apparatus for transmitting a continuous data stream in packetized form | |
| CN104704764A (en) | Pre-fill retransmission queue | |
| US20070177662A1 (en) | Traffic shaping between the DMT processor and data link layer processor of a line-card | |
| JP3233104B2 (en) | Subscriber network system and method for setting information in concentrator thereof | |
| US20020141411A1 (en) | Apparatus for line-concentrating and distributing PPP frame data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20100616 Termination date: 20130917 |