Clustering routing method for wireless sensor networks based on square optimal scale dual cluster heads
By adopting a cluster routing method based on square optimal scale dual cluster heads in wireless sensor networks, the problems of energy waste and uneven distribution of cluster heads in traditional designs are solved, and more efficient data transmission and network life extension are achieved.
Patent Information
- Application Number
- CN202210489759.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-05-07
- Publication Date
- 2024-11-12
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The inter-cluster routing design of traditional wireless sensor networks results in large amounts of energy loss and uneven distribution of cluster heads, resulting in high network energy consumption and affecting network life.
Using a wireless sensor network cluster routing method based on square optimal scale dual cluster heads, the optimal cluster head size is calculated through the energy model, dynamically compete for the radius of the cluster into the cluster, select an effective secondary cluster head, and optimize the inter-cluster routing relay nodes. Position and energy selection to achieve efficient data transmission.
It effectively extends the survival time of nodes, reduces network energy consumption, improves data transmission efficiency, and extends the energy life of the network.
Smart Images

Figure CN115087070B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a clustering routing method for reducing network energy consumption, in particular to a clustering routing method for a wireless sensor network based on a square optimal scale double cluster head.
[0002] The traditional inter-cluster routing design selects a single-hop method between the cluster head and the base station, and forwards data point-to-point with the base station. It does not make too much design on the inter-cluster routing, which will inevitably lead to a large amount of energy loss. Background Art
[0003] Wireless sensor networks have been widely used in various fields. The research on cluster routing has become an indispensable factor in extending the life of wireless sensor networks. Traditional cluster routing has problems such as uneven distribution of cluster heads, which leads to accelerated energy loss. Summary of the invention
[0004] The purpose of the present invention is to provide a clustering routing method for wireless sensor networks based on a square optimal scale dual cluster head. The method calculates the optimal cluster head scale under the dual cluster head according to the energy model, avoids blindly determining the ratio and causing unnecessary loss of network energy, improves the threshold formula, dynamically competes for the radius to enter the cluster, selects an effective secondary cluster head after clustering, performs data collection and transmission tasks, and optimizes the location and energy of the relay node selection of the inter-cluster routing. The method can effectively increase the survival time of the node and also prolong the network energy, thereby ensuring that more data is transmitted to the destination.
[0005] The objective of the present invention is achieved through the following technical solutions:
[0006] Based on the square optimal scale dual cluster head wireless sensor network cluster routing method, the method includes the optimal cluster head scale under dual cluster heads, cluster head selection, cluster formation, deputy cluster head selection, and inter-cluster routing design.
[0007] a. Determine the optimal number of cluster heads K opt :
[0008] (8)
[0009] From the final formula (8), we can see that the optimal number of cluster heads K opt Determination of M, N, E ELEC 、E DA , d0, after initializing the parameters, the optimal number of cluster heads under dual cluster head communication needs to be calculated, which can improve the fault tolerance of network deployment;
[0010] b. Cluster selection:
[0011] p is the optimal cluster head size ratio, and the calculation formula of U is (13):
[0012] (13)
[0013] c. Cluster formation:
[0014] The cluster head publishes cluster entry information announcements. After receiving the cluster entry data information, the node selects the cluster to join according to the competition radius, sends a cluster entry request to the cluster head, and waits for the cluster head to respond before joining the cluster. For nodes that receive cluster entry messages broadcast by multiple cluster heads at the same time, the node judges based on the distance energy function and sends a cluster entry request based on the function value.
[0015] d. Selection of deputy cluster heads:
[0016] After the clusters are established, the deputy cluster heads are elected in each cluster. The deputy cluster heads share the energy consumption of the individual cluster heads to maximize the energy utilization within the entire cluster, between clusters, and between cluster heads and base stations. Each cluster must elect the node with the best energy and the best distance from the cluster head among the nodes outside the cluster head to be the deputy cluster head.
[0017] e. Inter-cluster routing design:
[0018] The square area is used for dual cluster heads to forward data. The deputy cluster head obtains information from other nodes in the cluster, and performs data fusion and redundancy. Finally, the most concise and accurate information is transmitted to the head node in the cluster.
[0019] The square optimal scale dual cluster head wireless sensor network cluster routing method described above selects a suitable relay jump cluster head for information transmission between the cluster head and the base station, and finally transmits the data to the base station or directly transmits the collected information to the base station; the next hop of the cluster head has only two options, one is to send the next cluster head, and the other is to directly transmit to the base station.
[0020] The square optimal scale dual cluster head wireless sensor network cluster routing method, the next cluster head should be selected with high energy, that is, W1=E i / E0, the ratio of currently consumed remaining energy to initial energy is high.
[0021] The square optimal scale dual cluster head wireless sensor network cluster routing method, the next candidate forwarding node is closer to the cluster head, that is, W2=d relayNode / d CHavg ,The ratio of the next hop forwarding distance and the average distance from other cluster heads to the next hop is high.
[0022] The square optimal scale dual cluster head wireless sensor network cluster routing method, the selection of the cluster head close to the base station, according to the process steps of the method, if the reverse distance propagation is used, it will cause unnecessary loss, that is, W3=d iCHtoBs -dRelayNodetoBS ; Data packets have higher forwarding efficiency, that is, W4=SP i / SP i +RP i ; Select the function Select for this:
[0023] (16)
[0024] Where b1, b2, b 3, b4 is a weight factor and satisfies b1+b2+b3+b4=1.
[0025] The advantages and effects of the present invention are:
[0026] The present invention adopts a WSNs clustering routing method based on square optimal scale and double cluster heads (SOSCH). It is used in the monitoring area, and the optimal scale of the cluster head is determined according to factors such as the number of nodes and the distance between the relay forwarding node. In the cluster head election process, the cluster head election threshold formula in the LEACH (Low Energy Adaptive Clustering Hierarchy) protocol method is optimized, and the distance, the remaining energy consumed by the node and the node density conditions are compared with the randomly generated number. The ordinary node dynamically enters the cluster with the competition radius limited by energy, and after the cluster group is established, it will run for the deputy cluster head, share the data collection and fusion energy consumption of the main cluster head, and forward it to the cluster head; in the inter-cluster routing stage, the cluster head transmits data in a multi-hop and single-hop manner, and the energy and position selection of the transit node can improve the efficiency of inter-cluster data transmission. Through experimental analysis, this method can promote the life of the monitoring area network and effectively reduce the network energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 This is a flow chart of the network clustering routing method of the present invention. DETAILED DESCRIPTION
[0028] The present invention will be described in detail below with reference to the embodiments shown in the accompanying drawings.
[0029] The invention includes several completion processes including optimal cluster head scale under dual cluster heads, primary cluster head selection, cluster formation, secondary cluster head selection, and inter-cluster routing design;
[0030] Optimal cluster head size under dual cluster heads
[0031] The SOSCH method mentioned in the present invention adopts the concept of "round" in the LEACH protocol. The entire method process includes three parts: cluster head selection, deputy cluster head selection, cluster composition and inter-cluster routing. 2In the square network monitoring model, it is assumed that the nodes deployed in the model are evenly distributed and divided into K clusters. The N nodes are evenly distributed in the K clusters. From the energy consumption model, it can be seen that the total energy consumption in the system is generated by three types of nodes, namely ordinary nodes, sub-cluster heads, and main cluster heads. It is assumed that all data transmission is within the threshold d0. The loss of ordinary nodes mainly completes the consumption of the formula. Specifically, the energy loss of ordinary nodes in each cluster is E Ordnode It manifests as:
[0032] (1)
[0033] Where E ELEC is the energy consumption of sending and receiving 1 bit of data, ɛ fs Energy loss coefficient of the free space model, a is the packet size, d is toSCH is the distance from the common node to the vice cluster head in each cluster. The condition is uniformly distributed before, so the density of nodes in the entire area is , so the distance d from each node in each cluster to the cluster head can be calculated by integration toSCH Values:
[0034] (2)
[0035] The limited area of the double integral in the formula is to regard each cluster group as a circle with the cluster head as the center, and find d toSCH for:
[0036] (3)
[0037] For the secondary cluster head, each round requires a lot of consumption in receiving data from common nodes in the cluster group, fusing the data, and forwarding to the main cluster head. The cluster head is also a member of the cluster group, and the forwarding distance to the main cluster head can also be regarded as a formula. The energy consumption of the secondary cluster head is specifically expressed as follows:
[0038] (4)
[0039] Where E DA The energy loss for fusing 1 bit of data. Since dual cluster heads are used to complete network data transmission, the main cluster head needs to receive the member information of the secondary cluster head and forward it to the next hop selection node under the routing rule (other cluster heads or base stations except the main cluster head itself). The energy consumption of the main cluster head is specifically expressed as follows:
[0040] (5)
[0041] The scale design is that the propagation distance between all nodes that need to communicate is within the threshold d0, then d Realynode The following formula must be satisfied:
[0042] (6)
[0043] Where area C refers to the circular area with the central base station as the center and meeting the threshold d0. The consumption of the entire network is the energy consumption of all clusters E cluster The sum of E all :
[0044] (7)
[0045] Substitute equations (1), (4), and (5) into (7) and obtain the partial derivative for the optimal number of cluster heads K. When the function value of the equation is zero, the optimal number of cluster heads K is determined. opt :
[0046] (8)
[0047] From the final formula (8), we can see that the optimal number of cluster heads K opt Determination of M, N, E ELEC 、E DA , d0. After initializing the parameters, the optimal number of cluster heads under dual cluster head communication needs to be calculated, which can improve the fault tolerance of network deployment.
[0048] Cluster First Pick
[0049] The energy consumption model formula shows that in the process of data transmission, the distance factor plays an essential role in the overall consumption. The smaller the distance between data transmission nodes, the less energy is consumed. Conversely, the energy consumption will increase. As the number of nodes increases gradually, the communication capability of the nodes will weaken as the node energy decreases. The competition radius cannot be ignored. Considering the energy of the nodes, this paper adopts an M*M square monitoring area, and its area relationship is shown in formula (9):
[0050] (9)
[0051] (10)
[0052] The competition radius is affected by energy. Based on formula (10), the energy relationship is added to limit the size of the competition radius. The optimal competition radius is as shown in formula (11):
[0053] (11)
[0054] Where E0 represents the initial energy of the election node, E i is the current energy of the node.
[0055] In the process of cluster head election, in order to achieve better optimization, more energy saving, and avoid the situation of local cluster heads, a control function U is set, and the function is set in combination with the distance between the node and the base station node, as well as the node energy and node density factors. Because the cluster head node has a heavy task to complete, it needs to handle the data collection and fusion of the other nodes in the cluster, and communicate with other cluster heads. Therefore, the energy of the node is the primary consideration. The larger the cluster head energy, the more data it can process, and the network will not collapse prematurely due to cluster head energy problems; the closer the cluster head node is to the base station, the less energy it consumes in the energy consumption model; the more neighbor nodes around the node, the more dense the density of the cluster head is, so it can be determined that the distance is small and the energy consumption is reduced in a small space. The method proposes a new threshold formula for this:
[0056] (12)
[0057] Where: p is the cluster head probability, that is, the ratio of the number of clusters to the number of nodes in the entire network; r is the number of rounds; G represents the nodes that are not selected in the round cycle; p=K opt / N node , p is the optimal cluster head size ratio, and the calculation formula of U is (13):
[0058] (13)
[0059] In the formula, α, β, and γ are the weight coefficients of the control function, and α+β+γ=1. The three weight coefficients can be used to change the proportion of node energy, the distance between the node and the base station, and the number of neighbor nodes; E i (r) represents the energy of the current node; E avg (r) represents the average node energy; d max Interpreted as the distance between the node and the base station that is farthest away; d min Interpreted as the distance between the node and the closest node to the base station; d i Interpreted as the distance between the current node and the base station.
[0060] Cluster formation
[0061] The cluster head publishes cluster entry information announcements. Nodes that receive cluster entry data information choose to join the cluster according to the competition radius, and send cluster entry requests to the cluster head. They need to wait for the cluster head to respond before they can join the cluster. For nodes that receive cluster entry messages broadcast by multiple cluster heads at the same time, they judge based on the distance energy function and send cluster entry requests based on the function value. The distance energy function is as follows:
[0062] (14)
[0063] In the formula, a1 and a2 are weight factors, a1+a2=1; E CHres represents the cluster head node energy, d BStoi Represents the distance between the node and the base station, d CHtoi Represents the distance between the node and the cluster head. Because energy is the preferred factor, if the ratio of the current energy of the node to the initial energy is high, it means that it has consumed less energy before, so the current energy of the node is dominant and is capable of completing data forwarding and other tasks for the next cluster members; the ratio of the distance from the node to the base station to the distance between the cluster head and the node entering the cluster, the larger the ratio, the closer the node is to the cluster head, and the cluster head is capable of collecting node data with less energy consumption.
[0064] Selection of deputy cluster heads
[0065] After the cluster is established, the next step is to elect a sub-cluster head in each cluster. This method introduces a sub-cluster head to share the energy consumption of a single cluster head to maximize the energy utilization within the entire cluster, between clusters, and between the cluster head and the base station. Within each cluster, the node with the best energy and the best distance from the cluster head must be elected from the nodes outside the cluster head, namely the sub-cluster head. The sub-cluster head is used to collect data from the nodes in the cluster, and at the same time, the data must be fused and simplified before being forwarded to the cluster head. Therefore, the sub-cluster head must meet the optimal energy within the cluster and optimize the positions of other nodes. The restriction function DCH(E,d) is proposed, and the expression is as follows:
[0066] (15)
[0067] Where: d itoCH Indicates the distance from a node in the cluster to the cluster head; d avg represents the average distance from the nodes in the cluster to the cluster head; E i is the remaining energy of a node; E cavg represents the average residual energy of the nodes in the cluster; E CH Represents the residual energy of the cluster head node in the cluster; m1 and m2 are parameters. For the election of the deputy cluster head node, it is necessary to ensure that the distance between the deputy cluster head and the cluster head is optimal, the energy of the node is above the average energy loss residual value, and at the same time, there is not too much energy difference with the energy of the main cluster head, so as to ensure better collection of the deputy cluster head and reduce the energy consumption rate.
[0068] Inter-cluster routing design
[0069] The present invention adopts a square area to forward data with dual cluster heads. The deputy cluster head absorbs the information of other nodes in the cluster group, and performs data fusion and redundancy. Finally, the most concise and accurate information is transmitted to the head node in the cluster group. The information transmission between the cluster head and the base station selects a suitable relay jump cluster head, and finally transmits the data to the base station or directly transmits the collected information to the base station. There are only two options for the next hop of the cluster head, one is to the next cluster head, and the other is to directly transmit to the base station. For the selection of the next cluster head node, it is necessary to select a high energy, that is, W1=E i / E0, the ratio of the current remaining energy consumed to the initial energy is high; the next candidate forwarding node is close to the cluster head, that is, W2=d relayNode / d CHavg , the next-hop forwarding distance is higher than the average distance from other cluster heads to the next hop; at the same time, the cluster head close to the base station is selected in terms of geographical location. If the reverse distance propagation is used, it will cause unnecessary loss, that is, W3=d iCHtoBs -d RelayNodetoBS ; The in-and-out degree of the cluster head node affects the packet forwarding relationship, SP i For packets forwarded by cluster head nodes, RP i For cluster head nodes to receive data packets, if the ratio of forwarded data packets to total data packets is high, it can be proved that the data packets have higher forwarding efficiency, that is, W4=SP i / SP i +RP i ; Select the function Select for this:
[0070] (16)
[0071] Where b1, b2, b 3, b4 is the weight factor and satisfies b1+b2+b3+b4=1; the specific process is as follows Figure 1 shown.
