WO2015135120A1

WO2015135120A1 - End-to-end network qos control system, communication device and end-to-end network qos control method

Info

Publication number: WO2015135120A1
Application number: PCT/CN2014/073170
Authority: WO
Inventors: 王江胜
Original assignee: 华为技术有限公司
Priority date: 2014-03-11
Filing date: 2014-03-11
Publication date: 2015-09-17
Also published as: CN103975551B; CN103975551A

Abstract

The present invention is applicable to the technical field of communications, and provides an end-to-end network QoS control method. The method comprises the following steps: performing flow classification on an IP grouping to be sent at a sending end; performing packet multiplexing on the whole data block of the same class of flows after being classified; performing FEC redundancy encoding processing on the whole multiplexed data block; interleaving the encoded data block; and performing subpackage encapsulation on the interleaved data block to form an IP grouping packet to be sent to a transmission network. By means of the technical solution provided in the embodiments of the present invention, network packet loss suppression can be realized, and data transmission quality can be improved.

Description

End-to-end network QoS control system, communication equipment and

End-to-end network QoS control method

Technical field

The present invention belongs to the field of communication technologies, and in particular, to an end-to-end network QoS control system, a communication device, and an end-to-end network QoS control method. Background technique

IP (Internet Protocol) is the most widely used protocol in the network. Because the IP protocol is not connection-oriented, the reliability is poor, and the QoS (Quality of Service) of the application layer service cannot be guaranteed.

In order to improve the QoS for transmitting data using the IP protocol, FEC (Forward Error Correction) technology has been proposed in the prior art. FEC technology is often used for the transmission of the physical layer and the data link layer, and the FEC corrects the erroneous bit problem in transmission by redundant coding. In the case of an erroneous bit, FEC decoding can be used to identify the error and correct the error. FEC is often used for encoding a data segment (for example, a few hundred bits), and its error correction capability is limited under certain coding capabilities. For example, in the case where the R-S code of (255, 239) is FEC, the source data is used as a code block every 239 bits, and is encoded to have a length of 255 bits. Its error correction capability is 8 bits, that is, within the 255-bit code block, the error is less than 8 bits, which can be corrected by FEC technology.

Since the FEC adopts the redundancy coding mode, the data packet loss rate is increased. If the packet loss rate is too high, the data block cannot be recovered, and the data transmission quality is degraded. Summary of the invention

The object of the present invention is to provide an end-to-end network QoS control system, a communication device, and an end-to-end network QoS control method, which can improve data transmission quality. The present invention is implemented in this way, an end-to-end network QoS control system, including

a transmitting end communication device, configured to perform traffic classification on a network protocol IP data packet to be sent, and perform packet multiplexing on a payload (payload) of an IP data packet belonging to the same data stream; The payload performs forward error correction FEC encoding to obtain an encoded data block; the encoded data block is subjected to data block interleaving processing to obtain an interleaved message; and the interleaved packet is packet-packed Forming an IP packet message and transmitting it to the transmission network;

The air traffic classification includes: determining, by the IP data packet configured with the same traffic classification mark, the same data flow;

a receiving end communication device, configured to receive the IP packet message from the transport network; record a flow classification flag of a header of the IP packet message, and reassemble the IP packet according to the flow classification flag User data, obtaining a reassembled payload; deinterleaving the reassembled payload; performing FEC decoding processing on the deinterleaved payload; demultiplexing the decoded payload, and classifying the stream The tag reconstitutes the IP header to obtain the user IP packet.

Another object of the present invention is to provide a communication device, the communication device comprising:

a traffic classification module, configured to perform traffic classification on the Internet Protocol IP data packet to be sent, where the traffic classification includes: determining, by the IP data packet configured with the same traffic classification mark, the same data flow;

a packet multiplexing module, configured to perform packet multiplexing on a payload payload of an IP data packet belonging to the same data stream;

An encoding processing module, configured to perform forward error correction FEC encoding on the payload that performs packet multiplexing, to obtain an encoded data block;

An interleaving module, configured to perform data block interleaving processing on the encoded data block to obtain an interleaved message;

The packet encapsulation module is configured to perform packetization and encapsulation on the interleaved packet, and form an IP packet message to be sent to the transmission network, so that the receiving end receives and reassembles the IP packet packet, and obtains the reassembled payload. Performing deinterleaving and FEC decoding on the reassembled payload.

