A low-energy and privacy-preserving data aggregation method

Abstract

This invention discloses a low-energy, privacy-preserving data aggregation method for addressing the issues of unbalanced node load and high energy consumption in wireless sensor networks. The method comprises organizing the network into a binary tree-like structure, where non-leaf nodes are single sensors, and leaf nodes are a chain of multiple sensors connected in series, with only the nodes at the end of the chain being sliced ​​and encrypted. This invention improves the topology of wireless sensor networks, shortening the distance between parent and child nodes, reducing node energy consumption, balancing node loads, addressing the problem of premature node failure in the network, reducing data loss caused by collisions during data transmission, extending the network lifetime, and improving the accuracy of aggregation results.

Description

Technical Field

[0001] The present invention belongs to the field of wireless sensor network routing security, in particular to a data aggregation method with low energy consumption and privacy protection. Background Art

[0002] With the rapid development of the Internet of Things (IoT), wireless sensor networks are increasingly being used in military, environmental monitoring, healthcare, industrial production, traffic control, and other fields. However, due to the limited computing power, storage capacity, and energy consumption of nodes, and the fact that each node's energy consumption is closely related to the network's lifetime, sensor nodes must be as energy-efficient as possible to extend the network's lifespan. Furthermore, protocol design must account for individual node failures and implement fault-tolerance mechanisms while minimizing energy consumption. Therefore, protocols should be able to filter and process data within the network to reduce energy consumption. Data aggregation technology is an important means of reducing network transmission volume, lowering energy consumption, and improving security, and is widely used in wireless sensor networks. Its concept is to merge data from different sensor nodes, eliminating redundancy, reducing transmission times, and thus conserving energy.

[0003] Earlier, He et al. proposed the SMART algorithm based on the TAG algorithm. This algorithm uses the slice-mix-aggregate concept. Each node divides its data into J slices, sending J-1 slices to neighboring nodes and retaining 1 slice for itself. Data on leaf nodes is encrypted to protect data privacy, while data on non-leaf nodes is not protected. This algorithm protects data privacy to a certain extent, but the large number of messages exchanged due to slice distribution shortens the network lifespan. Building on this, Li et al. proposed the ESMART algorithm. This algorithm uses slice-mix technology and only slices data for leaf nodes, with the number of slices being random within a set range. Wang et al. proposed the PECDA algorithm. Similarly, data is sliced ​​only for leaf nodes and sent to neighboring nodes over a secure channel. Specifically, PECDA allows for continuous data aggregation and is scalable, improving node energy consumption and enhancing security. Zhou et al. proposed the EPDA algorithm, which organizes the network into a tree, connects the leaf nodes in the tree into chains, and only slices the data at the end of the chain, thus protecting data privacy and reducing the energy consumption of data aggregation.

[0004] In summary, the design of secure routing protocols has become a research hotspot in wireless sensor networks in recent decades. How to balance the energy consumption of nodes while taking into account the security of data transmission has become an urgent problem to be solved in this field. Summary of the Invention

[0005] To solve the technical problems mentioned in the above background art, the present invention proposes a data aggregation method with low energy consumption and privacy protection.

[0006] To achieve the above technical purpose, the technical solution of the present invention is as follows:

[0007] A data aggregation method with low energy consumption and privacy protection, comprising the following steps:

[0008] (1) The base station broadcasts a message to the nodes in the network, and organizes the nodes in the network into a tree structure according to certain rules.

[0009] (2) Based on the formed tree structure, reconstruct it according to the reconstruction algorithm to form a structure similar to a binary tree. The non-leaf nodes of the tree are single sensors, and the leaf nodes are a chain formed by connecting multiple sensors in series.

[0010] (3) The nodes in the network sense data. The leaf nodes cut the sensed data into J pieces, randomly encrypt J - 1 pieces and send them to neighbor nodes for mixing, and keep 1 piece.

[0011] (4) The tail node of the chain sends the encrypted data to the parent node for aggregation. The data reaches the head node of the chain along the linked list structure, and then reaches the base station along the binary tree structure.

[0012] Further, in step (1), the method for constructing the network is as follows:

[0013] (101) The non-leaf node (initially the non-leaf node is the base station) broadcasts a "Hi" message to neighbor nodes to invite them to join the network.

[0014] (102) After a node receives the "Hi" message, if the node has not joined the network, it replies with a "Join" message to express its willingness to join. If it receives multiple "Hi" messages, it preferentially replies to the "Hi" message that arrives first. [[ID=2,9]]

[0015] (103) After the non-leaf node receives the "Join" message, it calculates the distance d from the message sender. If d < d0 (d0 is a preset value), it replies with an "Accept" message to allow it to join the network. <00,00052>

[0016] (104) After a node receives the "Accept" message, it becomes the child node of the "Accept" message sender, joins the network, and is selected as a non-leaf node with a certain probability P. P is a preset value.