Claims
1. A clustering routing method for wireless sensor networks based on a square optimal scale dual cluster head, characterized in that: The method includes the optimal cluster head scale under dual cluster heads, the selection of the first cluster head, the formation of clusters, the selection of the deputy cluster heads, and the inter-cluster routing; Use M*M m 2 Square network monitoring model, assuming that the nodes deployed in the model are evenly distributed and divided into K clusters, N nodes are evenly distributed in K clusters. From the energy consumption model, it can be seen that the total energy consumption in the system is generated by three types of nodes, namely ordinary nodes, deputy cluster heads, and main cluster heads, and it is assumed that all data transmission is within the threshold d0 range; a. Determine the optimal number of cluster heads K opt : Where E ELEC is the energy consumption of sending and receiving 1 bit of data, ε fs Energy loss coefficient for the free space model, E DA The energy consumption required to fuse 1 bit of data; From the final formula (8), we can see that the optimal number of cluster heads K opt Determination of M, N, E ELEC 、E DA , d0, after initializing the parameters, the optimal number of cluster heads under dual cluster head communication needs to be calculated, which can improve the fault tolerance of network deployment; b. Cluster selection: p is the optimal cluster head size ratio, and the calculation formula of U is (13): In the formula, α, β, and γ are the weight coefficients of the control function, α+β+γ=1, and the proportion of node energy, the distance between the node and the base station, and the number of neighbor nodes can be changed by the three weight coefficients; E i (r) represents the energy of the current node; E avg (r) represents the average node energy; d max Interpreted as the distance between the node and the base station that is farthest away; d min Interpreted as the distance between the node and the closest node to the base station; d i Interpreted as the distance between the current node and the base station; c. Cluster formation: The cluster head publishes cluster entry information announcements. After receiving the cluster entry data information, the node selects the cluster entry according to the competition radius and sends a cluster entry request to the received cluster head. It needs to wait for the cluster head to respond before entering the cluster. For nodes that receive cluster entry messages broadcast by multiple cluster heads at the same time, they judge according to the distance energy function and send cluster entry requests according to the function value. The distance energy function is as follows: In the formula, a1 and a2 are weight factors, a1+a2=1; E CHres represents the cluster head node energy, d BStoi Represents the distance between the node and the base station, d CHtoi represents the distance between the node and the cluster head; E0 represents the initial energy; d. Selection of deputy cluster heads: After the clusters are established, the next step is to select a secondary cluster head in each cluster. The secondary cluster head is to share the energy consumption of the individual cluster head to maximize the energy utilization within the entire cluster, between clusters, and between the cluster head and the base station. Each cluster must elect the node with the best energy and the best distance from the cluster head among the nodes outside the cluster head to be the secondary cluster head. e. Inter-cluster routing design: The square area is used for dual cluster heads to forward data. The deputy cluster head obtains information from other nodes in the cluster, and performs data fusion and redundancy. Finally, the most concise and accurate information is transmitted to the head node in the cluster.
2. The square optimal scale dual cluster head wireless sensor network cluster routing method according to claim 1 is characterized in that: The information transmission between the cluster head and the base station selects a suitable relay to jump to the cluster head, and finally transmits the data to the base station or directly transmits the collected information to the base station; there are only two options for the next hop of the cluster head, one is to the next cluster head, and the other is to directly transmit to the base station.
3. The square optimal scale dual cluster head wireless sensor network cluster routing method according to claim 1, characterized in that: For the selection of the next cluster head node, a cluster head node with high energy should be selected, that is, W1 = E i / E0, current remaining energy consumption E i The ratio to the initial energy E0 is high.
4. The square optimal scale dual cluster head wireless sensor network cluster routing method according to claim 3, characterized in that: The next candidate forwarding node is closer to the cluster head, that is, W2 = d relayNode / d CHavg , the next hop forwarding distance d relayNode The average distance d from other cluster heads to the next hop CHavg The ratio is high.