Another object of the present invention is to provide a communication device, the communication device including a fragment reassembly module, configured to receive an IP packet message, where the IP packet message is obtained by packet multiplexing, FEC encoding, and interleaving processing by the transmitting end; and recording a traffic classification flag of the IP packet packet ;

An assembly module, configured to reassemble user data in the IP packet according to the flow classification flag, and obtain a restructured payload payload;

a deinterleaving module, configured to perform deinterleaving on the reassembled payload;

a decoding module, configured to perform FEC decoding processing on the deinterleaved payload;

The demultiplexing module is configured to demultiplex the decoded payload, and re-form the stream classification flag into an IP header to obtain an IP data packet.

Another object of the present invention is to provide an end-to-end network QoS control method, the method comprising the following steps:

The Internet Protocol IP data packet to be sent is classified into a traffic, where the traffic classification includes: determining, by using the same traffic classification tag, an IP data packet as the same data flow;

Performing packet multiplexing on the payload payload of the IP data packet belonging to the same data stream; performing forward error correction FEC encoding on the payload multiplexed by the packet to obtain the encoded data block;

Performing block interleaving processing on the encoded data block to obtain an interleaved message; packetizing and encapsulating the interleaved message to form an IP packet message to be sent to the transmission network, so that the receiving end receives the packet Recombining the IP packet, obtaining a reassembled payload, performing deinterleaving and FEC decoding on the reassembled payload.

By adopting the technical solution provided by the embodiment of the present invention, the data processed by the forward error correction coding is interleaved and corresponding demodulation and deinterleaving are performed to implement network packet loss suppression, thereby greatly improving the application quality of the Internet. Especially in the TCP transmission scenario, since there is no need to lose packet retransmission and slow down, it is extremely

DRAWINGS FIG. 1 is a schematic flowchart of an implementation process of an end-to-end network QoS control method according to Embodiment 1 of the present invention.

FIG. 2 is a schematic flow chart showing an implementation process of an end-to-end network QoS control method according to Embodiment 2 of the present invention.

FIG. 3 is a schematic diagram of a form of a data block provided by an embodiment of the present invention.

4 is a schematic structural diagram of a communication device according to Embodiment 3 of the present invention.

FIG. 5 is a schematic structural diagram of a communication device according to Embodiment 4 of the present invention.

6 and 7 are schematic diagrams showing the structure of an end-to-end network QoS control system according to Embodiment 5 of the present invention.

FIG. 8 is a schematic structural diagram of a transmitting end communication device according to Embodiment 6 of the present invention.

FIG. 9 is a schematic structural diagram of a receiving end communication device according to Embodiment 7 of the present invention. detailed description

The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

In the embodiment of the present invention, the technical solution described in the embodiment of the present invention is applied before the "slice reassembly" of the IP layer; that is, the embodiment of the present invention processes the IP packets that are not fragmented.

The following embodiments are used to improve the TCP throughput rate and improve the tolerance of the radio bearer to the packet loss of the transmission network.

Embodiment 1:

Referring to FIG. 1 , an implementation flow of an end-to-end network QoS control method according to Embodiment 1 of the present invention relates to a data sending end, including the following steps:

In the step S101, the IP data packet to be sent is subjected to traffic classification on the transmitting end, where the traffic classification includes determining IP data packets configured with the same traffic classification mark as the same data flow.

In the embodiment of the present invention, the foregoing traffic classification identifier includes a SIP (Session Initiazation Protocol, Session Initiation Protocol) parameters, DIP (Dynamic Inspection Protocol) parameters, PT (payload type) parameters, TOS (Service Type, Terms of Service) parameters. That is, packets with the same {SIP, DIP, PT, TOS} can be divided into one type of stream. The same class of flows will flow into the same processing module.

In the embodiment of the present invention, the traffic classification can be configured, and can be switched, and the granularity of the flow can be configured, and the following manners are respectively adopted:

• {SIP, DIP}

• {SIP, DIP, PT}

See {SIP, DIP, PT, TOS}

After the configured flow is processed by the sender in the embodiment of the present invention, the {SIP, DIP, TOS} field in the IP header in the sent packet will be filled with corresponding values according to different flows. If the TOS is not classified as a traffic class, the TOS value of the sent packet can be configured or specified by default (all 0s).

In step S102, the payloacK payload of the IP packet belonging to the same data stream is packet-multiplexed.

