Depth-first underwater acoustic sensor network clustering routing method

Abstract

The invention provides a depth-first underwater acoustic sensor network clustering routing method, and relates to the technical field of signal transmission. The method includes: acquiring depth data of a target area; performing height division according to the depth data to obtain a node network layer; performing sub-layer division according to the height of each layer of the node network layer to obtain an upper sub-layer and a lower sub-layer of each layer; determining a cluster head of each layer; and according to the network layer and the sub-layers where the current source node is located, selecting the cluster head of the upper sub-layer in the current network layer or the cluster heads in the two sub-layers of the upper network layer to communicate, and determining an optimal path for information transmission. According to the method, the problem that the service life of the grid is shortened due to increase of network hops and energy consumption in an underwater routing method in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of signal transmission, and in particular to a depth-first underwater acoustic sensor network clustering routing method. Background Art

[0002] With the increase of global marine activities, there is a growing demand for efficient and reliable underwater information collection systems in the fields of underwater environmental monitoring, disaster prevention, resource exploration, etc. As an important underwater information collection technology, Underwater Acoustic Sensor Networks (UASNs) have attracted extensive attention due to their unique advantages in deep-sea exploration and ocean data collection. However, the particularity of the underwater environment, such as signal attenuation, noise interference, low communication bandwidth, and limited energy supply, has brought great challenges to the design and optimization of underwater acoustic sensor networks.

[0003] In recent years, many studies have been devoted to improving the data transmission efficiency and network life of UASNs. At present, many routing protocols developed for UASNs have been proposed, but how to achieve low-energy consumption and efficient data collection is still an unresolved problem. Scholars have introduced different routing methods to reduce network energy consumption and extend the network lifetime as much as possible. Many routing methods are multi-hop routing, and each node transmits and receives data independently. This communication method will cause some nodes to frequently participate in data transmission and quickly exhaust energy, thereby causing uneven energy distribution in the network and even routing holes, thereby shortening the life of the entire network. The clustering method is a network topology optimization for the above problem. This method divides sensor nodes into multiple clusters. Each cluster is responsible for data fusion and centralized transmission and reception by a rotating cluster head (CH) node, which reduces the number of data transmissions and thus reduces the overall energy consumption of the network.

[0004] Unlike ordinary wireless sensor networks, underwater acoustic sensor networks have significant differences in design, implementation, and application. These differences are mainly due to the unique physical characteristics of the underwater environment. Underwater sensor networks mainly rely on sound waves for communication, and when sound waves propagate in water, they are affected by multipath effects and noise interference, resulting in signal distortion and reduced transmission reliability. In addition, the propagation speed of sound waves in water is about 1500m / s, which is much faster than the propagation speed of electromagnetic waves used in ordinary wireless sensor network communications (about 3×10 8 These characteristics make it more difficult to design routing protocols for underwater acoustic sensor networks.

[0005] The purpose of most UASNs is to transmit data collected by sensors at different depths in the underwater environment to sink nodes located on the water surface, that is, to follow the transmission direction from deep to shallow for the next step of analysis and processing. Many underwater routing methods, including some cluster routing methods, more or less have some cases of data transmission from shallow nodes to deep nodes. This measure can improve the delivery rate of data packets to a certain extent and avoid some routing holes, but in theory it is not efficient and will increase the number of network hops and energy consumption, thereby reducing the network life.

[0006] Network life, or network life cycle, is one of the key indicators for measuring the performance of UASNs. It is directly related to the reliability, stability and overall effectiveness of the network. A network with a long life can continuously collect and transmit underwater monitoring data without being affected by node failure caused by battery exhaustion, thus ensuring the continuity and integrity of the data. This is crucial for application scenarios that require long-term monitoring, such as marine environment monitoring, tsunami warning systems, etc. Summary of the invention

[0007] In order to overcome the deficiencies of the prior art, the purpose of the present invention is to provide a depth-first underwater acoustic sensor network clustering routing method, which solves the problem that the underwater routing method in the prior art increases the number of network hops and energy consumption, resulting in a reduction in grid life.

