CN106028357A - Novel underwater wireless sensor network point coverage control method - Google Patents

Abstract

The invention relates to a novel underwater wireless sensor network point coverage control method. In the method, node depth adjustment is taken as a feasible mobile mode. The method comprises the following steps that: firstly, nodes determine depths of the nodes in reference to a distance between a Sink node and a water surface vertex and the depth of a monitoring water area; secondly, the nodes determine perception radiuses of the nodes according to a distance between the Sink node and a water bottom vertex in reference to a minimum perception radius and a maximum perception radius; thirdly, communication radiuses of the nodes are increased greatly till neighbor nodes with smaller depths are inquired, and a node with a smallest coverage target quantity is selected from a set of the nodes to take the selected node as a cluster head; and lastly, a cluster head node makes inner-cluster nodes hibernate and wake up. Compared with an existing relevant method, the method disclosed by the invention is more feasible. Moreover, indexes such as network coverage, a communication rate, energy consumption and balance, and a network life cycle can be optimized comprehensively.

Description

A kind of new underwater wireless sensor network point overlapping control method

Technical field

The present invention relates to a kind of new underwater wireless sensor network point overlapping control method, belong to underwater wireless sensor The technical field that the network coverage controls.

Background technology

Underwater wireless sensor network (Underwater Wireless Sensor Networks, UWSN) is by some Have information gathering, store, process, monitoring network that the underwater sensor node of transfer function is formed in an ad-hoc fashion, energy Enough target waters is completed the tasks such as environmental monitoring, hazard prediction, resource detection, military monitoring, there is network structure and present three Tie up distributed, network design haves a wide reach and node density is sparse, node use sound wave as communication media, node easily by water ring Border impact and move, node energy is limited and is difficult to features such as supplementing.The Coverage Control of underwater wireless sensor network is asked Topic, refers on the premise of taking into full account These characteristics, in conjunction with concrete application demand, to the position of nodes, state, road By etc. carry out reasonable disposition, utilize limited Internet resources to obtain the most excellent network service.Pass as building underwater wireless One of basic fundamental of sensor network, Coverage Control is not only directly related with network monitor quality, and profound influence node Follow-up each quasi-protocol and the designs of algorithm such as location, data compression, time synchronized, have important research meaning, has caused a large amount of Correlational study.On the one hand, according to node motion ability it is assumed that existing underwater wireless sensor network cover control method can It is divided into static covering, the degree of depth to adjust covering, move freely covering.Static cover, refer to that node does not possess locomotivity, need to be by Manual type is deployed to specify region.The degree of depth adjusts and covers, and refers to that node possesses the locomotivity on vertical direction, and node can root In three-dimensional waters, degree of depth adjustment is carried out according to needs.Move freely covering, refer to that node possesses the locomotivity in all directions, joint Point can move freely as required in three-dimensional waters.The required human cost of static covering is too high, and when deployment personnel are difficult During to enter monitoring waters, the method cannot be suitable for；Moving freely covering excessively to idealize, the method generally need to be by high cost Autonomous robot under water, be difficult at present popularize；When node is connected to water surface float (or water-bed anchor) by rope, by certainly The electric machine regulation rope lengths that body is equipped with can realize the node degree of depth easily and adjust, and the degree of depth adjusts to cover and is acknowledged as working as Before the most practicable method.On the other hand, whether there is the difference of covering demand according to monitoring waters, existing underwater wireless passes Sensor network cover control method can be divided into region overlay and put Coverage Control.Region overlay controls to refer to monitor each region, waters There is not covering demand difference, network design purpose is to pursue whole monitoring waters is realized the highest coverage rate.Point covers Controlling to refer to that monitoring each region, waters exists covering demand difference, network design purpose is to pursue the isolated thing in monitoring waters Part (also known as impact point) realizes the highest coverage rate.For underwater wireless sensor network point Coverage Control problem, Aitsaadi etc., for the distribution density difference of zones of different impact point, propose cross hatch representation, determine node density, and complete Become corresponding node deployment；Xia etc. propose the some overlapping control method that a kind of population inspires, by simulation particle group's behavior, Node is made independently to tend to impact point, it is achieved node mates with impact point distribution density；Jiang etc. propose a kind of based on sub-clustering Point overlapping control method, defines and utilizes comprehensive contribution degree, it is desirable to the node that comprehensive contribution degree is little substitutes and in the network operation is By dead node, delay the decay of network covering property.But, above-mentioned all methods are required to node and can move freely, Lack practical feasibility；And be all difficult to take into account the network coverage, connected ratio, energy consumption size simultaneously and refer to equilibrium, life cycle etc. Target optimizes.

Summary of the invention

Technical problem

Therefore, it is necessary to for underwater wireless sensor network point Coverage Control demand, design one is practicable deeply Degree adjust overlapping control method, on the premise of taking into full account underwater wireless sensor network feature as far as possible, to the network coverage, Connected ratio, energy consumption size carry out complex optimum with indexs such as equilibrium, life cycles.

Technical scheme

Owing to existing underwater wireless sensor network point overlapping control method is all premised on node can move freely, lack Weary practical feasibility, the present invention, first for underwater wireless sensor network point Coverage Control demand, proposes one more practical The feasible degree of depth adjusts overlapping control method.The movement of node is no longer by equipment such as expensive autonomous robots under water, it is only necessary to By rope, node is connected with water surface float (or water-bed anchor), and regulates rope lengths by motor, complete on vertical direction The degree of depth adjust.

The underwater wireless sensor network point overlapping control method proposed mainly comprises the steps that node is carried out initially The degree of depth adjusts；Knot adjustment each perception radius；Knot adjustment each communication radius, carries out network cluster dividing；Adjust the network operation During the state of some node and position.

Node carries out ID adjustment and refers to after node is uniformly deployed in the water surface in monitoring waters by boats and ships, in conjunction with Self-position and the Euclidean distance of water surface center (i.e. Sink node position), determine self initial descent degree of depth, and distance is more Farther submerged depth is the biggest.

After node completes ID adjustment, each node is further combined with self-position with water surface center (i.e. Sink node position) Euclidean distance, determine self perception radius, distance more farther perception radius is the biggest.

After each node determines self perception radius, adjust self communication radius, until searching out neighbours' joint that the degree of depth is less Point, carries out perception to periphery monitoring region afterwards, determines that the target self covered is counted, and from the less neighbor node of the degree of depth Set selects coverage goal those at least of counting as a bunch head, complete network cluster dividing.

The covering contribution degree of each node in each bunch of head analytic manifold, in conjunction with dormancy with wake up mechanism up, it is desirable to there is invalid covering Node enter resting state, when some node during the network operation will dead time, leader cluster node can be by dormancy node Wake up up, delay the decay of network covering property.

Beneficial effect

The present invention is directed to underwater wireless sensor network point Coverage Control demand, it is proposed that a kind of practicable covering is controlled Method processed, has considered the network coverage, connected ratio, the optimization of energy consumption size and the indexs such as equilibrium, life cycle.Specifically In fact, the method for the invention has the advantages that

(1) breach existing underwater wireless sensor network point overlapping control method excessively to manage in terms of node motion ability The limitation of wanting, proposes to introduce in wireless sensor network point Coverage Control problem under water node degree of depth Adjusted Option first, Make proposed method more economically, easy, feasible.

(2) first, during ID adjusts and determines perception radius, each node need to consider self-position and the water surface The Euclidean distance of center (i.e. Sink node position), submerged depth and the perception radius of distance more farther are the biggest.This is permissible Guarantee in the three-dimension monitor network ultimately formed, with Sink node closer to Area Node density the biggest, perception radius is more Little；With Sink node further away from Area Node density the least, perception radius is the biggest.Due to Sink node closer to node Its information forwards pressure the biggest (i.e. communication energy consumption is the biggest), and the least perception radius will favorably reduce its perception energy consumption；Additionally, During network cluster dividing, forward pressure (i.e. communication energy consumption is bigger) owing to leader cluster node will undertake more information, therefore cover mesh Punctuate number those at least (i.e. perception energy consumption those at least) will be as a bunch head.Above-mentioned two measure all can improve network energy consumption equilibrium, prolongs Long network life cycle.

