CN103220216B - A kind of wireless body area network route method based on Region dividing - Google Patents

Abstract

The present invention provides a wireless body area network routing method based on area division, which is applicable to different wireless body area network application scenarios, and utilizes the characteristics of sufficient energy of the wireless body area network convergence node to divide the network monitoring area into multiple sub-areas Form a cluster, select the appropriate node in each cluster as the cluster head of the current argument and the cluster head of the next round according to the position information and energy information of the node in the human body, and select the appropriate node at the same time considering factors such as energy and communication costs. The node acts as the next hop node of the cluster head, and sends data to the sink node in a multi-hop manner between the cluster heads. The invention can effectively improve the information transmission efficiency of the wireless body area network node and the network life cycle.

Description

A Routing Method for Wireless Body Area Networks Based on Area Division

technical field

The invention relates to a wireless body area network clustering routing method based on area division, which mainly uses the area division method to improve the information transmission efficiency of wireless body area network nodes, and belongs to the cross technical field of computer network, wireless body area network and embedded system .

Background technique

Advances in microelectronics, computers, and wireless communication technologies have promoted the rapid development of low-power multifunctional sensors, enabling them to integrate multiple functions such as information collection, data processing, and wireless communication in a small volume. Wireless sensor networks are composed of The sensor nodes deployed in the monitoring area form a multi-hop self-organizing network system through wireless communication. The purpose is to cooperatively perceive, collect and process the information of the sensing objects in the network coverage area, and send it to the observer. Wireless sensor networks integrate the logical information world with the objective physical world, changing the way humans interact with nature, and are listed as one of the top ten technologies that will change the world in the 21st century. Wireless sensor network is a multi-disciplinary and comprehensive scientific research field, which can monitor, sense and collect information of various environments and detection objects in its distributed area in real time, and can also process these information, through wireless The mode is sent through the aggregation node, and then transmitted to the monitoring host or the user who needs to use the information.

Wireless sensor networks are developing rapidly and can be widely used in public security, ecological environmental protection, environmental detection, emergency command and biomedical fields. Sensor technology has been applied in the field of biomedicine for a long time, but it has not brought great convenience to people. Traditional medical monitoring uses wired sensors, which transmit physiological data and other information to the medical center to achieve the effect of monitoring. However, this medical monitoring system limits the mobility of monitored objects, and also affects people's normal life for those objects who need long-term medical monitoring and health care. The introduction of wireless technology has enabled rapid development in the field of modern healthcare. Medical sensors are worn on the human body or implanted in the human body to monitor the medical data of the patient, and transmit these medical data to the data aggregation center through low-power radio frequency technology, and process the data through the aggregation node or transmit it to medical monitoring Center, this type of sensor network is the wireless body area network. Compared with the usual sensor network, due to the introduction of human body media and the particularity of medical applications, this network has specific requirements in terms of scale, reliability, and continuity. Of course, energy saving is also the primary consideration of wireless body area networks. .

In the wireless body area network with cluster structure, the cluster head is responsible for data collection, aggregation processing and inter-cluster forwarding tasks, and its energy consumption is much greater than that of cluster member nodes. In order to balance the energy consumption of nodes in the network, the clustered wireless body area network generally operates in rounds, and the cluster head is rotated at the beginning of each round. Intra-cluster communication and inter-cluster routing; intra-cluster communication is generally based on time-division multiple access scheduling, and the cluster head allocates different communication time slots for each member node. Communication interference between adjacent clusters is generally avoided through two-layer TDMA scheduling or multiple channels; data forwarding between cluster heads generally uses carrier detection multiple access to avoid channel contention. In most clustered wireless body area networks, cluster members communicate with the cluster head in a single-hop manner, so there is generally no routing problem in intra-cluster communication, while data transmission between cluster heads mostly uses multi-hop methods, so cluster head routing Protocol is essential. Moreover, in the two-layer data collection mode of the clustered wireless body area network, the failure of a cluster head routing will cause all the sensing data of a cluster for a period of time to be lost. important.

Contents of the invention

Technical problem: The characteristics of the clustered wireless body area network make the cluster head routing design face many problems. At present, many cluster head selection methods will cause the following results: the distribution of the generated clusters in the network is not uniform enough, which increases the overall energy of the network. consumption, reducing the lifetime of the network. The object of the present invention is to provide a wireless body area network clustering routing method based on area division, which improves the cluster head selection, inter-cluster data transmission mode, etc., and solves the above problems.

