CN100399765C - A voice message transmitting method - Google Patents

A voice message transmitting method Download PDF

Info

Publication number
CN100399765C
CN100399765C CNB021546886A CN02154688A CN100399765C CN 100399765 C CN100399765 C CN 100399765C CN B021546886 A CNB021546886 A CN B021546886A CN 02154688 A CN02154688 A CN 02154688A CN 100399765 C CN100399765 C CN 100399765C
Authority
CN
China
Prior art keywords
token
voice message
message
voice
time
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
Application number
CNB021546886A
Other languages
Chinese (zh)
Other versions
CN1505377A (en
Inventor
康凯
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to CNB021546886A priority Critical patent/CN100399765C/en
Publication of CN1505377A publication Critical patent/CN1505377A/en
Application granted granted Critical
Publication of CN100399765C publication Critical patent/CN100399765C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Landscapes

  • Data Exchanges In Wide-Area Networks (AREA)
  • Telephonic Communication Services (AREA)

Abstract

The present invention relates to a method for sending a speech message, which belongs to the technical field of telecommunication. In the method, first, a real-time transmission priority queue is created at the outgoing interface of a network device for speech flow; the speech message enters the real-time transmission priority queue; the speech message in the real-time transmission priority queue is sent with priority. The method of the present invention has the advantages that an RTP priority queue is independent of other queues of QoS, can coexist with other QoS queues, can use various combination, such as FIFO and RTP, WFQ and RTP, etc., and has strong flexibility and expandability. The RTP priority queue ensures that the speech message is sent with absolute priority under the condition of congestion, and simultaneously, the RTP priority queue also limits the bandwidth of the speech message, namely that the bandwidth occupied by the speech flow is smaller than actual physical bandwidth, so that the normal transmission of other messages is ensured.

