Method and system for protecting position privacy of source node of underwater acoustic network

Abstract

The invention discloses an underwater acoustic network source node position privacy protection method and system, and belongs to the field of ocean development. Aiming at the problems of poor privacy protection strength and unstable operation of the existing underwater acoustic network source node, the invention provides an underwater acoustic network source node position privacy protection method, which comprises the following steps that: a source node sends original data to a target node hop by hop through a source node data distribution strategy; wherein the source node data distribution strategy is that original data in a source node is divided into a plurality of data fragments, each data fragment is distributed to one candidate source node in a plurality of candidate source nodes, and the original data reaches a target node through different and scattered paths. According to the method, the data fragments are dispersed in the whole network, and different data fragments are transmitted to the target node through paths of different candidate source nodes, so that the search range is expanded, the difficulty of identifying the source node is increased, and the security of the location privacy of the source node is improved; and the overall operation stability is high.

Description

Technical Field

[0001] The present invention belongs to the technical field of ocean development, and more specifically, relates to a method and system for protecting the location privacy of source nodes in an underwater acoustic network. Background Art

[0002] The ocean contains rich economic resources and energy resources. Reasonable exploration and development of ocean resources can benefit all mankind. In the 21st century, in order to seek development and break through resource barriers, countries regard ocean construction as one of the most important national strategies. To better develop and utilize ocean resources, underwater acoustic networks (UANs) have emerged as the times require. UANs have great practical application value in the fields of ocean resource exploration, ocean environmental monitoring, national defense security, ocean disaster monitoring and early warning, etc. In ocean resource exploration, UANs can explore the location information of resources such as oil and natural gas on the seabed, facilitating subsequent mining work. In ocean environmental monitoring, UANs can monitor parameters such as temperature, salinity, and pH in the ocean in real time, helping humans to promptly discover events such as ocean pollution and algal blooms, so as to quickly take countermeasures. In the field of national defense security, by deploying underwater sensor nodes in the ocean to construct a sensitive underwater acoustic sensor monitoring network, military equipment such as submarines and torpedoes can be discovered in a timely manner, and the location information of these equipment can be effectively submitted, thus protecting national defense security. In ocean disaster monitoring and early warning, UANs provide early warning information in a timely manner by monitoring ocean disasters such as tsunamis and ocean earthquakes, ensuring the safety of human life and property. Generally speaking, UANs provide new information for various fields through real-time, remote, and multi-parameter data collection, providing strong support for decision-makers, scientists, and national defense agencies, and promoting the sustainable development and security of the ocean field.

[0003] The research of UANs has entered a new stage. UANs are usually deployed in open and unmonitored areas, so they are vulnerable to various attacks, including active attacks and passive attacks. For data-driven UANs, once the source node is attacked, the entire network may be paralyzed. Therefore, some researchers have begun to pay attention to the security issue of the location privacy of source nodes in UANs. Due to the limitations of low bandwidth, high propagation delay, limited energy, high bit error rate, and node mobility in data communication in UANs, the current research on the location privacy of source nodes in UANs is still in its infancy.

[0004] For example, Chinese Patent Application No. CN202211492947.X, with a publication date of March 17, 2023, discloses a game-based source node location privacy protection method in an underwater acoustic sensor network. Its steps include: First, combining the Ekman drift model, dividing the ocean hierarchical structure and deploying corresponding underwater nodes and source nodes; Second, to ensure the timeliness of source data, the shortest route is selected as the data forwarding strategy. Considering the hole problem caused by insufficient remaining energy of nodes, an energy threshold is considered in the shortest route. Finally, considering a hybrid attack-type attacker, that is, the attacker simultaneously uses wormhole attacks and listening attacks, and traces back hop by hop to find the possible location of the source node. Combining game theory strategies and location privacy protection technologies, while ensuring the transmission of source data, reducing and quantifying the probability that the attacker finds the location of the source node. However, the deficiency of this patent is that this method relies on the Ekman drift model to divide the ocean hierarchical structure. However, in practical applications, the ocean currents, temperature gradients, and seabed topographies in different sea areas are significantly different, and the applicability is poor; and selecting the shortest route as the data forwarding strategy enables the attacker to use the deterministic characteristics of the shortest route for trajectory tracking. Especially in the static layer, the fixed position of the nodes may weaken the dynamic advantages of the game strategy, and the protection of the location privacy of the source node is less than satisfactory.

[0005] Another example is Chinese Patent Application No. CN201811637370.0, with a publication date of May 28, 2019. This patent discloses a source node location privacy protection method based on Sink and grid in WSN. The traditional scheme has the disadvantage of excessive communication overhead caused by using source node flooding to determine phantom nodes. The method of the present invention first performs network initialization and pre-deploys the nodes in the network. Then the source node receives a set of phantom nodes. If the first goal is low energy consumption, a single phantom node scheme is used. If the first goal is high security, a double phantom node scheme is used. According to different schemes, phantom nodes are selected and routing transmission is performed. The deficiency of this patent is that due to the limitations such as low bandwidth, high propagation delay, limited energy, high bit error rate, and node mobility in data communication in UANs, the source node location privacy protection method in WSNs is not applicable when facing UANs. Summary of the Invention

[0006] 1. Problems to be Solved

[0007] Aiming at the problems of poor privacy protection of the source node in the existing underwater acoustic network and unstable operation, the present invention provides an underwater acoustic network source node location privacy protection method and system. By dispersing data fragments throughout the network, different data fragments are transmitted to the target node through the paths of different candidate source nodes, expanding the search range, increasing the difficulty of identifying the source node, and then improving the security of the source node location privacy; and the overall operation stability is high.

[0008] 2. Technical Solution

[0009] To solve the above problems, the present invention adopts the following technical solution.

[0010] A method for protecting the location privacy of source nodes in an underwater acoustic network, where the source node sends the original data hop by hop to the target node through the source node data distribution strategy; wherein, the source node data distribution strategy is as follows:

[0011] The original data in the source node is divided into several data segments, and then each data segment is assigned to one candidate source node among several candidate source nodes, so that different data segments correspond to different candidate source nodes, realizing that the original data reaches the target node through different and scattered paths.

[0012] By adopting the above technical solution, through establishing the source node data distribution strategy, the original data in the source node is divided into different data segments, and the data segments are randomly assigned to different candidate source nodes, so that the data segments are scattered throughout the network, and different data segments will be transmitted to the target node through different paths; effectively avoiding the leakage of the source node location privacy caused by the attacker eavesdropping on a large number of data segments from the same direction in a relatively concentrated area; at the same time, expanding the attacker's search range, increasing the difficulty for the attacker to identify the source node, and then improving the security of the source node location privacy;

[0013] At the same time, due to constructing different and scattered diversified transmission paths, when resisting the attacks of attackers, it can not only resist the attacks of a single attacker, but also resist the simultaneous attacks of multiple attackers, further highlighting that this method has a strong ability to protect the location privacy of source nodes and realizing the stable operation of the underwater acoustic network.

[0014] Furthermore, several candidate source nodes are determined in the following way:

[0015] Establish a preliminary candidate source node pool: all nodes that meet the requirement of having a hierarchical level less than the two-hop neighbor nodes of the source node are used to establish a preliminary candidate source node pool;

[0016] Select a reference node: Select the node with the largest remaining energy in the preliminary candidate source node pool as the reference node, and remove the reference node from the preliminary candidate source node pool to obtain the remaining candidate source node pool;

[0017] Select candidate source nodes: Select nodes one by one from the remaining candidate source node pool as candidate source nodes. The selected candidate source nodes enter the selected candidate source node pool until the number of data segments is the same as the number of candidate source nodes in the selected candidate source node pool, and then stop selecting candidate source nodes. Among them, only when the sum of the distances between the currently selected node and each candidate source node in the selected candidate source node pool is the largest can it be used as a candidate source node and enter the selected candidate source node pool.

[0018] With the above technical solution, several data segments need several corresponding candidate source nodes for transmission. By considering the remaining energy of the candidate source nodes and the distance between the candidate source nodes, the candidate source nodes are selected to ensure that the distance between two different candidate source nodes is far enough to achieve a balance between balanced energy consumption and protecting the location privacy of the source nodes.

[0019] Furthermore, the source node maps the original data into several data segments based on the sharing mechanism of the Chinese Remainder Theorem. Specifically, it includes the following steps:

[0020] Pre - deployment: Set a group of integers (d1, d2, … d n ), and this group of integers satisfies: d1 < d2 <... < d n ; gcd(d i , d j ) = 1, i ≠ j; D = d1 × d2 ×... × d t , D1 = d n-t+2 × d n-t+3 ×... × d n , D > M > D1, where M is the original data.

[0021] Data segmentation: Divide the original data M into n data segments, and the data segments are represented as (d i , m i ), and the specific formula is as follows:

[0022]

[0023] Data recombination: After the source node sends n data segments, as long as the target node successfully receives any t data segments among the n data segments, it can recombine the original data M.

[0024] With the above technical solution, the original data is mapped into several data fragments through a sharing mechanism based on the Chinese Remainder Theorem, which has a low algorithm complexity. At the same time, this mechanism has good robustness to packet loss, thereby improving the reliability of data transmission in an underwater acoustic channel with a high bit error rate. Even if some data fragments are lost during the transmission process, as long as the target node receives more than t data fragments, it can successfully reconstruct the original data, greatly reducing the possibility of packet loss.

[0025] Furthermore, it also includes setting a sleep mechanism during the data transmission process from the source node to the target node: when the original data in the source node is sent to the receiving node through the sending node, the sending node and its high-level neighbor nodes are set to the sleep state; the sleep state means not participating in data transmission.

[0026] With the above technical solution, a dangerous node sleep mechanism is introduced. After the sending node sends a data packet, the sending node and its high-level neighbor nodes are set to the sleep state and no longer participate in data transmission, effectively preventing attackers from moving towards the source node by eavesdropping on consecutive data packets, thereby effectively resisting patient attacks or multi-attacker collaborative attacks, preventing the leakage of the source node's location privacy, and further improving the security of the source node's location privacy.

[0027] Furthermore, before the sending node sends data fragments, it also includes evaluating the states of all neighbor nodes of the sending node: when the receiving nodes among all neighbor nodes are in the idle state and the remaining neighbor nodes are not in the receiving state; the sending node will send data fragments.

[0028] With the above technical solution, by evaluating the states of neighbor nodes before the sending node sends data fragments, the sending node can send data packets to the receiving node only when the receiving node is in the idle state; in addition, to avoid interfering with the data reception of other neighbor nodes, the sending node can send data packets to the receiving node only when other neighbor nodes are not in the receiving state, reasonably scheduling the transmission of data packets to avoid packet collisions to the greatest extent, thereby minimizing packet loss caused by channel contention and packet collisions, and ensuring the reliability and robustness of data transmission;

[0029] Furthermore, it also includes designing a network model and an attack model for the underwater acoustic network. The network model adopts a three-dimensional model, underwater sensor nodes are randomly deployed in the underwater environment, and the target node is deployed on the water surface; underwater sensor nodes have the same functions and parameters; the attack model uses the method of multiple attackers for attacks.

[0030] With the above technical solution, since the underwater acoustic network is deployed in a three-dimensional underwater environment, a three-dimensional model is established, which is more in line with the actual application environment; and the attack model uses the method of multiple attackers to attack, which can approach the position of the source node more efficiently, thus accelerating the speed of the attacker to find the source node.

[0031] Furthermore, the specific attack process of the attack model is as follows:

[0032] Step 1: Randomly select one attacker from multiple attackers as the initial leading attacker, and the other attackers become following attackers;

[0033] Step 2: Assign positions to the following attackers;

[0034] Step 3: The following attackers collect the traffic information at their own positions and transmit the traffic information to the initial leading attacker;

[0035] Step 4: The initial leading attacker designates the following attacker that has collected the most traffic information as the new leading attacker, and the area where the new leading attacker is located becomes the hot spot area. The previous initial leading attacker and the remaining following attackers become the new following attackers;

[0036] Step 5: Based on the position of the new leading attacker, repeat Steps 2 to 4 until the position of the source node is found.

[0037] With the above technical solution, compared with determining the source node position by hop-by-hop backtracking of data packets, the multi-attacker cooperation strategy based on traffic analysis can find the source node position more efficiently and accurately, and improve the attack power of the attack model.

[0038] Furthermore, the position assignment in Step 2 specifically includes the following steps:

[0039] Step 21: Establish a three-dimensional coordinate system with the target node as the coordinate origin;

[0040] Step 22: Construct a sphere with the position of the leading attacker as the center of the sphere and the communication radius of the leading attacker as the radius of the sphere;

[0041] Step 23: Inscribe a regular polygon prism in the sphere. There are several contact points between the regular polygon prism and the sphere; the number of sides of the polygon in the regular polygon prism is one less than the number of attackers;

[0042] Step 24: Assign the following attackers one by one to the contact points where the bottom surface of the regular polygon prism contacts the sphere.

[0043] Further, before the source node sends data, it also includes network initialization: the destination node periodically broadcasts Hello packets so that each node in the network can obtain its level and a neighbor table containing two-hop neighbors; the neighbor table contains the node ID, level, location, remaining energy, and the status of neighbor nodes.

[0044] Adopting the above technical solution, each node can know the status of neighbor nodes and its own status attributes, simplify routing decisions, reduce redundant broadcasts, and reduce energy consumption; the periodic broadcast can reflect node movement, failure, or addition in real time, adapt to the dynamic changes in the underwater environment, and enhance the reliability and applicability of the entire process.

[0045] A system using the underwater acoustic network source node location privacy protection method described in any one of the above, includes:

[0046] Source node: used to generate original data;

[0047] SSP-MAC protocol module: used to send the original data in the source node to the destination node through the source node data distribution strategy;

[0048] Destination node: used to receive data.

[0049] Adopting the above technical solution, by splitting the original data into several data segments and correspondingly selecting different candidate source nodes for multi-path transmission to the destination node, the search range of the attacker is expanded, and the difficulty for the attacker to find the source node is greatly increased, thus ensuring the security of the source node location privacy. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a schematic diagram of the network model of the underwater acoustic network in the present invention;

[0051] Figure 2 It is a schematic diagram of the location distribution of the attacker in the attack model of the present invention;

[0052] Figure 3 It is a schematic diagram of the process of the present invention;

[0053] Figure 4 It is a schematic diagram of splitting data segments in the present invention;

[0054] Figure 5 It is a schematic diagram of the node network topology in the present invention;

[0055] Figure 6 It is a schematic diagram of the data transmission process in the present invention;

[0056] Figure 7 It is a schematic diagram of the relationship between the sleep duration and the safe time after adopting the method of the present invention;

[0057] Figure 8 Schematic diagram of the relationship between sleep duration and end - to - end delay after adopting the method of the present invention;

[0058] Figure 9 Schematic diagram of the relationship between depth and safety time under different n values;

[0059] Figure 10 Schematic diagram of the relationship between depth and end - to - end delay under different n values;

[0060] Figure 11 Schematic diagram of the relationship between depth and energy consumption fairness index under different n values;

[0061] Figure 12 Schematic diagram of the relationship between the number of nodes adopting different methods and safety time;

[0062] Figure 13 Schematic diagram of the relationship between the number of nodes adopting different methods and packet delivery ratio;

[0063] Figure 14 Schematic diagram of the relationship between the number of nodes adopting different methods and energy consumption fairness index. Detailed implementation manners

[0064] The present invention will be further described below in conjunction with specific embodiments and the accompanying drawings.

[0065] Before elaborating on the solution of this application, a brief introduction to UANs is given: The source nodes are usually deployed at the bottom of the ocean or lake, while the Sink nodes are deployed on the water surface. Therefore, data is transmitted from bottom to top. The attackers are initially located around the Sink nodes and aim to determine the location of the source nodes. Therefore, their attacks are carried out from top to bottom. At the same time, some terms are explained as follows:

[0066] Target node: That is, the Sink node, which is usually deployed on the water surface and is the convergence center of network data;

[0067] Source node: The node that produces the original data, and this original data needs to be transmitted to the Sink node;

[0068] Neighbor node: The node within the communication range of the current node and can communicate directly with the current node;

[0069] Node: Refers to a node that has a reachable path to the Sink node in the current network topology;

[0070] Sending node: In a specific data transmission process, the node currently responsible for forwarding data to other nodes;

[0071] Receiving node: In the process of transmission, the node that currently receives data from the sending node.

[0072] Given the limitations such as low bandwidth, high propagation delay, limited energy, high bit error rate, and node mobility in data communication in UANs, the current research on the location privacy of source nodes in UANs is still in its infancy. The location privacy of source nodes in UANs is of great significance for national underwater battlefield monitoring and the monitoring of key underwater protected areas. Therefore, the applicant provides a method for protecting the location privacy of source nodes in an underwater acoustic network, which specifically includes: the source node sends the original data hop by hop to the target node through the source node data distribution strategy; among them, the source node data distribution strategy is:

[0073] The original data in the source node is divided into several data segments, and then each data segment is assigned to one of several candidate source nodes, so that different data segments correspond to different candidate source nodes, realizing that the original data reaches the target node through different and scattered paths. Specifically, in this embodiment, each data segment is assigned to different candidate source nodes through random permutation, that is, for the divided data segments and an equal number of candidate source nodes, first the order of the candidate source nodes is randomly permuted, and then according to the permuted order, the data segments are sequentially and uniquely assigned to each candidate source node. Since random permutation assignment is relatively common in the prior art and does not belong to the core improvement point of this application, this application will not elaborate on it in detail.

[0074] In this embodiment, by establishing the source node data distribution strategy, the original data in the source node is divided into different data segments, and the data segments are randomly assigned to different candidate source nodes, so that the data segments are scattered throughout the network, and different data segments will be transmitted to the target node through different paths; effectively avoiding the leakage of the source node location privacy caused by the attacker eavesdropping on a large number of data segments from the same direction in a relatively concentrated area; at the same time, expanding the attacker's search range, increasing the difficulty for the attacker to identify the source node, and then improving the security of the source node location privacy;

[0075] At the same time, due to the construction of different and scattered diversified transmission paths, when resisting the attack of the attacker, it can not only resist the attack of a single attacker, but also resist the simultaneous attack of multiple attackers, further highlighting that this method has a strong ability to protect the location privacy of source nodes and realizing the stable operation of the underwater acoustic network.

[0076] In a specific embodiment, several candidate source nodes are determined in the following manner:

[0077] Step 1: Establish a preliminary candidate source node pool: Establish a preliminary candidate source node pool for all nodes that meet the requirement of being two-hop neighbor nodes with a level less than that of the source node; as Figure 4 shown, in an underwater acoustic network, each node is assigned a level. After the source node and its level are determined, for example, if the source node is in the sixth layer Figure 4 the green node in, then the nodes in the candidate source node pool are the neighbor nodes that are two hops away from the source node in the sixth layer, such as Figure 4 the red nodes in; Figure 4 the transmission direction of the message segment in is the transmission direction of the data segment;

[0078] Step 2: Select a reference node: Select the node with the largest remaining energy in the preliminary candidate source node pool as the reference node, and remove the reference node from the preliminary candidate source node pool to obtain the remaining candidate source node pool;

[0079] Step 3: Select candidate source nodes: Select nodes one by one in the remaining candidate source node pool as candidate source nodes. The selected candidate source nodes enter the selected candidate source node pool, and stop selecting candidate source nodes until the number of data segments is the same as the number of candidate source nodes in the selected candidate source node pool; among them, the currently selected node can be used as a candidate source node to enter the selected candidate source node pool only when the sum of the distances between it and each candidate source node in the selected candidate source node pool is the largest.

[0080] Specifically, that is to say, in this embodiment, in order to protect the location privacy of the source node by dispersing the data segments in the transmission network, a mechanism needs to be proposed for selecting candidate source nodes; specifically, a decentralized candidate source node selection algorithm can be used to implement the above method. The decentralized candidate source node selection algorithm takes into account the remaining energy of the candidate source nodes and the distances between the selected candidate source nodes: First, select the candidate source node with the largest remaining energy as the reference node. On this basis, select candidate source nodes one by one. The newly selected candidate source node needs to meet the requirement that the cumulative distance between it and the selected candidate source nodes is the largest, which can ensure that the distances between two different candidate source nodes are far enough. The decentralized candidate source node selection algorithm is shown in Table 1 below:

[0081] Table 1 Decentralized candidate source node selection algorithm

[0082]

[0083] Specifically, the source node selects candidate source nodes equal in number to the number of data segments according to the decentralized candidate source node selection algorithm, and randomly assigns each data segment to a different candidate source node. Therefore, these information segments can be dispersed throughout the network, and different information segments will be transmitted to the Sink node through different paths, expanding the search scope of the attacker and thus enhancing the security of the source node location privacy. In addition, this algorithm preferentially selects the candidate source node with the largest remaining energy as the reference node and takes into account the distance between candidate source nodes on this basis to achieve a balance between balancing energy consumption and protecting the source node location privacy.

[0084] In a specific embodiment, the source node maps the original data into a number of data segments based on the sharing mechanism of the Chinese Remainder Theorem; specifically, it includes the following steps:

[0085] Pre-deployment: Set a group of integers (d1, d2,... d n ), which are used for the segmentation of the original data, and this group of integers satisfies: d1 < d2 <... < d n ; gcd(d i , d j ) = 1, i ≠ j; D = d1 × d2 × … × d t , D1 = d n-t+2 × d n-t+3 ×... × d n , D > M > D1, where M is the original data;

[0086] Data segmentation: Segment the original data M into n data segments, and the data segments are represented as (d i , m i ), and the specific formula is as follows:

[0087]

[0088] Data recombination: After the source node sends n data segments, as long as the target node successfully receives any t data segments among the n data segments, it can recombine the original data M; the specific recombination method is as follows: The t information segments received by the Sink node can be expressed as Construct a system of modular congruence equations, where each equation is used to calculate the remainder of m i modulo d i , and the solution x of this system of congruence equations satisfies all congruence equations. Subsequently, according to Formula 1 and Formula 2, the original information M is recombined, and the system of modular congruence equations is as follows:

[0089]

[0090] Formula 1 is: Formula 2 is: M ≡ x (mod D2).

[0091] Specifically, in this embodiment, a sharing mechanism based on the Chinese Remainder Theorem is adopted to map the original data into several data segments. First, compared with the classical Shamir secret sharing scheme, the sharing mechanism method based on the Chinese Remainder Theorem has a lower algorithm complexity. Second, this sharing mechanism method has good robustness to packet loss, thereby improving the reliability of data transmission in an underwater acoustic channel with a high bit error rate. Even if some information segments are lost during the transmission process, as long as the Sink node receives more than t data segments, it can successfully reconstruct the original data.

[0092] In a specific embodiment, it further includes setting a sleep mechanism during the data transmission process from the source node to the target node: after the original data in the source node is sent to the receiving node through the sending node, the sending node and its high-level neighbor nodes are set to the sleep state; the sleep state means not participating in data transmission.

[0093] Specifically, in this embodiment, a sleep mechanism is introduced to prevent the sending node and its high-level neighbor nodes from participating in packet forwarding, thereby avoiding the risk of revealing the source node location. At the same time, in order to improve the performance of the entire underwater acoustic network, the sleep duration of the sending node and its high-level neighbor nodes is determined according to the network scale. It should be noted here that the nodes set to the sleep state are called dangerous nodes, and the selection of dangerous nodes is very important. It does not arbitrarily select nodes as dangerous nodes, but is considered through multiple factors, as follows:

[0094] After the underwater acoustic network is initialized, each node connected to the Sink node will be configured with a level and record the information of neighbor nodes in the neighbor table; as Figure 5 shown, the one-hop neighbor table of node 5 is shown in Table 2. To protect the privacy of the source node location, the nodes in the network may be in the following four states: sending state: the neighbor node is sending a packet to other nodes; receiving state: the neighbor node is receiving a packet from other nodes; idle state: at this time, the neighbor node does not participate in any transmission; sleep state: the neighbor node neither participates in transmission nor listens to the channel;

[0095] Table 2 One-hop neighbor table of node 5

[0096]

[0097] Suppose node 5 is the sending node, and nodes 1 and 2 are both lower-level neighbor nodes of node 5. Therefore, the probability that either node 1 or node 2 is selected as the receiving node is high; the neighbor nodes at the same level as node 5 are nodes 3 and 4, and the probability that they are selected as the receiving node is lower; the higher-level neighbor nodes are nodes 6 and 7, and they cannot become the receiving nodes. In the attack model with multi-attacker cooperation, the attackers will gradually move towards the hot spot area during the attack. Suppose the attackers can monitor the traffic from node 5. Then, according to the top-down attack pattern of the attackers, it can be determined that one of the attackers is located at node 1 or node 2. After node 5 sends a data packet, there are the following two situations for the attack progress of the attackers: First, the attackers cannot determine the new hot spot area based on the detected traffic. At this time, if node 5 continues to participate in data forwarding, it will speed up the attackers' positioning of the new hot spot area (node 5); Second, the attackers determine the new hot spot area (node 5) based on the detected traffic and move to that area. At this time, if the higher-level neighbor nodes of node 5 (such as node 6 or node 7) continue to forward data packets, the attackers will soon discover the new hot spot area and get closer to the source node. Therefore, both the sending node and its higher-level neighbor nodes are regarded as dangerous nodes. To resist passive attacks, after the sending node sends a data packet, the dangerous nodes will enter the sleep state, which can prevent the attackers from moving towards the source node by eavesdropping on consecutive data packets, thus effectively resisting patient attacks or attacks with multi-attacker cooperation.

[0098] In a specific implementation, before the sending node sends a data segment, it also includes evaluating the states of all neighbor nodes of the sending node: when the receiving nodes among all neighbor nodes are in the idle state and the remaining neighbor nodes are not in the receiving state; the sending node will send the data segment.

[0099] Specifically, in this embodiment, the sending node determines the status of neighbor nodes to forward data fragments, which can effectively avoid data packet collisions. Among them, each node updates the status of neighbor nodes in the neighbor table in real time by listening to the packets transmitted in the channel. Specifically, during the process of the sending node using the handshake mechanism to transmit data packets to the receiving node, after hearing a frame, the node updates the status of neighbor nodes according to the algorithm. When the node hears an xRTS from another neighbor node, it sets the status of this neighbor node in its neighbor table to the sending state. xRTS represents an RTS frame sent by the sending node to other nodes rather than the current node. The explanations of xCTS and xACK are similar to that of xRTS. When a node hears an xCTS, it sets the status of this neighbor node in its neighbor table to the receiving state. After receiving the ACK from the receiving node, the sending node sends a DAN frame, and the neighbor nodes that receive the DAN frame change the status of the sending node in their neighbor tables to the sleep state. At the same time, after receiving the DAN from the sending node, the high-level neighbor nodes of the sending node forward the DAN to their respective neighbor nodes to notify other nodes that the lower-layer neighbor nodes of the sending node enter the sleep state.

[0100] That is to say, to avoid collisions at the receiving node, the sending node can send data packets to the receiving node only when the receiving node is in the idle state. In addition, to avoid interfering with the data reception being carried out by other neighbor nodes, the sending node can send data packets to the receiving node only when other neighbor nodes are not in the receiving state. Before transmitting data packets, a pair of handshake frames (RTS / CTS) need to be exchanged between the sending node and the receiving node, and the data transmission process is as Figure 6 shown.

[0101] In a specific embodiment, it also includes designing a network model and an attack model for the underwater acoustic network. The network model adopts a three-dimensional model, and underwater sensor nodes are randomly deployed in the underwater environment, while the target node is deployed on the water surface; the underwater sensor nodes have the same functions and parameters; the attack model adopts the method of multiple attackers for attack.

[0102] Since UANs are deployed in a three-dimensional underwater environment and WSNs are deployed in a two-dimensional scenario, the "Panda-Hunter" model of WSNs cannot be directly applied to UANs. This embodiment combines the three-dimensional UANs model with the "Panda-Hunter" model to form a model for protecting the source location privacy of UANs, as Figure 1 shown. In this model, underwater sensor nodes are randomly deployed in the three-dimensional UANs, and a Sink node is deployed on the water surface. The underwater sensor nodes transmit the sensed data to the Sink node in a multi-hop transmission manner. The following assumptions are made in this network model:

[0103] (1) Underwater sensor nodes are randomly deployed in three-dimensional UANs, and the Sink node is deployed on the water surface; except for the Sink node, all sensor nodes have the same functions and parameters, including initial energy, monitoring range, fixed transmission power, gain, etc.;

[0104] (2) Once a node generates and sends a data packet, that node becomes the source node, and the location near the source node is considered the location of the panda; meanwhile, the initial location of the attacker is near the Sink node;

[0105] (3) Except for the initial energy, the attacker has the same functions and parameters as other sensor nodes; specifically, the attacker has infinite energy, and the attacker only performs passive attacks and local attacks;

[0106] (4) The data packets transmitted in the network are encrypted and cannot be decrypted by the attacker.

[0107] The attack model adopts the multi-attacker attack method because it is considered that: currently, the source node location privacy protocols for UANs all adopt the patience attack model. However, the attack ability of this patience attack model is significantly weaker than the strong attack model in WSNs. Due to the limited energy and communication range of UANs, the strong attack model in WSNs is not applicable to UANs. In WSNs, a small number of attackers are distributed in various regions of the network and can communicate with each other to monitor the traffic of the entire network. However, attackers in UANs face significant challenges in terms of computing power, communication ability, energy supply, long-term operation, deployment and maintenance, etc., and their capabilities are not much different from ordinary underwater sensor nodes. Therefore, attackers in UANs do not have the same level of powerful communication ability as attackers in WSNs. Unless the number of attackers is particularly large, it is difficult for a small number of attackers scattered in different underwater regions to establish effective communication. Therefore, the attack models in the existing source node location privacy protection protocols for UANs all assume that there is only one attacker in the network.

[0108] This embodiment proposes an attack model of multi-attacker cooperation. The initial positions of multiple attackers are around the Sink node, and a "leader-follower" attack mode is established among the multiple attackers for cooperative attacks to find the location of the source node, so as to achieve more efficient and accurate attacks.

[0109] In a specific implementation manner, the specific attack process of the attack model is as follows:

[0110] Step 1: Randomly select one attacker from multiple attackers as the initial leader attacker, and the other attackers become follower attackers;

[0111] Step 2: Assign positions to the follower attackers;

[0112] Step 3: Follow the attacker to collect the traffic information of its own location and transmit the traffic information to the initial leading attacker;

[0113] Step 4: The initial leading attacker designates the follower attacker that has collected the most traffic information as the new leading attacker. The area where the new leading attacker is located becomes the hot spot area, and the previous initial leading attacker and the remaining follower attackers become the new follower attackers;

[0114] Step 5: Based on the location of the new leading attacker, repeat Steps 2 to 4 until the location of the source node is found.

[0115] In a specific embodiment, the location allocation in Step 2 specifically includes the following steps:

[0116] Step 21: Establish a three-dimensional coordinate system with the target node as the coordinate origin;

[0117] Step 22: Construct a sphere with the location of the leading attacker as the center of the sphere and the communication radius of the leading attacker as the radius of the sphere;

[0118] Step 23: Inscribe a regular polygon prism in the sphere. There are several contact points between the regular polygon prism and the sphere; the number of sides of the polygon in the regular polygon prism is one less than the number of attackers;

[0119] Step 24: Allocate the follower attackers one by one at the contact points where the bottom surface of the regular polygon prism touches the sphere.

[0120] To better facilitate the understanding of the attack model in this application, the following specific examples are given:

[0121] In this example, the attack model includes a "leader-follower" attack mode with five attackers, and the specific process is as follows:

[0122] Step 1: Randomly select an attacker among these attackers as the initial leading attacker, and the other attackers become follower attackers;

[0123] Step 2: Allocate positions for the four follower attackers: Establish a three-dimensional coordinate system with the Sink node as the coordinate origin; Construct a sphere with the location O of the leading attacker as the center of the sphere and the communication radius R of the leading attacker as the radius of the sphere; At the same time, inscribe a cube in the sphere; As Figure 2 shown, the cube has eight contact points with the sphere, that is, the eight vertices of the inscribed cube;

[0124] Since the data packets in UANs are always transmitted from bottom to top, in this embodiment, the positions of the four follower attackers are respectively initialized as the four bottom vertices of the inscribed cube ( Figure 2A, B, C, and D in it), so that the attacker can monitor network traffic in a larger range. Assuming that the initial coordinates of the leading attacker are (0, 0, 0), the coordinates of A, B, C, and D can be calculated according to formula (1-1):

[0125]

[0126] where θ represents the polar angle, that is, the angle between the positive direction of the z-axis and the radius axis where the vertex is located; represents the azimuth angle, that is, the angle between the projection of the vertex on the plane and the positive direction of the x-axis; The exact coordinates of A, B, C, and D depend on their respective azimuth angles and the polar angle θ; The polar angles θ of A, B, C, and D are all π / 4; The azimuth angles of A, B, C, and D are π / 4, 3π / 4, 5π / 4, and 7π / 4 respectively; The initial coordinates of the leading attacker are (0, 0, 0); However, as the attacker continuously implements the attack, the attacker gradually moves towards the source node; Therefore, the coordinates of the attacker are constantly changing. Based on this, assuming that the coordinates of the leading attacker in the coordinate system with the Sink node as the origin are (X, Y, Z); The coordinates of A, B, C, and D are shown in formulas (1-2) to (1-5) respectively:

[0127]

[0128] Step 3: Multiple attackers collect traffic information. In the "leader-follower" attack mode, the leading attacker will dispatch four follower attackers to points A, B, C, and D respectively to collect traffic information, and each follower attacker will transmit the collected traffic information to the leading attacker;

[0129] Step 4: Select a new leading attacker. The current leading attacker designates the follower attacker that has collected the most traffic information as the new leading attacker, and the area where the new leading attacker is located becomes the hot spot area, and the previous leading attacker and the other three follower attackers become the new follower attackers;

[0130] Step 5: Reconstruct the sphere with the position of the new leading attacker as the center of the sphere, and repeat steps 2 to 4 until the position of the source node is found.

[0131] In a specific embodiment, before the source node sends data, it also includes initializing the underwater acoustic network: The target node periodically broadcasts Hello packets, so that each node in the network can obtain its level and the neighbor table including two-hop neighbors; The neighbor table includes node ID, level, position, remaining energy, and the status of neighbor nodes.

[0132] Specifically, in this embodiment, when an underwater node hears a Hello packet, the level of the node and the information in the neighbor table are updated according to the heard Hello packet. In fact, the level value of any node is the minimum number of hops from the node to the Sink node. Each node can know the status of neighbor nodes and its own status attributes, simplify routing decisions, reduce redundant broadcasts, and reduce energy consumption; the periodic broadcast can reflect the node movement, failure or addition in real time, adapt to the dynamic changes of the underwater environment, and enhance the reliability and applicability of the whole process.

[0133] In a specific embodiment, a system using the method for protecting the location privacy of the source node in an underwater acoustic network as described in any one of the above, includes: a source node: for generating original data; an SSP-MAC protocol module: for sending the original data in the source node to a target node through a source node data distribution strategy; a target node: for receiving data.

[0134] The system divides the original data into several data segments, and correspondingly selects different candidate source nodes for multi-path transmission to the target node, expanding the search range of the attacker and greatly increasing the difficulty for the attacker to find the source node, thereby ensuring the security of the source node location privacy.

[0135] Embodiment 1

[0136] To further facilitate the understanding of the present application, as Figure 3 shown, the method for protecting the location privacy of the source node in an underwater acoustic network of the present application includes the following steps:

[0137] S1: Initialize the underwater acoustic network;

[0138] S2: The source node generates data;

[0139] S3: The source node divides the original data into n data segments according to the source node data distribution strategy, and selects different candidate source nodes for each data segment;

[0140] S4: Any sending node will evaluate the status of neighbor nodes before sending a data segment: if the receiving node is in an idle state while other neighbor nodes are not in a receiving state, the sending node sends the data segment; if the receiving node and other neighbor nodes are not in the above state, the sending node listens to the signal until the requirements are met and then sends the data segment;

[0141] S5: After the data segment is successfully transmitted, set the sending node and the high-level neighbor nodes of the sending node to the sleep state;

[0142] S6: When the Sink node receives more than t data segments, reconstruct the original data; t < n.

[0143] In this method, first, the source node data distribution strategy and the sleep mechanism are combined to effectively prevent multiple attackers from collaborating to identify hotspots and locate the source node; moreover, the message sharing mechanism and the decentralized candidate source node selection algorithm in the data distribution strategy ensure that consecutive data packets reach the Sink node through different and decentralized paths, increasing the difficulty for attackers to identify hotspots; the introduction of the sleep mechanism places the nodes that may be eavesdropped by attackers in the sleep state, preventing dangerous nodes from further participating in data packet forwarding; therefore, it is difficult for attackers to identify hotspots, making this method resistant to attacks launched by a multi-attacker collaborative attack model; at the same time, it can also resist attacks from a patient attack model: the method in this embodiment constructs decentralized and diverse transmission paths, and the nodes that may be eavesdropped by attackers do not participate in data forwarding, which prevents attackers from receiving consecutive data packets and moving towards the source node.

[0144] Secondly, this method reduces packet loss from two aspects: on the one hand, in the source node data distribution strategy, a lightweight (t,n)-threshold message sharing method based on the Chinese Remainder Theorem is used. As long as the Sink node receives a part of the data fragments, it can successfully reconstruct the original data. Therefore, even if some information fragments are lost due to node movement and high bit error rate in UANs, this method can still complete the reconstruction of the original information, greatly reducing the possibility of packet loss; on the other hand, this method evaluates the status of neighbor nodes before the sending node sends data packets, thereby reasonably scheduling the transmission of data packets and minimizing packet loss caused by channel contention and data packet collisions; in summary, this method reduces packet loss caused by high bit error rate, node movement, and data packet collisions, thereby improving the reliability and robustness of data transmission in UANs.

[0145] Meanwhile, to further verify the effect of this embodiment, the method for protecting the location privacy of the source node in the underwater acoustic network in this embodiment is also called the SSP-MAC protocol. To better evaluate the performance of the SSP-MAC protocol, MATLAB is used as a simulation tool for simulation experiments. At the same time, the SSP-MAC protocol is compared with the MAC mechanism in the EECOR protocol, the DBR-MAC protocol, and the CFTSLP-TSA protocol: among them, the EECOR and DBR-MAC protocols contain MAC protocols designed for UANs, and the CFTSLP-TSA protocol is a MAC protocol designed for protecting the location privacy of the source node in UANs (the specific compositions of the other three protocols are prior arts and will not be elaborated in detail in this application). The parameters used in the simulation experiments are shown in Table 3:

[0146] Table 3 Simulation Parameters

[0147]

[0148] The simulation experiment evaluates the performance of the SSP-MAC protocol from three performance metrics: safety time, packet delivery ratio, and energy consumption fairness index. Safety time refers to the time interval between the source node starting to send a data packet and the attacker successfully finding the location of the source node. The larger the time interval, the longer it takes for the attacker to find the source node, and the more secure the location privacy of the source node. The packet delivery ratio is the ratio of the number of data packets successfully received by the Sink node to the number of data packets sent by the source node. As shown in Equation 1-6:

[0149]

[0150] Among them, N source represents the number of data packets sent by the source node, and N sink represents the number of data packets successfully received by the Sink node.

[0151] The energy consumption fairness index reflects the degree of balance of the remaining energy of nodes in the network; the unbalanced remaining energy is usually caused by unfair channel occupancy. Nodes that occupy the channel for a long time consume much more energy than other nodes, and this imbalance in energy consumption will shorten the lifespan of the entire network. Therefore, the energy consumption fairness index is one of the important evaluation metrics for measuring network performance. The larger the energy consumption fairness index, the more balanced the energy consumption of the entire network. The definition of the energy consumption fairness index is shown in Equation (1-7):

[0152]

[0153] Among them, N represents the number of nodes in the network, and e i represents the remaining energy of node i.

[0154] The specific verification process is as follows:

[0155] 1. Verify the impact of the attack model and sleep duration on performance:

[0156] Figure 7 shows the relationship between the sleep duration and the safety time of the SSP-MAC protocol in the face of two attack models. From Figure 7It can be seen that in the face of a collaborative attack by multiple attackers, the safety period of the SSP-MAC protocol is reduced by at least about 200s, which shows that the attack model proposed in this application has a stronger attack capability than the patient attack model. In addition, the safety time increases with the increase in sleep duration. This is because the longer the sleep duration, the later the dangerous node participates in data forwarding, and the slower the attacker moves toward the source node. What is explained here is: one attack model is to use the traditional patient single attacker attack model, which assumes that there is only one attacker in the network, and the attacker uses passive attacks such as eavesdropping attacks and backtracking attacks to find the location of the source node: specifically, the attacker performs a backtracking attack to move toward the source node based on the information obtained from the eavesdropping attack; the attacker will only track the location of the node that sent the data packet if he eavesdrops on the data packet, otherwise the attacker will remain stationary; the attacker will continue the iterative process of "eavesdropping-tracking" until the location of the source node is found; one attack model is to use the multi-attacker attack model in this application; Figure 7 and Figure 8 The line where the middle triangle is located is the traditional patient single attacker attack model; the line where the star is located is the multi-attacker attack model in this application.

[0157] Figure 8 The relationship between the sleep duration and the end-to-end delay of the SSP-MAC protocol facing two attack models is given. As the sleep duration increases, the number of nodes that can participate in data transmission in the network decreases, making it more difficult for data packets to find forwarding nodes, thereby increasing the end-to-end delay. In addition, it can be seen from the figure that the end-to-end delay of the SSP-MAC protocol facing the two attack models is not much different.

[0158] Based on the above analysis, the following conclusions are drawn: attacks by multiple attackers in collaboration show stronger attack capabilities than patient attacks.

[0159] 2. The impact of parameter n, i.e. the number of data segments split, on performance

[0160] To defend against passive attacks and protect the privacy of the source node location, the original data is mapped into n data fragments. Figure 9 The relationship between depth and security time under different n values ​​in the face of multi-attacker coordinated attacks is given. Figure 9 It can be seen that as the value of n increases, the security time of the SSP-MAC protocol gradually increases. When the value of n increases, the number of nodes involved in data forwarding increases, making the transmission path more diverse, increasing the difficulty for attackers to track the location of the source node, and thus extending the security time.

[0161] Figure 10 The relationship between depth and end-to-end delay under different n values ​​is given. Figure 10It can be seen that regardless of the value of n, as the depth increases, the end-to-end delay increases. In addition, as the value of n increases, the Sink node needs to receive more data fragments to reconstruct the original data, resulting in a gradual increase in the end-to-end delay.

[0162] Figure 11 shows the relationship between depth and energy consumption fairness index for different values of n. From Figure 11 it can be seen that as the depth increases, the change in the energy consumption fairness index is not significant. In contrast, the energy consumption fairness index increases with the increase of the value of n. Since the value of n is positively correlated with the number of nodes participating in data transmission, therefore, as the value of n increases, more nodes participate in data transmission, and the energy consumption becomes more balanced, resulting in a gradual increase in the energy consumption fairness index.

[0163] In summary, from Figures 9 to 11 it can be seen that as the value of n increases, the security time, end-to-end delay, and energy consumption fairness index also increase. In order to balance network performance such as security time, end-to-end delay, and energy consumption fairness index, the value of n is set to 7 in the subsequent simulation experiments.

[0164] 3. Comparative Experiments

[0165] To evaluate the performance of the SSP-MAC protocol, the SSP-MAC protocol, EECOR protocol, DBR-MAC protocol, and CFTSLP-TSA protocol are compared. Since the EECOR protocol, DBR-MAC protocol, and CFTSLP-TSA protocol are all existing technologies, this application only gives the protocol sources and will not elaborate on the above three protocols in detail.

[0166] The reference for the EECOR protocol is Rahman M A, Lee Y, Koo I. EECOR: An energy-efficient cooperative opportunistic routing protocol for underwater acoustic sensor networks[J]. IEEE Access, 2017, 5: 14119 - 14132.

[0167] The reference for the DBR-MAC protocol is i C, Xu Y, Diao B, et al. DBR-MAC: A depth-based routing aware MAC protocol for data collection in underwater acoustic sensor networks[J]. IEEE Sensors Journal, 2016, 16(10): 3904 - 3913.

[0168] The reference for the CFTSLP-TSA protocol is Han G, Liu Y, Wang H, et al. A collision-free-transmission-based source location privacy protection scheme in UANs under time slot allocation[J]. IEEE Internet of Things Journal, 2022, 10(2): 1546-1557.

[0169] Since other protocols adopt a patience attack model, the SSP-MAC protocol also conducts simulation experiment evaluations based on the patience attack model:

[0170] Figure 12 The relationship between the number of nodes and the security time of four protocols is given. It can be seen from the figure that the security period of the SSP-MAC protocol is at least 700 seconds longer than that of the DBR-MAC protocol, 400 seconds longer than that of the EECOR protocol, and 50 seconds longer than that of the CFTSLP-TSA protocol, which proves the effectiveness of the SSP-MAC protocol in protecting the location privacy of source nodes. Since the DBR-MAC protocol does not adopt any source node location privacy protection strategy, its security time is the shortest among the four protocols. The EECOR protocol selects forwarding nodes based on the remaining energy of nodes, and data packets may reach the Sink node through different paths, expanding the search scope of attackers. Therefore, its security time is longer than that of the DBR-MAC protocol. The CFTSLP-TSA protocol uses false data packets to cover up real traffic and uses multi-path technology to expand the search scope of attackers. The SSP-MAC protocol places dangerous nodes in the sleep state and adopts a source node data distribution strategy, which not only slows down the tracking speed of attackers but also expands the search scope of attackers. Therefore, Figure 12 it can be seen that the security times of the SSP-MAC and CFTSLP-TSA protocols are significantly longer. When the number of nodes is less than 400, the security time of the SSP-MAC protocol is significantly longer than that of the CFTSLP-TSA protocol. However, when the number of nodes exceeds 400, although the security time of the SSP-MAC protocol still has a slight advantage, the difference in security time between the SSP-MAC and CFTSLP-TSA protocols becomes less obvious.

[0171] Different from the other three protocols, the security time of the SSP-MAC protocol does not increase with the increase in the number of nodes. In the SSP-MAC protocol, transmitting data fragments requires different candidate source nodes and paths, thus involving more nodes in data transmission. When the number of nodes is less than 400, the small number of nodes in the network leads to long detours of data packets from the source node to the Sink node, increasing the difficulty for attackers to trace the source node. Therefore, the security time gradually decreases as the number of nodes increases. When the number of nodes exceeds 400, due to the sufficient number of nodes, the detour problem becomes less obvious. As the number of nodes increases, the diversity of paths and forwarding nodes increases, and the security time also gradually increases.

[0172] Although it can be seen that the SSP-MAC protocol has the longest security time, this simulation experiment is based on a patience attack model and cannot intuitively demonstrate the advantages of SSP-MAC compared with other protocols in resisting the attack model of multi-attacker cooperation. In the face of the attack model of multi-attacker cooperation, the security time of the four protocols must be less than Figure 12 the results shown. However, combining Figure 12 and Figure 7 and Figure 12 it can be seen that the security time of the SSP-MAC protocol under the attack model of multi-attacker cooperation is still better than the security time of the DBR-MAC and EECOR protocols under the patience attack model. Therefore, the SSP-MAC protocol still has a high level of source node location privacy security when facing multi-attacker cooperative attacks.

[0173] Figure 13 shows the relationship between the number of nodes and the packet delivery ratio of the four protocols. From Figure 13It can be seen that as the number of nodes increases, the packet delivery ratios of all protocols show an upward trend. As the number of nodes increases, more nodes are eligible to participate in data packet forwarding. Therefore, the packet delivery ratios of the four protocols gradually increase and tend to be stable. In the EECOR protocol, a holding time is set for the forwarding nodes to schedule the transmission of data packets and avoid collisions. However, since setting the holding time cannot well avoid data packet collisions, its packet delivery ratio is the lowest among the four protocols. The DBR-MAC protocol adopts a handshake and adaptive backoff mechanism, reducing the probability of data packet collisions. Therefore, compared with the EECOR protocol, the DBR-MAC protocol has a higher packet delivery ratio. The CFTSLP-TSA protocol uses different time slots to transmit real data packets and false data packets to avoid collisions. Therefore, its packet delivery ratio performs well. The SSP-MAC protocol avoids collisions through a transmission mechanism based on the states of neighbor nodes. In addition, in the SSP-MAC protocol, the Sink node only needs to receive some data fragments to reconstruct the original data, thus solving the problem of data packet loss caused by the high bit error rate of the underwater acoustic channel. Therefore, in the case of high node density (the number of nodes is greater than 400), the SSP-MAC protocol has the highest packet delivery ratio. However, when the node density is low (the number of nodes is less than 400), the sending node may not be able to find the next-hop node, resulting in the loss of some information fragments. Therefore, in a low-density network, the SSP-MAC protocol has no obvious advantage in terms of packet delivery ratio.

[0174] Figure 14 The relationship between the number of nodes and the energy consumption fairness index of the four protocols is given. It can be seen from the figure that the energy consumption fairness indexes of the three protocols, namely SSP-MAC, EECOR, and CFTSLP-TSA, all reach 95%, while the energy consumption fairness index of the DBR-MAC protocol is about 85%. In addition, the SSP-MAC protocol has the highest energy consumption fairness index, proving its advantage in energy consumption balance. Specifically, due to the data distribution strategy of the source node in the SSP-MAC protocol, more scattered nodes participate in data transmission. Therefore, its energy consumption fairness index is the highest. The EECOR protocol selects forwarding nodes based on the remaining energy, making the transmission paths diverse. The CFTSLP-TSA protocol realizes diverse data transmission paths through a forwarding node selection algorithm. Therefore, both the EECOR and CFTSLP-TSA protocols perform well in terms of energy consumption fairness index, but the CFTSLP-TSA protocol has a better performance in terms of energy consumption fairness index. On the contrary, the DBR-MAC protocol selects forwarding nodes according to the depth of candidate nodes, resulting in relatively single paths. Therefore, as the number of nodes increases, the energy consumption fairness index of the DBR-MAC protocol gradually decreases, and its energy consumption fairness index is the lowest among the four protocols.

[0175] After a series of simulation experiments, the effectiveness of the SSP-MAC protocol in protecting the location privacy of source nodes, avoiding packet collisions, alleviating packet loss caused by high bit error rates, and balancing energy consumption is verified. Specifically, compared with the existing MAC protocols in UANs, the SSP-MAC protocol has obvious advantages in terms of security time, packet delivery rate, and energy consumption fairness index. The simulation experiments show that the SSP-MAC protocol can not only achieve high source node location privacy security but also solve the problems of packet collisions, packet loss, and unbalanced energy consumption in UANs. Table 4 clearly shows the performance of the SSP-MAC protocol in various performance indicators.

[0176] Table 4 Performance indicators of the SSP-MAC protocol

[0177]

[0178] In summary, this application first designs an attack model with multi-attacker cooperation to endow attackers with stronger attack capabilities, making it more applicable to UANs; then, the source node data distribution strategy protects the location privacy of source nodes through a message sharing mechanism and a decentralized candidate source node selection algorithm; at the same time, this strategy solves the packet loss problem caused by node mobility and high bit error rates and achieves energy consumption balance; furthermore, the sleep mechanism is introduced to protect the location privacy of source nodes, and the transmission mechanism based on the neighbor node status can effectively avoid packet collisions. Compared with other methods, the method of this application has a longer security time, a higher packet delivery rate, and a higher energy consumption fairness index. In short, this application not only realizes the protection of source node location privacy and energy consumption balance but also greatly alleviates the problems caused by packet loss.

[0179] The examples described in the present invention are only descriptions of the preferred embodiments of the present invention, and do not limit the concept and scope of the present invention. Without departing from the design idea of the present invention, various deformations and improvements made by those skilled in the art to the technical solutions of the present invention should fall within the protection scope of the present invention.

Claims

1. A method for protecting the location privacy of source nodes in an underwater acoustic network, characterized in that: The source node sends the original data hop by hop to the destination node through the source node data distribution strategy; wherein, the source node data distribution strategy is as follows: The original data in the source node is divided into several data segments, and then each data segment is assigned to one of several candidate source nodes, so that different data segments correspond to different candidate source nodes, realizing that the original data reaches the destination node through different and scattered paths.

2. The method for protecting the location privacy of a source node in an underwater acoustic network according to claim 1, wherein: The determination of several candidate source nodes is carried out in the following way: Establish a preliminary candidate source node pool: all nodes that meet the requirement of having a hierarchical level less than the two-hop neighbor nodes of the source node are used to establish a preliminary candidate source node pool; Select a reference node: Select the node with the largest remaining energy in the preliminary candidate source node pool as the reference node, and remove the reference node from the preliminary candidate source node pool to obtain the remaining candidate source node pool; Select candidate source nodes: Select nodes one by one in the remaining candidate source node pool as candidate source nodes. The selected candidate source nodes enter the selected candidate source node pool until the number of candidate source nodes in the selected candidate source node pool is the same as the number of data segments. Stop selecting candidate source nodes; among them, when the sum of the distances between the currently selected node and each candidate source node in the selected candidate source node pool is the largest, it can be used as a candidate source node and enter the selected candidate source node pool.

3. A method for protecting the location privacy of source nodes in an underwater acoustic network according to claim 1 or 2, characterized in that: The source node maps the original data into several data segments based on the sharing mechanism of the Chinese Remainder Theorem; specifically, it includes the following steps: Pre - deployment: Set a group of integers (d1, d2, … d n ), and this group of integers satisfies: d1 < d2 <... < d n ; gcd(d i , d j ) = 1, i ≠ j; D = d1 × d2 × … × d t , D1 = d n-t+2 × d n-t+3 ×... × d n , D > M > D1, where M is the original data; Data segmentation: The original data M is segmented into n data segments, and the data segments are represented as (d i , m i ), and the specific formula is as follows: Data recombination: After the source node sends n data segments, as long as the destination node successfully receives any t of the n data segments, the original data M can be recombined.

4. A method for protecting the location privacy of source nodes in an underwater acoustic network according to claim 1, characterized in that: It also includes setting a sleep mechanism during the data transmission process from the source node to the destination node: when the original data in the source node is sent to the receiving node through the sending node, the sending node and its higher-level neighbor nodes are set to the sleep state; the sleep state means not participating in data transmission.

5. A method for protecting the location privacy of source nodes in an underwater acoustic network according to claim 4, characterized in that: Before sending data segments, the sending node also includes evaluating the states of all its neighbor nodes: when the receiving nodes among all neighbor nodes are in the idle state and the rest of the neighbor nodes are not in the receiving state; the sending node will send data segments.

6. The method for protecting the location privacy of the source node in an underwater acoustic network according to claim 1, wherein: It also includes designing the network model and attack model of the underwater acoustic network. The network model adopts a three-dimensional model, underwater sensor nodes are randomly deployed in the underwater environment, and the destination node is deployed on the water surface; underwater sensor nodes have the same functions and parameters; the attack model uses the method of multiple attackers for attack.

7. A method for protecting the location privacy of source nodes in an underwater acoustic network according to claim 6, characterized in that: The specific attack process of the attack model is as follows: Step 1: Randomly select one attacker from multiple attackers as the initial leading attacker, and the other attackers become following attackers; Step 2: Assign positions to the following attackers; Step 3: The following attackers collect the traffic information at their locations and transmit the traffic information to the initial leading attacker; Step 4: The initial leading attacker designates the follower attacker that has collected the most traffic information as the new leading attacker. The area where the new leading attacker is located becomes the hot spot area, and the previous initial leading attacker and the remaining follower attackers become the new follower attackers; Step 5: Based on the position of the new leading attacker, repeat Steps 2 to 4 until the position of the source node is found.

8. A method for protecting the location privacy of the source node in an underwater acoustic network according to claim 7, characterized in that: The specific steps for allocating positions in Step 2 are as follows: Step 21: Establish a three-dimensional coordinate system with the target node as the coordinate origin; Step 22: Construct a sphere with the position of the leading attacker as the center of the sphere and the communication radius of the leading attacker as the radius of the sphere; Step 23: Inscribe a regular polygon prism in the sphere. There are several contact points between the regular polygon prism and the sphere; the number of sides of the polygon in the regular polygon prism is one less than the number of attackers; Step 24: Allocate the follower attackers one by one at the contact points where the bottom surface of the regular polygon prism touches the sphere.

9. A method for protecting the location privacy of source nodes in an underwater acoustic network according to claim 1, characterized in that: Before the source node sends data, it also includes initializing the underwater acoustic network: The target node periodically broadcasts Hello packets so that each node in the network obtains its level and a neighbor table containing two-hop neighbors; the neighbor table contains the node ID, level, position, remaining energy, and the status of neighbor nodes.

10. A system using the method for protecting the location privacy of an underwater acoustic network source node as described in any one of claims 1-9, characterized in that: It includes: Source node: Used to generate the original data; SSP-MAC protocol module: Used to send the original data in the source node to the target node through the source node data distribution strategy; Target node: Used to receive data.

Images