CN109347604B - Multi-hop network communication method and system based on batched sparse codes - Google Patents
Multi-hop network communication method and system based on batched sparse codes Download PDFInfo
- Publication number
- CN109347604B CN109347604B CN201811256029.0A CN201811256029A CN109347604B CN 109347604 B CN109347604 B CN 109347604B CN 201811256029 A CN201811256029 A CN 201811256029A CN 109347604 B CN109347604 B CN 109347604B
- Authority
- CN
- China
- Prior art keywords
- packets
- node
- batch
- recoding
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0056—Systems characterized by the type of code used
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
- H04L1/0009—Systems modifying transmission characteristics according to link quality, e.g. power backoff by adapting the channel coding
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
- H04L1/0015—Systems modifying transmission characteristics according to link quality, e.g. power backoff characterised by the adaptation strategy
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/06—Protocols specially adapted for file transfer, e.g. file transfer protocol [FTP]
Abstract
The invention relates to a multi-hop network communication method and a system based on batched sparse codes, wherein the method comprises the following steps: a transport layer in a source node receives data files from a data source, each data file comprising a plurality of data packets; when a file sending request is received, a transmission layer in the source node encodes the data packet by using an outer code in batch sparse codes to generate a plurality of batches, wherein the batches comprise a plurality of encoded packets, and are sent to a network layer; when the network layer receives the packets in the batch, whether the current node is a target node is detected, if not, the network layer recodes the packets in the same batch and sends the recoded packets to the next node; and if the current node is detected to be the destination node, sending the packet to a transmission layer in the destination node, and decoding the received batch by the transmission layer in the destination node to recover the data file.
Description
Technical Field
The invention relates to a multi-hop network communication method and a multi-hop network communication system, in particular to a multi-hop network communication method and a multi-hop network communication system based on batch sparse coding.
Background
Multi-hop network communication is a key technology in a plurality of application fields such as wireless sensor networks, underwater acoustic communication networks, satellite networks and the like. In the existing underwater wireless communication technology, the transmission range of an underwater sound channel is longest. However, since the path loss of the underwater acoustic channel increases exponentially with transmission distance, and the effective bandwidth decreases with increasing transmission distance, the communication distance of a practical acoustic modem is typically within 2 km. One way to achieve efficient, secure long-range (e.g., tens of kilometers) underwater communications is to implement multi-hop transmission by adding multiple intermediate nodes between the source node and the target node. With appropriate multi-hop transmission, the bandwidth between each neighboring node may be greater than the direct channel bandwidth from the source node to the target node, and the total transmission power of all nodes may be substantially lower than the total transmission power of the direct transmission from the source node to the target node.
However, there are many challenges to efficient multi-hop network communications (e.g., media access control). Due to the characteristics of high error rate, large propagation delay and the like of the wireless communication link, the packet loss phenomenon in the wireless communication link is not negligible, and the conventional schemes such as retransmission and fountain codes cannot effectively solve the packet loss problem. When the existing multi-hop wireless network communication solution is faced with the disadvantage of low throughput and/or long time delay when the network multi-hop frequency is larger.
Due to packet loss accumulation on each hop, only through intermediate nodesForwarding is not efficient. Considering a linear topology network consisting of a source node, a destination node and a sequence of consecutively connected intermediate nodes, assuming that each network link can send one packet per unit time and the packet loss rate is 0.2, if each intermediate node forwards only the received packet, the throughput of the line network with l hops is 0.8lEven if the value of l is small, throughput decreases at a very fast rate. For example, when l is 3, the throughput drops to 0.512, and when l is 10, the throughput drops to 0.107. If the feedback is immediate, reliable and free of communication cost, the transmission of the line network can be completed by the inter-hop retransmission in the prior art, but due to the characteristics of prolonged transmission time of a wireless communication link, high packet loss rate, half duplex and the like, the inter-hop retransmission is far from optimally solving the defects. Researchers in the field use hop-by-hop Forward Error Correction (FEC) techniques (including fountain codes and random linear codes) to solve the feedback problem of inter-hop retransmissions in the prior art. Hop-by-hop forward error correction requires full decoding and re-encoding at each intermediate node, thus creating additional computational effort and transmission delay in each hop. Furthermore, each intermediate node needs to buffer all received encoded packets for decoding, so the buffer size is at least the file size.