In the embodiment of the present invention, packet multiplexing combines multiple packets of the same stream into one. To segment different snippets, a multiplex header is used to indicate different message blocks in the payload.

In the packet-multiplexed payload, the {SIP, DIP, PT, TOS} of the multiplexed message is retained and filled in the finally transmitted message. A multiplex header is added to the message to isolate each original block and contains part of the information (such as length) of the original packet. The number of multiplexed packets is related to the expected network packet loss rate. For example, a 1% packet loss rate requires about 100 packet multiplexing.

Mode of multiplexing the header 1 : Configure an independent multiplexing header in front of each of the message blocks, and then concatenate them. :3⁄4 port is shown in the table below:

Multiplex header message block multiplex header message block multiplex header message block

1 1 2 2 NN The format of the multiplexer header is:

When PT=UDP, the following format is used:

Bytel Byte2 Byte3 Byte4

Length PT Reserved

UDP Header

UDP payload

When PT=TCP, the following format is used:

In the multiplexing mode of mode 1, the message is parsed in a serial manner.

Multiplex header mode 2: The different message blocks are concatenated, and a unified multiplexing header is configured for the concatenated message blocks, as shown in the following table:

Bytel Byte2 Byte3 Byte4

HeadLength PT Reserved

Packet Point 1

Packet Point 2

Packet Point N

Header + Payload 1

Header + Payload 2

Header + Payload 4 In the multiplexing mode of mode 2, each sub-block can be quickly found according to the unified header. The multiplexed data block size can be configured. Generally, the multiplexed data block size is configured according to the statistical packet loss condition and the transmission PMTU (Path MTU, Path Maximum Transmission Unit) value. For example, when the packet loss rate is 1% and PMTU=1000 Byte, the multiplex block size is configured to be 100 Kbyte. After the block size reaches 100KByte, the new packet will no longer be reused. The multiplexed packet size will be used as a parameter to follow the packet to the processing flow below. The multiplexed packet will enter the encoding module as a complete data block.

In step S103, the payload subjected to packet multiplexing is FEC-encoded to obtain an encoded data block.

In the embodiment of the present invention, the strength of the redundancy coding is related to the expected packet loss rate of the network. For example, a 1% packet loss rate is encoded with 10% error correctable decoding capability.

In the embodiment of the present invention, the manner of FEC coding is not limited, and it should be included in the scope of protection of the present invention as long as it meets the requirements. In the implementation of the present invention, RS coding is taken as an example. Generally, the coding of the RS decoding capability is selected by more than 5 times the packet loss rate of the bearer network. Taking the RS coding of (255, 239) as an example, the bit error rate with a bit error rate of 6.25% can be decoded. Therefore, the (255, 239) code can be used for the bearer network coding with a packet loss rate of 1% to resist packet loss. . After FEC encoding, the data block will be larger than the multiplexed data block. The entire data block after encoding, as a whole, enters the interleaving module below.

In step S104, the encoded data block is subjected to data block interleaving processing to obtain an interleaved message.

In the embodiment of the present invention, the step of interleaving the encoded data block includes performing interleaving processing in units of bit Bits, and using the encoded data block as an interleaver depth. Specifically: the entire data block after the FEC processing is scrambled. The main purpose is to completely disrupt all the bits. In the case of packet loss in the bearer network, in the de-interleaved data, the bits lost by the packet can be distributed in the entire FEC coded data block.

In step S105, the interleaved packet is packet-encapsulated, and the IP packet is sent to the transmission network, so that the receiving end receives and reassembles the IP packet, and obtains the reassembled payload. Performing deinterleaving and FEC decoding on the reassembled payload.

In the embodiment of the present invention, the interleaved packet is encapsulated and encapsulated, and the IP header, the UDP header, and the packet header are added to form an IP packet to be sent. The IP packet can be IP packetized and sent to the transport network. In order to improve efficiency, data packetization and IP fragmentation can be combined, that is, when the packet is packetized, the end-to-end PMTU is sub-packaged to avoid the delay caused by fragmentation reorganization and the transmission efficiency caused by fragmentation.

In the embodiment of the present invention, after encoding and interleaving, a large data block is formed. For example, a data block with a size of 100Kbyte after multiplexing, after encoding and interleaving, will reach a size of about 120Kbytes.