[0017] (105) If the node is selected as a non-leaf node, it repeats steps (101)-(104) until all the nodes within the communication range of the non-leaf nodes in the network have joined the network. [[ID=3,7]]

[0018] (106) A node that has not joined the network (a node without a parent node) broadcasts an “RHi” message to nodes within the communication range to request to join the network.

[0019] (107) After receiving the “RHi” message, the node joining the network replies with an “R Join” message.

[0020] (108) After receiving the “R Join” message, the node replies with an “Accept” message.

[0021] (109) After the node receives the "Accept" message, if it is not a non-leaf node, it is selected as a non-leaf node and executes steps (101)-(105).

[0022] Furthermore, in step (2), the reconstruction algorithm is as follows:

[0023] (201) The child nodes of each non-leaf node Node in the network are divided into a leaf node list and a non-leaf node list. Let LastNode = Node. Perform steps (202)-(212) for the leaf node list and the non-leaf node list respectively.

[0024] (202) For the node list List, LastNode broadcasts a “Rebuild” message to the surrounding nodes.

[0025] (203) The nodes in the List accept the message and reply to the LastNode with a “Join” message.

[0026] (204) If LastNode receives the "Join" message, it replies with an "Accept message" to the sender of the message that arrived first. Otherwise, it sends a "Check" message to Node and executes (207).

[0027] (205)Node i After receiving the "Accept" message, it becomes the child node of the message sender. Let LastNode = Node i , and sends a "JoinNetwork" message to Node, telling Node to delete it from the List.

[0028] (206) Repeat (202)-(205) until the LastNode does not receive the "Join" message.

[0029] (207)Node randomly selects a node in the List that is not added to the network j , send a "Head" message to it, and Node jDelete it from the list and send a "Change Head" message to the chain head node.

[0030] (208) After receiving the message, the chain head node sends a "Join" message to Node j .

[0031] (209)Node j After receiving the "Head" message, it becomes a child node of the Node node.

[0032] (210)Node j After receiving the "Join" message, it replies to the sender of the message with an "Accept" message allowing it to join the network.

[0033] (211) Node receives Node j After sending the "Accept" message, it becomes a Node j 's child nodes.

[0034] (212) Repeat steps (207)-(211) until the List is empty.

[0035] Furthermore, in step (3), the data slicing method is as follows:

[0036] (301) Nodes in the network sense and collect data.

[0037] (302) The leaf nodes in the network cut the collected data into J pieces, of which J-1 pieces are randomly sent to neighboring nodes and 1 piece is retained.

[0038] (303) After receiving the slice data, the node mixes the slice data with its own data.

[0039] Furthermore, in step (4), the data aggregation method is as follows:

[0040] (401) The leaf nodes in the network send their data to the parent node.

[0041] (402) The node aggregates the received data and sends the aggregation result to its parent node.

[0042] (403) Repeat step (402) until the data reaches the head node along the linked list structure.

[0043] (404) The chain head node sends the data to its father node.

[0044] (405) The node aggregates the received data and sends the aggregation result to its parent node.

[0045] (406) Repeat step (405) until the data reaches the base station along the binary tree structure.

[0046] The beneficial effects brought about by adopting the above technical solution are:

[0047] (1) This method adopts the “slice assembly” technology and only slices the leaf nodes, which reduces the overall network communication overhead while protecting privacy, reduces data loss caused by collisions during data transmission, and improves the accuracy of the aggregation results.

[0048] (2) A topology structure that combines a binary tree structure and a linked list structure is adopted, so there is no need to use routing discovery to collect data. In addition, this topology structure shortens the distance between parent nodes and child nodes, balances the load of each node, reduces the energy consumption of the node, solves the problem of premature failure of some nodes in the network, and prolongs the survival time of the network. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic diagram of the network topology of the present invention;

[0050] Figure 2 It is a schematic diagram of establishing a tree network topology structure of the present invention;

[0051] Figure 3 It is a schematic diagram of repairing the tree network topology structure of the present invention;

[0052] Figure 4 is a schematic diagram of topological reconstruction of the present invention;

[0053] Figure 5 is a flowchart of topology reconstruction of the present invention;

[0054] Figure 6 is a schematic diagram of the slicing and mixing process of the present invention;

[0055] Figure 7 is a schematic diagram of the polymerization process of the present invention; DETAILED DESCRIPTION

[0056] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings.

[0057] A low-power wireless sensor network routing control method based on trust evaluation includes the following steps:

[0058] Step 1: Non-leaf nodes (initially, only the BS is a non-leaf node in the network) invite nodes within their communication range to join the network by broadcasting "Hi" messages to these nodes. After receiving the "Hi" message, if a node has not joined the network (a node without a parent node is considered not joined the network, except for the BS) and its distance from the sender node of the message is less than d0, it replies with a "Join" message to the sender of the message.

[0059] In Figure 1 , the non-leaf node (BS) broadcasts the "Hi" message to its neighbor nodes (1, 6, 7). When other nodes (1, 6, 7) receive the "Hi" message, they calculate the distance d from the sender node of the message. If d < d0, they reply with a "Join" message to request to join. After receiving the "Join" message from a node, the non-leaf node sends an "Accept" message to the node to allow it to join. After receiving the "Accept" message, the node joins the network (becomes a child node of the sender of the "Accept" message) and elects itself as a non-leaf node with a certain probability P (P is a preset value). If elected as a non-leaf node, it broadcasts the "Hi" message to its neighbor nodes to continue constructing the network, enabling as many nodes as possible to join the network, and finally forming an aggregation tree. Then, the nodes that have not joined the network (nodes without a parent node) broadcast messages to request to join the network within their communication range. As Figure 2 shown, taking node 12 as an example, node 12 sends a "R Hi" message to its neighbor nodes 7, 10, 11 to request to join the network. Since nodes 7, 11 have already joined the network (have a parent node), they reply with a "R Join" message to allow node 12 to join the network. Node 12 replies with an "Accept" message to the sender node 10 of the first-arriving "R Join" message (the closer the distance between the two nodes, the earlier the message arrives) and becomes a child node of node 10, thus joining the network.

[0060] Step 2: The child nodes of each non-leaf node N i in the network are divided into a leaf node list LeafList i and a non-leaf node list noLeafList i . Subsequently, LeafList i and noLeafList i are reconstructed. The specific reconstruction process is as Figure 4As shown in the figure, the leaf node list reconstruction process of node 1 (the head of the chain) is used as an example. Node 1 randomly selects a node (node ​​4) and sends it a "Head" message. After receiving the message, node 4 broadcasts a "Rebuild" message to its neighboring nodes (nodes 3 and 5). After receiving the "Rebuild" message, the node (which does not have a parent node during the reconstruction phase) replies with a "Join" message to node 4, expressing its willingness to join. Node 4 prioritizes replying to the first message (from node 3) with an "Accept" message. Subsequently, node 3 becomes a child of node 4 and continues broadcasting "Rebuild" messages to its neighboring nodes. As this process continues, node 2 becomes a child of node 3. After a period of broadcasting, node 2 receives no response from other nodes and sends a "Check" message to node 1. After receiving the "Check" message, node 1 broadcasts a "Check" message to check for any nodes that have not yet joined the chain. Finding that node 5 has not yet joined, node 1 sends a "Head" message to node 5 and a "Change Head" message to the node immediately below the head of the chain (node ​​4). After receiving the "Head" message, node 5 replies with a "Join" message to the chain head (node ​​1). Node 4 receives the "Change Head" message and sends a "Join" message to node 5, requesting to become a child of node 5. After receiving the message, nodes 1 and 5 reply with an "Accept" message. This process continues until all child nodes of non-leaf nodes have joined the chain. The complete flow chart of the second stage is shown below. Figure 5 shown.

[0061] Step 3: The tail node (leaf node) divides the sensed data into J pieces and retains 1 piece. J-1 pieces are encrypted with the shared key and sent to the neighbor node N. i , and wait for an ACK message. If no ACK message is received, N i Delete it from the neighbor node list and resend the slice message to other neighbor nodes. Figure 6 As shown in the figure, taking node 7 as an example, node 7 splits the data into d7,7, d7,6, and d7,10. d7,7 is retained, and d7,6 and d7,10 are sent to nodes 6 and 10 respectively. At the same time, the node also receives the slice d12,7 from node 12. Considering factors such as distance, to ensure that all slices are received, the node needs to wait for a period of time, decrypt all received encrypted data slices using the shared key, and mix them with the data it perceives.

[0062] Step 4: The tail node (leaf node) of the chain encrypts the data and sends it to the parent node for aggregation, and waits for an ACK message. If no ACK message is received, the node is detected to be faulty and replaced with the grandparent node, and the message is resent. After receiving the data, the node waits for a while and aggregates the received data with its own data, and then sends it to the parent node, and finally reaches the BS along the aggregation tree. Figure 7 As shown, the leaf node 12 stores the data S 12 Send it to its father node 11. After waiting for a while (to ensure that all data from other nodes are received), node 11 aggregates the received data with its own data, and the result is S 11 +S 12 , and then sends the new aggregation result to the father node 10. The data follows the chain structure to the chain head (12→11→10), and then follows the tree structure to reach BS (10→6→1→BS).

Claims

1. A low-energy and privacy-preserving data aggregation method, characterized in that: It includes the following steps: (1) The base station broadcasts a message to the nodes in the network, organizing the nodes in the network into a tree structure according to certain rules; (2) Based on the formed tree structure, it is reconstructed according to the reconstruction algorithm to form a structure similar to a binary tree. The non-leaf nodes of the tree are single sensors, and the leaf nodes are a chain formed by connecting multiple sensors in series; (3) The nodes in the network sense data. The leaf nodes cut the sensed data into J pieces, randomly send J - 1 pieces for mixing to neighbor nodes, and keep 1 piece; (4) The tail node of the chain sends the encrypted data to the parent node for aggregation. The data reaches the head node of the chain along the linked list structure and then reaches the base station along the binary tree structure.

2. A low-energy and privacy-preserving data aggregation method according to claim 1, characterized in that: The specific process of step (1) is as follows: (101) The non-leaf nodes broadcast a "Hi" message to neighbor nodes to invite them to join the network; (102) After a node receives the "Hi" message, if the node has not joined the network, it replies with a "Join" message to express its willingness to join. If it receives multiple "Hi" messages, it preferentially replies to the "Hi" message that arrives first; (103) After the non-leaf node receives the "Join" message, it calculates the distance d from the message sender. If d < d0 (d0 is a preset value), it replies with an "Accept" message to allow it to join the network; (104) After a node receives the "Accept" message, it becomes the child node of the "Accept" message sender, joins the network, and is selected as a non-leaf node with a certain probability P (P is a preset value); (105) If the node is selected as a non-leaf node, repeat steps (101)-(104) until all nodes within the communication range of the non-leaf nodes in the network have joined the network; (106) The nodes that have not joined the network broadcast a "R Hi" message to the nodes within their communication range to request to join the network; (107) After a node that has joined the network receives the "R Hi" message, it replies with a "R Join" message; (108) After a node receives the "RJoin" message, it replies with an "Accept" message; (109) After a node receives the "Accept" message, if it is not a non-leaf node, it is selected as a non-leaf node and executes steps (101)-(105).

3. The low-energy and privacy-preserving data aggregation method according to claim 1, characterized in that: The steps of the reconstruction algorithm in step (2) are as follows: (201) The sub-nodes of each non-leaf node Node in the network are divided into a leaf node list and a non-leaf node list. Let LastNode = Node, and steps (202)-(212) are respectively executed on the leaf node list and the non-leaf node list; [[ID=1 ​ ​ (205)Node i After receiving the "Accept" message, it becomes the child node of the message sender, and sets LastNode = Node i , and sends a "JoinNetwork" message to Node, telling Node to delete it from the List; (206) Repeat (202)-(205) until the LastNode does not receive the "Join" message; (207)Node randomly selects a node in the List that is not added to the network j , send a "Head" message to it, and Node j Delete from the list and send a "Change Head" message to the chain head node; (208) After receiving the message, the chain head node sends a "Join" message to Node j ; (209)Node j After receiving the "Head" message, it becomes a child node of the Node node; (210)Node j After receiving the "Join" message, it replies to the sender of the message with an "Accept" message to allow it to join the network; (211) Node receives Node j After sending the "Accept" message, it becomes a Node j 's child nodes; (212) Repeat steps (207)-(211) until the List is empty.

4. The low-energy and privacy-preserving data aggregation method according to claim 1, characterized in that: The specific process in step (3) is as follows: (301) Nodes in the network sense and collect data; (302) The leaf nodes in the network cut the collected data into J pieces, of which J-1 pieces are randomly sent to neighboring nodes and 1 piece is retained; (303) After receiving the slice data, the node mixes the slice data with its own data.

5. The low-energy and privacy-preserving data aggregation method according to claim 1, characterized in that: The specific process in step (4) is as follows: (401) The leaf nodes in the network send their data to their father nodes; (402) The node aggregates the received data and sends the aggregation result to its parent node; (403) Repeat step (402) until the data reaches the head node along the linked list structure; (404) The chain head node sends the data to its father node; (405) The node aggregates the received data and sends the aggregation result to its parent node; (406) Repeat step (405) until the data reaches the base station along the binary tree structure.

Images