Therefore, there is a need for a multi-hop network communication method that can achieve lower transmission delay, lower computation amount, and lower storage cost of intermediate nodes.
Disclosure of Invention
The invention aims to solve the defects of the prior art and provides a multi-hop network communication method based on batch sparse coding, so that high-throughput and low-delay data transmission in a multi-hop network is realized.
In order to achieve the above object, the present invention provides a multi-hop network communication method based on batch sparse codes, which includes the following steps: s100, a transmission layer in a source node receives data files from a data source, wherein each data file comprises a plurality of data packets; s200, when a file sending request is received, a transmission layer in the source node encodes the data packet by using an outer code in batch sparse codes to generate a plurality of batches, wherein the batches comprise a plurality of encoded packets, and are sent to a network layer; s300, when the network layer receives the packets in the batch, detecting whether the current node is a destination node, if not, performing the step S400, and if so, performing the step S500; s400, the network layer recodes the packets in the same batch and sends the recoded packets to the network layer of the next node to return to execute the step S300; s500, the packet is sent to a transmission layer in a destination node, and the transmission layer in the destination node decodes the received batch to recover the data file.
Preferably, step S100 further comprises dividing the data file into equal-length data packets.
Preferably, step S400 further includes applying system re-encoding in the re-encoding process, wherein after the network layer receives the encoded packet, the received encoded packet is used as a re-encoded packet of the current node, an additional re-encoded packet is generated by random linear encoding, and the additional re-encoded packet and the received encoded packet are sent to the link layer of the current node.
Preferably, the method further includes setting the number of the re-encoded packets of the batch transmitted by the intermediate node to be the same as the number of the re-encoded packets of the source node.
Preferably, the batched sparse codes in step S200 include an outer code implemented in a transport layer of the node and an inner code implemented in a network layer of the node, wherein the outer code is a matrix fountain code and the inner code includes a random linear coding on the intermediate node.
Preferably, the batch sparse coding in the step S200 to generate a plurality of batches further comprises the following sub-steps: s210, distributing Ψ ═ by contrast (Ψ)1,...,ΨK) Sampling and returning degree di(ii) a S220, uniformly and randomly selecting d from all K input packetsiAn input packet Bi(ii) a S230, forming diFull random matrix G of xMiBatch X generated from multiple input packetsiIs represented by Xi＝BiGiWhere M is the batch size.
Preferably, the decoding the received batch in step S500 further includes decoding based on a belief propagation algorithm, and when the decoding based on the belief propagation algorithm is stopped, marking an input packet that is not decoded as inactive and replacing the input packet into the batch as a decoded packet to resume the decoding process based on the belief propagation algorithm.
According to another aspect of the present invention, a batched sparse coding based multi-hop network communication system is provided, which includes a data source module for sending data files to a source node, wherein each data file includes a plurality of data packets; an encoding module to encode the data packets with an outer code at a transport layer in a source node to generate a plurality of batches, the batches including a plurality of encoded packets, the batches being sent to a network layer; the recoding module is used for recoding the plurality of coded packets in the source node and the intermediate node; a storage module for buffering or storing the encoded packet in each node; and the decoding module is used for decoding the received batch at a transmission layer in the destination node so as to recover the data file.
The invention has the beneficial effects that: in a multi-hop network, batch sparse codes are utilized to realize high-throughput and low-delay data transmission, reliable transmission required by inter-hop retransmission is not required, and the method has low calculation and storage cost and the like.
Drawings
Fig. 1 shows a flowchart of a batched sparse code based multi-hop network communication method according to an embodiment of the present invention.
Fig. 2 shows a schematic diagram of a multi-hop wireless network according to an embodiment of the invention.
Fig. 3 shows a structural distribution diagram of each node according to an embodiment of the present invention.
Fig. 4 shows a block diagram of a batched sparse code based multi-hop network communication system according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, in one or more embodiments of the present invention, a batched sparse code based multi-hop network communication method disclosed by the present invention includes the following steps: a transport layer in a source node receives data files from a data source, each data file comprising a plurality of data packets; when a file sending request is received, a transmission layer in the source node encodes the data packet by using an outer code in batch sparse codes to generate a plurality of batches, wherein the batches comprise a plurality of encoded packets, and are sent to a network layer; when the network layer receives the packets in the batch, whether the current node is a target node is detected, if not, the network layer recodes the packets in the same batch and sends the recoded packets to the next node; and if the current node is detected to be the destination node, sending the packet to a transmission layer in the destination node, and decoding the received batch by the transmission layer in the destination node to recover the data file.