In the embodiment of the present invention, the Payload packetization module is used to perform packetization and encapsulation on the interleaved packet, and the Payload packetization module is responsible for the 120Kbyte data block according to "PMTU-IP header length - UDP header length - packet header The length of the length is divided into small data blocks. Each small block will be sent as an IP packet.

After Payload subcontracting and IP encapsulation, the format of the message is as follows:

IP header

UDP header

Subheader

Among them, the SIP, DIP, and TOS in the IP header are derived from the configuration of the traffic classification. The remaining fields are populated in the standard IP format. The PT field is padded with UDP.

In the embodiment of the present invention, the SIP and DIP of the UDP header are used as the port number. Other domains are standard IP mode is filled.

The packet header has the following format:

Bytel Byte2 Byte3 Byte4

Total Length

SN where Total Length is the size of the entire data block after interleaving. The SN is the starting position of the block of data in the entire block of data.

In this embodiment, the transmission efficiency is improved by dividing the large data block formed after the coding interleaving into small data blocks.

The end-to-end network QoS control method provided by the embodiment of the present invention performs network interleaving processing and corresponding demodulation and deinterleaving by performing forward error correction encoded data, thereby implementing network packet loss suppression and greatly improving Internet application. quality. Especially in the TCP transmission scenario, since there is no need to reduce packet retransmission and slowdown, the TCP throughput rate is greatly improved and the data transmission quality is improved. Embodiment 2:

Referring to FIG. 2, an implementation flow of an end-to-end network QoS control method according to Embodiment 2 of the present invention relates to a data receiving end, which includes the following steps:

In step S201, at the receiving end, receiving an IP packet, wherein the IP packet is obtained by packet multiplexing, FEC encoding, and interleaving by the transmitting end; and recording the "3⁄4" of the IP packet The flow classification is marked as "^.

Specifically, the reassembling the user data in the IP packet includes: assembling a plurality of the IP packet files into a complete payload according to the packet header information in the IP packet.

Optionally, the flow classification flag of each flow may be recorded, including {SIP, DIP, PT, TOS} of the IP header, for reorganizing the user data in the flow. For example, after receiving the message, record the flow classification tag of the header. The message is concatenated according to the totallength and SN in the packet header.

Optionally, the packet can be received according to a general IP packet receiving method. If fragmentation is carried out, Then the fragment is reorganized and transferred to the specified UDP (User Datagram Protocol) port. It can be understood that if the PMTU can be installed on the sender for packetization, the fragment will be reassembled at the sender.

In step S202, the packet header is reassembled to assemble a plurality of IP packets into a complete payload. Optionally, in the process of reorganizing the user data in the IP packet message, if data packet loss occurs, the data lost during the data transmission may be filled with random data.

If a certain data block is not received within a specified time (preconfigurable, generally twice the delay of the bearer network), the data block is considered to be lost. Fill the lost bit with random data and fill this data block.

Reorganize and drop the padded data block, which has the following format:

Data block 1

Data block 3

Data block N where data block 3 is a random padding caused by packet loss.

In step S203, the reassembled payload is deinterleaved.

In the embodiment of the present invention, deinterleaving is an order in which the data blocks are reversed according to the interleaving algorithm, and the data blocks are changed back to the order in which they are not interleaved.

After de-interleaving, the form of the data block is similar to that shown in Figure 3. It can be seen that after de-interleaving, the randomly padded bits after the packet loss are dispersed at the respective small points of the entire big data block.

In step S204, FEC decoding is performed on the deinterleaved payload.

In the embodiment of the present invention, decoding is performed using a standard FEC decoding method. Due to UDP and TCP Each has its own checksum checksum, so the decoded data block will be handed over to the application layer without verification. Guaranteed by the application layer is correct.

In step S205, demultiplexing is performed according to the multiplexing header, and the flow classification flag is reconstituted into an IP header to obtain an IP data packet.

For example, the {SIP, DIP, PT, TOS} of the pre-recorded stream can be reconstituted into an IP header to obtain an IP packet.

In the embodiment of the present invention, the step of performing demultiplexing according to the multiplexing head is specifically: after receiving the FEC error-corrected decoded data block, using the multiplexing header therein, recovering each sub-message. After populating the corresponding IP headers, submit them to the various applications on the application layer.