[0008] To achieve the above object, the present invention provides the following solutions:

[0009] A depth-first clustering routing method for underwater acoustic sensor networks, comprising:

[0010] Obtain depth data of the target area;

[0011] Performing height division according to the depth data to obtain a node network layer;

[0012] Divide the node network layer into sub-layers according to the layer heights of each layer to obtain an upper sub-layer and a lower sub-layer of each layer;

[0013] Determine the cluster heads of each layer;

[0014] According to the network layer and sublayer where the current source node is located, it chooses to communicate with the cluster head of the upper sublayer in the current network layer or the cluster heads in the two sublayers of the previous network layer to determine the optimal path for information transmission.

[0015] Preferably, determining the cluster heads of each layer includes:

[0016] Determine a routing table and a neighbor set of each node according to the node network layer;

[0017] Determine candidate cluster heads for each layer in the node network layer according to the routing table and the neighbor set of each node;

[0018] A threshold of the candidate cluster heads is calculated and a cluster head is determined according to the threshold.

[0019] Preferably, the number of layers of the node network layer is expressed as:

[0020]

[0021] Among them, LN is the total number of layers of network division, H area is the height of the monitoring area, i.e., the depth, H layer is the height of each layer.

[0022] Preferably, the expression of the neighbor set is:

[0023]

[0024] Among them, x i ,y i ,z i are the axis coordinate values ​​of node i, x j y j ,z j are the axis coordinates of node j, N i is the neighbor set, and R is the communication radius.

[0025] Preferably, the expression of the threshold is:

[0026]

[0027] T i is the threshold, a, b, c are the first weight factor, the second weight factor and the third weight factor respectively, E res is the remaining energy of the node, E0 is the initial energy of the node, N nei is the number of neighbors of the node, N nei-max is a fixed value, D i The depth of the node's current layer.

[0028] Preferably, the expression of the depth of the current layer of the node is:

[0029] D i =Dpt i oeLh layer ;

[0030] Among them, Dpt i is the depth of node i.

[0031] Preferably, the calculating the threshold of the candidate cluster head and determining the cluster head according to the threshold comprises:

[0032] Determine the threshold of the current candidate cluster head;

[0033] Determine a corresponding timer according to the threshold;

[0034] Determine whether the current candidate cluster head is elected as the cluster head according to the timer.

[0035] The present invention discloses the following technical effects:

[0036] The present invention provides a depth-first underwater acoustic sensor network clustering routing method, including: obtaining the depth data of the target area; performing height division according to the depth data to obtain a node network layer; performing sub-layer division according to the layer height of each layer of the node network layer to obtain an upper sub-layer and a lower sub-layer of each layer; determining the cluster head of each layer; according to the network layer and sub-layer where the current source node is located, selecting to communicate with the cluster head of the upper sub-layer in the current network layer or the cluster heads in the two sub-layers of the previous network layer to determine the optimal path for information transmission. The present invention first layers the network, and then selects candidate cluster heads and cluster heads according to parameters such as the residual energy, number of neighbors, and depth of the node at different levels. The sensor node will select the cluster head to enter the cluster and transmit data according to its own situation. In order to further optimize the routing, we divide each layer into upper and lower sub-layers. The node will select the best cluster head according to the cluster head priority and actual situation, so as to achieve the purpose of reducing the number of network hops and extending the network life. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0038] Figure 1 A flow chart of a depth-first underwater acoustic sensor network cluster routing method provided by an embodiment of the present invention;

[0039] Figure 2 A schematic diagram of the UASN structure provided in an embodiment of the present invention;

[0040] Figure 3 A comparative schematic diagram provided for an embodiment of the present invention, wherein: Figure 3 (a) is a schematic diagram of signal transmission of the existing method. Figure 3 (b) Schematic diagram of signal transmission of the present invention;

[0041] Figure 4 A schematic diagram of the Hello message structure provided in an embodiment of the present invention;