In accordance with one or more embodiments of the present invention, the disclosed method is applied to multihop network communications, as shown in fig. 2, which illustrates a model of a multihop network. The network has l hops, where node v0Is a source node and a node vlIs the destination node. The invention discloses a multi-hop network communication method based on batch sparse codes, which comprises operations in a network layer and a transmission layer, and is based on the following assumptions on a lower link layer and a physical layer:
there are only direct communication links between successive nodes, e.g. the network in fig. 2, node vi-1Only with node viDirect communication, i is a positive integer between 1 and l.
Each network link transports packets of the same length.
The transmission between the network links is not reliable, but all errors are detectable, so the network layer can only process correctly received packets.
The existence of a system parameter epsilon representing the maximum packet loss rate, in the link (v)i,vi-1) Packet loss rate on epsiloni≤ε。
Based on the above assumptions, the method disclosed herein operates independently of operations at the link and physical layers, and the transmission performance of the method is related to the average number of packets transmitted by the network.
According to one or more embodiments of the present invention, as shown in fig. 3, the communication method of the present invention performs transmission of packets at a transport layer and a network layer. Batch Sparse code (BATS code) short for BATS code, include outer code and inner code, wherein the outer code is a matrix fountain code and is realized on the transport layer, can produce a plurality of batches unlimited according to the data packet received, each batch includes the data packet (code packet) after the code of certain number M; the inner code comprises linear random coding on the intermediate node and is realized in a network layer, and is used for processing all packets in the same batch to process the packet loss phenomenon occurring in each hop. When M is 1, the outer code is a fountain code and the encoding process at the intermediate node is converted to a packet transmission process. The batch sparse codes retain the salient features of fountain codes, particularly their rate-free nature and low encoding/decoding complexity, and thus the batch sparse codes are able to achieve optimal throughput with smaller values of M in a multi-hop network.
Referring to fig. 3, although there are a transport layer and a network layer at each network node, the functions they perform at different nodes are different. The application layer of the source node has one or more data sources that transmit data files to the transport layer, which when a file transfer request is received by the transport layer at the source node encodes the data files with an outer code to generate batches that are further transmitted to the network layer. The network layer is an internal code part for realizing batch sparse codes, and after the network layer at a source node or an intermediate node receives the batch, the network layer recodes the same batch of packets and sends the recoded packets to the network node of the next hop through a link layer. After the network layer at the destination node receives the batch, it sends the packets in the batch to the transport layer at the destination node, which decodes the received packets to recover the file sent from the data source, which, when the decoded file is successfully obtained, passes the file to the application layer as shown in fig. 3, and may optionally send a notification to the source node confirming receipt of the file. For a wireless network based on batch sparse codes, the performance of the wireless network is not dependent on any other global network state information except the network length l and the maximum packet loss rate epsilon. Therefore, the wireless network based on the batch sparse codes does not need to acquire packet loss rate information between nodes, so that all link information of intermediate nodes does not need to be collected.
Further, in accordance with one or more embodiments of the present invention, a method for encoding a batch sparse code in a multi-hop network is presented. Specifically, K is the number of input packets, each packet being a column vector containing T elements in the base field, and a set of packets is equivalent to a matrix formed by juxtaposing the packets in the set. M is the batch size, 1,2 for i, the ith batch XiThe method comprises the following steps:
a contrast distribution Ψ ═ (═ Ψ)1,...,ΨK) Sampling and returning degree di；
Uniformly and randomly selecting d from all K input packetsiAn input packet Bi；
Form diFull random matrix G of xMiBatch X generated from multiple input packetsiIs represented by Xi＝BiGi。
The batch sparsely encoded inner code is formed by linear network encoding, i.e., re-encoding. The re-encoding is applied to the encoded packets belonging to the same batch so that the end-to-end transition of each batch from the source node to the destination node is a linear operation. HiIs the batch transformation matrix of the ith batch, YiIs the ith batch of outgoing or received packets at the destination node,
Yi＝XiHi＝BiGiHi(1)
wherein HiThe number of rows of (1) is M, HiNumber of columns of (i) th batch receptionThe number of packets of (a) is variable and limited for different batches, i.e. if no packets are received in the batch, Y isi(Hi) Is an empty matrix of 0 columns.