Therefore, step S206 may be further included, and the IP data packet is sent to the corresponding target application layer. The end-to-end network QoS control method provided by the embodiment of the present invention performs network interleaving processing and corresponding demodulation and deinterleaving by performing forward error correction encoded data, thereby implementing network packet loss suppression and greatly improving Internet application. quality. Especially in the TCP transmission scenario, since there is no need to reduce packet retransmission and slowdown, the TCP throughput rate is greatly improved and the data transmission quality is improved. Embodiment 3:

FIG. 4 is a schematic structural diagram of a communication device according to an embodiment of the present invention. For the convenience of description, only the parts related to the embodiment of the present invention are shown. The communication device includes a stream classification module 101, a packet multiplexing module 102, an encoding processing module 103, an interleaving module 104, and a packet encapsulation module 105. The communication device may be a software unit, a hardware unit, or a combination of hardware and software.

The communication device can be used as a data transmission end in the QoS control method in the embodiment shown in FIG. 1 and FIG. 2, and can be an RNC.

The traffic classification module 101 is configured to perform traffic classification on the Internet Protocol IP data packet to be sent, where the traffic classification includes determining, as the same data flow, an IP data packet configured with the same traffic classification identifier.

The packet multiplexing module 102 is configured to perform packet multiplexing on the payload payload of the IP data packet belonging to the same data stream. The encoding processing module 103 is configured to perform forward error correction FEC encoding on the payload that has undergone packet multiplexing, to obtain an encoded data block.

The interleaving module 104 is configured to perform data block interleaving processing on the encoded data block to obtain a woven message.

The packet encapsulation module 105 is configured to perform packetization and encapsulation on the interleaved packet, and form an IP packet packet to be sent to the transmission network, so that the receiving end receives and reassembles the IP packet packet, and obtains the reassembled payload. De-interleaving and FEC decoding of the reassembled payload.

As an embodiment of the present invention, the flow classification module 101 may be specifically configured to determine an IP data packet that is configured with the same session initiation protocol parameter SIP, dynamic monitoring protocol parameter DIP, payload type parameter PT, and service type parameter TOS. For the same data stream.

As an embodiment of the present invention, the packet multiplexing module 102 may be specifically configured to add a multiplexing header to the payload, where the multiplexing header is used to indicate different >3⁄4 text blocks in the payload.

The packet multiplexing module may be further configured to: configure an independent multiplexing header in front of each of the message blocks, and serially connect the message blocks of the independent multiplexing header; or, serially connect the different A message block, and a unified multiplexing header is configured for the concatenated message block.

The packet multiplexing module may be further configured to: retain, in a packet multiplexed payload, a flow parameter of the IP data packet, where the flow parameter includes {SIP, DIP, PT, TOS}.

As another embodiment of the present invention, the packet encapsulation module may be specifically configured to perform packetization by using an end-to-end path maximum transmission unit PMTU.

As another embodiment of the present invention, the interleaving module may be specifically configured to perform interleaving processing by using a bit Bit as a unit, and using the encoded data block as an interleaver depth.

The end-to-end network QoS control method provided by the embodiment of the present invention performs network interleaving processing and corresponding demodulation and deinterleaving by performing forward error correction encoded data, thereby implementing network packet loss suppression and greatly improving Internet application. quality. Especially in the TCP transmission scenario, since there is no need to lose packet retransmission and slow down, the TCP throughput rate is greatly improved, and the data transmission quality is improved. Embodiment 4:

FIG. 5 is a schematic structural diagram of a communication device according to an embodiment of the present invention. For the convenience of description, only parts related to the embodiment of the present invention are shown. The communication device includes: a fragment reassembly module 201, an assembly module 202, a deinterleave module 203, a decoding module 204, a demultiplexing module 205, and a sending module 206. The communication device may be a software unit, a hardware unit or a unit combining software and hardware.

The communication device can be used as a data receiving end in the QoS control method in the embodiment shown in FIG. 1 and FIG. 2, and can be a Node B.

The fragment reassembly module 201 is configured to receive an IP packet, where the IP packet is obtained by packet multiplexing, FEC encoding, and interleaving, and records a flow classification of the header of the IP packet. mark.

The flow classification flag includes a session initiation protocol parameter SIP, a dynamic monitoring protocol parameter DIP, a payload type parameter PT, and a service type parameter TOS.

The assembling module 202 is configured to reassemble user data in the IP packet according to the flow classification flag, and obtain a reassembled payload.

The deinterleaving module 203 is configured to perform deinterleaving on the reassembled payload.

The decoding module 204 is configured to perform FEC decoding processing on the deinterleaved payload.

The demultiplexing module 205 is configured to demultiplex the decoded payload, and reconstitute the stream classification tag into an IP header to obtain an IP data packet.

A sending module 206 is further included for transmitting the reacquired IP data packet to the corresponding target application layer.

As an embodiment of the present invention, the fragment reassembly module 201 is specifically configured to: perform fragmentation and reassembly on the IP packet, and forward the packet obtained by fragment reassembly to a specified user data packet protocol UDP port. .

The fragment reassembly module 201 may be further configured to assemble a plurality of the IP packet files into a complete payload according to the packet header information in the IP packet message. In the process of fragment reassembly, if data packet loss occurs, random data can be used to fill the bits lost during data transmission.

It can be understood that the communication device shown in the embodiment of FIG. 4 and FIG. 5 can cooperate with the end-to-end network QoS control method shown in the embodiment of FIG. 1 and FIG. 2, and the detailed function description of each module can refer to the method embodiment. The relevant content will not be described here.

The end-to-end network QoS control method provided by the embodiment of the present invention performs network interleaving processing and corresponding demodulation and deinterleaving by performing forward error correction encoded data, thereby implementing network packet loss suppression and greatly improving Internet application. quality. Especially in the TCP transmission scenario, since there is no need to reduce packet retransmission and slowdown, the TCP throughput rate is greatly improved and the data transmission quality is improved. Embodiment 5:

Referring to FIG. 6 and FIG. 7, FIG. 7 is an end-to-end network QoS control system according to Embodiment 5 of the present invention. The system includes: a transmitting end communication device 100 and a receiving end communication device 200.

The transmitting end communication device 100 is configured to perform traffic classification on the network protocol IP data packet to be sent, and perform packet multiplexing on the payload payload of the IP data packet belonging to the same data stream; The payload performs forward error correction FEC encoding to obtain the encoded data block; the encoded data block is subjected to data block interleaving processing to obtain an interleaved packet; and the interleaved packet is packet-encapsulated, Forming an IP packet message and transmitting it to the transmission network;

a receiving end communication device 200, configured to receive the IP packet message from the transport network; record a flow classification flag of a header of the IP packet message, and reassemble the IP packet according to the flow classification flag User data, obtaining a reassembled payload; deinterleaving the reassembled payload; performing FEC decoding processing on the deinterleaved payload; demultiplexing the decoded payload, and decoding the stream The classification tag reconstitutes the IP header to obtain the user IP packet.

The transmitting end communication device 100 may be the communication device in the embodiment shown in FIG. 4, and the receiving end communication device 200 may be the communication device in the embodiment shown in FIG. 5, which may be executed in the embodiment shown in FIG. 1 or FIG. End-to-end network QoS control method for the end-to-end network QoS control system Example 6:

Referring to FIG. 8, the communication device provided in Embodiment 6 of the present invention includes: a processor 61, a memory 62, and a network interface 63. among them,

The processor 61 is configured to execute a program.

In an embodiment of the invention, the program may include program code, the program code including computer operating instructions.

Processor 61 may be a central processing unit CPU or one or more integrated circuits configured to implement embodiments of the present invention.

The memory 62 is used to store a program.

Memory 62 may include random access memory and may also include non-volatile memory.

The network interface 63 is configured to send the IP packet to be sent to the transport network.

In the embodiment of the present invention, the network interface is a network card. The processor 61 performs the following method:

And classifying the Internet Protocol IP data packet to be sent, where the traffic classification includes: determining IP data packets configured with the same traffic classification mark as the same data flow; and IP data packets belonging to the same data flow The payload payload is subjected to packet multiplexing; the payload of the packet multiplexing is subjected to forward error correction FEC encoding to obtain a coded data block; and the encoded data block is subjected to data block interleaving processing to obtain an interleaved The packet is encapsulated and encapsulated, and the IP packet is sent to the transmission network, so that the receiving end receives and reassembles the IP packet, and obtains the reassembled payload. The reassembled payload is deinterleaved and FEC decoded. Example 7: And network interface 73. among them,

The processor 71 is configured to execute a program.

Processor 71 may be a central processing unit CPU or one or more integrated circuits configured to implement embodiments of the present invention.

The memory 72 is used to store a program.

Memory 72 may include random access memory and may also include non-volatile memory.

The network interface 73 is configured to receive a packet sent by the sending device.

