Underwater acoustic network security online fountain code transmission method based on component feedback

Abstract

The invention discloses an underwater acoustic network security online fountain code transmission method based on component feedback, relates to the technical field of underwater acoustic network security, and is used for solving the problem of low overall secrecy capability of a system. A sender node, a receiver node and an eavesdropper node are set to transmit data, an underwater data transmission scheme based on an online fountain code is optimized by minimizing a transmission range, shortening transmission time and controlling illegal node decoding three paths, and confidentiality of underwater data transmission is realized by using component feedback and XOR encryption. According to the invention, the data transmission efficiency can be ensured while the underwater data transmission security is realized.

Description

Technical Field

[0001] This invention relates to the field of underwater acoustic network security technology, and more specifically, to a method for transmitting underwater acoustic network fountain codes based on component feedback. Background Technology

[0002] Underwater acoustic networks (UANs) are widely used in marine environmental monitoring, seabed exploration, underwater communication, and underwater robot navigation. Due to their special deployment environment, they often rely on underwater acoustic channels for data transmission. However, underwater acoustic channels have characteristics such as narrow bandwidth, large latency, strong attenuation, and significant multipath effects, which lead to problems such as high bit error rate and unstable communication delay in underwater data communication. At the same time, due to the lack of physical isolation in the underwater communication environment, the data transmission process is susceptible to eavesdropping, decoding, or interference by illegal nodes, which seriously threatens the information security and reliability of the system.

[0003] Currently, especially when the location information of illegal nodes is unknown or the quality of the underwater acoustic channel fluctuates continuously, it is difficult to simultaneously guarantee the security of underwater transmission and the accuracy of decoding. Online fountain codes (OFC), as an efficient error correction coding technique, possesses high-efficiency error correction capabilities and flexible coding characteristics, giving it significant advantages in the application of underwater acoustic networks (UANs). Therefore, this paper proposes a secure online fountain code transmission method for underwater acoustic networks based on component feedback. Summary of the Invention

[0004] This invention provides a network security online fountain code transmission method for underwater acoustic networks based on component feedback. Based on random graph theory and online fountain code technology, it proposes COFCNC to optimize the encoding and feedback mechanism, reducing the probability of unauthorized node eavesdropping while minimizing overhead. Furthermore, the invention's XOR encryption scheme enhances the confidentiality of underwater data transmission, ensuring both security and data transmission efficiency.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for online fountain code transmission over a network based on component feedback in underwater acoustic networks, comprising the following components: Security system model establishment: Construct an underwater point-to-point security system model and set up three functional nodes for encrypted data transmission; Optimization goals: Improve underwater data transmission security by simultaneously minimizing transmission range, shortening transmission time, and controlling unauthorized node decoding. Online fountain code design: Design a component-based, phaseless online fountain code (COFCNC) to optimize coding and feedback mechanisms; Encryption mechanism: Use XOR encryption scheme to encrypt the transmission of COFCNC.

[0006] In a preferred embodiment, when constructing the underwater point-to-point transmission system model, a sender node, a receiver node, and an eavesdropper node are defined and labeled as Alice, Bob, and Eve, respectively. Alice sends encrypted data input by the user, Bob receives the encrypted data, and Eve eavesdrops on and decodes the encrypted data. Alice divides the confidential file into equal-length raw packets, then encodes these raw packets using an encoder, and sends these encoded data packets to Bob until Bob receives enough encoded packets to decode all the raw packets. During this process, Eve attempts to eavesdrop on the channel to obtain the encoded packets Alice sends to Bob. As long as Eve cannot decode the original file before Bob recovers all the raw data packets, the underwater transmission remains secure.

[0007] In a preferred embodiment, let The minimum effective transmission power of the sender. The signal-to-noise ratio threshold for receiving data packets. For the attenuation model of underwater acoustic signals, Given the total noise in the underwater acoustic environment, the transmitter's transmission power can be obtained as follows: When the recipient provides feedback, feedback The sender is given optimized transmission power to ensure that the coded packets sent by the sender can be received by legitimate nodes and that the eavesdropping range of illegitimate nodes is minimized.

[0008] In a preferred embodiment, the sender encoding process is as follows: Step 1. When the sender selects a set of raw packets for encoding. Size ( When a given threshold is met, the sender from The sender randomly selects two original packets, encodes them, and sends the encoded data packets to the receiver. Upon receiving an ACK packet from the receiver, the sender proceeds to step 2; when... At that time, the sender from the set The sender randomly selects an original packet and sends the selected packet to the receiver. After receiving a response packet from the receiver representing a component, the sender proceeds to step 3. Step 2. The sender adjusts the transmission power based on the information in the received ACK packet, and then... The sender randomly selects and sends a 1-degree packet. When it receives feedback from the receiver regarding the component representative, the sender proceeds to step 3. Step 3. The sender retrieves data from the set. Remove all original packages that are not represented by components. Update to the current component representative set, Update to the number represented by the current component, then jump to step 1. Step 4. The sender repeats the above steps until all original packets have been recovered. The receiver's decoding process is as follows: Step 1. Upon receiving a 2-degree encoded packet, the receiver stores the packet in its buffer and establishes an edge between the two corresponding vertices in the decoding graph. Upon receiving a 1-degree packet, the corresponding vertex in the decoding graph turns green, indicating that the original packet has been decoded, and any edges connected to it (if any) are removed. When the decoding graph forms a size of... ( When the largest connected component i-LCC is reached, the receiver jumps to step 2; when When the i-LCC is decoded, the receiver jumps to step 3. When, if received After one degree of encoding, some original packets are still orange, so the receiver jumps to step 3. Step 2. The receiver sends an ACK packet containing the optimal transmission power to the sender and then proceeds to step 1; Step 3. The receiver sends the component representative's feedback to the sender and then proceeds to Step 1; Step 4. The receiver repeats the above steps until all original packets have been recovered. In a preferred embodiment, the process of optimizing the feedback strategy using dual feedback information is as follows: The feedback strategy includes: 1) sending an ACK packet during i-LCC establishment; 2) sending a component-represented feedback packet during i-LCC decoding. First, a small number of coded packets are sent to preempt the channel, and successful preemption is confirmed by receiving feedback information. During the i-th round of COFCNC establishment, the receiver processes the second-degree coded packets. After successful i-LCC establishment, the following operations are performed. 1) The receiver switches to send mode and sends. One ACK packet; 2) Then, the receiver switches to receive mode and sets a timeout period. Before the timeout, it can continue to process the received 2-degree encoded packets. 3) If no 1-degree encoded packet is received, the receiver sends... One ACK packet.

[0009] This loop repeats until a 1-degree encoded packet is received. The receiver then decodes the 1-degree encoded packet. After successfully decoding the i-LCC, the following steps are performed: 1) The receiver switches to send mode and sends. Each component represents a feedback package; 2) Then, the receiver switches to receive mode and sets a timeout, continuing to process the 1-degree encoded packet before the timeout. 3) If no second-degree encoded packet is received, the receiver will send a new packet based on the latest decoding diagram. Each component represents a feedback package.

[0010] This loop repeats until a 2-degree encoded packet is received.

[0011] In a preferred embodiment, the component representative feedback package is configured as follows: We consider a connected component as a unit: an unrecovered original packet. Component representatives are of two types. One type is for connected components of size 1, where the component representative is the unrecovered original packet. The other type is for connected components of size greater than 1. For connected components of size greater than 1, an original packet of degree 1 is randomly selected from that component as the component representative. Only one original packet can serve as the component representative in each connected component. In the feedback packet containing the component representative's state, the state field has... One bit, one bit per original packet. This indicates the component representative status of the i-th original package. The value of the i-th bit represents the status of the i-th original package. A value of "0" indicates that the corresponding original package is either not a component representative or has been restored. A value of "1" indicates that the corresponding original package has not yet been restored and is a component representative. When there is... When there are multiple components, the component represents the status field in the feedback package. The bit is filled with "1". The bits are filled with "0".

[0012] In a preferred embodiment, the XOR encryption process is as follows: The ACK returned by the legitimate receiver contains a binary sequence of the same length as the original packet. After receiving the ACK packet, the sender needs to send a 1-degree data packet. Before that, the sender first processes this random number... and from A randomly selected original packet is XORed. The XOR-encoded packet is then sent to the receiver. The designated legitimate node decodes the received XOR-encoded packet and reconstructs the original packet. If an illegitimate node fails to eavesdrop on the feedback channel from Bob to Alice, it cannot obtain the binary random number. Therefore, it is impossible to decode the 1-degree XOR encoded packet. This invention is a network security online fountain code transmission method based on component feedback for underwater acoustic networks, which has the following technical effects and advantages: This invention constructs an underwater point-to-point security system model, defining sender nodes, receiver nodes, and eavesdropping nodes for data transmission. It optimizes the underwater data transmission scheme based on online fountain codes by simultaneously minimizing transmission range, shortening transmission time, and controlling unauthorized node decoding. Component feedback and XOR encryption are used to achieve confidentiality in underwater data transmission, ensuring both security and efficiency. Simulation experiments demonstrate that the proposed COFCNC exhibits good performance in terms of security, overhead, and maximum buffer usage, making it more suitable for resource-constrained UANs. The secure transmission method of this invention is suitable for underwater network environments with long delays, low bit rates, high bit error rates, and dynamically changing topologies. This invention demonstrates ideal performance and shows promising application prospects. Attached Figure Description

[0013] Figure 1 This is a flowchart illustrating the implementation of an online fountain code transmission method for underwater acoustic networks based on component feedback, as per the present invention.

[0014] Figure 2 This is a system model diagram of an online fountain code transmission method for underwater acoustic networks based on component feedback, according to the present invention.

[0015] Figure 3 This is a decoding diagram of a method for transmitting online fountain codes in underwater acoustic networks based on component feedback, according to the present invention.

[0016] Figure 4 This is a flowchart illustrating the encoding process of an online fountain code transmission method for underwater acoustic networks based on component feedback, as described in this invention.

[0017] Figure 5 This is a flowchart illustrating the decoding process of an online fountain code transmission method for underwater acoustic networks based on component feedback, as described in this invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be understood that the described embodiments are merely illustrative and not limiting of the present invention.

[0019] Example 1: A method for online fountain code transmission over a network based on component feedback in underwater acoustics, such as... Figure 1 As shown, it includes the following steps: Security system model establishment: Construct an underwater point-to-point security system model and set up three functional nodes for encrypted data transmission; Optimization goals: Improve underwater data transmission security by simultaneously minimizing transmission range, shortening transmission time, and controlling unauthorized node decoding. Online fountain code design: Design COFCNC to optimize encoding and feedback mechanisms; Encryption mechanism: Use XOR encryption scheme to encrypt the transmission of COFCNC.

[0020] The specific implementation is as follows: When constructing the underwater point-to-point transmission system model, a sender node, a receiver node, and an eavesdropper node are defined and labeled as Alice, Bob, and Eve, respectively. Figure 2 As shown in the diagram, in this system model, Alice divides the confidential file into equal-length raw packets, then encodes these raw packets using an encoder, and sends these encoded data packets to Bob until Bob receives enough encoded packets to decode all the raw packets. During this process, Eve attempts to eavesdrop on the channel to obtain the encoded packets Alice sends to Bob. As long as Eve cannot decode the original file before Bob recovers all the raw data packets, the underwater transmission remains secure.

[0021] This invention fully considers the underwater acoustic channel model, including underwater acoustic signal attenuation, underwater acoustic signal propagation speed, and underwater acoustic network environmental noise. By controlling the transmission power, it reduces the receiving range of data packets, thereby lowering the probability of being eavesdropped on by illegal nodes; it reduces the overall transmission time, improves communication efficiency, and shortens the communication time between legitimate nodes, thus reducing the time eavesdroppers can spend eavesdropping on data; it optimizes the encoding mechanism, reducing the transmission of encoded packets that help illegal nodes establish connectivity components, thereby lowering the probability of them recovering data packets.

[0022] This invention fully utilizes the feedback characteristics of underwater acoustic channels, dynamically adjusting the coding strategy based on real-time feedback from components, thereby effectively enhancing the security and stability of underwater point-to-point communication while ensuring transmission efficiency. By setting up a three-node structure of sender node, receiver node, and eavesdropper node, combined with a joint design mechanism of XOR encryption and online fountain codes, it becomes difficult for unauthorized nodes to reconstruct the original information before obtaining all data packets, significantly improving anti-eavesdropping capabilities and information confidentiality. Furthermore, the proposed optimization strategy achieves coordinated control over transmission range, transmission time, and the risk of unauthorized decoding, providing a practical solution for secure real-time data transmission in underwater communication environments.

[0023] Different transmission powers result in different reception ranges for encoded data packets. The smaller the reception range, the lower the probability of the encoded data packet being eavesdropped on by an eavesdropping node. (Receiver feedback) Optimizing the transmission power for the sender ensures that legitimate nodes can receive the encoded packets sent by the sender, while minimizing the eavesdropping range for illegitimate nodes. Once the optimization target is set, according to the OFC encoding / decoding mechanism, after an illegal node receives a useful encoded packet, there are two possibilities: i) When it receives a 1-degree encoded packet, it will directly decode the original packet and eliminate any edges connected to it (if any). This increases the decoding rate of the illegal node; ii) It will establish new edges, which reduces the number of connected components in the decoding graph and increases the size of the connected components. In this case, although the decoding rate of the illegal node remains unchanged, the increased size of the connected components will significantly improve its decoding rate once those components are decoded. Therefore, reducing the transmission of 1-degree encoded packets can prevent illegal nodes from decoding the original packet and reduce their decoding rate. Furthermore, repeatedly selecting the same original packet for encoding increases the likelihood of it being leaked to illegal nodes. Therefore, avoiding the same original packet being selected for encoding multiple times reduces the likelihood of illegal nodes establishing edges in the decoding graph, thus slowing down their decoding process. The COFCNC mechanism of this invention eliminates the completion phase, which can significantly reduce the transmission of 1-degree encoded packets. Furthermore, the degree represented by each feedback component is 1, which helps avoid the original packet being encoded multiple times, reduces the likelihood of illegal nodes establishing larger connectivity components, and slows down the decoding process of illegal nodes. The specific steps for setting up COFCNC are as follows: The sender from the set Select the original package. Initialize the collection. The size is In the receiver's decoding diagram, there are An unconnected orange vertex represents An unrecovered original package.

[0024] The sending and receiving encoding process is as follows: Figure 4 As shown, the details are as follows: Step 1. When the sender selects a set of raw packets for encoding. Size ( When a given threshold is met, the sender from The sender randomly selects two original packets, encodes them, and sends the encoded data packets to the receiver. Upon receiving an ACK packet from the receiver, the sender proceeds to step 2; when... At that time, the sender from the set The sender randomly selects an original packet and sends the selected packet to the receiver. After receiving a response packet from the receiver representing a component, the sender proceeds to step 3. Step 2. The sender adjusts the transmission power based on the information in the received ACK packet, and then... The sender randomly selects and sends a 1-degree packet. When it receives feedback from the receiver regarding the component representative, the sender proceeds to step 3. Step 3. The sender retrieves data from the set. Remove all original packages that are not represented by components. Update to the current component representative set, Update to the number represented by the current component, then jump to step 1. Step 4. The sender repeats the above steps until all original packets have been recovered. The receiver's decoding process is as follows Figure 5 As shown, the details are as follows: Step 1. Upon receiving a 2-degree encoded packet, the receiver stores the packet in its buffer and establishes an edge between the two corresponding vertices in the decoding graph. Upon receiving a 1-degree packet, the corresponding vertex in the decoding graph turns green, indicating that the original packet has been decoded, and any edges connected to it (if any) are removed. When the decoding graph forms a size of... ( When the largest connected component i-LCC is reached, the receiver jumps to step 2; when When the i-LCC is decoded, the receiver jumps to step 3. When, if received After one degree of encoding, some original packets are still orange, so the receiver jumps to step 3. Step 2. The receiver sends an ACK packet containing the optimal transmission power to the sender and then proceeds to step 1; Step 3. The receiver sends the component representative's feedback to the sender and then proceeds to Step 1; Step 4. The receiver repeats the above steps until all original packets have been recovered.

[0025] It is important to note that the feedback packet plays a central role in the design and performance optimization of online fountain codes. It allows the sender to dynamically adjust its encoding strategy based on the receiver's decoding status, thereby achieving optimal encoding and low overhead. Implementing effective feedback in real-world UANs is a key factor affecting communication efficiency and reliability, especially in underwater environments with long propagation delays, high bit error rates, and limited bandwidth. Furthermore, underwater acoustic modems typically operate in half-duplex mode, where receiving and transmitting cannot occur simultaneously; an effective feedback mechanism should be able to prevent interference between the sender and receiver.

[0026] This invention proposes two types of feedback information. The first type consists of a simple ACK frame, used to notify the sender whether the i-LCC has been successfully established. An ACK frame contains a 1-bit acknowledgment field. A value of "0" indicates that the i-LCC has not yet been established, and a value of "1" indicates that the i-LCC has been established. The second type of feedback packet is used to report the status of the current component representative. A component representative feedback frame contains a k-bit component representative status field. A "0" in the i-th bit indicates that the i-th original packet is not a component representative, and a "1" in the j-th bit indicates that the j-th original packet is a component representative. Once the status of the decoding graph meets the feedback conditions, the receiving node will send the corresponding feedback information. ACK packets can reduce bandwidth consumption, while component representative feedback packets provide fine-grained information, enabling more precise adjustments to the encoding strategy and accelerating the overall decoding process. The combined use of the two feedback types can reduce bandwidth consumption, improve communication efficiency, and enhance adaptability to underwater acoustic environments.

[0027] The COFCNC feedback strategy includes: 1) sending an ACK packet during i-LCC establishment; 2) sending a component representative feedback packet during i-LCC decoding. Let... and These represent the error rates of the ACK packet and the component-represented feedback packet, respectively. In actual underwater transmission, a small number of coded packets should first be sent to preempt the channel, and the successful preemption is confirmed by receiving feedback information. During the i-th round of COFCNC establishment, the receiver processes the 2-degree coded packets. After successful i-LCC establishment, the following operations will be performed. 1) The receiver switches to send mode and sends. One ACK packet; 2) Then, the receiver switches to receive mode and sets a timeout period. Before the timeout, it can continue to process the received 2-degree encoded packets. 3) If no 1-degree encoded packet is received, the receiver sends... One ACK packet.

[0028] This loop repeats until a 1-degree encoded packet is received. The receiver then decodes the 1-degree encoded packet. After successfully decoding the i-LCC, the following steps are performed: 1) The receiver switches to send mode and sends. Each component represents a feedback package; 2) Then, the receiver switches to receive mode and sets a timeout, continuing to process the 1-degree encoded packet before the timeout. 3) If no second-degree encoded packet is received, the receiver will send a new packet based on the latest decoding diagram. Each component represents a feedback package.

[0029] This loop repeats until a 2-degree encoded packet is received.

[0030] The feedback mechanism of this invention allows the receiver to process the received encoded packets before the timeout, thus maintaining feedback robustness while avoiding retransmission of encoded packets and feedback packets.

[0031] This invention considers optimizing the feedback strategy to reduce redundant data packets and loops, and updates the set by removing decoded original packets. To reduce the number of sets The number of original packets belonging to the same connected component. The component representing the feedback packet is defined as follows: First, we treat a connected component as a unit: an unrecovered original packet, which can significantly narrow down the selection of original packets for encoding and avoid selecting original packets from the same connected component.

[0032] It should be noted that the range of raw packets used for encoding differs under different mechanisms. As shown in a decoding diagram in Figure 3, the range of raw packets used for encoding in the traditional OFC mechanism is all... ( ( ) original packets. However, the range of original packets selected for encoding in this invention is... This can significantly narrow down the range of raw packets to be encoded, increase the probability of using useful encoded packets, and avoid the possibility of selecting decoded raw packets or packets in the same connected component, thereby avoiding redundant data packets and loops.

[0033] Component representatives are of two types. One type is for connected components of size 1, where the representative is the unrecovered original packet. The other type is for connected components of size greater than 1. For connected components of size greater than 1, an original packet with degree 1 is randomly selected from that component as the component representative. Only one original packet can be used as the component representative for each connected component.

[0034] In the feedback package containing the component's representative state, the state fields are: One bit, one bit per original packet. This indicates the component representative status of the i-th original package. The value of the i-th bit represents the status of the i-th original package. A value of "0" indicates that the corresponding original package is either not a component representative or has been restored. A value of "1" indicates that the corresponding original package has not yet been restored and is a component representative. When there is... When a component is involved, the status field in the feedback data packet... The bit is filled with "1". The bits are filled with "0". The specific steps for encrypting the transmission of online fountain codes using the XOR encryption scheme are as follows: The ACK returned by the legitimate receiver contains a binary sequence of the same length as the original packet. After receiving the ACK packet, the sender needs to send a 1-degree data packet. Before that, the sender first processes this random number... and from A randomly selected original packet is XORed. The XOR-encoded packet is then sent to the receiver. A designated legitimate node decodes the received XOR-encoded packet and reconstructs the original packet. If an illegitimate node fails to eavesdrop on the feedback channel from Bob to Alice, it cannot obtain the binary random number. Therefore, it is impossible to decode the 1-degree XOR encoded packet.

Claims

1. A method for online fountain code transmission over a network based on component feedback in underwater acoustics, characterized in that: The design of component-based online fountain code without completion phase (COFCNC) eliminates the completion phase of traditional online fountain codes, significantly reducing the transmission of 1-degree packets. The optimized feedback mechanism avoids the original packet being encoded multiple times, reducing the probability of illegal nodes decoding and establishing large connected components, and slowing down the decoding process of illegal nodes.

2. The method for online fountain code transmission over a network based on component feedback for underwater acoustic networks according to claim 1, characterized in that: The COFCNC encoding / decoding process is as follows: The sender's encoding process is as follows: Step 1. When the sender selects a set of raw packets for encoding. Size ( When a given threshold is met, the sender from The sender randomly selects two original packets, encodes them, and sends the encoded data packets to the receiver; after receiving an ACK packet from the receiver, the sender proceeds to step 2; when At that time, the sender from the set The sender randomly selects an original packet and sends the selected data packet to the receiver; after receiving the component representative feedback packet from the receiver, the sender jumps to step 3. Step 2. The sender adjusts the transmission power based on the information in the received ACK packet, and then... The sender randomly selects and sends a 1-degree packet; when it receives a component representative from the receiver, it jumps to step 3. Step 3. The sender retrieves data from the set. Remove all original packages that are not represented by components. Update to the current component representative set, Update to the number represented by the current component, then jump to step 1; Step 4. The sender repeats the above steps until all original packets have been recovered; The receiver's decoding process is as follows: Step 1. Upon receiving a 2-degree encoded packet, the receiver stores the packet in its buffer and establishes an edge between the two corresponding vertices in the decoding graph; upon receiving a 1-degree packet, the corresponding vertex in the decoding graph turns green, indicating that the original packet has been decoded, and any edges connected to it (if any) are removed; when a size of [missing information] is formed in the decoding graph... ( When the largest connected component i-LCC is reached, the receiver jumps to step 2; when When the i-LCC is decoded, the receiver jumps to step 3; when When, if received After a 1-degree encoded packet, some original packets are still orange, so the receiver jumps to step 3; Step 2. The receiver sends an ACK packet containing the optimal transmission power to the sender and then proceeds to step 1; Step 3. The receiver sends the component representative's feedback to the sender and then proceeds to Step 1; Step 4. The receiver repeats the above steps until all original packets have been recovered.

3. The method for online fountain code transmission over a network based on component feedback for underwater acoustic networks according to claim 2, characterized in that: The process of optimizing the feedback strategy using dual feedback information is as follows: The feedback strategy includes: 1) sending an ACK packet when i-LCC is established; 2) sending a component representing a feedback packet when i-LCC is decoded; first, a small number of coded packets are sent to preempt the channel, and the successful preemption is confirmed by receiving feedback information; in the i-th round of COFCNC establishment, the receiver processes the 2-degree coded packets; after i-LCC is successfully established, the following operations are performed; 1) The receiver switches to send mode and sends. One ACK packet; 2) Then, the receiver switches to receive mode and sets a timeout period. Before the timeout, it can continue to process the received 2-degree encoded packets. 3) If no 1-degree encoded packet is received, the receiver sends... One ACK packet; This loop repeats until a 1-degree encoded packet is received; then, the receiver decodes the 1-degree encoded packet; after successfully decoding the i-LCC, the following operations are performed: 1) The receiver switches to send mode and sends. Each component represents a feedback package; 2) Then, the receiver switches to receive mode and sets a timeout, continuing to process the 1-degree encoded packet before the timeout. 3) If no second-degree encoded packet is received, the receiver will send a new packet based on the latest decoding diagram. Each component represents a feedback package; This loop repeats until a 2-degree encoded packet is received.

4. The method for online fountain code transmission in underwater acoustic networks based on component feedback according to claim 3, characterized in that: The component representative feedback package is configured as follows: We consider a connected component as a unit: an unrecovered original packet; there are two types of component representatives: one is for connected components of size 1, where the component representative is the unrecovered original packet; the other is for connected components of size greater than 1; for connected components of size greater than 1, an original packet of degree 1 is randomly selected in that component as the component representative; only one original packet can be used as the component representative in each connected component. In the feedback package containing the component's representative state, the state fields are: One bit, one bit per original packet; This indicates the component representative status of the i-th original package; the value of the i-th bit indicates the status of the i-th original package; a value of "0" indicates that the corresponding original package is either not a component representative or has been restored; a value of "1" indicates that the corresponding original package has not been restored and is a component representative; when there is... When there are multiple components, the component represents the status field in the feedback package. The bit is filled with "1". The bits are filled with "0".

Images