Technical solution: In the clustering routing method based on area division in the present invention, the clustering wireless body area network is a distributed network composed of self-organized sensor nodes that are worn on the human body or implanted in the human body and have unique identifiers , the entire network is logically divided into clusters, each cluster includes cluster head nodes and non-cluster head nodes, and the sink node can communicate wirelessly with all sensor nodes and send and receive human body information.

The wireless body area network routing method based on area division includes the following steps:

1) After all the nodes of the wireless body area network are started, each sensor node will set the pre-defined variable of the selected identification variable of the cluster head and the selected identification variable of the next round of cluster heads to 0, and transmit its own identification, current energy and position to the sink node Coordinate information; all nodes of the wireless body area network include aggregation nodes and sensor nodes, and all sensor nodes form clusters, each cluster has its own nodes in the cluster, and these nodes in the cluster include cluster heads and non-cluster head nodes, wherein A certain sensor node acts as the cluster head, and other sensor nodes are non-cluster head nodes.

2) After the sink node receives the sensor node information, it stores the information, divides the entire network area evenly into n sub-areas according to the preset optimal number of cluster heads n, and assigns an identifier to each sub-area;

3) After the sub-area division is completed, the aggregation node divides each sensor node into its corresponding sub-area according to its position in the human body, forms a corresponding cluster, and sends the identification of the sub-area to the node;

4) The converging node calculates the centroid coordinates of each sub-area in each sub-area, and the distance between the nodes of each sub-area and the centroid;

5) The converging node selects the sensor node closest to its centroid and with the smallest identity in each sub-area as the cluster head in the first round, and sets the selected identification variable of the cluster head of the sensor node to 1; selects the sensor node closest to the centroid As the cluster head of the next round, the selected identification variables of the next round of the cluster head of the sensor node are all set to 1;

6) The aggregation node broadcasts information to all sensor nodes in the network. This information includes the belonging cluster of each sensor node in the first round, whether it is the first round of cluster head information, and whether it is the next round of cluster head information;

7) All sensor nodes store the information after receiving the information from the sink node, and check whether they are set as the first-round cluster head according to the information. If they are cluster heads, they allocate time slots for non-cluster head nodes in their own clusters , generate a time slot allocation table, then send the time slot allocation table information to non-cluster head nodes in the cluster, and then enter step 8);

8) Each non-cluster head node continuously collects the monitored information in its own time slot according to the time slot allocation table and sends the perceived data to its cluster head as needed, and closes the communication module when it is not in its own time slot;

9) Each cluster head collects the data of all non-cluster head nodes in the cluster, and performs data fusion on these data, and also receives data forwarded by other cluster heads;

10) Each cluster head sends the fused data in the cluster or the data forwarded by other cluster heads to the sink node, and the operations adopted by each cluster head are as follows:

101) Broadcasting own identification, current energy consumption, cluster head identity information, receiving and recording corresponding information sent by adjacent cluster heads;

102) Select other cluster heads one by one in the cluster head set of neighbors, and calculate the communication cost of transmitting data to the sink node through the selected cluster head;

103) Among the cluster heads whose communication cost is less than the communication cost of directly transmitting data to the sink node, select the cluster head with the largest energy value and the smallest identification as the relay node, and forward the data to this cluster head; if there is no qualified As the relay node, the cluster head directly transmits the data to the sink node;

104) By selecting a relay node in other cluster heads, the data is transmitted to the sink node in a multi-hop data transmission mode;

105) Before sending the next data, communication interaction obtains the energy of the current relay node, if it is found that the energy of the relay node is too low to undertake the forwarding task, then go to step 103);

11) The network cycle has not ended, enter step 8); the network cycle has ended, enter step 12);

12) The aggregation node decides whether to perform the next round of cluster head selection according to the completion of the task. If the task has been completed or the energy loss of each sensor node causes the network to fail to operate normally, it will broadcast the end of network operation information, otherwise broadcast the next round of cluster head selection message , go to step 13).

13) The network receives the next round of cluster head selection message broadcast by the aggregation node. When the next round of cycle time comes, the network starts the next round of cluster head selection. The specific operations are as follows:

131) Each sensor node first checks its own status. If the sensor node is the cluster head in the current round of tasks, it broadcasts the message of becoming a cluster head to the wireless body area network, and waits for other sensor nodes to join as non-cluster head nodes; The node is not a cluster head in this round of tasks, so it waits for each cluster head to send broadcast information. After receiving the cluster head broadcast information, it selects the cluster head closest to itself and with the smallest node ID in all cluster head sets to join the corresponding cluster, and sends The cluster head sends the request joining information including the current remaining energy of the node and the node identification, and waits for the cluster head to send the time slot allocation table information.

132) After each cluster head receives the request to join information sent by other sensor nodes, it regards it as a non-cluster head node, records the information of these non-cluster head nodes, and then according to the received non-cluster head nodes in the human body location and The remaining energy information calculates the average energy value of all non-cluster head nodes, and forms a candidate cluster head set with the nodes whose remaining energy is greater than or equal to the average energy value, and selects the closest to the current cluster head node from the candidate cluster head set. The smallest node acts as the cluster head of the next round, and sends the time slot allocation information and the election information of the next round of cluster heads to all non-cluster head nodes in the cluster.

133) After each non-cluster head node receives the time slot allocation information from its cluster head and the next round of cluster head election information, if it is the next round of cluster head, set its next round of cluster head election identification variable to is 1, if you are not the next round of cluster heads, set your own next round of cluster head elected identification variable to 0, and go to step 8).

Beneficial effects: the routing method based on area division proposed by the present invention is suitable for wireless body area networks, and has the following beneficial effects:

(1) The method of the present invention makes full use of the characteristics that the energy of the aggregation node is not limited, and divides the network monitoring area into a plurality of cluster areas before the first round of cluster selection, and then according to the position information and energy information of the node in the human body, in each cluster Selecting a suitable node as the cluster head in the next round, the average energy efficiency of the node is significantly improved.

(2) Considering energy and inter-cluster routing cost and other factors, select the appropriate node as the next hop node of the cluster head, and transmit data between the cluster heads to the sink node in a multi-hop manner, and the network life cycle is significantly improved.

Description of drawings

FIG. 1 is a schematic structural diagram of a clustered wireless body area network;

Fig. 2 is a schematic flowchart of a routing method for wireless body area network based on area division.

Detailed ways

1. Architecture

In a specific implementation, all nodes in the wireless body area network of the present invention are randomly distributed in a square area, as shown in Figure 1, and the network satisfies the following conditions:

(1) All nodes are distributed in the wireless body area network according to the needs of the human body, and they are all stationary.

(2) Each node in the network knows its own position coordinates and can know its current energy.

(3) All nodes except the sink node are homogeneous and isomorphic, and the initial energy is equal. Except the sink node's energy is unlimited, the energy of all other nodes is limited.

(4) The power of the node is large enough to communicate directly with other nodes in the network.

(5) The node can calculate the distance to the sending node according to the received signal strength.

(6) The node power is controllable, and the transmission power can be adjusted according to the distance from the receiving node.

(7) The node continuously monitors the surrounding area, and sends the collected data to the cluster head at its own time.

(8) The data collected by nodes that are very close to each other has data correlation, so the data of adjacent nodes has great redundancy, and the redundant data is useless to the observer.

2. Method flow

A more detailed description of the specific implementation of the present invention is given below:

1. Cluster head selection

Step 1: Each node on the body in the wireless body area network transmits its own information to the sink node i, and this information includes the identification of the node, the energy of the node and the position coordinates of the node.

Step 2: The sink node i stores the information after receiving the node information, and then according to the pre-selected optimal number of cluster heads, runs the area division algorithm to divide the entire wireless body area network area into n sub-domains according to the location of each node in the human body .

Step 3: After completing the area division, the sink node i divides each node into its corresponding area according to its position in the human body, and marks the area identification.

Step 4: The converging node i calculates the centroid coordinates of each sub-area in each sub-area, and then calculates the distance between the nodes (j, k, l...) of each sub-area and the centroid.

Step 5: Convergence node i selects the node closest to its centroid in each human body area as the cluster head of the first round, and selects the node next closest to the centroid as the cluster head of the next round.

Step 6: The aggregation node i broadcasts information to all nodes in the wireless body area network. This information includes the belonging cluster of each node in the first round, the status of the node (member node or cluster head node) and the next round of cluster head nodes. Node ID.

Step 7: After receiving the information from the sink node, node j stores it, and checks whether it is set as the first-round cluster head according to the information. If it is a cluster head, it allocates time slots for its members in the cluster and generates time slots allocation table, and then send the time slot allocation table to the member nodes in the cluster.

Step 8: The cluster head node CH broadcasts its own information to the network, establishes a routing table according to the received information of the adjacent cluster head node CH1, and then enters the stable data transmission stage.