In the embodiment of the present invention, the network interface may be a network card. The processor 71 performs the following method:

Receiving an IP packet message, wherein the IP packet message is obtained by packet multiplexing, FEC encoding, and interleaving processing by the transmitting end; recording a traffic classification flag of a header of the IP packet packet; and according to the traffic classification flag, Recombining the user data in the IP packet message, acquiring the reassembled payload; deinterleaving the reassembled payload; performing FEC decoding processing on the deinterleaved payload; performing the decoded information problem Demultiplexing, and re-forming the stream classification tag into an IP header to obtain an IP packet.

The processor 71 can further transmit the reacquired IP data packet to the target application layer.

For a detailed description of the communication device shown in the sixth embodiment and the seventh embodiment, reference may be made to the related content of other embodiments of the present invention, and details are not described herein.

In summary, the embodiment of the present invention implements network packet loss suppression by using interleaving and FEC, and greatly improves the quality of Internet applications. In the TCP transmission scenario, TCP throughput is greatly improved because packet loss retransmission and slowdown are not required.

It will be understood by those skilled in the art that all or part of the steps of implementing the above embodiments may be performed by a program to instruct related hardware, and the program may be stored in a computer readable form. In the storage medium, the storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, or the like. The above is only the preferred embodiment of the present invention, and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection of the present invention. Within the scope.

Claims

claims

1. An end-to-end network service quality QoS control system, characterized by including: a sending end communication device for flow classification of network protocol IP data packets to be sent, and classifying IP data belonging to the same data flow The payload of the packet is multiplexed; the payload that has been multiplexed is subjected to forward error correction FEC encoding to obtain the encoded data block; the encoded data block is subjected to data block interleaving processing to obtain The interleaved message; The interleaved message is sub-packaged and encapsulated to form an IP packet message and sent to the transmission network;

Wherein, the flow classification includes determining IP data packets configured with the same flow classification mark as the same data flow;

A receiving end communication device, configured to receive the IP packet message from the transmission network; record the flow classification mark of the header of the IP packet message, and reassemble the IP packet message according to the flow classification mark. User data, obtain the reorganized payload; perform deinterleaving processing on the reorganized payload; perform FEC decoding processing on the deinterleaved payload; demultiplex the decoded payload, and classify the flow The tags are reassembled into IP headers to obtain user IP packets.

2. A communication device, characterized by: including:

The flow classification module is used to perform flow classification on the Internet protocol IP data packets to be sent, wherein the flow classification includes determining IP data packets configured with the same flow classification tag as the same data flow;

A packet multiplexing module, used to packet multiplex the payloads of IP data packets belonging to the same data flow;

An encoding processing module, used to perform forward error correction (FEC) encoding on the packet multiplexed payload to obtain an encoded data block;

An interleaving module, used to perform data block interleaving processing on the encoded data blocks to obtain interleaved messages;

The subpackaging and encapsulating module is used to subpackage and encapsulate the interleaved message to form an IP packet message and send it to the transmission network, so that the receiving end receives and reorganizes the IP packet message and obtains the reorganized payload. The reorganized payload is deinterleaved and FEC decoded.

3. The device according to claim 2, characterized in that, the packet multiplexing module is specifically configured to add a multiplexing header to the payload, and the multiplexing header is used to mark different packets in the payload. text block.

4. The device according to claim 3, characterized in that the packet multiplexing module is specifically configured to configure an independent multiplexing header in front of each of the message blocks, and configure the independent multiplexing in series. header message block; or,

The different message blocks are concatenated, and a unified multiplexing header is configured for the concatenated message blocks.

5. The device according to claims 2-4, characterized in that the packet multiplexing module is specifically configured to retain the flow parameters of the IP data packet in the packet multiplexed payload.

6. The device according to claims 2-5, characterized in that the packetization and encapsulation module is specifically used to packetize using the end-to-end path maximum transmission unit PMTU.

7. The device according to claim 2, wherein the interleaving module is specifically configured to perform interleaving processing in bit units, and use the entire encoded data block as the interleaving depth.

8. The device according to claims 2-7, characterized in that the traffic classification module is specifically configured to be configured with the same session initiation protocol SIP parameters, dynamic monitoring protocol DIP parameters, payload type PT parameters and services IP packets of type TOS parameter are identified as the same data flow.