[0042] Figure 5 A schematic diagram of an expected clustering solution provided by an embodiment of the present invention;

[0043] Figure 6 A schematic diagram of the optimized cluster head broadcast message structure provided by an embodiment of the present invention;

[0044] Figure 7 A schematic diagram of routing selection of optimization strategies in different situations provided by an embodiment of the present invention, wherein: Figure 7 (a) is a schematic diagram of cluster head routing selection within the communication range of both the upper and lower sub-layers. Figure 7 (b) is a schematic diagram of cluster head routing selection when there are invalid nodes. Figure 7 (c) is a schematic diagram of cluster head routing selection when the source node a is in the lower sublayer and there is no cluster head in the communication range of its upper layer. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0047] like Figure 1 As shown, the present invention provides a depth-first underwater acoustic sensor network clustering routing method, comprising:

[0048] Step 100: Acquire depth data of the target area;

[0049] Step 200: Perform height division according to the depth data to obtain a node network layer;

[0050] Step 300: Divide the node network layer into sub-layers according to the layer height of each layer to obtain an upper sub-layer and a lower sub-layer of each layer;

[0051] Step 400: Determine the cluster heads of each layer;

[0052] Step 500: According to the network layer and sublayer where the current source node is located, select to communicate with the cluster head of the upper sublayer in the current network layer or the cluster heads in the two sublayers of the previous network layer to determine the optimal path for information transmission.

[0053] Specifically, Figure 2As shown in the figure, the network scenario considered in this paper is a three-dimensional underwater acoustic sensor network. The network is divided into different levels according to the depth. The sensor nodes forward the data packets from the deeper levels to the shallower levels and finally reach the sink node.

[0054] like Figure 3 As shown in the figure, for the routing method applied to UASN, reducing energy consumption and improving data packet delivery rate are the main design goals. In view of the particularity of the underwater environment, acoustic communication is widely used in underwater sensor networks. The propagation speed of sound signals in water is much lower than the propagation speed of radio waves in the air, and it will be interfered by underwater noise and multipath effects, which will aggravate the challenges of node energy consumption and data transmission efficiency. Therefore, designing and improving UASN routing methods is necessary for marine exploration and underwater environment monitoring.

[0055] The superior performance of clustering routing methods in reducing network energy consumption and reducing the number of data transmissions has been proven. At present, many clustering routing methods designed for underwater environments have been proposed. They perform well in reducing the number of data transmissions, but there is a problem with these routing methods, that is, some nodes with shallow depths will send data to the cluster head with deeper depths because they are members of the cluster. This way of data transmission will increase the number of network hops and the number of data transmissions. Our goal is to prevent this from happening and improve the efficiency of data transmission.

[0056] Furthermore, the determining of the cluster heads of each layer includes:

[0057] Determine a routing table and a neighbor set of each node according to the node network layer;

[0058] Determine candidate cluster heads for each layer in the node network layer according to the routing table and the neighbor set of each node;

[0059] A threshold of the candidate cluster heads is calculated and a cluster head is determined according to the threshold.

[0060] The distance from itself to the sink node is estimated at a certain time. Then the network is layered. Since each node has a unique ID and is equipped with a depth sensor, nodes at different depths can be divided into different layers according to the depth data sensed by it, so as to proceed to the next step. When the layer height of the network is clear, the total number of layers of the network can be known.

[0061]

[0062] Where LN is the total number of layers in the network, H area is the height of the monitoring area, i.e., the depth, H layer is the height of each layer.

[0063] After obtaining the network layer information, we can calculate the layer number of each node

[0064]

[0065] Where L i is the layer number of node i, Dpt i is the depth of node i. The total number of layers LN and the node layer number information L i It will be recorded in the storage of node N for subsequent cluster head election and path selection.

[0066] After the network layering is completed, the sink node will broadcast the activation information to all nodes in the network. After receiving the information, the sensor node will broadcast the Hello message to the neighboring nodes within its communication range. The message structure is as follows: Figure 4 The message content includes node ID, remaining energy, depth, node type, and node level. Their meanings are as follows:

[0067] Node ID: unique identifier of the sensor node, used to distinguish different nodes;

[0068] Remaining energy: the current available energy of the node. When it reaches 0, the node dies;

[0069] Depth: The underwater depth of the node, used for node stratification and finding neighbor nodes;

[0070] Node type: used to distinguish nodes of different roles, divided into common nodes, candidate cluster heads and cluster heads;

[0071] Node level: calculated from the node underwater depth, used to indicate the node's level;

[0072] Each node then calculates the number of neighbors within its communication range based on the number of messages received from different nodes, calculates the distance based on the time difference between sending and receiving messages, and integrates all the above information into a neighbor table, which will be updated and maintained at an appropriate time in the future. It should be noted that in the method proposed in this paper, neighbors refer to nodes within the communication range of the current node and located at a deeper level than the current node.

[0073] Let the node set be N = {1,2,...,i,...,k}, k is the number of nodes, and the number of layers of node i is L i , the coordinates are (x i ,y i ,z i ), the communication radius is R, then the neighbor set of node i can be expressed as:

[0074]

[0075] Candidate cluster heads are nodes that have the conditions to become cluster heads. They are used to optimize the cluster head selection method and narrow the range of cluster head selection. At each layer of the network, candidate cluster heads are determined based on the node residual energy, number of neighbors, and depth. The more residual energy, the more neighbors, and the shallower the depth, the higher the probability that the node will be selected as the cluster head node. Each node i calculates a threshold T based on the above information. i , this threshold is used to determine the probability of a node being elected as a cluster head. i Before that, we need to calculate the relative depth D of node i. i , which refers to the depth of the node in its current layer, is calculated as follows

[0076] D i =Dpt i oeLh layer (4)

[0077] Among them, D i is the relative depth of the node, which is positive (the water surface depth is 0, and the deeper the depth, the greater the depth value), then T i The calculation is as follows:

[0078]

[0079] Among them, E res is the node residual energy, E0 is the node initial energy, N nei is the number of neighbors of the node, N nei-max is a fixed value that is greater than the number of neighbors of each node. a, b, and c are weight factors, c is a negative number, and a+b+c=1. i After that, each node generates a random number n between 0 and 1. i >n, node i is selected as a candidate cluster head in this round, otherwise node i serves as an ordinary sensor node in this round.

[0080] Further, the calculating the threshold of the candidate cluster head and determining the cluster head according to the threshold includes:

[0081] Determine the threshold of the current candidate cluster head;

[0082] Determine a corresponding timer according to the threshold;

[0083] Determine whether the current candidate cluster head is elected as the cluster head according to the timer.

[0084] When a node is elected as a candidate cluster head, it broadcasts the election information to its neighbors. Its message structure is the same as that of the Hello message to update the neighbor table. Since only the basic parameter information of each node is comprehensively considered when selecting candidate cluster heads, the location information between candidate cluster heads is not considered, which may cause the coverage areas of two or more candidate cluster head nodes to overlap, thus increasing the energy consumption of redundant cluster heads and reducing the network communication efficiency. Each candidate cluster head sets a timer TM according to its threshold i The larger the threshold, the smaller the timer, as follows

[0085]

[0086] where α is the first adjustment factor

[0087] When all candidate cluster heads have assigned their own timers, the timers start counting down, and the timer of the candidate cluster head with a larger threshold will reach zero earlier. When the timer of a certain candidate cluster head C1 reaches zero, it will be elected as the cluster head and make the following judgments according to the number of other candidate cluster heads C2 within the communication range and at the same layer

[0088] If there are no other candidate cluster heads within the communication range, C1 is immediately elected as the cluster head

[0089] If there is one or more other candidate cluster heads within the communication range, C1 is elected as the cluster head, reads the distances to each candidate cluster head in the neighbor table, and makes different judgments according to the following formula

[0090] r = R * ω (7)

[0091] where r is the radius threshold used by the current candidate cluster head for making judgments, R is the node communication radius, and ω is the adjustment factor. When the distance d between them is < r, the timer of C2 is destroyed and its candidate cluster head qualification is cancelled because it is too close to the cluster head and there is too much overlap in their coverage areas; when the distance d between them is > r, C2 maintains its candidate cluster head status and waits for the timer to reach zero for the next operation

[0092] The current threshold T of the cluster head is considered during the clustering and cluster head selection phases i . When all cluster head nodes are determined, the current threshold T is calculated according to formula (5) i , and the election information is broadcast to other nodes within the communication range. During a fixed message reception waiting time t, ordinary sensor nodes may receive multiple broadcast messages, but they will only process the messages from nodes at a higher level than them and, after comparing the threshold T of the message broadcaster i , select T iThe largest cluster head sends a cluster entry application message to it, and the cluster head feeds back an Ack message after receiving the message. In the method we proposed, selecting the cluster head is actually selecting the next hop transmission node. The expected clustering scheme is as follows Figure 5 .

[0093] The cluster heads in the clustering method will have a greater load than ordinary sensor nodes, so it is necessary to rotate the cluster heads at the right time to balance the energy consumption between nodes and avoid the premature death of some nodes due to excessive energy consumption, thereby reducing the impact on the entire network. Cluster heads will be rotated in the following two situations:

[0094] The cluster head fails and cannot complete the existing task;

[0095] The cluster head energy is lower than the threshold E th .

[0096] Energy threshold E th It changes dynamically according to the remaining energy of neighboring nodes around the cluster head, and is calculated as shown in formula (8).

[0097]

[0098] in is the average residual energy of the cluster head neighbor nodes, ρ is the adjustment factor, and its value is (0,1). When ρ is larger, the energy consumption between nodes is more balanced, but the frequency of cluster head replacement will increase; when ρ is smaller, the frequency of cluster head replacement can be reduced, but the energy consumption between nodes will be more unbalanced.

[0099] The cluster head will periodically broadcast the remaining energy information to the neighboring nodes in the same layer, and the neighboring nodes will reply with their own remaining energy information after receiving it. When the cluster head energy is lower than the threshold or fails to respond, it will give priority to selecting the nodes that lost in the previous cluster head election. The selection basis is still the node threshold T i , T i The node with the largest value is given priority. If there is no candidate cluster head competing with the cluster head before, T is selected from the neighbors at the same layer. i The largest node is selected as the new cluster head. When the new cluster head is selected, the old cluster head is delisted and becomes a common sensor node.

[0100] In underwater acoustic sensor networks, the number of network hops and packet delivery ratio (PDR) of routing methods are two important indicators for evaluating the quality of the methods. The number of network hops refers to the number of intermediate nodes that data passes through during transmission from the source node to the destination node. Fewer hops usually means a more direct path, which can reduce the possibility of delays and packet loss. PDR refers to the ratio of the number of packets successfully transmitted to the destination node to the total number of packets sent by the source node, usually expressed as a percentage. A higher PDR means that most packets are able to successfully reach the destination, indicating that the network reliability is higher.

[0101] In order to reduce the number of network hops and improve PDR, we propose a routing optimization strategy based on layering. In this method, each layer is evenly divided into two sub-layers according to the layer height. When a node is elected as a cluster head, it will i The value, the layer and sub-layer information are broadcast to neighboring nodes. All nodes will store the cluster head broadcast messages of the two sub-layers of the previous layer and the sub-layer of the same layer in the table for selecting the best cluster head or the optimal path. The message structure of the cluster head broadcast is as follows: Figure 6 .

[0102] The sublayers are divided according to the relative depth of the node in the hierarchy. When the relative depth of the node is less than half of the layer height, it belongs to the upper sublayer, represented by a value of 1; otherwise, it belongs to the lower sublayer, represented by a value of 2.

[0103]

[0104] The addition of sublayers makes the cluster head broadcast information received by sensor nodes divided into upper and lower sublayers. In order to further improve the possibility of nodes finding cluster heads, we added cluster heads located in the same layer as the nodes in the optimization strategy, provided that the cluster heads are located in the upper sublayer and the sensor nodes are located in the lower sublayer of the same layer. The priority of sensor nodes in selecting cluster heads is

[0105] The upper sublayer of the upper layer

[0106] The lower sublayer of the upper layer

[0107] The upper sublayer of the same layer (if any)

[0108] Use a diagram to describe Figure 7 , a complete layer is divided into an upper sublayer and a lower sublayer, node a is the source node, and A, B, and C are the cluster heads near it. Figure 7 In (a), when the upper and lower sub-layers of the source node a have cluster heads within the communication range, a will give priority to the cluster head of the upper sub-layer, which can avoid unnecessary transmission times and reduce the number of network hops; Figure 7In (b), cluster head A becomes an invalid node due to energy exhaustion or failure, and can no longer effectively forward messages for a. At this time, a will choose cluster head B located in the lower sublayer of the upper layer to forward the information, and then B will forward it upward to cluster head C to avoid data loss and improve PDR. Figure 7 In (c), the source node a is in the lower sublayer, and there is no cluster head in the communication range in its upper layer. At this time, a will choose cluster head A in the upper sublayer of the same layer to transmit data, and A will also give priority to cluster head B in the upper sublayer one level higher than it when transmitting, rather than cluster head C in the lower sublayer of the upper layer.

[0109] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0110] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A depth-first clustering routing method for underwater acoustic sensor networks, characterized in that: include: Obtain depth data of the target area; Performing height division according to the depth data to obtain a node network layer; Divide the node network layer into sub-layers according to the layer heights of each layer to obtain an upper sub-layer and a lower sub-layer of each layer; Determine the cluster heads of each layer; According to the network layer and sublayer where the current source node is located, choose to communicate with the cluster head of the upper sublayer in the current network layer or the cluster heads in the two sublayers of the previous network layer to determine the optimal path for information transmission.

2. A depth-first clustering routing method for underwater acoustic sensor networks according to claim 1, characterized in that: Determining the cluster heads of each layer includes: Determine a routing table and a neighbor set of each node according to the node network layer; Determine candidate cluster heads for each layer in the node network layer according to the routing table and the neighbor set of each node; A threshold of the candidate cluster heads is calculated and a cluster head is determined according to the threshold.

3. A depth-first clustering routing method for underwater acoustic sensor networks according to claim 1, characterized in that: The expression of the number of layers of the node network layer is: Among them, LN is the total number of layers of network division, H area is the height of the monitoring area, i.e., the depth, H layer is the height of each layer.

4. A depth-first clustering routing method for underwater acoustic sensor networks according to claim 2, characterized in that: The expression of the neighbor set is: Among them, x i ,y i ,z i are the axis coordinate values ​​of node i, x j y j ,z j are the axis coordinates of node j, N i is the neighbor set, and R is the communication radius.

5. A depth-first clustering routing method for underwater acoustic sensor networks according to claim 3, characterized in that: The expression of the threshold is: T i is the threshold, a, b, c are the first weight factor, the second weight factor and the third weight factor respectively, E res is the remaining energy of the node, E0 is the initial energy of the node, N nei is the number of neighbors of the node, N nei-max is a fixed value, D i The depth of the node's current layer.

6. A depth-first clustering routing method for underwater acoustic sensor networks according to claim 5, characterized in that: The expression of the depth of the node at the current layer is: D i =Dpt i modH layer ; Among them, Dpt i is the depth of node i.

7. A depth-first clustering routing method for underwater acoustic sensor networks according to claim 5, characterized in that: The step of calculating the threshold of the candidate cluster heads and determining the cluster heads according to the threshold comprises: Determine the threshold of the current candidate cluster head; Determine a corresponding timer according to the threshold; Determine whether the current candidate cluster head is elected as the cluster head according to the timer.

Images