The batched transformation matrix is determined by re-encoding in conjunction with the network topology between the source node and the destination node and is recovered at the destination node by appending a coefficient vector to each encoded packet. Suppose HiI ═ 1,2.. h ═ (h) with the same rank distribution0,h1,...,hM) Wherein a rank distribution h is used to evaluate decoding performance, and a reachability of a batch sparse code is at a desired rank
As an upper boundary.
Further, in accordance with one or more embodiments of the present invention, a method for decoding a batch sparse code in a multi-hop network is presented. Assuming that n batches are received at the destination node, decoding is performed using a belief propagation algorithm, for equation (1), GiAnd HiAs is known, the generator matrix G and the transformation matrix H are considered decodable if the rank of (GH) is equal to its degree. Belief propagation algorithm decoding involves multiple iterations, in the first of which all decodable batches are decoded (by solving the associated system of linear equations (1)) and the input packets contained in those decodable batches are recovered. In each of the following iterations, the undecoded batch is first updated: for each undecoded batch, all recovered incoming packets contained in the batch are replaced into the associated linear system, and the batch is decremented accordingly. Some batches can be decoded after the update and the incoming packets in these decodable batches are recovered. In this implementation, belief propagation algorithm decoding will stop when there are no decodable batches.
Specifically, the batch sparse code has the following characteristics, when θ >0 and η ∈ (0,1), satisfied:
Ω(x；h,Ψ)+θln(1-x)>0,0≤x≤η,
wherein the content of the first and second substances,
when K tends to be infinite, for
The belief propagation algorithm decoding of an individual batch can recover η K input packets.
In one or more embodiments, for a given rank distribution h, the degree distribution is derived by solving an optimization problem as follows
s.t.Ω(x；h,Ψ)+θln(1-x)>0,0≤x≤η,
Wherein
And η is close to 1.
In the above embodiment, when the value of K is large, the degree distribution obtained by the above optimization model can obtain a good decoding effect. In one or more preferred embodiments, where the value of K is relatively small, such as 256 or 512, the belief propagation algorithm decoding tends to stop when decoding most of the incoming packets, although decoding can continue through gaussian elimination decoding in the prior art, but the computational complexity becomes very high. This problem is solved in the proposed method by using an inactive decoding (inactive decoding) method, where when the belief propagation algorithm decoding stops, the undecoded input packets are marked inactive and replaced into the batch as decoded packets to restore the belief propagation algorithm decoding process. The deactivated decoding reduces computational complexity and improves the success rate of confidence propagation algorithm decoding.
Further, in one or more of the present inventionIn one embodiment, the encoded packets are re-encoded at the source node or intermediate node, having Mk(Mk≧ M) as the node vkK 0, 1.. 1., the number of re-encoded packets in the batch sent by l-1, node v assuming that each lost packet in the network link is represented by an all-zero packetiL-1 receives M, i is 0,1i-1A packet including all zero packets for indicating a lost packet, and M-1M. Then at node viThe recoding of (A) is by M in the base domaini-1×MiMatrix phiiGiven, in the Link (v)i,vi+1) In batch MiEach recoded packet can pass through one Mi×MiDiagonal matrix E ofi+1Modeling is performed where the diagonal elements in the diagonal matrix are Bernoulli random variables, and then the batch transformation matrix H can be represented as: h ═ phi0E1Φ1E2...Φl-1El. Thus, phiiDeciding how to locate at node vkGenerating a recoded packet. Wherein the recoding method is random linear recoding, phiiIs a completely random matrix over the base domain. When random linear re-encoding is used, each transmitted packet is a linear combination of all received packets belonging to the same batch, so the network layer must receive all packets in the batch before transmitting the first re-encoded packet, at the intermediate node vkThe recoding delay of (A) is O (M)k-1T), and the total re-encoding delay is
In one or more preferred embodiments of the present invention, in order to reduce the re-encoding delay and computational complexity of the above-described random linear re-encoding, a system re-encoding method (systematic re-encoding) is employed for re-encoding, all received packets are used as re-encoded packets and additional packets are generated as re-encoded packets by random linear re-encoding. When the system recoding method is used, the network layer of the intermediate node sends out the packets immediately after receiving the packets, and the random linear recoding is carried out on all other packets without receiving the packets. In the system re-encoding method, the re-encoding delay is O ((l-1) T). For a larger base domain (e.g., q ═ 256), the decoding performance of applied system re-encoding is similar to that of random linear re-encoding, but the system re-encoding method can greatly reduce the re-encoding complexity and delay.