9. A communication device, characterized by, including,

Fragmentation and reassembly module, used to receive IP packet messages, wherein the IP packet messages are obtained by the sending end through packet multiplexing, FEC encoding and interleaving processing; record the flow classification mark of the >¾ header of the IP packet message ;

An assembly module, configured to reorganize the user data in the IP packet message according to the flow classification mark, and obtain the reorganized payload payload;

A deinterleaving module, used to deinterleave the reorganized payload;

Decoding module, used to perform FEC decoding processing on the deinterleaved payload;

The demultiplexing module is used to demultiplex the decoded payload, reassemble the flow classification mark into an IP header, and obtain the IP data packet.

10. The device according to claim 9, characterized in that the fragmentation and reorganization module is specifically used to:

The IP packet message is fragmented and reassembled, and the fragmented and reassembled message is forwarded to the designated User Data Message Protocol UDP port.

11. The device according to claim 9 or 10, characterized in that the fragmentation reorganization module is specifically used to:

According to the packet header information in the IP packet message, multiple IP packet messages are concatenated and assembled into a complete payload.

12. The device according to claims 9-11, characterized in that the fragmentation reorganization module is specifically used to:

When data packet loss occurs, random data is used to fill the bits lost during data transmission.

13. The device according to claims 9-12, characterized in that the fragmentation reorganization module is specifically used to:

Record the traffic classification mark of the header of the IP packet message, including session initiation protocol SIP parameters, dynamic monitoring protocol DIP parameters, payload type PT parameters and service type TOS parameters.

14. An end-to-end network QoS control method, characterized in that the method includes the steps of: performing flow classification on Internet Protocol IP data packets to be sent, wherein the flow classification includes: configuring the same flow classification tag The IP data packets are determined to be the same data flow;

Perform packet multiplexing on the payload of the IP data packet belonging to the same data flow; perform forward error correction FEC encoding on the payload that has been packet multiplexed to obtain an encoded data block;

Perform data block interleaving processing on the encoded data blocks to obtain interleaved messages; subpackage and encapsulate the interleaved messages to form IP packets and send them to the transmission network, so that the receiving end can receive and The IP packet is reorganized to obtain a reorganized payload, and the reorganized payload is deinterleaved and FEC decoded.

15. The method according to claim 14, characterized in that: the data will belong to the same Packet multiplexing of the payload of the streamed IP data packet includes adding a multiplexing header to the payload, and the multiplexing header is used to mark different text blocks in the payload.

16. The method according to claim 15, characterized in that, adding a multiplexing header to the payload includes configuring an independent multiplexing header before each message block, and configuring the multiplexing header in series. Message blocks with independent multiplexing headers; or,

17. The method according to claims 14-16, characterized in that,

Packet multiplexing the payload of the IP data packet belonging to the same data flow includes retaining the flow parameters of the IP data packet in the packet multiplexed payload.

18. The method according to claims 14-17, characterized in that said sub-packaging and encapsulating the interleaved messages includes sub-packaging using the end-to-end path maximum transmission unit PMTU.

19. The method according to claim 14, characterized in that, performing data block interleaving processing on the encoded data blocks includes: performing interleaving processing in units of bits, and interleaving the encoded data blocks in units of bits. The entire data block is used as the interleaving depth.

20. The method according to claims 14-19, characterized in that,

The traffic classification tag includes session initiation protocol SIP parameters, dynamic monitoring protocol DIP parameters, payload type PT parameters and service type TOS parameters.

21. An end-to-end network QoS control method, characterized in that the method includes: receiving IP packet messages, wherein the IP packet messages are obtained by the sending end through packet multiplexing, FEC coding and interleaving processing;

Record the flow classification mark of the header of the IP packet;

According to the traffic classification mark, reorganize the user data in the IP packet message and obtain the reorganized payload;

Perform deinterleaving processing on the reorganized payload;

Perform FEC decoding on the deinterleaved payload;

The decoded information header is demultiplexed and the flow classification mark is reassembled into an IP header to obtain Get IP packets.

22. The method according to claim 21, characterized in that: receiving IP packet messages includes:

The IP packet message is fragmented and reassembled by the sending end, and the fragmented and reassembled message is forwarded to the designated user data message protocol UDP port.

23. The method according to claim 21 or 22, characterized in that said reorganizing the user data in the IP packet message includes:

24. The method according to claims 21-23, characterized in that said reorganizing the user data in the IP packet context includes:

25. The method according to claims 21-24, characterized in that,