Description

A kind of method that sends voice message
Technical field
The present invention relates to a kind of method that sends voice message, belong to the telecommunication technology field.
Background technology
The application of Internet Protocol telephone (Voice over IP is hereinafter to be referred as VoIP) more and more widely, voice flux is the real-time flow, and is very responsive to postponing.In order to make the VoIP operate as normal, (Quality of Service is hereinafter to be referred as the QoS) network that ensures that has service quality must be provided, guarantee that voice flux has higher priority than other non real-time flows, when network takes place when congested, guarantee that the voice message first priority sends.
According to the demand of different flow on the network, adopting corresponding QoS queue mechanism is the effective technology of current solution network service quality.When network sends when congested,, the message buffering in the formation in respective queue, again according to different queue mechanisms, is taked corresponding send mode by the various queue mechanisms of QoS.The QoS queue mechanism mainly contains fifo queue (First in First out, hereinafter to be referred as FIFO), Priority Queues (Priority Queuing, hereinafter to be referred as PQ), custom queuing (Custom Queuing, hereinafter to be referred as CQ), Weighted Fair Queuing (Weighted FairQueuing, hereinafter to be referred as WFQ) etc., at present, utilize the PQ queue mechanism, can solve the preferential problem that sends of voice message in theory.As shown in Figure 1, the PQ formation has 4 formations, be respectively high, in, normal, low, wherein high queue priority is the highest, low queue priority is minimum.By allowing voice message go into high-priority queue, can realize the preferential transmission of voice message.
1) when outgoing interface is used the current QoS formation, can only use a certain formation in the QoS formation, promptly or be PQ, or WFQ.PQ have only height, in, normal, low four priority queries, voice message has taken high-priority queue, all the other flows have only 3 formations to use, and that is to say all the other flows to be divided into 3 classifications.Feng Fu network traffics relatively, 3 classifications obviously lack autgmentability and flexibility.If this moment, outgoing interface had a large amount of stream, the demand of using the WFQ formation is arranged, the preferential transmission of using PQ formation solution voice message on interface obviously is inappropriate.
2) on low speed chain circuit, for example outgoing interface is the serial ports of 64Kbps, it is very normal that voice flux surpasses 64Kbps, if adopting the PQ formation to solve voice flux preferentially sends, so, because the preferential transmission of voice flux, can be with the bandwidth absorb, make other network traffics can not get sending, this situation can not put up with.
Summary of the invention
The objective of the invention is to propose a kind of method that sends voice message, (Real-Time Transport Protocol by the identification RTP, hereinafter to be referred as RTP) port numbers, the RTP Priority Queues of a strictness is provided for voice message, guarantee that when network congestion strict other non real-time flows of distinguishing realize that the first priority of voice message sends.
The method of the transmission voice message that the present invention proposes comprises the steps:
1, the outgoing interface at the network equipment is that voice flux is created a real-time Transmission Priority Queues;
2, make voice message enter the real-time Transmission Priority Queues;
3, the voice message in the transmission real-time Transmission Priority Queues.
In the said method, can limiting network equipment outgoing interface on the bandwidth of voice flux, for example making its bandwidth is 1~75% of actual physics flow, the method for restriction voice flux comprises the steps:
When 1, voice message arrives, add token in token bucket, the token number of interpolation is: the time that the permission of voice message arrives by Mean Speed and current message deducts the product of difference of the time of last message arrival;
2, the token number in the inspection token bucket makes its highest number of tokens Bc that is no more than token bucket, and unnecessary token abandons;
3, the bit number of voice message is compared with the token number in the token bucket,, from token number, deduct the bit number of voice message simultaneously, otherwise voice message is abandoned if the bit number of voice message then makes voice message enter the RTP Priority Queues less than token number.
In the said method, can carry out the transmission rate restriction, comprise the steps: the voice message in the formation
When 1, sending voice message from formation, add token in token bucket, the token number of interpolation is: the time that the permission of voice message arrives by Mean Speed and current message deducts the product of difference of the time of last message arrival;
2, the token number in the inspection token bucket makes its highest number of tokens Bc that is no more than token bucket, and unnecessary token abandons;
3, the bit number of voice message is compared with the token number in the token bucket,, from token number, deduct the bit number of voice message simultaneously, otherwise send the message in the non real-time transmit queue if the bit number of voice message less than token number, then makes voice message send.
The method that the assurance voice message first priority that the present invention proposes sends, its advantage is as follows:
1, the RTP Priority Queues is independent of other formations of QoS, can coexist with other QoS formations, can adopt multiple combinations such as FIFO and RTP, WFQ and RTP, and very strong flexibility and autgmentability are arranged.
2, the first priority under guaranteeing congestion situation of RTP Priority Queues sends in the voice message, and also the bandwidth to voice message limits, and promptly the shared bandwidth of voice flux is less than actual physical bandwidth, thereby has guaranteed the normal transmission of other messages.
Description of drawings
Fig. 1 sends the voice message flow chart with the Priority Queues mode in the prior art.
Fig. 2 be RTP voice message Priority Queues of the present invention join the team, go out group flow chart.
Embodiment
The method that the assurance voice message first priority that the present invention proposes sends, at first the outgoing interface at the network equipment is that voice flux is created a real-time Transmission Priority Queues; Make voice message enter the real-time Transmission Priority Queues; Send the voice message in the real-time Transmission Priority Queues.
The network equipment in the said method can be switch.