Further, in one or more embodiments of the invention, the number of re-encoded packets per batch sent at the intermediate node is equal to the number of re-encoded packets at the source node, i.e., M0＝M1＝...＝Ml-1. Preferably, after the intermediate node receives the re-encoded packets from the previous node, comparing whether the number of the received re-encoded packets is equal to the number sent by the previous node, if not, generating additional packets as the re-encoded packets in the same batch through the random linear re-encoding, so that the number of the re-encoded packets sent by the current node is equal to the number sent by the previous node.
Referring to fig. 4, according to one or more embodiments of the present invention, a batched sparse code based multi-hop network communication system is further proposed, which includes a data source module for sending data files to a source node, where each data file includes a plurality of data packets; an encoding module to encode the data packets with an outer code at a transport layer in a source node to generate a plurality of batches, the batches including a plurality of encoded packets, the batches being sent to a network layer; the recoding module is used for recoding the plurality of coded packets in the source node and the intermediate node; a storage module for buffering or storing the encoded packet in each node; and the decoding module is used for decoding the received batch at a transmission layer in the destination node so as to recover the data file.
Further, in one or more embodiments of the present invention, in order to reduce the re-encoding delay and computational complexity of the above-described random linear re-encoding in the aforementioned communication system, a system re-encoding method is employed for re-encoding, all received packets are used as re-encoded packets and additional packets are generated as re-encoded packets by random linear re-encoding. When the system recoding method is used, the network layer of the intermediate node sends out the packets immediately after receiving the packets, and the random linear recoding is carried out on all other packets without receiving the packets.
Further, in one or more embodiments of the present invention, the decoding module in the aforementioned communication system uses a belief propagation algorithm for decoding, the belief propagation algorithm decoding tends to stop when decoding a smaller number of input packets, and the method of deactivating decoding is used to solve the problem when the belief propagation algorithm decoding stops, and an input packet that is not decoded is marked as deactivated and is replaced into a batch as a decoded packet to resume the belief propagation algorithm decoding process. The deactivated decoding reduces computational complexity and improves the success rate of confidence propagation algorithm decoding.
It should be recognized that the embodiments of the present application can be realized and implemented by computer hardware, a combination of hardware and software, or by computer instructions stored in a non-transitory computer readable memory. The method may be implemented in a computer program using standard programming techniques, including a non-transitory computer readable storage medium configured with the computer program, where the storage medium so configured causes a computer to operate in a specific and predefined manner, according to the method and figures described in the detailed description. Each program may be implemented in a high level procedural or object oriented programming language to communicate with a computer system. However, the program(s) can be implemented in assembly or machine language, if desired. In any case, the language may be a compiled or interpreted language. Furthermore, the program can be run on a programmed application specific integrated circuit for this purpose.
Further, the method may be implemented in any type of computing platform operatively connected to a suitable interface, including but not limited to a personal computer, mini computer, mainframe, workstation, networked or distributed computing environment, separate or integrated computer platform, or in communication with a charged particle tool or other imaging device, and the like. Aspects of the application may be embodied in machine-readable code stored on a non-transitory storage medium or device, whether removable or integrated into a computing platform, such as a hard disk, optically read and/or write storage medium, RAM, ROM, or the like, such that it is readable by a programmable computer, which when read by the storage medium or device can be used to configure and operate the computer to perform the procedures described herein. Further, the machine-readable code, or portions thereof, may be transmitted over a wired or wireless network. The applications described herein include these and other different types of non-transitory computer-readable storage media when such media include instructions or programs that implement the above steps in conjunction with a microprocessor or other data processor. The present application also includes the computer itself when programmed according to the methods and techniques described herein.
While the disclosed technology is susceptible to various modifications and alternative constructions, certain embodiments thereof have been shown in the drawings and have been described above in detail. It should be understood, however, that there is no intention to limit the application to the specific form or forms disclosed; on the contrary, the intention is to cover all modifications, alternative constructions, and equivalents falling within the spirit and scope of the application, as defined in the appended claims.