2. The next round of cluster head selection

Step 1: After each node receives the broadcast information of the next round of cluster head selection deadline sent by the aggregation node, when the next round of cycle deadline arrives, they first check their status, if node a is the cluster head in this round of task role , it broadcasts the message that it becomes the cluster head to the wireless body area network, and waits for the member nodes in the cluster to join.

Step 2: If node b is a non-cluster head node, it waits for each cluster head to send broadcast information. After receiving the cluster head broadcast information, it selects the cluster head closest to itself in all cluster head sets as the cluster head that it wants to join, and then sends to The cluster head sends join request information. This information contains the current remaining energy and identity of the node, and then waits for the cluster head to send the slot allocation information.

Step 3: After cluster head a receives the request to join information sent by the node, it records the information of each node. Then calculate the average energy value of the cluster member nodes according to the received node's position in the human body and the remaining energy information, and form a candidate cluster head node set with the nodes whose remaining energy is greater than or equal to the average energy value. Select the node closest to the current cluster head node from the candidate cluster head node set as the official cluster head for the next round. Incidentally, when sending the time slot allocation table, the information is sent to the member nodes in the cluster.

Step 4: After member node j receives the time slot allocation table and the next round of cluster selection information from the cluster head, it sets its next round of cluster head flag to 1 or 0, and then according to the time slot allocation table in its own time slot allocation table. Send the perception data to the cluster head as needed in the gap.

3. Routing establishment

Step 1: Each cluster head node CH (1, 2, 3..) transmits data to the sink node by selecting a relay node in a multi-hop data transmission mode when sending data to the sink node.

Step 2: In specific implementation, when selecting a relay node, if the energy of the node is low, it will no longer act as a relay node, and select the next relay node through rotation.

Step 3: In the specific implementation, it is assumed that the node communication uses a free space model, and the communication cost between nodes is proportional to the square of the distance between nodes. If node b wants to transmit data to sink node i, it firstly transmits data in its own Each cluster head node is selected one by one from the set of neighboring cluster head nodes, and the communication cost of transmitting data to the sink node via this node is calculated. The nodes whose transmission path cost through the relay node is less than the direct transmission cost form the candidate node set. Then compare the energy value of each node with the energy value of this node in the candidate node set in turn, and take the node greater than or equal to the energy value of this node as the official relay cluster head. If there are multiple cluster heads with high energy that meet the conditions, the one with the largest energy value is selected as the relay cluster head. If the candidate cluster head set is empty or there is no qualified cluster head in the candidate cluster head node set, the data will be directly transmitted to the sink node.

4. Data transmission

In a specific implementation, after the cluster head is determined in each round, the cluster head allocates time slots for each member in the cluster according to the number of members in the cluster. Then, the cluster head sends the allocated time slot table to the members in the cluster, and the cluster head broadcasts its own information to establish inter-cluster routing paths. After the cluster establishment phase is completed, the path between the cluster head and the cluster head is established, and the member nodes in the cluster continue to collect the surrounding environment information, and then collect the collected data in a one-hop manner in the time slot allocated to them. Send to the cluster head of the cluster where the member node is located. Nodes turn off the communication module when they are not in their own transmission time to reduce energy consumption. After the cluster head receives the data of all the nodes in the cluster, it fuses the collected data, and then transmits the fused data to the sink node. In the time period of a round, the stable data transmission phase is longer than the cluster establishment phase. The data collected in each cluster is one frame, and multi-frame data collection is performed in each round of transmission, which will reduce the number of cluster establishments, use more energy for data transmission, and improve energy utilization.

In the specific implementation, after the data transmission runs for a period of time, the aggregation node decides whether to perform the next round of cluster head selection according to the completion of the task. If the task has been completed or the energy loss of the node causes the network to fail to operate normally, the network end information will be broadcast, otherwise broadcast The time limit for the next round of cluster head selection.

Claims (1)

1., based on a wireless body area network route method for Region dividing, it is characterized in that the step that the method comprises is:

1) after all nodes of wireless body area network start, predefined bunch of head is elected to marking variable by each sensor node, a next round bunch head is elected to marking variable and is set to 0, transmits oneself mark, present energy and location coordinate information to aggregation node; The all nodes of described wireless body area network comprise aggregation node, sensor node, all the sensors node forms sub-clustering, each sub-clustering has oneself bunch interior nodes, these bunch of interior nodes comprises a bunch head, non-leader cluster node, wherein some sensor nodes are as a bunch head, and other sensor nodes are non-leader cluster nodes;

2) aggregation node to receive these information storage after sensor node information, according to the optimum bunch head number n preset, whole network area is divided into n sub regions equably, distributes a mark to each subregion;

3) after completing sub-zone dividing, in the subregion that each sensor node is divided in corresponding to it in the position of human body according to it by aggregation node, form corresponding sub-clustering, and send the mark of subregion to node;

4) aggregation node is fallen into a trap in each sub regions respectively and is calculated the center-of-mass coordinate of each sub regions, distance between the node of each sub regions and barycenter;

5) aggregation node its barycenter of chosen distance in each sub regions nearest, identify bunch head of minimum sensor node as the first run, a bunch head for this sensor node is elected to marking variable and is set to 1; The next round of this sensor node bunch head, as bunch head of next round, is elected to marking variable and is all set to 1 by the secondary near sensor node of chosen distance barycenter;

6) aggregation node is to all the sensors node broadcasts information in network, this packets of information contain each sensor node the first run home cluster, whether be the first run bunch header, whether be next round bunch header;

7) all the sensors node is stored after receiving the information coming from aggregation node, the first run bunch head whether is set to according to this information check oneself, if bunch head, be just oneself bunch in non-leader cluster node distribution time slot, generate Schedule, then the non-leader cluster node in Schedule information being sent to bunch, then enters step 8);

8) each non-leader cluster node also send to its bunch of hair the data perceived, closed communication module time not in the time slot of oneself as required according to the Schedule information that persistent collection is monitored in the time slot of oneself;

9) each bunch of head collects the data of the non-leader cluster node in all bunches, and carries out data fusion to these data, also receives other bunch of head in addition and forwards the data of coming;

10) each bunch of head by bunch in merge after data or other bunch of head forward the data come and send to aggregation node, the operation that each bunch of head adopts is as follows:

101) broadcast the mark of oneself, current energy consumption, bunch head identity information, receive and record the corresponding informance that adjacent cluster hair sends here;

102) in the neighbours' bunch head set of oneself, select other each bunch of head one by one, and calculate through the communication cost of selected bunch head to aggregation node transmission data;

103) communication cost be less than directly transfer data to aggregation node time communication cost bunch head in, select energy value maximum, identify minimum bunch head as via node, by this data retransmission this bunch of head; If there is no qualified bunch of head as via node, then directly transfer data to aggregation node;

104) by selecting via node in other bunch of head, the data mode of multi-hop is adopted to send data to aggregation node;

105) in transmission once before data, communication interaction obtains the energy of current hop node, if find this via node energy lower and can not bear forwarding task, then forwards step 103 to);

11) network one cycle of taking turns does not terminate, and enters step 8); Network one cycle of taking turns terminates, enters step 12);

12) according to task performance, aggregation node determines that whether carrying out next round bunch head selects, if task has completed or each sensor node energy loss causes network normally not run, then radio network end of run information, otherwise broadcast next round bunch head selects message, enters step 13);

13) network receives the next round bunch head selection message of aggregation node broadcast transmission, and when the time limit in next round cycle arrives, network starts next round bunch head and selects, and concrete operations are as follows:

131) first each sensor node checks the state of oneself, if this sensor node is a bunch head at epicycle task role, just oneself becomes the message of bunch head to wireless body area network broadcast, waits for that other sensor nodes add as non-leader cluster node; If node is not a bunch head at epicycle task role, just wait for that each bunch of hair send broadcast message, after have received bunch head broadcast message, bunch head that chosen distance oneself is nearest in all bunches of head set, node identification is minimum adds corresponding sub-clustering, send to this bunch of hair comprise node current remaining, the request of node identification adds information, waits for that a bunch hair send Schedule information;

132) each bunch of head receives after request that other sensor nodes send adds information, it can be used as non-leader cluster node, record the information of these non-leader cluster nodes, then calculate the average energy value of all non-leader cluster nodes in human body position and dump energy information according to these the non-leader cluster nodes received, node dump energy being greater than or equal to the average energy value forms bunch head set of a candidate, distance current cluster head node is selected nearest from bunch head set of this candidate, identify minimum node as next round bunch head, and by time slot allocation information, next round bunch head be elected to information to send to bunch in all non-leader cluster node,

133) each non-leader cluster node receives time slot allocation information that its bunch of hair come and after a next round bunch head is elected to information, if oneself be next round bunch head, the next round of oneself a bunch head is elected to marking variable and is set to 1, if oneself be not next round bunch head, the next round of oneself a bunch head is elected to marking variable and is set to 0, enter step 8).