In the said method, voice flux on can limiting network equipment outgoing interface, for example making its bandwidth is 1~75% of actual physics flow, the method of restriction voice flux comprises: when voice message arrives, add token in token bucket, the token number of interpolation is: the time that the permission of voice message arrives by Mean Speed and current message deducts the product of difference of the time of last message arrival; Check the token number in the token bucket, make its highest number of tokens Bc that is no more than token bucket, unnecessary token abandons; The bit number of voice message is compared with the token number in the token bucket,, from token number, deduct the bit number of voice message simultaneously, otherwise voice message is abandoned if the bit number of voice message then makes voice message enter the RTP Priority Queues less than token number.
In the said method, can carry out the transmission rate restriction to the voice message in the formation, method for limiting is a token bucket algorithm, comprise: when from formation, sending voice message, add token in token bucket, the token number of interpolation is: the time that the permission of voice message arrives by Mean Speed and current message deducts the product of difference of the time of last message arrival; Check the token number in the token bucket, make its highest number of tokens Bc that is no more than token bucket, unnecessary token abandons; The bit number of voice message is compared with the token number in the token bucket,, from token number, deduct the bit number of voice message simultaneously, otherwise send the message in the non real-time transmit queue if the bit number of voice message less than token number, then makes voice message send.
In the said method, create a RTP Priority Queues separately, can coexist with other QoS formations for voice flux, as shown in Figure 2, to enrich the formation control of interface.
RTP Priority Queues among the present invention is when the voice message first priority sends under guaranteeing congestion situation, adopted on the exit port of the network equipment bandwidth to carry out method for limiting to voice message, be the shared bandwidth of voice flux less than actual physical bandwidth, thereby guaranteed the normal transmission of other flows.
Voice message is User Datagram Protoco (UDP) (being called for short a UDP) message, and its destination slogan is between 16384~32767.In voice flux, the destination slogan is that the message of even number is a data message, and the destination slogan is that the message of odd number is the control message, and the first priority that only needs to guarantee voice data message sends, and just can guarantee speech quality.Therefore,, make the destination slogan between 16384~32767 and for the UDP message of even number, go into the RTP Priority Queues, like this, just guaranteed that the first priority of voice message sends by identification protocol type and destination slogan.
When the outgoing interface of the network equipment takes place will guarantee that to the voice message of delay sensitive first priority sends when congested.If voice flux is not carried out bandwidth constraints, when voice flux surpassed interface bandwidth, other flows on the network can not get sending at all, and this is insupportable.Therefore, be necessary for the certain bandwidth of the flow of other except that voice assurance on the network.By in the bandwidth of going into formation time limit system voice flux, guarantee the bandwidth of other network traffics.
On the column direction of joining the team, be by token bucket (TockenBucket) algorithm to the qualification of speech bandwidth.Token number in the token bucket is token formation speed * token rise time TI (the Time Interval time interval).The capacity of token bucket be in the token bucket highest number of tokens (Conformed burst size, unit: bit) be Bc, committed information rate CIR (Committed Information Rate, unit: bps) be voice message allow to pass through Mean Speed.Token number in the token bucket can not surpass the capacity Bc of token bucket, and unnecessary token is dropped.
The concrete steps of bandwidth constraints are as follows:
A. during system start-up, token bucket is full state.
B. under the congested state of the outgoing interface of the network equipment, when voice message arrives, at first calculate in token bucket, to add how many token numbers.Token number=token formation speed * token rise time the TI that adds.Wherein, the token formation speed is exactly committed information rate CIR, and token rise time TI deducts the time that a message arrives time that current message arrives.Token number in the token bucket can not surpass the capacity Bc of token bucket, and unnecessary token is dropped.
C. check whether the token number in the token bucket is enough.If enough, then go into the RTP Priority Queues, consume simultaneously and the identical token number of voice message size; Otherwise, with packet loss.
By above-mentioned token bucket algorithm, set certain bandwidth, its span is (1~75) % of interface actual physical bandwidth, has guaranteed at least 25% bandwidth for other flows like this.
The RTP formation has oneself independently queue control block (QCB), is used with other queue control block (QCB)s of QoS.If what default situation lower interface was used is the WFQ formation, when interface congestion, if RTP formation this moment and WFQ formation all have message, the then unconditional message that sends earlier in the RTP formation, the message in the RTP formation sends and finishes, the message in sending out the WFQ formation.Because going into formation is the bandwidth that has limited voice flux, so, needn't worry the situation that the RTP Priority Queues takes bandwidth.
The present invention can also not limit when going into formation, when the outgoing interface of the network equipment is congested, so long as voice flux is just gone into the RTP Priority Queues, and carry out rate limit when dequeue.The method of restriction is identical with the method for limiting of above-mentioned bandwidth, when promptly from formation, sending voice message, add token in token bucket, the token number of interpolation is: the time that the permission of voice message arrives by Mean Speed and current message deducts the product of difference of the time of last message arrival; Check the token number in the token bucket, make its highest number of tokens Bc that is no more than token bucket, unnecessary token abandons; The bit number of voice message is compared with the token number in the token bucket,, from token number, deduct the bit number of voice message simultaneously, otherwise send the message in the non-RTP Priority Queues if the bit number of voice message less than token number, then makes voice message send.

Claims (5)

1. a method that sends voice message is characterized in that this method comprises the steps:
(1) outgoing interface at the network equipment is that voice flux is created a real-time Transmission Priority Queues, and the bandwidth of described real-time Transmission Priority Queues is less than actual physical bandwidth;
(2) port numbers of identification RTP makes voice message enter the real-time Transmission Priority Queues;
(3) voice message in the transmission real-time Transmission Priority Queues.
2. the method for claim 1 is characterized in that the network equipment is a switch described in the step (1).
3. the method for claim 1 is characterized in that step (2) also is included in voice message and enters before the real-time Transmission Priority Queues, in the enterprising lang sound of the outgoing interface of network equipment flow restriction.
4. method as claimed in claim 3 is characterized in that the voice flux restriction comprises the steps:
When (1) voice message arrives, add token in token bucket, the token number of interpolation is: the time that the permission of voice message arrives by Mean Speed and current message deducts the product of difference of the time of last message arrival;
(2) token number in the inspection token bucket makes its highest number of tokens Bc that is no more than token bucket, and unnecessary token abandons;
(3) bit number of voice message is compared with the token number in the token bucket,, from token number, deduct the bit number of voice message simultaneously, otherwise voice message is abandoned if the bit number of voice message then makes voice message enter the RTP Priority Queues less than token number.
5. the method for claim 1 is characterized in that described (3) step comprises that also the voice message in the formation is carried out transmission rate to be limited, and comprises the steps:
When (1) sending voice message from formation, add token in token bucket, the token number of interpolation is: the time that the permission of voice message arrives by Mean Speed and current message deducts the product of difference of the time of last message arrival;
(2) token number in the inspection token bucket makes its highest number of tokens Bc that is no more than token bucket, and unnecessary token abandons;
(3) bit number of voice message is compared with the token number in the token bucket,, from token number, deduct the bit number of voice message simultaneously, otherwise send the message in the non real-time transmit queue if the bit number of voice message less than token number, then makes voice message send.
CNB021546886A 2002-12-04 2002-12-04 A voice message transmitting method Expired - Fee Related CN100399765C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CNB021546886A CN100399765C (en) 2002-12-04 2002-12-04 A voice message transmitting method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CNB021546886A CN100399765C (en) 2002-12-04 2002-12-04 A voice message transmitting method

Publications (2)

Publication Number Publication Date
CN1505377A CN1505377A (en) 2004-06-16
CN100399765C true CN100399765C (en) 2008-07-02

Family

ID=34235552

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB021546886A Expired - Fee Related CN100399765C (en) 2002-12-04 2002-12-04 A voice message transmitting method

Country Status (1)

Country Link
CN (1) CN100399765C (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100420249C (en) * 2005-03-22 2008-09-17 中国科学院计算技术研究所 Method for guarantee service quality of radio local network
CN100461965C (en) * 2006-07-31 2009-02-11 华为技术有限公司 High-speed down block business attachment managing method, base station and system
CN105871839A (en) * 2016-03-30 2016-08-17 上海斐讯数据通信技术有限公司 WIFI voice message sending method, receiving method, sending device and receiving device
CN110085234A (en) * 2019-04-29 2019-08-02 苏州狗尾草智能科技有限公司 Access automatic speech recognition system

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0706297A1 (en) * 1994-10-07 1996-04-10 International Business Machines Corporation Method for operating traffic congestion control in a data communication network and system for implementing said method
WO2002041598A2 (en) * 2000-11-17 2002-05-23 Infineon Technologies North America Corp. Priority handling of voice over data in a voice-over-internet protocol processor
CN1352862A (en) * 1999-01-14 2002-06-05 艾利森电话股份有限公司 Priority transmission for various types of speech in network traffic

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0706297A1 (en) * 1994-10-07 1996-04-10 International Business Machines Corporation Method for operating traffic congestion control in a data communication network and system for implementing said method
CN1352862A (en) * 1999-01-14 2002-06-05 艾利森电话股份有限公司 Priority transmission for various types of speech in network traffic
WO2002041598A2 (en) * 2000-11-17 2002-05-23 Infineon Technologies North America Corp. Priority handling of voice over data in a voice-over-internet protocol processor

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
. 多业务企业网QOS保障技术方案.现代电信科技,第5卷. 2001
. 多业务企业网QOS保障技术方案.现代电信科技,第5卷. 2001 *
40-41. 流量控制和流量整型的比较.计算机系统应用,第11卷. 2002
40-41. 流量控制和流量整型的比较.计算机系统应用,第11卷. 2002 *
45-46. 多业务企业网QOS保障技术方案.现代电信科技,第5卷. 2001
45-46. 多业务企业网QOS保障技术方案.现代电信科技,第5卷. 2001 *

Also Published As

Publication number Publication date
CN1505377A (en) 2004-06-16

Similar Documents

Publication Publication Date Title
US7720065B2 (en) Method and apparatus for biasing of network node packet prioritization based on packet content
Roberts Internet traffic, QoS, and pricing
CA2464601C (en) Methods and systems for configuring voice over internet protocol network quality of service
US20050281277A1 (en) Establishing traffic priorities in a voice over IP network
CN101815032B (en) Method for classifying and isolating information based on integrated network security service architecture
US7050447B2 (en) Multi-level expedited forwarding per hop behavior
US20060268692A1 (en) Transmission of electronic packets of information of varying priorities over network transports while accounting for transmission delays
CN101692648B (en) Method and system for queue scheduling
EP1187399B1 (en) Method and a device for controlling source specific data flow
JP2007201965A (en) Packet repeating apparatus having transmission control function
JP2006506845A (en) How to select a logical link for a packet in a router
US20130272118A1 (en) System and Method for Automatically Adapting Audio Packet Marking in a Packet Network
WO2022022224A1 (en) Data packet scheduling method and related apparatus
ES2237190T3 (en) PROCEDURE FOR THE PREPARATION OF A STABLE SERVICE QUALITY FOR DATA SERVICES WITHIN A PACKAGE SWITCHING NETWORK.
CN100399765C (en) A voice message transmitting method
Ahmed et al. Implementing MPEG-4 video on demand over IP differentiated services
CN101212417A (en) Time granularity based internet QoS assurance method
JP2005528823A (en) A method for commonly controlling the bandwidth of a group of individual information flows
JP4087279B2 (en) BAND CONTROL METHOD AND BAND CONTROL DEVICE THEREOF
CN1798106B (en) Method and apparatus for scheduling transmission-link capacity between packet-switched traffic flows
Wang et al. A novel marking mechanism for packet video delivery over DiffServ Networks
Sharafali et al. Performance Evaluation of MPLS TE Signal Protocols for Voice Applications with QoS Implementation
Cobb An in-depth look at flow aggregation for efficient quality of service
Baraković et al. Priority level configuration for signaling service class
Baldi et al. Time driven priority router implementation and first experiments

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
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20080702

Termination date: 20151204

EXPY Termination of patent right or utility model