Claims (7)
1. A multi-hop network communication method based on batch sparse codes is characterized by comprising the following steps:
s100, a transmission layer in a source node receives data files from a data source, wherein each data file comprises a plurality of data packets;
s200, when a file sending request is received, a transmission layer in the source node encodes the data packet by using an outer code in batch sparse codes to generate a plurality of batches containing a plurality of encoded packets, and the batches are sent to a network layer;
s300, when the network layer receives the coded packets in the batch, detecting whether the current node is a destination node, if not, performing the step S400, and if so, performing the step S500;
s400, the network layer recodes the coded packets in the same batch, and system recoding is applied in the recoding process, wherein after the network layer receives the coded packets, the received coded packets are used as the recoding packets of the current node, whether the number of the received recoding packets is equal to the number sent by the previous node or not is compared, if not, extra recoding packets are generated by random linear coding and used as the recoding packets in the same batch, the extra recoding packets and the received coded packets are sent to the link layer of the current node, and the recoded packets are sent to the network layer of the next node to return to execute the step S300;
s500, the packet is sent to a transmission layer in a destination node, and the transmission layer in the destination node decodes the received batch to recover the data file.
2. The method according to claim 1, wherein the step S100 further comprises dividing the data file into equal-length data packets.
3. The method of claim 1, further comprising setting the number of re-encoded packets of the batch transmitted by the intermediate node to be the same as the number of re-encoded packets of the source node.
4. The method of claim 1, wherein the batched sparse codes in step S200 comprise an outer code implemented in a transport layer of a node and an inner code implemented in a network layer of the node, wherein the outer code is a matrix fountain code and the inner code comprises a random linear coding on an intermediate node.
5. The method of claim 1, wherein the batch sparse coding in step S200 to generate a plurality of batches further comprises the steps of:
s210, distributing Ψ ═ by contrast (Ψ)1,...,ΨK) Sampling and returning degree di；
S220, uniformly and randomly selecting d from all K input packetsiAn input packet Bi；
S230, forming diFull random matrix G of xMiBatch X generated from multiple input packetsiIs represented by Xi＝BiGi(ii) a Where M is the batch size.
6. The method according to claim 1, wherein the decoding of the received batch in step S500 further comprises decoding based on a belief propagation algorithm, and when the decoding based on the belief propagation algorithm is stopped, marking an undecoded input packet as inactive and replacing it as a decoded packet into the batch to resume the belief propagation algorithm-based decoding process.
7. A batched sparse coding based multi-hop network communication system, comprising:
a data source for sending data files to a source node, wherein each data file comprises a plurality of data packets;
an encoder for encoding the data packets with an outer code at a transport layer in a source node to generate a plurality of batches, the batches comprising a plurality of encoded packets, the batches being sent to a network layer;
a recoder for recoding the multiple encoded packets in the source node and the intermediate node, and applying system recoding in the recoding process, wherein after the network layer receives the encoded packets, the received encoded packets are used as the recoding packets of the current node, whether the number of the received recoding packets is equal to the number sent by the previous node is compared, if not, additional recoding packets are generated by random linear coding and used as the recoding packets in the same batch, and the additional recoding packets and the received encoded packets are sent to the link layer of the current node;
a memory for buffering or storing the encoded packet in each node;
a decoder for decoding the received batch at a transport layer in the destination node to recover the data file.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811256029.0A CN109347604B (en) | 2018-10-26 | 2018-10-26 | Multi-hop network communication method and system based on batched sparse codes |
| PCT/CN2018/115189 WO2020082442A1 (en) | 2018-10-26 | 2018-11-13 | Multihop network communication method and system based on batch sparse codes |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811256029.0A CN109347604B (en) | 2018-10-26 | 2018-10-26 | Multi-hop network communication method and system based on batched sparse codes |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109347604A CN109347604A (en) | 2019-02-15 |
| CN109347604B true CN109347604B (en) | 2021-01-19 |
Family
ID=65311946
