CN1567849A - A real time flow buffering and jitter eliminating method for decreasing additive time delay - Google Patents
A real time flow buffering and jitter eliminating method for decreasing additive time delay Download PDFInfo
- Publication number
- CN1567849A CN1567849A CN 03149379 CN03149379A CN1567849A CN 1567849 A CN1567849 A CN 1567849A CN 03149379 CN03149379 CN 03149379 CN 03149379 A CN03149379 A CN 03149379A CN 1567849 A CN1567849 A CN 1567849A
- Authority
- CN
- China
- Prior art keywords
- real
- time
- stream
- read
- channel
- 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.)
- Granted
Links
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention relates to a method of eliminating the vibration of real-time flow buffer memory which reduces added time delay, firstly receiving real-time flow data packet, sending a write request for the data packet to queue control center and sending a read/write request for the data packet to read/write arbitrator at regular intervals, and the arbitrator arbitrates the read request signal and the write request signal; after this, respectively starting read and write courses: if it is a write request, determine storage channel of the data packet; if it is a read request, read corresponding channel information, etc, read the data packet from shared buffer memory and store the data packet in output buffer memory; checking the output buffer memory: if there is a complete real-time flow data packet, send the data packet. Its advantage: on the premise of not reducing the real-time flow processing rate and the capacity of real-time flow channel, reduces additional time delay to the minimum and the length of the added time delay can be controlled.
Description
Technical field
The present invention relates to a kind of real-time stream buffer memory jitter removing method that reduces additional delay, belong to, comprise transmitting realtime stream technical field on IP network and the Ethernet in Packet Based Network.
Background technology
On IP network during transmitting realtime stream, because the uncertainty of network delay, can introduce the time jitter of real-time stream and out of order.In order to guarantee the quality of stream in real time, need send each bag at interval according to regular time then at the bag of the receiving terminal buffer memory some that flows in real time, be that cost is eliminated time jitter and out of order with the additional delay, this method is called " buffer memory goes to tremble ".Realize that the key technology that buffer memory goes to tremble is a queue scheduling.
In the prior art, the chain table method that general employing realizes based on software carries out buffer memory and goes to tremble.Fig. 1 is that it realizes block diagram.Each real-time circulation road safeguards that a buffer memory removes to tremble chained list, and the content of chained list node is information such as the time, buffer address of each real-time streaming data bag.The handling process of real-time streaming data bag is as follows:
1, when input receives a real-time streaming data bag, to shared buffer memory management application real-time streaming data bag metadata cache.
2, shared buffer memory management is this real-time streaming data memory space that is responsible for assigning, and returns the buffer address of this memory space of sensing.
3, after input obtains the buffer address of memory space, the real-time streaming data bag is written in the shared buffer memory goes, then information such as the time of this real-time streaming data bag, buffer address are passed to real-time streaming data bag queue scheduling.
4, queue scheduling at first generates a new real-time streaming data packet node, and the transmitting time order according to the real-time streaming data bag is inserted into the correct position that buffer memory removes to tremble chained list to this node then.
5, real-time streaming data bag queue scheduling is safeguarded the real-time streaming data bag transmitting time of each passage, and when the transmitting time of some passages arrived, queue scheduling removed to tremble the real-time streaming data bag of searching transmission the chained list from this passage buffer storage.If inquired the real-time streaming data bag that meets the transmission condition, then export and from shared buffer memory, read the real-time streaming data bag and send.
6, queue scheduling is deleted this node, and the memory space of this real-time streaming data bag is reclaimed in the shared buffer memory management.
The shortcoming of said method is:
(1) chained list safeguards more complicated.Normally insert the real-time streaming data bag, need this moment the whole chained list of traversal just to find the insertion position of real-time streaming data bag at the end of chained list.If out of order real-time streaming data bag then needs former and later two real-time streaming data bags and the sequence number that is inserted into the real-time streaming data bag in the comparison chained list, just can find the insertion position of this real-time streaming data bag.No matter be which kind of situation, the chain table handling is more complicated all.Because the node number of chained list and the uncertainty of real-time streaming data bag transmitting time, the time that real-time streaming data bag insertion operation is spent also has very big uncertainty.
(2) the real-time streaming data bag transmitting time of each passage is independent of each other, when a plurality of passages send the transmission request simultaneously, is difficult to guarantee that each real-time streaming data bag can both be responded timely, and the passage of Fa Songing can be introduced bigger time-delay at last.If it is unequal that adjacent two real-time streaming data bags of some passages send the request responding time, will introduce additional " shake ", the size of shake depends on the passage number that proposes the transmission request simultaneously and the processing procedure of queue scheduling.The processing procedure of queue scheduling and processing time all have very big uncertainty.
Therefore the said method complex disposal process has very big uncertainty again, single processor the port number that can handle all little, generally between the passage of hundreds of～1,000, make the single channel cost than higher.
Summary of the invention
The objective of the invention is shortcoming at the prior art existence, a kind of real-time stream buffer memory jitter removing method that reduces additional delay is proposed, the buffer memory of removing in the prior art in processing procedure removes to tremble chained list, make channel information and real-time streaming data bag data space by all passage mean allocation, and the ablation process of simplification real-time streaming data bag, the unified voice packet transmitting time of safeguarding all passages is set up in real time related between the stream memory channel and read channel.When passage generation user is switched, determine and the corresponding memory channel of write access voluntarily according to read channel, to reduce the additional time delay in the stream transmission in real time.
The real-time stream buffer memory jitter removing method that reduces additional delay that this method proposes comprises following each step:
1, receives stream in real time, and real-time stream is deposited in the input-buffer;
2,, and provide the real-time circulation road of writing of request number if existing complete real-time stream in the input-buffer then sends to the read-write arbitration and writes real-time stream request;
3, transmitting time is safeguarded with regular time and is sent the stream request in real time of reading to the read-write arbitration at interval, provides circulate the in real time Taoist monastic name and the transmission position of stream in real time of reading of request to reading real-time stream simultaneously;
4, the read-write arbitration is arbitrated reading request signal and written request signal;
5, after arbitration,, then determine the memory channel of stream in real time if write real-time stream request;
6, write real-time stream flow process reads write request from channel information real-time circulation road information, calculate and to write the writing position of real-time stream, and should flow in real time and be written to the shared buffer memory from input-buffer;
7, after arbitration, if read stream request in real time, then from channel information, read the channel information under the stream in real time read of request, and according to given reading circulate in real time Taoist monastic name and in real time stream send the position, from shared buffer memory, read corresponding real-time stream, it is deposited in the output buffers;
8, check output buffers,, then send stream in real time if find to have complete real-time stream.
Transmitting time in the said method safeguards that flow process comprises following each step:
(1) in 20 milliseconds, sends 0th the real-time stream that sends position of 0 passage to the 32K-1 passage;
(2) in next 20 milliseconds, send 1st the real-time stream that sends position of 0 passage to the 32K-1 passage;
(3) in another 20 milliseconds, send 2nd the real-time stream that sends position of 0 passage to the 32K-1 passage;
(4) the rest may be inferred, in the end sends 15th the real-time stream that sends position of 0 passage to the 32K-1 passage in 20 milliseconds;
(5) repeat above-mentioned steps 1 to 4.
Read-write arbitration in the said method may further comprise the steps:
(1) after writing real-time stream flow process end, at first judged whether to read stream request in real time,, then read real-time flow operation if having;
(2) if do not read stream request in real time, then further judged whether to write real-time stream request,, then write real-time flow operation if having:
(3) after reading flow process end in real time, at first judged whether to write real-time stream request,, then write real-time flow operation if having;
(4) if do not write real-time stream request, then judged whether to read stream request in real time, if having, then read real-time flow operation.
Determine in the said method that the process of stream memory channel may further comprise the steps in real time:
(1) sets up real-time circulation road and take mapping table between the memory channel in label table, write access and the shared buffer memory;
(2) judge whether to take place the user from the header packet information of real-time stream and switch,, then from above-mentioned mapping table, read and write access number corresponding memory channel number,, then carry out following (3) step if switch if do not switch;
(3) number be initial address with read channel, take from above-mentioned real-time circulation road and search idle channel the label table, with first idle channel of finding as memory channel.
Can also in above-mentioned (1) step, set up the inverse mapping table between the memory channel in write access and the shared buffer memory simultaneously; In above-mentioned (3) step, when the generation user is switched, if find idle channel, then memory channel is effective, if search less than idle channel, then memory channel is invalid, and the information that write access number and memory channel is whether effective writes in the inverse mapping table between the memory channel in above-mentioned write access and the shared buffer memory; When reading to flow request in real time, from above-mentioned inverse mapping table, read the whether effective information of memory channel,, then read stream end in real time,, then read the affiliated channel information of stream in real time of reading of request if effectively if memory channel is invalid.
In the said method, write real-time stream flow process and may further comprise the steps:
(1) judges according to flowing header packet information in real time, select replacement algorithm or the timestamp algorithm computation writing position of stream in real time; Wherein, if use the replacement algorithm, then calculate stream writing position in real time, if stab algorithm service time, then calculate stream writing position in real time earlier, judge again whether real-time stream abandons, if abandon this real-time stream, then judge whether to force to write real-time stream with the replacement algorithm according to continuous packet loss sign;
(2) according to real-time stream memory channel, the real-time stream writing position of aforementioned calculation and the real-time stream length in the header packet information determined, real-time stream is write in the shared buffer memory;
(3) timestamp that will flow in real time, in real time flow length, in real time flow effective information, in real time the stream shaking reduction buffer memory degree of depth writes in the channel information.
Above-mentioned writing in the real-time stream flow process, in order to the replacement algorithm that calculates real-time stream writing position be, will work as prepass the transmission position and current write real-time stream go tremble buffer memory degree of depth sum as real-time stream writing position.
Above-mentioned writing in the real-time stream flow process, as follows in order to the computational process of the timestamp algorithm that calculates real-time stream writing position:
(1) calculate the relative writing position of stream in real time:
Relative writing position=(timestamp of the timestamp+80-passage of stream in real time)/160,
If this real-time stream is first real-time stream of stream section in real time, writing position=(timestamp of the timestamp+80-passage of stream in real time)/160+ (going of flowing in real time tremble the buffer memory degree of depth-passage go to tremble the buffer memory degree of depth) relatively then;
(2) judge whether this real-time stream abandons:
If the relative writing position of above-mentioned real-time stream less than 0, then further judges, if on prepass one in real time the relative writing position sum of writing position and the stream in real time of stream then abandon this real-time stream less than the transmission position of working as prepass;
If the relative writing position of above-mentioned real-time stream more than or equal to 0, then further judges, if when prepass last one in real time the relative writing position sum of writing position and the stream in real time of stream then abandon this real-time stream greater than transmission position and 16 sums of working as prepass;
(3) calculate the writing position of stream in real time: the real-time stream that does not abandon after above-mentioned the 2nd step is judged, then the writing position of stream equals the last writing position and the relative writing position sum that should flow in real time of stream in real time when prepass in real time.
Above-mentioned writing in the real-time stream flow process, packet loss sign deterministic process is continuously: the initial value of packet loss sign is 0 continuously, whenever receive a stream in real time, the packet loss sign is to moving to left one, if this real-time stream is dropped continuously, then packet loss sign 1 inserts 1 continuously, otherwise packet loss sign 1 inserts 0 continuously, if each of packet loss sign all is 1 continuously, shows that continuous n real-time stream all is dropped, then use the replacement algorithm computation writing position of stream in real time, and this real-time stream is deposited in the shared buffer memory.
In the shared buffer memory of said method, the storage capacity of stream in real time is equally divided into 32K " OK " by 32K passage, and the 0th row is wherein distributed to passage 0, the 1st row is distributed to passage 1, the 2nd row is distributed to passage 2 ..., the capable passage 32K-1 that distributes to of 32K-1, the capacity of each row is 256 * 64; The capacity of channel information is equally divided into 32K " OK " by 32K passage, and the 0th capable passage 0, the 1 row of distributing to is wherein distributed to passage 1, the 2 capable passage 2 of distributing to ..., the capable passage 32K-1 that distributes to of 32K-1, the capacity of each row is 8 * 32.
Read in real time that flow process may further comprise the steps described in the said method:
(1) safeguards the channel number that flow process is given according to transmitting time, utilize the allocation tables of channel information to calculate the channel information initial address, from channel information, read real-time stream effective information, if it is invalid, then finish to read real-time stream, if effectively, then read real-time stream length information;
(2) according to above-mentioned given channel number and the position of stream transmission in real time, and the above-mentioned real-time stream length information of reading, from shared buffer memory, read corresponding stream in real time, deposit in the output buffers; Remove the effective information of above-mentioned real-time stream in the channel information simultaneously.
The real-time stream buffer memory jitter removing method that reduces additional delay that the present invention proposes has the following advantages:
1, in this method, save the buffer memory in the prior art and removed to tremble chained list, safeguard that by transmitting time the flow process unification safeguards all transmitting times of stream in real time, and adjacent two transmission time intervals that flow are in real time fixed in each passage, have only a bag of a passage to file a request at every turn, thereby avoided in the prior art owing to a plurality of passages propose to read the additional dither that the voice packet request is introduced simultaneously.
What 2, the present invention proposed determines voluntarily that according to read channel the method for stream memory channel can be reduced to minimum with additional delay under the prerequisite that does not reduce real-time stream processing speed and real-time circulation road capacity in real time, and the controlled amount of additional delay, can determine the passage seek scope according to actual needs at work, adjust the additional delay size.
3, the real-time stream buffer memory of the present invention's proposition goes to tremble implementation method, and different length, difference are removed to tremble the real-time stream of the buffer memory degree of depth, and its processing time does not have too big difference basically, and handling property is stable.
4, the parallel of the mean allocation of storage space volume and each processing procedure carried out, and the buffer memory that this method can be handled removes to tremble port number greater than other any existing method, thereby the single channel cost is low.And, increasing the capacity of memory as long as improve the hardware handles frequency, the method for the present invention's design can realize at an easy rate that more the buffer memory of major path number goes to tremble.
5, the transplantability of the inventive method is better, applicable to packing real-time stream processing at interval arbitrarily, as 5 ms intervals, 10 ms intervals, 30 ms intervals, 40 ms intervals etc.
Description of drawings
Fig. 1 be prior art realize that based on chained list the real-time stream that buffer memory goes to tremble handles block diagram.
Fig. 2 is that the buffer memory that the present invention proposes removes to tremble flow chart.
Fig. 3 is a transmitting time service sequence schematic diagram among the embodiment of this method.
Fig. 4 is read-write arbitration process figure in this method.
Fig. 5 determines the flow chart of stream memory channel in real time in this method.
Fig. 6 is the flow chart of an embodiment of this method.
Fig. 7 writes real-time stream flow chart in this method.
Fig. 8 is a continuous packet loss sign schematic diagram in this method.
Fig. 9 is a shared buffer memory structural representation in this method.
Figure 10 reads to flow in real time flow chart in this method.
Figure 11 is an operation chart of determining the voice flow memory channel in the real-time example of this method.
Embodiment
As shown in Figure 2, this method at first receives real-time stream, and real-time stream is deposited in the input-buffer; If existing complete real-time stream in the input-buffer then sends to the read-write arbitration and writes real-time stream request, and provide the real-time circulation road of writing of request number; Transmitting time safeguards with regular time and sends the stream request in real time of reading to the read-write arbitration at interval, provides circulate the in real time Taoist monastic name and the transmission position of stream in real time of reading of request to reading real-time stream simultaneously; The read-write arbitration is arbitrated reading request signal and written request signal; After arbitration,, then determine the memory channel of stream in real time if write real-time stream request; Write real-time stream flow process reads write request from channel information real-time circulation road information, calculate and to write the writing position of real-time stream, and should flow in real time and be written to the shared buffer memory from input-buffer; After arbitration, if read stream request in real time, then from channel information, read the channel information under the stream in real time read of request, and according to given reading circulate in real time Taoist monastic name and in real time stream send the position, from shared buffer memory, read corresponding real-time stream, it is deposited in the output buffers; Check output buffers,, then send stream in real time if find to have complete real-time stream.
Among the embodiment of said method, transmitting time safeguards that flow process can send 0th the real-time stream that sends position of 0 passage to 32K one 1 passages as shown in Figure 3 in 20 milliseconds; In next 20 milliseconds, send 1st the real-time stream that sends position of 0 passage to the 32K-1 passage; In another 20 milliseconds, send 2nd the real-time stream that sends position of 0 passage to the 32K-1 passage; The rest may be inferred, in the end sends 15th the real-time stream that sends position of 0 passage to the 32K-1 passage in 20 milliseconds.
The flow process of read-write arbitration after writing real-time stream flow process end, has at first judged whether to read stream request in real time as shown in Figure 4 in the said method, if having, then reads real-time flow operation; As if reading stream request in real time, then further judged whether to write real-time stream request, if having, then write real-time flow operation: after reading flow process end in real time, at first judged whether to write real-time stream request,, then write real-time flow operation if having; If do not write real-time stream request, then judged whether to read stream request in real time, if having, then read real-time flow operation.
In the said method, the flow process of determining stream memory channel is in real time at first set up real-time circulation road and is taken mapping table between the memory channel in label table, write access and the shared buffer memory as shown in Figure 5; Judge whether to take place user's switching from the header packet information of real-time stream, if do not switch, then from above-mentioned mapping table, read and write access number corresponding memory channel number, if switch, number be initial address then with read channel, take from above-mentioned real-time circulation road and to search idle channel the label table, with first idle channel of finding as memory channel.Can also be at the inverse mapping table of setting up between the memory channel of setting up when real-time circulation road takies mapping table between the memory channel in label table and write access and the shared buffer memory in write access and the shared buffer memory; When the generation user is switched, if find idle channel, then memory channel is effectively, if search less than idle channel, then memory channel is invalid, and the information that write access number and memory channel is whether effective writes in the inverse mapping table between the memory channel in above-mentioned write access and the shared buffer memory; When reading to flow request in real time, read the whether effective information of memory channel from above-mentioned inverse mapping table, if memory channel is invalid, then read stream end in real time, if effectively, then read the affiliated channel information of stream in real time of reading of request, its process as shown in Figure 6.
In the said method, write real-time stream flow process as shown in Figure 7, at first judge, select replacement algorithm or the timestamp algorithm computation writing position of stream in real time according to flowing header packet information in real time; Wherein, if use the replacement algorithm, then calculate stream writing position in real time, if stab algorithm service time, then calculate stream writing position in real time earlier, judge again whether real-time stream abandons, if abandon this real-time stream, then judge whether to force to write real-time stream with the replacement algorithm according to continuous packet loss sign; Real-time stream memory channel, the real-time stream writing position of aforementioned calculation and the real-time stream length in the header packet information according to determining write real-time stream in the shared buffer memory; With the timestamp of stream in real time, in real time flow length, in real time flow effective information, in real time the stream shaking reduction buffer memory degree of depth writes in the channel information.Replacement algorithm wherein is, will work as prepass the transmission position and current write real-time stream go tremble buffer memory degree of depth sum as stream writing position in real time; The computational process of timestamp algorithm wherein is:
(1) calculate the relative writing position of stream in real time:
Relative writing position=(timestamp of the timestamp+80-passage of stream in real time)/160,
If this real-time stream is first real-time stream of stream section in real time, writing position=(timestamp of the timestamp+80-passage of stream in real time)/160+ (going of flowing in real time tremble the buffer memory degree of depth-passage go to tremble the buffer memory degree of depth) relatively then;
(2) judge whether this real-time stream abandons:
If the relative writing position of above-mentioned real-time stream less than 0, then further judges, if on prepass one in real time the relative writing position sum of writing position and the stream in real time of stream then abandon this real-time stream less than the transmission position of working as prepass;
If the relative writing position of above-mentioned real-time stream more than or equal to 0, then further judges, if when prepass last one in real time the relative writing position sum of writing position and the stream in real time of stream then abandon this real-time stream greater than transmission position and 16 sums of working as prepass;
(3) calculate the writing position of stream in real time:
The real-time stream that does not abandon after above-mentioned the 2nd step judged, then the writing position of stream equals the last writing position and the relative writing position sum that should flow in real time of stream in real time when prepass in real time.
Above-mentioned continuous packet loss sign deterministic process as shown in Figure 8, if the initial value of packet loss sign is 0 continuously, whenever receive a stream in real time, the packet loss sign is to moving to left one, if this real-time stream is dropped continuously, then packet loss sign 1 inserts 1 continuously, otherwise packet loss sign 1 inserts 0 continuously, if each of packet loss sign all is 1 continuously, shows that continuous n real-time stream all is dropped, then use the replacement algorithm computation writing position of stream in real time, and this real-time stream is deposited in the shared buffer memory.
In the said method, the structure of shared buffer memory as shown in Figure 9, the capacity that storage is flowed in real time is equally divided into 32K " OK " by 32K passage, the 0th row is wherein distributed to passage 0, the 1st row is distributed to passage 1, the 2nd row is distributed to passage 2 ..., the capable passage 32K-1 that distributes to of 32K-1, the capacity of each row is 256 * 64; The capacity of channel information is equally divided into 32K " OK " by 32K passage, and the 0th capable passage 0, the 1 row of distributing to is wherein distributed to passage 1, the 2 capable passage 2 of distributing to ..., the capable passage 32K-1 that distributes to of 32K-1, the capacity of each row is 8 * 32.
In the said method, read in real time flow process as shown in figure 10, at first safeguard the channel number that flow process is given according to transmitting time, utilize the allocation tables of channel information to calculate the channel information initial address, from channel information, read real-time stream effective information,, then finish to read real-time stream if invalid, if effectively, then read real-time stream length information; According to above-mentioned given channel number and the position of stream transmission in real time, and the above-mentioned real-time stream length information of reading, from shared buffer memory, read corresponding stream in real time, deposit in the output buffers; Remove the effective information of above-mentioned real-time stream in the channel information simultaneously.
Be example with the real-time voice stream in the real-time stream below, introduce content of the present invention in detail.
Among the embodiment of the inventive method, adopt the shared buffer memory of the SDRAM of 16M * 64 as voice flow.Sdram size is equally divided into 32K " OK " by 32K passage, and the 0th row SDRAM distributes to memory channel 0, the 1 row SDRAM and distributes to memory channel 1, the 2 row SDRAM and distribute to memory channel 2......, and the capable SDRAM of 32K-1 distributes to memory channel 32K-1.The capacity of each row SDRAM is 16M * 64/32K=512 * 64.The address assignment of SDRAM is as follows:
| Address realm (24 hexadecimal addresses) | Function is distributed |
| 000000～0001FF | The voice flow of |
| 000200～0003FF | The voice flow of |
| 000400～0005FF | The voice flow of buffer memory passage 2 |
| 000600～0007FF | The voice flow of buffer memory passage 3 |
| …… | …… |
| FFFC00～FFFDFF | The voice flow of buffer memory passage 32K-2 |
| FFFE00～FFFFFF | The voice flow of buffer memory passage 32K-1 |
In the present embodiment, adopt the SSRAM memory channel information of 1M * 32.The SSRAM capacity is equally divided into 32K " OK " by 32K passage, and the allocation tables of channel information is as shown in the table among the SSRAM:
| Address realm (20 hexadecimal addresses) | Function is distributed |
| 00000～0001D | The channel information of the |
| 0001E | Mapping table between the memory channel in |
| 0001F | Inverse mapping table between the |
| 00020～0003D | The channel information of the |
| 0003E | Mapping table between the memory channel in |
| 0003F | Inverse mapping table between the |
| …… | …… |
| FFFC0～FFFDD | The channel information of memory channel 32K-2 in the shared buffer memory |
| FFFDE | Mapping table between the memory channel in write access 32K-2 and the shared buffer memory |
| FFFDF | Inverse mapping table between the memory channel 32K-2 in write access and the shared buffer memory |
| FFFE0～FFFFD | The channel information of memory channel 32K-1 in the shared buffer memory |
| FFFFE | Mapping table between the memory channel in write access 32K-1 and the shared buffer memory |
| FFFFF | Inverse mapping table between the memory channel 32K-1 in write access and the shared buffer memory |
Number corresponding memory channel of mapping table record write access between the memory channel in write access and the shared buffer memory number, data format is as follows:
| The position | Implication | Value |
| 31～16 | Keep and | 0 |
| 14～0 | | 0～32767 |
Number corresponding write access of inverse mapping table record memory channel between the memory channel in write access and the shared buffer memory number and the whether effective information of memory channel, data format is as follows:
| The position | Implication | Value |
| 31～16 | Keep and use | ????0 |
| 15 | The information that whether effective memory channel is | ????1/0 |
| 14～0 | Write access number | ????0～32767 |
Transmitting time in the present embodiment safeguards that flow process sends the voice flow of 32K passage according to memory channel number ascending order, at first in 20 milliseconds, send the voice flow of the 0th position of memory channel 0～32K-1, in next 20 milliseconds, send the voice flow of the 1st position of memory channel 0～32K-1 then, in next 20 milliseconds, send the voice flow of the 2nd position of memory channel 0～32K-1 again, ... the last voice flow that sends the 15th position of memory channel 0～32K-1 in 20 milliseconds, repeat since the 0th position to send then.
In the present embodiment, the process of read-write arbitration is: after writing the end of voice flow flow process, at first judged whether to read the voice flow request, if having, then read the voice flow operation; If do not read the voice flow request, then further judged whether to write the voice flow request, if having then write voice flow operation.After reading the voice flow flow process and finishing, at first judged whether to write the voice flow request, if having then write the voice flow operation; If do not write the voice flow request then judged whether to read the voice flow request, if having then read the voice flow operation.
In the present embodiment, read the voice flow operation, at first safeguard the given read channel of flow process number according to transmitting time, utilize the allocation tables of channel information to calculate the address of the write access and the inverse mapping table between the memory channel in the shared buffer memory of this read channel correspondence, from above-mentioned inverse mapping table, read the whether effective information of memory channel, if memory channel is invalid, then reading voice flow finishes, if effectively, then utilize the allocation tables of channel information to calculate the channel information initial address of this passage correspondence, from channel information, read the voice flow effective information, and read the voice flow operation.
In the present embodiment, the process of determining the voice flow memory channel is, at first the header packet information by voice flow judges whether to take place user's switching, if the user is taken place switches, with read channel number is initial address, take from above-mentioned real-time circulation road and to search idle channel the label table, first idle channel of finding as memory channel, and number is write this memory channel in the mapping table between the memory channel in above-mentioned write access and the shared buffer memory.Switch if the user is taken place, then read the mapping table between write access and memory channel and write access number corresponding memory channel number.After finishing, the operation of determining the voice flow memory channel begins to write the voice flow operation.
In the present embodiment, the operation chart that the voice flow memory channel takes place to determine when the user is switched as shown in figure 11.As shown in Figure 11, the current time read channel points to passage 1, position indicator pointer points to position 0, this moment, write access 0 generation user was switched, take in the label table at the voice flow passage and to search 256 passages backward from passage 1, if this moment, path 10 0 was an idle channel, then path 10 0 is exactly the memory channel of write access 0, and this voice flow will be read out after (100/32K-1) * 20 millisecond=0.06 millisecond.
In the present embodiment, the process of writing voice flow is: at first judge according to the voice flow header packet information, select the writing position of replacement algorithm or timestamp algorithm computation voice flow; Wherein, if use replacement algorithm, the then writing position of computing voice stream; If stab algorithm service time, the writing position of computing voice stream and judge whether voice flow abandons then if abandon this voice flow, then judges whether to write voice flow with the pressure of replacement algorithm according to packet loss sign continuously.If this voice flow is not dropped, then, voice flow is write in the shared buffer memory according to memory channel, voice flow writing position, voice flow length; Simultaneously the timestamp of voice flow writing position, voice flow, voice flow length, voice flow effective information, voice stream shaking reduction buffer memory degree of depth etc. are write in the channel information, the voice flow writing position that wherein writes is the voice flow writing position of passage, and the timestamp of the voice flow that writes is the timestamp of passage; What the going of the voice flow that writes trembled that the buffer memory degree of depth is passage goes to tremble the buffer memory degree of depth.
Above-mentioned writing in the voice flow flow process, the algorithm of the writing position of computing voice stream has two kinds, and a kind of is the replacement algorithm, and another kind is the timestamp algorithm.
The current speech of replacement algorithm computation stream writing position is: when the transmission position of prepass and current write voice flow go to tremble buffer memory degree of depth sum.
The computational process of timestamp algorithm is as follows:
1) the relative writing position of computing voice stream: the timestamp that the timestamp of voice flow is deducted passage, obtained voice flow and when the time tolerance of the last voice flow of prepass, this difference is divided by 160 (20 milliseconds, it is the voice flow time interval of passage), its result is rounded up to nearest integer, has just obtained the relative writing position of voice flow.Be relative writing position=(timestamp of the timestamp of voice flow+80-passage)/160.If this voice flow is first voice flow of a voice segments, then relatively writing position also will consider to tremble the variation of the buffer memory degree of depth, i.e. writing position=(timestamp of the timestamp of voice flow+80-passage)/160+ (the going of voice flow tremble the buffer memory degree of depth-passage go to tremble the buffer memory degree of depth) relatively.
2) judge whether this voice flow abandons.
If the relative writing position of voice flow illustrates that less than 0 this voice flow is an out of order bag.If the writing position of a voice flow and the relative writing position sum of voice flow are less than when the transmission position of prepass on the prepass, show that this voice flow comes to such an extent that " evening " very much, its transmitting time is over and done with, then abandons realize voice and flows.
If the relative writing position of voice flow is more than or equal to 0, if when the relative writing position sum of the writing position of the last voice flow of prepass and voice flow greater than transmission position and 16 sums of working as prepass, show that then this voice flow comes too " early ", the storage limit that has exceeded this passage then abandons this voice flow.
3) writing position of computing voice stream
If do not abandon this voice flow, the writing position of this voice flow=(when the relative writing position of writing position+this voice flow of the last voice flow of prepass) then.
Claims (12)
1, a kind of real-time stream buffer memory jitter removing method that reduces additional delay is characterized in that this method comprises following each step:
(1) receives stream in real time, and real-time stream is deposited in the input-buffer;
(2), and provide the real-time circulation road of writing of request number if existing complete real-time stream in the input-buffer then sends to the read-write arbitration and writes real-time stream request;
(3) transmitting time is safeguarded with regular time and is sent the stream request in real time of reading to the read-write arbitration at interval, provides circulate the in real time Taoist monastic name and the transmission position of stream in real time of reading of request to reading real-time stream simultaneously;
(4) the read-write arbitration is arbitrated reading request signal and written request signal;
(5) after arbitration,, then determine the memory channel of stream in real time if write real-time stream request;
(6) write real-time stream flow process reads write request from channel information real-time circulation road information, calculate and to write the writing position of real-time stream, and should flow in real time and be written to the shared buffer memory from input-buffer;
(7) after arbitration, if read stream request in real time, then from channel information, read the channel information under the stream in real time read of request, and according to given reading circulate in real time Taoist monastic name and in real time stream send the position, from shared buffer memory, read corresponding real-time stream, it is deposited in the output buffers;
(8) check output buffers,, then send stream in real time if find to have complete real-time stream.
2, the method for claim 1 is characterized in that wherein said transmitting time safeguards that flow process comprises following each step:
(1) in 20 milliseconds, sends 0th the real-time stream that sends position of 0 passage to the 32K-1 passage;
(2) in next 20 milliseconds, send 1st the real-time stream that sends position of 0 passage to the 32K-1 passage;
(3) in another 20 milliseconds, send 2nd the real-time stream that sends position of 0 passage to the 32K-1 passage;
(4) the rest may be inferred, in the end sends 15th the real-time stream that sends position of 0 passage to the 32K-1 passage in 20 milliseconds;
(5) repeat above-mentioned steps 1 to 4.
3, the method for claim 1 is characterized in that wherein said read-write arbitration may further comprise the steps:
(1) after writing real-time stream flow process end, at first judged whether to read stream request in real time,, then read real-time flow operation if having;
(2) if do not read stream request in real time, then further judged whether to write real-time stream request,, then write real-time flow operation if having:
(3) after reading flow process end in real time, at first judged whether to write real-time stream request,, then write real-time flow operation if having;
(4) if do not write real-time stream request, then judged whether to read stream request in real time, if having, then read real-time flow operation.
4, the method for claim 1 is characterized in that wherein determining that the process of stream memory channel may further comprise the steps in real time:
(1) sets up real-time circulation road and take mapping table between the memory channel in label table, write access and the shared buffer memory;
(2) judge whether to take place the user from the header packet information of real-time stream and switch,, then from above-mentioned mapping table, read and write access number corresponding memory channel number,, then carry out following (3) step if switch if do not switch;
(3) number be initial address with read channel, take from above-mentioned real-time circulation road and search idle channel the label table, with first idle channel of finding as memory channel.
5, method as claimed in claim 4 is characterized in that also comprising:
(1) in (1) step, sets up the inverse mapping table between the memory channel in write access and the shared buffer memory simultaneously;
(2) in (3) step, when the generation user is switched, if find idle channel, then memory channel is effective, if search less than idle channel, then memory channel is invalid, and the information that write access number and memory channel is whether effective writes in the inverse mapping table between the memory channel in above-mentioned write access and the shared buffer memory;
When (3) reading to flow request in real time, from above-mentioned inverse mapping table, read the whether effective information of memory channel,, then read stream end in real time,, then read the affiliated channel information of stream in real time of reading of request if effectively if memory channel is invalid.
6, the method for claim 1 is characterized in that wherein saidly writing real-time stream flow process and may further comprise the steps:
(1) judges according to flowing header packet information in real time, select replacement algorithm or the timestamp algorithm computation writing position of stream in real time; Wherein, if use the replacement algorithm, then calculate stream writing position in real time, if stab algorithm service time, then calculate stream writing position in real time earlier, judge again whether real-time stream abandons, if abandon this real-time stream, then judge whether to force to write real-time stream with the replacement algorithm according to continuous packet loss sign;
(2) according to real-time stream memory channel, the real-time stream writing position of aforementioned calculation and the real-time stream length in the header packet information determined, real-time stream is write in the shared buffer memory;
(3) timestamp that will flow in real time, in real time flow length, in real time flow effective information, in real time the stream shaking reduction buffer memory degree of depth writes in the channel information.
7, method as claimed in claim 6 is characterized in that wherein saidly in order to the replacement algorithm that calculates real-time stream writing position being, will work as prepass the transmission position and current write real-time stream go tremble buffer memory degree of depth sum as stream writing position in real time.
8, method as claimed in claim 6 is characterized in that wherein said computational process in order to the timestamp algorithm that calculates real-time stream writing position is as follows:
(1) calculate the relative writing position of stream in real time:
Relative writing position=(timestamp of the timestamp+80-passage of stream in real time)/160,
If this real-time stream is first real-time stream of stream section in real time, writing position=(timestamp of the timestamp+80-passage of stream in real time)/160+ (going of flowing in real time tremble the buffer memory degree of depth-passage go to tremble the buffer memory degree of depth) relatively then;
(2) judge whether this real-time stream abandons:
If the relative writing position of above-mentioned real-time stream less than 0, then further judges, if on prepass one in real time the relative writing position sum of writing position and the stream in real time of stream then abandon this real-time stream less than the transmission position of working as prepass;
If the relative writing position of above-mentioned real-time stream more than or equal to 0, then further judges, if when prepass last one in real time the relative writing position sum of writing position and the stream in real time of stream then abandon this real-time stream greater than transmission position and 16 sums of working as prepass;
(3) calculate the writing position of stream in real time:
The real-time stream that does not abandon after above-mentioned the 2nd step judged, then the writing position of stream equals the last writing position and the relative writing position sum that should flow in real time of stream in real time when prepass in real time.
9, method as claimed in claim 6, it is characterized in that wherein said continuous packet loss sign deterministic process is: the initial value of packet loss sign is 0 continuously, whenever receive a stream in real time, the packet loss sign is to moving to left one continuously, if this real-time stream is dropped, then packet loss sign 1 inserts 1 continuously, otherwise packet loss sign 1 inserts 0 continuously, if each of packet loss sign all is 1 continuously, show that continuous n real-time stream all is dropped, then use the replacement algorithm computation writing position of stream in real time, and this real-time stream is deposited in the shared buffer memory.
10, the method for claim 1, it is characterized in that in the wherein said shared buffer memory, the storage capacity of stream in real time is equally divided into 32K " OK " by 32K passage, the 0th row is wherein distributed to passage 0, the 1st row is distributed to passage 1, the 2nd row is distributed to passage 2 ..., the capable passage 32K-1 that distributes to of 32K-1, the capacity of each row is 256 * 64; The capacity of channel information is equally divided into 32K " OK " by 32K passage, and the 0th capable passage 0, the 1 row of distributing to is wherein distributed to passage 1, the 2 capable passage 2 of distributing to ..., the capable passage 32K-1 that distributes to of 32K-1, the capacity of each row is 8 * 32.
11, the method for claim 1 is characterized in that wherein saidly reading in real time that flow process may further comprise the steps:
(1) safeguards the channel number that flow process is given according to transmitting time, utilize the allocation tables of channel information to calculate the channel information initial address, from channel information, read real-time stream effective information, if it is invalid, then finish to read real-time stream, if effectively, then read real-time stream length information;
(2) according to above-mentioned given channel number and the position of stream transmission in real time, and the above-mentioned real-time stream length information of reading, from shared buffer memory, read corresponding stream in real time, deposit in the output buffers; Remove the effective information of above-mentioned real-time stream in the channel information simultaneously.
12, the method for claim 1 is characterized in that wherein said real-time stream is real-time voice stream.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB031493793A CN1297099C (en) | 2003-06-25 | 2003-06-25 | A real time flow buffering and jitter eliminating method for decreasing additive time delay |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB031493793A CN1297099C (en) | 2003-06-25 | 2003-06-25 | A real time flow buffering and jitter eliminating method for decreasing additive time delay |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1567849A true CN1567849A (en) | 2005-01-19 |
| CN1297099C CN1297099C (en) | 2007-01-24 |
Family
ID=34472524
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB031493793A Expired - Fee Related CN1297099C (en) | 2003-06-25 | 2003-06-25 | A real time flow buffering and jitter eliminating method for decreasing additive time delay |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1297099C (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100446504C (en) * | 2005-05-23 | 2008-12-24 | 华为技术有限公司 | Shared storage convertion device and method for line frame protocol data under broadband CDMA system |
| CN105306958B (en) * | 2014-06-30 | 2018-08-17 | 惠州市伟乐科技股份有限公司 | A kind of method and device of IP-based transmission stream bit rate smoothing processing |
| CN109977495A (en) * | 2019-03-09 | 2019-07-05 | 闽南理工学院 | A kind of system and method for the subjective dynamics of vehicle of acquisition based on Internet of Things |
| CN110232293A (en) * | 2019-05-22 | 2019-09-13 | 东南大学 | Based on maximum delay subchain and the minimum delay cascade APUF circuit of subchain |
| CN113377049A (en) * | 2021-06-11 | 2021-09-10 | 西安应用光学研究所 | DDR control method based on FPGA low-delay video image cache |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2001296248A1 (en) * | 2000-09-12 | 2002-03-26 | Innovative Communications Technologies, Inc. | Bit synchronizer and internetworking system and method |
| US6748000B1 (en) * | 2000-09-28 | 2004-06-08 | Nokia Networks | Apparatus, and an associated method, for compensating for variable delay of a packet data in a packet data communication system |
| JP3659183B2 (en) * | 2001-03-12 | 2005-06-15 | 岩崎通信機株式会社 | Real-time packet delay buffer control method |
| DE60137656D1 (en) * | 2001-04-24 | 2009-03-26 | Nokia Corp | Method of changing the size of a jitter buffer and time alignment, communication system, receiver side and transcoder |
| JP3636348B2 (en) * | 2001-09-12 | 2005-04-06 | 日本電気株式会社 | Voice packet delay fluctuation absorbing apparatus and absorbing method |
| CN1173526C (en) * | 2001-10-26 | 2004-10-27 | 宁波大学 | Method of realizing real time communication between digital communication network and public telephone network using speech sound card |
-
2003
- 2003-06-25 CN CNB031493793A patent/CN1297099C/en not_active Expired - Fee Related
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100446504C (en) * | 2005-05-23 | 2008-12-24 | 华为技术有限公司 | Shared storage convertion device and method for line frame protocol data under broadband CDMA system |
| CN105306958B (en) * | 2014-06-30 | 2018-08-17 | 惠州市伟乐科技股份有限公司 | A kind of method and device of IP-based transmission stream bit rate smoothing processing |
| CN109977495A (en) * | 2019-03-09 | 2019-07-05 | 闽南理工学院 | A kind of system and method for the subjective dynamics of vehicle of acquisition based on Internet of Things |
| CN109977495B (en) * | 2019-03-09 | 2022-09-23 | 闽南理工学院 | Internet of things-based optimization system and method for obtaining subjective vehicle dynamics |
| CN110232293A (en) * | 2019-05-22 | 2019-09-13 | 东南大学 | Based on maximum delay subchain and the minimum delay cascade APUF circuit of subchain |
| CN113377049A (en) * | 2021-06-11 | 2021-09-10 | 西安应用光学研究所 | DDR control method based on FPGA low-delay video image cache |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1297099C (en) | 2007-01-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1246778C (en) | Information processing system and distribution method of storage device | |
| CN1311363C (en) | Methods and apparatus for generating a content address to indicate data units written to a storage system proximate in time | |
| CN1102770C (en) | Power estimator for microprocessor | |
| CN1223156C (en) | Method and apparatus for interleaved non blocking pakcet buffer | |
| CN1162785A (en) | Direct memory access controlling device | |
| CN1991810A (en) | A dma controller with multiple intra-channel software request support | |
| CN1306773C (en) | Method and system for maintenance of packet order using caching | |
| CN111159002B (en) | Data edge collection method, edge collection equipment and system based on grouping | |
| US20110307677A1 (en) | Device for managing data buffers in a memory space divided into a plurality of memory elements | |
| CN1794208A (en) | Mass storage device and method for dynamically managing a mass storage device | |
| CN100343826C (en) | Method for implementing memory management | |
| WO2013078583A1 (en) | Method and apparatus for optimizing data access, method and apparatus for optimizing data storage | |
| US10049035B1 (en) | Stream memory management unit (SMMU) | |
| CN1869979A (en) | Buffer store management method | |
| CN1297099C (en) | A real time flow buffering and jitter eliminating method for decreasing additive time delay | |
| CN1969262A (en) | Interface device for debugging and/or tracing a computer system comprising one or multiple masters and one or multiple slaves working together | |
| CN1609850A (en) | System and method for resizing a database | |
| CN1214581C (en) | Large capacity realtime stream processing method for removing dithering in using buffer memory | |
| CN105577806A (en) | Distributed cache method and system | |
| CN110297810A (en) | A kind of stream data processing method, device and electronic equipment | |
| CN1920796A (en) | Cache method and cache system for storing file's data in memory blocks | |
| CN1390331A (en) | IP base LSI dssighing system and desighing method | |
| CN1208725C (en) | Method and device for high-speed buffering of buffer | |
| CN1188929A (en) | Method for transmitting data, and apparatus for transmitting data and medium | |
| CN100338586C (en) | Methods and apparatus for allocating memory |
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 | ||
| ASS | Succession or assignment of patent right |
Owner name: SHENZHEN HAISI SEMICONDUCTOR CO., LTD. Free format text: FORMER OWNER: HUAWEI TECHNOLOGY CO., LTD. Effective date: 20081010 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20081010 Address after: HUAWEI electric production center, Bantian HUAWEI base, Longgang District, Shenzhen Patentee after: Haisi Semiconductor Co., Ltd., Shenzhen Address before: Bantian HUAWEI headquarters office building, Longgang District, Shenzhen, Guangdong Patentee before: Huawei Technologies Co., Ltd. |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20070124 Termination date: 20170625 |