WO2011110130A2

WO2011110130A2 - Method for adjusting packet redundancy, device and network system thereof

Info

Publication number: WO2011110130A2
Application number: PCT/CN2011/073255
Authority: WO
Inventors: 胡国杰
Original assignee: 华为技术有限公司
Priority date: 2011-04-25
Filing date: 2011-04-25
Publication date: 2011-09-15
Also published as: CN102845008A; CN102845008B; WO2011110130A3

Abstract

A method for adjusting packet redundancy includes the following steps: the occurrence times of each packet loss event in a packet sequence which has been sent from a sender to a receiver are obtained; according to the occurrence times of each packet loss event, a redundancy to be used is determined, wherein the redundancy is the minimum redundancy among the redundancies which make the packet loss ratio of the packet sequence meet the requirement of packet loss; the sender is controlled to send, by using the redundancy to be used, packets to the receiver. A transmission device includes: an acquirement unit, for obtaining the occurrence times of each packet loss event in a packet sequence which has been sent from a sender to a receiver; a redundancy determination unit, for determining a redundancy to be used according to the occurrence times of each packet loss event, wherein the redundancy is the minimum redundancy among the redundancies which make the packet loss ratio of the packet sequence meet the requirement of packet loss; and a transmission unit, for sending packets to the receiver by using the redundancy to be used.

Description

Message redundancy adjustment method, related equipment and network system

The present invention relates to the field of communications technologies, and in particular, to a packet redundancy adjustment method, related device, and network system. Background technique

Packet-switched networks based on the Internet Protocol (IP) have been booming in recent decades and become a global communications network. IP networks have the advantages of low cost. With the development of communication, traditional telecommunication services including voice services are increasingly using IP networks. 7 IP networks with real-time voice services are also called packet voices. Voice over IP, called "VoIP" network. For example, H.323 networks, Next Generation Network (NGN), etc. are all VoIP networks.

A VoIP network can carry services with high real-time requirements, such as real-time voice, real-time video, and real-time data. Generally, the Real-Time Transfer Protocol (RTP) is used as a bearer protocol for real-time services. RTP is a transport protocol for multimedia data streams over IP networks that is defined to operate in a one-to-one or one-to-many transmission with the goal of providing time information and media stream synchronization. The typical application of RTP is based on the user datagram protocol ("UDP"), but it can also work on other protocols such as asynchronous transfer mode ("ATM"). The format of the voice packet on a typical VoIP network is: physical link layer + IP protocol layer + UDP protocol layer + RTP protocol layer + voice data.

RTP itself only guarantees the transmission of real-time data, but since the IP network is for an unreliable network with no connection, the RTP packet may be lost during transmission. However, the real-time service requires the quality of service ("Qos") to meet certain requirements. For example, the general requirement is that the packet loss rate is less than 1%, the delay is less than 100 milliseconds (ms), and the network jitter is less than 20 ms. Therefore, in order to guarantee the Q _{0S of the} real-time service, a redundant mechanism is needed to recover the lost data during the transmission.

The basic principle of the redundancy mechanism defined in RFC 2198 is: Each current RTP packet carries the content of the previous or the first few RTP packets. When an RTP packet is lost, it can be carried from the subsequent RTP packet. Recover lost RTP packets in redundancy. For example, when sending, if there is no redundancy The packet transmission sequence is RTP1, RTP2, RTP3, and RTP4, where RTPi represents the i-th RTP packet. If each RTP message carries a redundant packet, the message transmission sequence is RTP1, RTP2/RTP1, RTP3/RTP2, RTP4/RTP3; if each RTP message carries two redundant packets, the message transmission sequence It is RTP1, RTP2/RTP1, RTP3/RTP2/RTP1, RTP4/RTP3/RTP2. The part after the first "/" in the message is called the redundancy package, and the part before the first "/" is called If there is no packet loss during the transmission, the receiver only uses the primary packet as a valid service packet and discards the redundant packet. If there is a packet loss during the transmission, when each RTP packet carries a redundant packet, If the message RTP2/RTP1 is lost, the receiver will resolve the main packet RTP2 from the message RTP3/RTP2 to form a complete packet sequence RTPI, RTP2, RTP3.

The basic principle of the redundancy mechanism defined by the RFC2733 protocol is: Add redundant messages carrying Forward Error Correction (FEC) information to the transmitted message sequence, and the receiver can use redundancy. The packet recovers the lost RTP packet. In a specific implementation, a plurality of RTP packets may be extracted from the media data stream, and XORed, and a redundant packet containing FEC information may be generated, and the redundant packet may be used by the receiving end to recover any An RTP message used to generate it. For example, the sequence of message transmission when there are no redundant messages is a, b, c, d. The sequence of sending packets with redundant packets is a, b, f(a, b), c, d, f(c, d), where f(a, b) is a redundant message, indicating The result of the exclusive OR of the text a and the message b. On the receiving side, if no packet is lost, the redundant packets f(a, b) and f(c, d) are directly discarded; if the packet b is lost, the packet a and the packet f(a, b) are sent. After the XOR, the message b can be recovered, thereby realizing the packet loss compensation. If the message transmission sequence is a, f(a, b), b, f(b, c), c, f(c, d), d, the loss of two consecutive texts can be compensated. On the receiving side, if the messages b, c are lost, then a and f(a, b) XOR can be restored to b, and b and f(b, c) XOR can be restored (.

When the RTP redundancy mechanism defined in RFC 2198 is used, the prior art solution generally determines the number of redundant packets carried in each RTP packet during call setup, and each packet carries a fixed redundancy packet during the call process. When using the RTP redundancy mechanism defined by RFC2733, the prior art scheme generally determines the redundant error correction mode used by the call when the call is established, and uses the redundant error correction mode during the call.

The prior art has the following disadvantages:

Because the fixed redundancy is used in the prior art solution to compensate, if the network transmission environment deteriorates during the call, the packet loss rate will increase and the QoS requirement of the service cannot be met. Summary of the invention

The embodiment of the invention provides a packet redundancy adjustment method, a related device and a network system, which can reduce the service packet loss rate and ensure the service quality.

In view of this, the embodiments of the present invention provide:

A packet redundancy adjustment method includes:

Obtaining the number of occurrences of each packet loss event in the packet sequence sent by the sender to the receiver; determining the redundancy to be used according to the number of occurrences of each packet loss event, wherein the redundancy to be used is The packet loss ratio of the message sequence satisfies the minimum redundancy among the redundancy requirements of the packet loss requirement;

The control sender sends a message to the receiving end by using the redundancy to be used.

A transmitting device, comprising:

An obtaining unit, configured to acquire the number of occurrences of each packet loss event in the message sequence that the sending device has sent to the receiving device;

a redundancy determining unit, configured to determine a redundancy to be used according to the number of occurrences of each packet loss event, where the redundancy to be used is to enable the packet loss rate of the packet sequence to meet the packet loss requirement. The smallest redundancy among the redundancy;

And a sending unit, configured to send a message to the receiving device by using the redundancy to be used.

A network system, comprising: the above sending device, and a receiving device,

The receiving device is configured to receive the message sequence and the message sent by the sending device, and report the number of occurrences of each packet loss event in the message sequence to the sending device.

In the embodiment of the present invention, according to the number of occurrences of each packet loss event in the packet sequence, it is determined that the minimum redundancy among the redundancy that can satisfy the packet loss requirement of the packet sequence is the redundancy to be used, and then The control sender sends a message to the receiving end by using the redundancy to be used. Since the packet loss rate after the redundancy compensation to be used satisfies the packet loss requirement, the quality of the service can be ensured; and the redundancy to be used at the transmitting end is the smallest redundancy among the redundancy that satisfies the packet loss requirement. Therefore, the bandwidth required for service transmission can be reduced, and the waste of bandwidth can be avoided. DRAWINGS

In order to more clearly illustrate the technical solution of the embodiment of the present invention, the following needs to be made in the embodiment. BRIEF DESCRIPTION OF THE DRAWINGS The drawings in the following description are merely illustrative of some embodiments of the present invention, and those of ordinary skill in the art may also The drawings obtain other figures.

FIG. 1 is a flowchart of a method for adjusting packet redundancy according to an embodiment of the present invention;

2 is a flowchart of a message redundancy adjustment method for performing budget based on redundancy from large to small according to an embodiment of the present invention;

FIG. 3 is a flowchart of a message redundancy adjustment method for performing budget based on redundancy from small to large according to an embodiment of the present invention; FIG.

4 is a structural diagram of a transmitting device according to an embodiment of the present invention;

FIG. 5 is a structural diagram of another sending device according to an embodiment of the present invention;

6 is a structural diagram of still another transmitting device according to an embodiment of the present invention;

7 is a structural diagram of a transmitting device with a timer according to an embodiment of the present invention;

FIG. 8 is a structural diagram of a network system according to an embodiment of the present invention. detailed description

Referring to FIG. 1 , an embodiment of the present invention provides a packet redundancy adjustment method, where the method includes: 101. Obtain a number of occurrences of each packet loss event in a packet sequence that has been sent by a sending end to a receiving end. The execution body of each step of the embodiment of the present invention may be a sender or a receiver. The packet loss event in the packet sequence in the step may include: losing one packet event, dropping two packet events, dropping three packet events, and dropping four packet events. Among them, the loss of one packet event is that only one valid service packet is lost when the event occurs; the loss of two packet events is that two valid service packets are continuously lost when the event occurs; the three packet events are lost continuously when the event occurs. One valid service package; the loss of 4 packet events is the continuous loss of 4 valid service packets when the event occurs. The lost i packet event in the embodiments of the present invention refers to an event in which one valid service packet is lost.