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811256029.0A Active CN109347604B (en) | 2018-10-26 | 2018-10-26 | Multi-hop network communication method and system based on batched sparse codes |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109347604B (en) |
| WO (1) | WO2020082442A1 (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110113131B (en) * | 2019-04-24 | 2021-06-15 | 香港中文大学（深圳） | Network communication method and system based on batch coding |
| CN110460407B (en) * | 2019-07-02 | 2022-04-15 | 香港中文大学（深圳） | Communication method, mobile terminal and computer storage medium |
| CN110430011B (en) * | 2019-07-09 | 2020-04-24 | 武汉大学 | BATS code coding method based on regular variable node degree distribution |
| CN110943800B (en) * | 2019-11-06 | 2023-04-28 | 厦门亿联网络技术股份有限公司 | Data packet sending method, device and system, storage medium and electronic device |
| CN113497669B (en) * | 2020-03-20 | 2023-07-11 | 华为技术有限公司 | Method and device for transmitting coded data packet, electronic equipment and storage medium |
| CN114172619A (en) * | 2021-12-08 | 2022-03-11 | 电子科技大学 | Network communication method based on distributed batch sparse codes |
| CN115811381B (en) * | 2022-11-11 | 2024-04-19 | 香港中文大学（深圳） | Network communication method, network communication device, electronic apparatus, and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101208867A (en) * | 2004-09-30 | 2008-06-25 | 飞思卡尔半导体公司 | System and method for ultra wideband communications using multiple code words |
| CN103250463A (en) * | 2010-11-23 | 2013-08-14 | 香港中文大学 | Subset coding for communication systems |
| CN104798317A (en) * | 2012-11-16 | 2015-07-22 | 华为技术有限公司 | Systems and methods for sparse code multiple access |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| ATE470993T1 (en) * | 2005-12-22 | 2010-06-15 | Microsoft Corp | OPTIMIZATIONS FOR NETWORK CODING AND NETWORK DECODING |
| US9503979B2 (en) * | 2012-11-30 | 2016-11-22 | Google Technology Holdings LLC | Delivering data to a wireless station |
| CN107786298B (en) * | 2016-08-25 | 2020-04-28 | 华为技术有限公司 | Communication method and communication device based on opportunistic network coding |
-
2018
- 2018-10-26 CN CN201811256029.0A patent/CN109347604B/en active Active
- 2018-11-13 WO PCT/CN2018/115189 patent/WO2020082442A1/en active Application Filing
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101208867A (en) * | 2004-09-30 | 2008-06-25 | 飞思卡尔半导体公司 | System and method for ultra wideband communications using multiple code words |
| CN103250463A (en) * | 2010-11-23 | 2013-08-14 | 香港中文大学 | Subset coding for communication systems |
| CN104798317A (en) * | 2012-11-16 | 2015-07-22 | 华为技术有限公司 | Systems and methods for sparse code multiple access |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109347604A (en) | 2019-02-15 |
| WO2020082442A1 (en) | 2020-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109347604B (en) | Multi-hop network communication method and system based on batched sparse codes | |
| US8693501B2 (en) | Subset coding for communication systems | |
| EP2357731B1 (en) | Systematic decoding of chain reaction codes | |
| KR102349209B1 (en) | Systems and methods for multi-stage soft input decoding | |
| KR101751497B1 (en) | Apparatus and method using matrix network coding | |
| JP4260804B2 (en) | Retransmission control method and communication apparatus | |
| US10880202B2 (en) | Joint fountain coding and network coding for loss-tolerant information spreading | |
| US11424861B2 (en) | System and technique for sliding window network coding-based packet generation | |
| Yang et al. | BATS Codes: Theory and practice | |
| CN101459430B (en) | Encoding method and apparatus for low density generation matrix code | |
| US10425192B2 (en) | Loss-resilient protocols for communication networks | |
| CN110113131B (en) | Network communication method and system based on batch coding | |
| CN111917512A (en) | High-efficiency coding design method based on BATS code in multi-hop transmission system | |
| CN106998242B (en) | Unequal protection erasure coding method for space communication distributed dynamic network topology | |
| US9116827B2 (en) | System and method for optimizing luby transform (LT) codes to facilitate data transmission over communication network | |
| Rudow et al. | Learning-augmented streaming codes are approximately optimal for variable-size messages | |
| Shi et al. | Zigzag decodable online fountain codes with high intermediate symbol recovery rates | |
| Yang et al. | Multi-hop underwater acoustic networks based on BATS codes | |
| CN110430011B (en) | BATS code coding method based on regular variable node degree distribution | |
| JP2014239285A (en) | Data transmission unit and method | |
| CN110460340B (en) | Self-adaptive construction and decoding method based on random convolutional network error correcting code | |
| CN114172619A (en) | Network communication method based on distributed batch sparse codes | |
| Giri et al. | Performance analysis of feedback-based network-coded systems for broadcast | |
| Hung et al. | Content distribution network for streaming using multiple Galois fields | |
| CN110958078A (en) | Low-delay stream code packet transmission method for high-loss link |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |