CN101345674B - A 3D Coverage Control Method for Sensor Networks - Google Patents

Abstract

The invention relates to a three-dimensional coverage control method of a sensor network. In the current overlay method, dormant nodes are generally not processed, or active/sleep state transitions are directly adopted in turn, which cannot achieve high network performance. In the present invention, in the monitoring area, a plurality of sensor nodes are randomly set under the condition of satisfying the distribution density requirement, so that the redundant nodes are in a dormant state, and after waiting for the energy of the active nodes to be exhausted, a staged wake-up strategy is implemented for the dormant nodes. Determine the state of any node at a certain moment until the energy of all nodes in the entire sensor network is exhausted. By adopting the method of the present invention, the energy consumption of the network can be balanced, the network performance can be improved, and the same number of sensor nodes can be arranged in the three-dimensional monitoring area. The network performance of the method is higher than that of the direct wake-up method in turn.

Description

A 3D Coverage Control Method for Sensor Networks

technical field

The invention belongs to the technical field of three-dimensional environment monitoring of a wireless sensor network, and in particular relates to a three-dimensional coverage control method of a sensor network based on redundant node sleep and staged wake-up strategies.

Background technique

Wireless sensor network (WSNs) is an autonomous network system composed of a large number of low-cost, low-power tiny sensor nodes with sensing, computing and communication capabilities. It is an intelligent system that can independently complete various monitoring tasks according to the environment. Military, automotive electronics, industrial control, environmental monitoring, medical and health care, smart home and other fields have good application prospects, especially in event monitoring and target tracking in unattended or harsh environments. Due to the small size of sensor nodes and limited battery energy resources in WSNs, how to ensure the effective use of sensor node energy in a large-scale network environment has become a research topic that has attracted much attention, which directly affects the life cycle of the entire network.

The wireless sensor network used in environmental monitoring generally has a wide deployment range, and the environment is harsh and unattended for a long time. Therefore, a large number of spatially densely distributed nodes are required to cooperate to complete the entire monitoring task. The energy consumption of the nodes and the energy consumption of the network Longevity is a matter of great concern. The dense distribution of nodes means that there is a large redundancy in the network, and making the redundant nodes dormant will play a great role in saving the energy consumption of the entire network. However, letting too many nodes sleep without principle will inevitably affect the accuracy, integrity and communication quality of the data collected by the entire monitoring network. In the existing technology, the problem of network energy saving is studied. Under the condition of ensuring the initial network coverage (that is, the coverage achieved by the entire network when all nodes are active after the node deployment is completed), if the sensing range of a sensor node is reduced by When covered by the sensing range of other active nodes, the node can be in a dormant state. However, this method does not consider the life cycle of the entire network, that is, it does not involve how to deal with nodes in a dormant state. Another type of method in the prior art starts from the network life cycle and directly executes a unified active/dormant state decision after deploying nodes, such as WPCS coverage strategy and cyclic state change method. Neither of these two types of methods can achieve high network performance.

Contents of the invention

The purpose of the present invention is to address the deficiencies in the prior art, to provide a sensor network three-dimensional coverage control method based on redundant node sleep and phased wake-up strategies, to improve the overall performance of the sensor network, that is, on the basis of making redundant nodes sleep , implement a phased wake-up strategy.

In order to achieve the above object, technical method of the present invention is achieved like this:

(1) Model the monitoring area with a cube to determine the three-dimensional coordinate system of the monitoring area; in the monitoring area, randomly set multiple sensor nodes under the condition of meeting the distribution density requirements, and determine the x, y, and z axes of each sensor node three-dimensional coordinates;

(2) For the sensor nodes distributed in the monitoring area, considering the redundancy of the overall network, according to the contribution of the sensor nodes to the network coverage, make the redundant nodes in the dormant state, and the nodes in the active state are used to complete the current Network coverage tasks;

(3) After the energy of the nodes in the active state in step (2) is exhausted, implement a phased wake-up strategy for the nodes in the dormant state to determine the state of any node at a certain moment until the energy of all nodes in the entire sensor network is exhausted. exhausted.

In step (2), the method for making the redundant node in the dormant state is as follows:

(21) assume that each sensor is active;

(22) Calculate the cooperative perception strength of each grid point according to the sensor perception model, and calculate the initial network coverage.

Due to the randomness of sensor node deployment, the network may not achieve full coverage (that is, the network coverage is 100%) at the initial deployment, and the subsequent redundant node sleep strategy is based on the premise of ensuring the network coverage. The initial coverage of the network can be improved by increasing the number of deployed nodes.

If a grid point is located in the common sensing area of multiple sensor nodes, the cooperative sensing strength of the grid point is the sum of the sensing strengths of these sensor nodes at the grid point; network coverage refers to the number of covered grid points and The ratio of the total number of grid points in the 3D domain.

The method of calculating the cooperative perception strength of each grid point and the initial network coverage adopts existing mature methods, such as the literature Baiwei Yang, Hongyi Yu, Hong Li, Huifeng Hou.A coverage-preserving density control algorithm based-on cooperation in wireless sensor networks. Wireless Communication, Networking and Mobile Computing. 2006, methods disclosed in 1-4.

(23) For all grid points, according to the currently active sensor nodes, update the cooperative sensing strength of the grid points within the effective sensing range of the sensor, judge the decision parameters of each sensor node, and calculate the contribution of each node to the network perception contribution.

The effective sensing range of the sensor is determined by the sensor sensing model. If a sensor node is in a dormant state, and the cooperative sensing intensity of all grid points within its effective sensing range is greater than a certain threshold, the decision parameter is 1, which means that the sensor node can be in a dormant state; The perceptual contribution of the network refers to the sum of the perceptual contributions of the node to all grid points in the network, and the perceptual contribution of a node to a certain grid point refers to the difference between the node’s perceptual strength of the grid point and the grid point’s The ratio of perceived strength of cooperation.

(24) If there is a sensor node with a decision parameter of 1 in the network, select the node that contributes the least to the perception of the network, make it in a dormant state, set the sensory intensity of the node to 0, and update the network coverage. Each cycle selects a sensor node to implement sleep until there is no sensor node whose decision parameter is 1 in the network.

In step (3), the method for implementing the phase-by-stage wake-up strategy to the nodes in the dormant state is as follows:

(31) Assume that each node is in a dormant state;

(32) Divide the lifetime T of each sensor node into N stages. Define an energy vector E, whose elements represent the remaining energy of each node, and initialized as E[N…N], that is, the remaining energy of each sensor node is initialized as N.

(33) In each time period T/N, the sensor nodes are cyclically awakened, and a sensor node is randomly activated in each cycle. This node must ensure that it contributes to network coverage and is not blocked during this time period. Repeatedly set to the active state, the element E(i) corresponding to the node in the energy vector is decremented by 1, and the network coverage is updated. The loop termination condition is that initial network coverage is reached, or all nodes are active.

(34) After waiting for the end of the previous stage, make the state of each node be dormant (except for nodes with exhausted energy), enter the next time stage, repeat step (33), and cycle until the energy of each sensor node until all are exhausted.

The invention proposes a three-dimensional coverage control method of a sensor network based on redundant node sleep and staged wake-up strategies. In the monitoring area, randomly set up multiple sensor nodes under the condition of satisfying the distribution density requirements, make the redundant nodes in a dormant state, wait for the energy of the active nodes to be exhausted, implement a phased wake-up strategy for the dormant nodes, and determine any node The state at a certain moment until the energy of all nodes in the entire sensor network is exhausted. By adopting the method of the present invention, the network performance of the sensor network can be improved, and the same number of sensor nodes are arranged in the three-dimensional monitoring area, the wake-up strategy in stages is better than the wake-up strategy in no stages, and the network of the method of awakening the redundant nodes after dormancy first The performance is higher than the network performance of the direct wake-up polling method.

Description of drawings

Fig. 1 is a flow chart of the three-dimensional coverage control method of the sensor network of the present invention.

Detailed ways

The core idea of the present invention is: in the monitoring area, a plurality of sensor nodes are randomly set under the condition of satisfying the distribution density requirement, so that the redundant nodes are in the dormant state, and after the energy of the active nodes is exhausted, the nodes in the dormant state are distributed. The stage wake-up strategy determines the state of any node at a certain moment until the energy of all nodes in the entire sensor network is exhausted.

The present invention will be described in further detail below in conjunction with the accompanying drawings. This method is divided into three parts:

1. Use a cube to model the monitoring area, and determine the three-dimensional coordinate system of the monitoring area; in the monitoring area, set up multiple sensor nodes under the condition of meeting the distribution density requirements, and determine the three-dimensional x, y, and z axes of each sensor node coordinate;

2. For the sensor nodes distributed in the three-dimensional monitoring area, considering the redundancy of the overall network, according to the contribution of the sensor nodes to the network coverage, make the redundant nodes in the dormant state, and the nodes in the active state are used to complete the current Network coverage tasks;

3. After the energy of the nodes in the active state is exhausted in step 2, implement a phased wake-up strategy for the nodes in the dormant state to determine the state of any node at a certain moment until the energy of all nodes in the entire sensor network is exhausted until.

The specific implementation of this method follows the following process, see accompanying drawing 1:

(1) In the monitoring area, randomly set multiple sensor nodes under the condition of satisfying the distribution density requirements;

(2) Assume that each sensor is in an active state; calculate the cooperative perception strength of each grid point according to the sensor perception model, and obtain the initial network coverage from this. Due to the randomness of sensor node deployment, the network may not achieve full coverage during the initial deployment, and the subsequent redundant node sleep strategy is based on the premise of ensuring the network coverage. The initial coverage of the network can be improved by increasing the number of deployed nodes.

If a grid point is located in the common sensing area of multiple sensor nodes, the cooperative sensing strength of the grid point is the sum of the sensing strengths of these sensor nodes at the grid point; the network coverage is the number of covered grid points and the The ratio of the total number of grid points in the three-dimensional area;

(3) For all grid points, according to the currently active sensor nodes, update the cooperative sensing strength of the grid points within the effective sensing range of the sensor, judge the decision parameters of each sensor node, and calculate the contribution of each node to the network perception contribution.

The effective sensing range of the sensor is determined by the sensor sensing model; if a sensor node is in a dormant state, and the cooperative sensing strength of all grid points within its effective sensing range is greater than a certain threshold, its decision parameter is 1, which means The sensor node can be in a dormant state; the sensory contribution of a node to the network is the sum of the sensory contributions of the node to all grid points in the network, and the sensory contribution of a node to a grid point refers to the sensory contribution of the node to the network. The ratio of the perceived strength of a grid point to the cooperative perceived strength of that grid point;

(4) If there is a sensor node with a decision parameter of 1 in the network, select the node that contributes the least to the perception of the network, make it in a dormant state, set the sensory intensity of the node to 0, and update the network coverage. Each cycle selects a sensor node to implement sleep until there is no sensor node whose decision parameter is 1 in the network.

(5) After waiting for the energy of the active nodes selected in step (4) to run out, for the sensor nodes in the dormant state, divide the lifetime T of each sensor node into N stages. Define an energy vector E, whose elements represent the remaining energy of each node, and initialized as E[N…N], that is, the remaining energy of each sensor node is initialized as N.

(6) In each time period T/N, the sensor nodes are cyclically awakened, and a sensor node is randomly activated in each cycle, and the node guarantees that it contributes to the network coverage and is not repeated in this time period Set to the active state, the element E(i) corresponding to the node in the energy vector is decremented by 1, and the network coverage is updated. The loop termination condition is that initial network coverage is reached, or all nodes are active.

(7) After waiting for the time of the previous stage to end, make each node in a dormant state (except for the nodes whose energy is exhausted), enter the next time stage, and repeat step (6). Loop until the energy of each sensor node is exhausted.

In short, the present invention proposes a three-dimensional coverage control method for sensor networks based on redundant node sleep and phased wake-up strategies: in the monitoring area, randomly set a plurality of sensor nodes under the condition of meeting the distribution density requirements, so that the redundant nodes are in sleep After waiting for the energy of the active nodes to run out, implement a phased wake-up strategy for the nodes in the dormant state to determine the state of any node at a certain moment until the energy of all nodes in the entire sensor network is exhausted. It should be noted that the adoption of different sensor perception models and other methods does not deviate from the spirit and scope of the technical method of the present invention.

Claims (1)

1. three-dimensional overlapping control method of sensor network is characterized in that this method may further comprise the steps:

(1), determines the three-dimensional system of coordinate of monitored area to monitored area square modeling; In the monitored area, a plurality of sensor nodes are set at random satisfying under the distribution density requirement condition, determine x, y, the z axle three-dimensional coordinate of each sensor node;

(2) guaranteeing under the situation that initial network covers, adopt the redundant node dormancy strategy: to being distributed in the sensor node in the monitored area, consider from the redundancy of overall network, according to the contribution of sensor node to the network coverage, make redundant node be in resting state, the node that is in active state is used for finishing current network coverage task;

(3) after the node energy that is in active state in the step (2) exhausts, the node that is in resting state carried out wake strategy stage by stage up, determine any node state at a time, till the energy of all nodes all exhausts in the whole sensor network;

It is as follows to make redundant node be in the method for resting state in the step (2):

(21) suppose that each transducer all is in active state;

(22), and calculate the initial network coverage according to the cooperative sensing intensity of each grid point of sensor senses Model Calculation; If certain grid point is positioned at the public perception zone of a plurality of sensor nodes, then the cooperative sensing intensity of this grid point is the perceptive intensity sums of these sensor nodes at this grid point; The described network coverage is the ratio that the grid that is capped is counted and counted with grid total in 3D region;

(23) to all grid points, according to the current sensor node that is in active state, within the effective sensing range of transducer, upgrade the cooperative sensing intensity of grid point, judge the decision parameter of each sensor node, and calculate of the perception contribution of each node network;

Described node is the perception contribution sum of this node to all grid points in the network to the perception contribution of network; Described node is that this node is to the perceptive intensity of this grid point and the ratio of the cooperative sensing intensity of this grid point to the perception contribution of the grid point in the network;

Wherein the effective sensing range of transducer is determined by the sensor senses model; After if a certain sensor node is in resting state, the cooperative sensing intensity of all grid points in its effective sensing range is all greater than a certain threshold value, and then its decision parameter is 1, represents that this sensor node can be in resting state;

(24) if having the decision parameter in the network is 1 sensor node, then therefrom choose the minimum node of perception contribution, make it be in resting state, the perceptive intensity of this node is made as 0, upgrade the network coverage network; Each circulation is chosen a sensor node and is implemented dormancy, do not determine that in network parameter is 1 sensor node till;

It is as follows in the step (3) node that is in resting state to be carried out the method for waking strategy stage by stage up:

(31) suppose that each node all is in resting state;

(32) the life-span T with each sensor node is divided into N stage, defines an energy vector E, and its element is represented the dump energy of each node, and is initialized as E[N ... N], promptly the dump energy of each sensor node is initialized as N;

(33) in each time phase T/N, circulation wakes sensor node up, make a sensor node be in active state randomly in each circulation, this node guarantees that it has contribution and be not repeated to be changed to active state in this time phase the network coverage, element E (i) corresponding to this node in the energy vector subtracts 1, upgrades the network coverage; The loop termination condition is to have reached the initial network covering, or all nodes all are in active state;

(34) wait on last stage time up after, making the state of each node is resting state all, enters next time phase, repeating step (33), circulation is till the energy of each sensor node all exhausts.