The step may be: the sending end sends the message sequence to the receiving end by using the specific redundancy, and the number of occurrences of each packet loss event in the message sequence reported by the receiving end is obtained; wherein, the specific redundancy It may be 0, that is, the transmitting end sends the sequence of the packet without performing packet loss compensation, and the specific redundancy may also be 1, 2 or the like.

For the RFC2198 protocol, the redundancy is carried in each RTP packet allowed by the sender. The number of redundant packets; For the RFC2733 protocol, the redundancy indicates the forward error correction mode, and the forward error correction mode with the redundancy of 1 can compensate for one packet event; the redundancy is 2 for the forward direction. The error correction mode can compensate for 2 packet events; the forward error correction mode with redundancy 3 can compensate for 3 packet events; the forward error correction mode with redundancy 4 can compensate for 4 packet events. . The packet transmission sequence of the forward error correction mode indicated by the redundancy degree may be: a, b, f(a, b), c, d, f(c, d); The sequence of message transmission in the forward error correction mode indicated by 2 may be: a, f(a, b), b, f(b, c), c, f(c, d), d; The sequence of message transmission in the forward error correction mode indicated by 3 may be: a, b, f(a, b, c), c, f(a, c, d), f(a, b, d), d Where a, b, c, and d in the above-mentioned respective message transmission sequences are valid service packets, f(a, b), f(b, c), f(c, d), f(a, b, c), f(a, c, d), f(a, b, d) are redundant packets.

102. Determine, according to the number of occurrences of each packet loss event, the redundancy to be used, where the redundancy to be used is the smallest redundancy that enables the packet loss rate of the packet sequence to meet the packet loss requirement. Redundancy.

The packet loss requirement is smaller than the preset packet loss threshold; the packet loss rate of the packet sequence is the sum of packet loss rates caused by each packet loss event.

If the packet loss event in the message sequence sent by the sender to the receiver is only "lost one packet event", the total packet loss rate (ie, the packet loss rate of each packet loss event) is not performed even if packet loss compensation is not performed. The sum is larger, and the compensation method with redundancy of 1 can also ensure that the total packet loss rate after compensation is 0. In this case, even if the compensation method with redundancy 2 or redundancy 3 is used, there is no compensation method. Any meaning, it will waste bandwidth. Based on this consideration, the redundancy to be used determined in this step is the smallest redundancy among the redundancy of the packet loss rate and the packet loss requirement.

The step 102 may be implemented in the following manner: According to the number of occurrences of each packet loss event, the maximum redundancy is decremented, and the determining step is performed on the redundancy one by one, and the determining step is to determine the redundancy. Whether the packet loss rate of the compensated packet sequence satisfies the packet loss requirement; and the minimum redundancy that can satisfy the packet loss requirement is determined as the redundancy to be used. The calculating step of the packet loss rate of the packet sequence after the redundancy compensation includes: for the i packet event, when i is greater than the redundancy j, the number of lost packets is calculated = ( ij ) * lost i The number of occurrences of the packet event; according to the number of lost packets of each packet loss event greater than j and the number of valid service packets in the message sequence, the packet sequence lost by the redundancy j is obtained. Package rate. Specifically, according to each packet loss event greater than j The number of lost packets and the number of valid service packets in the sequence of messages, the packet loss rate of each packet loss event greater than j is obtained, and then the sum of packet loss rates of each packet loss event greater than j is obtained, and the sum is The packet loss rate of the packet sequence compensated by the redundancy j; or, first, the sum of the number of packet loss events of each packet loss event greater than j, and dividing the sum by the message sequence The number of valid service packets is obtained, and the packet loss rate of the message sequence compensated by the redundancy j is obtained. Wherein, starting from the maximum redundancy, performing a determining step on the redundancy one by one, indicating that the determining step is performed from the maximum redundancy, does not mean that the determining step is performed for all the redundancy, and The decision step is stopped after finding the redundancy to be used.

103. The control sending end sends the message to the receiving end by using the redundancy to be used.

As described above, the execution body of each step of the embodiment of the present invention may be a sender or a receiver. If the sender of the entity is executed, the sender directly receives the redundancy to be used in the step 103. If the executor is the receiving end, the receiving end returns the indication information indicating the redundancy to be used to the sending end, and after receiving the indication information, the transmitting end uses the redundancy to be used. Send a message to the receiver.

Optionally, step 101, step 102, and step 103 may be performed periodically, that is, step 101, step 102, and step 103 are performed every predetermined period to adjust the redundancy in time according to the transmission environment. That is, when the cycle time arrives, the sender sends the message sequence for the test to the receiver with a certain degree of redundancy, and then obtains the number of occurrences of each packet loss event in the message sequence for the test reported by the receiver. And determining the redundancy to be used according to the number of occurrences of each packet loss event; optionally, when the cycle time arrives, the sender does not send the message sequence for testing, but directly utilizes the specific redundancy. Sends a sequence of data packets to be transmitted during the call to the receiving end, such as an RTP packet sequence, and obtains the number of occurrences of each packet loss event in the packet sequence reported by the receiving end.

Optionally, the step 101, the step 102, and the step 103 are performed when the packet loss ratio is greater than or equal to the packet loss threshold, that is, when the current packet loss rate of the packet sequence is greater than or equal to the packet loss threshold, the step is performed. 101. Step 102 and step 103, in order to adjust the redundancy in time when the transmission environment is deteriorated, and ensure the quality of the service transmission.

In the embodiment of the present invention, according to the number of occurrences of each packet loss event in the packet sequence, the minimum redundancy among the redundancy that can make the packet loss rate of the packet sequence meet the packet loss requirement is the redundancy to be used. The control sender then sends a message to the receiving end using the redundancy to be used. Since the packet loss rate after the redundancy compensation to be used satisfies the packet loss requirement, the quality of the service can be guaranteed; The redundancy used is the smallest redundancy among the redundancy that satisfies the packet loss requirement. Therefore, the bandwidth required for service transmission can be reduced, and the waste of bandwidth can be avoided.

In order to make the technical solutions provided by the embodiments of the present invention more clear, the following embodiments provide a detailed description of the technical solutions provided by the embodiments of the present invention:

Referring to FIG. 2, an embodiment of the present invention provides a method for adjusting packet redundancy, which is

As an example, the RTP packet of the RFC 2198 protocol is used to calculate the sum of the packet loss rates of events according to the redundancy from large to small. The method includes:

201. When the call is established, the sending end determines the maximum redundancy and the packet loss threshold of the current service. The packet loss threshold is determined by the type of the current service. Generally, the real-time voice service has a packet loss threshold of 1%, and the real-time data service has a packet loss threshold of 0%.

Among them, the maximum redundancy is the maximum redundancy allowed by the sender. The maximum redundancy is usually preset in the product, and the maximum redundancy of different models can be different.

202. In the start of the call, the sending end does not use the packet loss compensation mode, and directly sends an RTP message sequence to the receiving end.

In this step, the sender uses a redundancy of zero.

203. The receiving end determines the number of occurrences of each packet loss event in the RTP packet sequence according to the sequence number of the received RTP packet, and reports the real-time transport control protocol (Real-Time Transport Control Protocol) to the transmitting end. a packet, the RTCP packet carries the number of occurrences of each packet loss event, and the sender obtains a prediction result of the network environment according to the RTCP packet, wherein the prediction result of the network environment includes the number of occurrences of each packet loss event, The total packet loss rate in the RTP message sequence.

The RTP protocol and the RTCP protocol are used together. The RTP protocol is used to transmit service packets. The RTCP protocol is used to monitor and feedback the transmission of RTP packets on the IP network. During the call progress, the receiver periodically sends The sender feeds back the RTCP packet.

The RTP packet has a unique sequence number. In this step, the receiving end determines the number of occurrences of each packet loss event according to the sequence number of the received RTP packet, for example, in the RTP packet sequence sent by the sender. There are 10 RTP messages, and the sequence numbers are 1-10. If the RTP message sequence number received by the receiver is 1, 2, 3, 5, 8, or 10, the receiver does not receive the sequence. The RTP packet with the number 4, the RTP packet with the sequence number of 6, the RTP packet with the sequence number of 7, and the RTP packet with the sequence number of 9, the receiver can determine that one packet event is lost in the packet sequence. of The number of occurrences is the second (that is, one packet loss event is the RTP packet with the sequence number 4 lost, and the other one packet event is the RTP packet with the sequence number 9 lost). The number of occurrences of the two packet events is lost. It is 1 time (that is, the RTP message with sequence number 6 and the RTP message with sequence number 7 are lost).

Optionally, the RTCP packet may also carry the number of packet loss caused by each packet loss event, the packet loss rate caused by each packet loss event, and the total number of packet loss in the RTP packet sequence.

The existing RTCP packet includes: a cumulative number of packets lost field, an extended highest sequence number received field, and a draft-specific extension (profile). a specific extensions field, where the accumulated packet loss number field identifies the total number of lost packets in a period of time. In this embodiment, it is assumed that the RTP packet sequence is transmitted during the time period, and the accumulated packet loss number field identifies the above. The total number of lost packets in the RTP packet sequence; the extended received highest sequence number field identifies the sequence number of the packet loss statistics. The embodiment of the present invention extends the draft-specific extension field, and the number of occurrences of each packet loss event is carried in the draft-specific extension field. Optionally, the number of packet loss caused by each packet loss event may also be carried, and each packet loss event is caused. The rate of packet loss and so on.

204: The sending end determines whether the timer expires. If yes, step 205 is performed, and if no, step 211 is performed.

The duration of the preset timer of the sending end, when the timer expires, indicates the time when the transmitting end adjusts the redundancy.

205. The sender sets the redundancy j = the maximum redundancy N.

The redundancy j is the redundancy of the packet loss rate to be predicted.

The sum of the packet loss ratios of the packet loss events is compared with the packet loss rate of each packet loss event, and the sum of the packet loss ratios of the packet loss events is smaller than the packet loss threshold. If yes, go to step 207; , go to step 210.

The specific process of summing the packet loss ratios of each packet loss event after the budget is compensated by the redundancy j includes: For the i packet event, when i is greater than the redundancy j, the number of lost packets is calculated = ( ij ) * The number of occurrences of i packet loss events; firstly, according to the number of packet loss events of each packet loss event greater than j and the number of valid service packets in the RTP message sequence, the packet loss rate of each packet loss event greater than j is obtained. Then, the sum of the packet loss ratios of the packet loss events greater than j is obtained; or, the sum of the number of packet loss events of each packet loss event greater than j is first obtained, and the sum is divided by the number of valid service packets in the RTP packet sequence. , get the sum of the packet loss rate of each packet loss event. 207. The sending end determines whether j is equal to 0. If yes, go to step 208. If no, go to step 209.

208. The sending end determines that the redundancy degree is the redundancy to be used, and performs step 211.

209. The sending end sets j = j-l , and returns to step 206.

210. When j is not the maximum redundancy, the sender determines that j+1 is the redundancy to be used. When j is the maximum redundancy, the sender determines j as the redundancy to be used.

211. The sending end determines whether the currently performed call service ends. If no, go to step 212. If yes, end the process.

212. The sending end sends the RTP packet to the receiving end by using the redundancy to be used.

It should be noted that when the next timer expires, the sender does not use any packet loss compensation mode.

(ie, the redundancy is 0 at this time), directly send the RTP message sequence to be transmitted to the receiving end, and repeat steps 205-212.

In the embodiment of the present invention, according to the number of occurrences of each packet loss event in the RTP packet sequence, the sum of the packet loss rates of the packet loss events after budget compensation according to the redundancy degree is large and small, and according to the budget result The redundancy to be used by the transmitting end is determined, where the redundancy to be used by the transmitting end is the smallest redundancy among the redundancy of the packet loss rate and the redundancy of the packet loss threshold. Since the packet loss rate of the RTP packet sent by the subsequent redundancy is smaller than the packet loss threshold, the quality of the service can be guaranteed; since the redundancy is the smallest redundancy among the redundancy smaller than the packet loss threshold. Therefore, the bandwidth required for service transmission can be reduced, and the waste of bandwidth can be avoided.

In order to make the above embodiment more clear, an example will be described as follows:

Assume that the number of occurrences of one packet event is 3516 times, the number of occurrences of one packet loss is 531 times, and the number of occurrences of three packet events is 156 times, in the case of the RTP message sequence sent by the receiving end in step 203, If the number of occurrences of the four packet events is 15 times, the number of occurrences of each packet loss event is sent, and the test result of the network environment can be obtained. The test results of the network environment include: the number of occurrences of each packet loss event, and each packet loss. The number of packet loss caused by the event, the packet loss rate caused by each packet loss event, the total packet loss rate, and the probability of each packet loss event occupying the total packet loss event. The number of occurrences of each packet loss event and each packet loss event. The number of lost packets, the packet loss rate caused by each packet loss event, the total packet loss rate, and the probability of each packet loss event occupying the total packet loss event are respectively obtained by the received message in a period of time. In the embodiment, it is assumed that the RTP packet sequence is transmitted during the period of time, and the number of occurrences of each packet loss event and each The number of packet loss caused by the packet loss event, the packet loss rate caused by each packet loss event, the total packet loss rate, and the probability of each packet loss event occupying the total packet loss event are: The packet loss event in the above RTP packet sequence The number of occurrences, the number of lost packets caused by each packet loss event, the packet loss rate caused by each packet loss event, the total packet loss rate, and the probability of each packet loss event accounting for the total packet loss event.

The specific values are shown in Table 1 below.

Table 1

Assume that the packet loss threshold of the current call service is 3%. After the packet loss occurs, the packet loss rate does not need to be compensated to 0. The total packet loss rate of each packet loss event is less than the packet loss after packet loss compensation. The threshold is OK.

For the test results of the above network environment, consider using the redundancy of 4 to compensate, and after the redundancy is 4, the packet loss rate is 0; then consider using the redundancy of 3 to compensate. When the redundancy is 3, the packet loss caused by one packet loss, two packet events, and three packet events can be compensated. For the packet loss caused by the loss of four packet events, Make up After reimbursing 3 packets and 1 packet is not compensated, the compensation is 3, and the packet loss rate is 15/30000. It can be seen that the packet loss rate is less than the loss. Package threshold. Consider the method of using the redundancy of 2 to compensate. When the redundancy is 1, the packet loss caused by losing one packet event and dropping two packet events can be compensated. For the loss of three packets. In the packet loss caused by the event, 2 packets are compensated, and 1 packet is not compensated. In the packet loss caused by the 4 packet event, 2 packets are compensated, and 2 packets are not compensated. The number of lost packets caused by 3 packet loss events is 105, and the number of lost packets caused by 4 packet loss events is 30. The total packet loss rate is (105+30) / 30000. It can be seen that the total packet loss rate is less than the packet loss. Threshold; consider using the method of redundancy 1 to compensate. After the compensation is 1 mode, the packet loss caused by one packet event can be compensated. For the loss of 2 packet events, compensation One packet, and one packet are not compensated; for the loss of three packet events, one packet is compensated, and two packets are not compensated. For the loss of 4 packets: 1 package is compensated, and 3 packages are not compensated. As shown in Table 2 below, for the loss of one packet event, the total number of lost packets after the redundancy is 1 is 531. For the loss of two packet events, the redundancy is 1. The total number of lost packets after compensation is 210. For the loss of 3 packet events, the total number of lost packets after the redundancy is 1 is 45; then the redundancy is 1 The total number of lost packets caused by the compensated packet loss events is 786. It can be seen that the packet loss rate is 0% due to the loss of one packet event, and the packet loss rate caused by the loss of two packet events is 1.77%, and the packet loss caused by three packet events is lost. The rate is 0.70%, and the packet loss rate caused by a packet loss event is 0.15 %. The sum of the packet loss rates caused by each packet loss event is 2.62%, which is less than 3%. Therefore, the compensation method with redundancy of 1 is used. Can guarantee the quality of the business.

Redundancy 1

Compensate for the loss of 1

Number of events, number of lost packets, packet loss rate, method of compensation, probability of failure, packet loss after reimbursement

Packet loss rate

Lost 1 package event 3516 3516 84.38% 11.72% 0 0.00%

Lost 2 package events 531 1062 12.74% 3.54% 531 1.77%

Lost 3 package events 105 315 2.52% 1.05% 210 0.70%

Lost 4 package events 15 60 0.36% 0.20% 45 0.15%

Total packet loss event 4167

Total no loss event 25047

Total packet loss 4953

Total number of packages 30,000

Total packet loss rate 16.51%

Table 2

Referring to FIG. 3, another embodiment of the present invention provides a packet redundancy adjustment method. The method uses an RTP packet conforming to the RFC2198 protocol as an example, and the packet loss rate of each event is budgeted according to the redundancy from small to large. And, the method specifically includes:

Steps 301-303 are the same as steps 201-203, and are not described herein again.

304. The sending end determines whether the timer expires. If yes, step 305 is performed. If no, step 310 is performed.

305. The sender sets the redundancy j = 0.

In another alternative embodiment, the sender can set the redundancy j = 1.

306, the sender's budget is compensated by the redundancy j, the sum of the packet loss rate of each packet loss event, comparing whether the sum of the packet loss rate of each packet loss event is greater than or equal to the packet loss threshold, and if so, executing step 307; If no, go to step 309.

The specific process of the sum of the packet loss ratios of the packet loss events is similar to the step 305, and is not described here.

307, j = j+l.

308. The sender determines whether j is equal to N. If no, returns to step 306; if yes, step 309 is performed.

309. The sender determines that the redundancy j is the redundancy to be used. Steps 310-311 are the same as steps 211-212, and are not described herein again.

In the embodiment of the present invention, according to the number of occurrences of each packet loss event in the RTP packet sequence, according to the redundancy from small to large, the sum of packet loss rates of the packet loss events after budget compensation, and according to the budget result, The redundancy to be used by the transmitting end is determined, wherein the redundancy to be used by the transmitting end is the smallest redundancy among the redundancy of the packet loss rate and the redundancy of the packet loss threshold. The sum of the packet loss ratios of the RTP packets sent by the subsequent redundancy is smaller than the packet loss threshold, which ensures the quality of the service; since the redundancy is the smallest redundancy among the redundancy smaller than the packet loss threshold. Therefore, the bandwidth required for service transmission can be reduced, and the waste of bandwidth can be avoided. Referring to FIG. 4, an embodiment of the present invention provides a sending device, including:

The obtaining unit 10 is configured to acquire the number of occurrences of each packet loss event in the message sequence that the sending device has sent to the receiving device.

The redundancy determining unit 20 is configured to determine, according to the number of occurrences of each packet loss event, the redundancy to be used, where the redundancy to be used is to enable the packet loss rate of the packet sequence to meet the packet loss requirement. The smallest redundancy among the redundancy; wherein the packet loss requirement may be less than a predetermined packet loss threshold.

The sending unit 30 is configured to send a message to the receiving device by using the redundancy to be used. The obtaining unit 10 is specifically configured to send the message sequence to the receiving device by using a specific redundancy, and obtain the number of occurrences of each packet loss event in the message sequence reported by the receiving device. Among them, the specific redundancy may be 0 or other values, which does not affect the implementation of the present invention.

As shown in FIG. 5, the redundancy determining unit 20 may specifically include: a determining unit 21, configured to perform decrementing from the maximum redundancy according to the number of occurrences of each packet loss event, and perform a determining step on the redundancy one by one, The determining step is to determine whether the packet loss rate of the packet sequence after the redundancy compensation is used meets the packet loss requirement; and the determining unit 22 is configured to determine the minimum redundancy that can satisfy the packet loss requirement. For redundancy to be used.

As shown in FIG. 6, the determining unit 21 may specifically include: a packet loss rate obtaining sub-unit 23, configured to decrement from the maximum redundancy according to the number of occurrences of each packet loss event, and perform redundancy on a redundancy basis as follows Operation: For the event of losing i packets, when i is greater than the redundancy j, the number of lost packets is calculated = ( ij ) * the number of occurrences of i packet events; the number of packets lost according to each packet loss event greater than j The number of valid service packets in the sequence of the message, and the packet loss rate of the message sequence compensated by the redundancy j is obtained; and the determiner The unit 24 is configured to determine whether the packet loss rate of the packet sequence compensated by the redundancy j satisfies the packet loss requirement. The packet loss rate of the packet sequence compensated by the redundancy j is obtained according to the number of packet loss events of each packet loss event greater than j and the number of valid service packets in the packet sequence. The implementation is similar to the method embodiment and will not be mentioned here.

As shown in FIG. 7, in a specific implementation, the sending device further includes: a timer 40, configured to trigger, when the timing time arrives, the acquiring unit to send the packet to the receiving device by using a specific redundancy. sequence. The message sequence may be a sequence of messages used for testing, or a sequence of messages to be transmitted between the sending device and the receiving device, such as an RTP message sequence, which does not affect the implementation of the present invention; The message sent by 30 is an RTP message that needs to be transmitted between the sending device and the receiving device. The minimum redundancy among the redundancy that enables the packet loss rate of the message sequence to meet the packet loss requirement is the redundancy to be used, and then the transmitting unit uses the redundancy to be used to the receiving device. Send a message. Since the packet loss rate after the redundancy compensation is used to meet the packet loss requirement, the service quality can be guaranteed; since the redundancy is the smallest redundancy among the redundancy requirements for meeting the packet loss requirement, the service transmission can be performed. The required bandwidth is reduced to avoid wasting bandwidth. Referring to FIG. 8, an embodiment of the present invention provides a network system, including: a sending device 50 and a receiving device 60, where

The sending device 50 is configured to obtain the number of occurrences of each packet loss event in the message sequence that the sending device 50 has sent to the receiving device 60, and determine the redundancy to be used according to the number of occurrences of each packet loss event, where to be used The redundancy is the smallest redundancy in the redundancy that enables the packet loss rate of the packet sequence to meet the packet loss requirement; and the packet is sent to the receiving device 60 by using the redundancy to be used;

The receiving device 60 is configured to receive the message sequence and the message sent by the sending device 50, and report the number of occurrences of each packet loss event in the message sequence to the sending device.

The structure and function of the specific sending device may be the same as the structure and function of the sending device in the embodiment shown in FIG. 4 to FIG. 7 , and details are not described herein again.

In the embodiment of the present invention, the sending device determines, according to the number of occurrences of each packet loss event in the packet sequence, that the redundancy of the packet sequence can meet the packet loss requirement, and the minimum redundancy is the redundancy to be used. The redundancy is then sent to the receiving device using the redundancy to be used. Since the packet loss rate after the redundancy compensation is used to meet the packet loss requirement, the service quality can be guaranteed; since the redundancy is the smallest redundancy among the redundancy requirements for meeting the packet loss requirement, the service transmission can be performed. The required bandwidth is reduced to avoid wasting bandwidth.

A person skilled in the art can understand that all or part of the steps of implementing the foregoing embodiments may be performed by a program to instruct related hardware, and the program may be stored in a computer readable storage medium, such as a read only memory. Disk or disc, etc.

The packet redundancy adjustment method, the related device, and the network system provided by the embodiments of the present invention are described in detail. The principles and implementation manners of the present invention are described in the specific examples. The description of the above embodiments is only The method for understanding the present invention and its core idea; at the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific embodiment and the scope of application. The description should not be construed as limiting the invention.

Claims

Rights request

A method for adjusting packet redundancy, which is characterized by comprising:

2. The method of claim 1 wherein

Obtaining the number of occurrences of each packet loss event in the message sequence that has been sent to the receiving end by the sending end includes: sending, by the sending end, the message sequence to the receiving end by using the specific redundancy, and acquiring the message sequence reported by the receiving end The number of occurrences of each packet loss event.

The method according to claim 1, wherein the redundancy to be used is determined according to the number of occurrences of each packet loss event, including:

Determining from the maximum redundancy according to the number of occurrences of each packet loss event, performing a determining step on the redundancy one by one, the determining step is determining the packet loss rate of the message sequence after the redundancy compensation is used. Whether the packet loss requirement is met; the minimum redundancy that can satisfy the packet loss requirement is determined as the redundancy to be used.

The method according to claim 3, wherein the calculating the packet loss rate of the message sequence after the redundancy compensation comprises:

For the i packet event, when i is greater than the redundancy j, the number of lost packets is calculated = ( i-j ) * The number of occurrences of i packet events is lost;

The packet loss rate of the text sequence compensated by the redundancy j is obtained according to the number of lost packets of each packet loss event greater than j and the number of valid service packets in the packet sequence.

The method according to claim 1, characterized in that said steps are performed every predetermined period.

6. A transmitting device, comprising:

a redundancy determining unit, configured to determine a redundancy to be used according to the number of occurrences of each packet loss event, where the redundancy to be used is to enable the packet loss rate of the packet sequence to meet the packet loss requirement. Redundancy Minimum redundancy;

7. The network device according to claim 6, wherein

The acquiring unit is configured to send the message sequence to the receiving device by using a specific redundancy, and obtain the number of occurrences of each packet loss event in the message sequence reported by the receiving device.

8. The transmitting device according to claim 6, wherein

The redundancy determining unit includes:

a judging unit, configured to perform decrementing from the maximum redundancy according to the number of occurrences of each packet loss event, and perform a judging step on the redundancy one by one, the judging step is to determine the packet after the redundancy compensation is used Whether the packet loss rate of the sequence meets the packet loss requirement;

A determining unit is configured to determine a minimum redundancy that satisfies the packet loss requirement as a redundancy to be used.

9. The transmitting device according to claim 8, wherein:

The determining unit includes:

The packet loss rate obtaining sub-unit is configured to perform decrementing from the maximum redundancy according to the number of occurrences of each packet loss event, and perform the following operations on redundancy one by one: for i packet events, when i is greater than redundancy j, the number of lost packets is calculated = ( ij ) * the number of occurrences of i packet events; the number of lost packets of each packet loss event greater than j and the number of valid service packets in the message sequence are obtained. The packet loss rate of the message sequence after the redundancy j is compensated;

The determining subunit is configured to determine whether the packet loss rate of the packet sequence compensated by the redundancy j satisfies the packet loss requirement.

The transmitting device according to claim 6, further comprising:

And a timer, configured to trigger the acquiring unit to send the sequence of the message to the receiving device by using a specific redundancy when the timing time arrives.

A network system, comprising: the transmitting device according to claims 6-10, and a receiving device,