(3) it is different from conventional method node perceived radius and communication radius employing isomorphism processing mode, the present invention couple Node perceived radius and communication radius all use isomery processing mode.Relative to perception radius isomorphism, perception radius isomery mode More flexible, can effectively reduce coverage hole, improve the network coverage, network energy consumption balance can be improved again, extend network raw Cycle of deposit；Relative to communication radius isomorphism, communication radius isomery mode then can effectively ensure that being smoothed out of network cluster dividing, carry High network connectivity efficiency.

(4) present invention proposes combine dormancy and wake up mechanism up, and carrying out the state of some node during the network operation must Adjust.The introducing of dormancy mechanism will be prevented effectively from invalid covering, and dormancy node directly can be waken up up by himself leader cluster node, The communication energy consumption brought is less, contributes to extending network life cycle.

Accompanying drawing explanation

Accompanying drawing 1 is towards a underwater wireless sensor network schematic diagram for covering demand

Accompanying drawing 2 method implementing procedure figure

Accompanying drawing 3 initial period water surface top view

Accompanying drawing 4 node perceived model schematic

5 bunches of heads of accompanying drawing select schematic diagram

Accompanying drawing 6 invalid overlay node schematic diagram

It is embodied as

It is embodied as being further described in more detail to the present invention below in conjunction with the accompanying drawings.

Fig. 1 is towards a underwater wireless sensor network schematic diagram for covering demand, and monitoring waters generally uses cube Modeling, x-axis direction ultimate range is x_m, y-axis direction ultimate range is y_m, z-axis direction ultimate range is z_m；Node can be in the degree of depth The enterprising line position in direction adjusts, and network coverage object is the impact point of random uneven distribution in monitoring waters.The present invention will be by According to Fig. 2 describe flow process, node location, state etc. are adjusted, it is achieved to the network coverage, connected ratio, energy consumption size with The complex optimum of the indexs such as equilibrium, life cycle.

1, for ease of collecting the information that each sensor node is collected, Sink node is typically secured to water surface center, its Coordinate is represented byAnd distance D of Sink node and water surface summit can be calculated as follows₁。

$D_1=\sqrt{{\left(\frac{x_m}{2}\right)}^2+{\left(\frac{y_m}{2}\right)}^2}---\left(1\right)$

In the initial period, remaining sensor node is deployed on the water surface the most equably, as it is shown on figure 3, the most arbitrarily pass Sensor node s_iCoordinate be represented by s_i(x_i,y_i, 0), node s_iCalculate the distance of self and Sink node, with reference to Sink node Distance D with water surface summit₁With monitoring Larger water depths z_m, it is calculated as follows own depth h (s_i), and by the vertical degree of depth Adjustment moves to correspondence position.

$\left\{\begin{array}{c}h\left(s_i\right)=\frac{D(s_i,Sink)}{D_1}\×z_m\\ D(s_i,Sink)=\sqrt{{\left(x_i-\frac{x_m}{2}\right)}^2+{\left(y_i-\frac{y_m}{2}\right)}^2}\end{array}\right.---\left(2\right)$

2, after all the sensors node completes ID adjustment, each node should arrange self perception according to self-position Radius.Take distance D of Sink node and water-bed summit₂As reference distance, this distance can be calculated as follows.

$D_2=\sqrt{{\left(\frac{x_m}{2}\right)}^2+{\left(\frac{y_m}{2}\right)}^2+{z_m}^2}---\left(3\right)$

As shown in Figure 4, node uses centre of sphere sensor model, for node s_i, when impact point and its distance are in its perception half Footpath R_s(s_iTime in the range of), this impact point can be detected (t in such as figure by this node₁), otherwise impact point cannot be detected by this node (t in such as figure₂).Additionally, due to sensor node need by acquired monitoring information with single-hop or multi-hop mode send to , there is different information from Sink node apart from different nodes and forward pressure in water surface Sink node, will apart from the nearest node Undertaking the heaviest information and forward pressure, its communication energy consumption is relatively large.Therefore, for improving network energy consumption equilibrium, the present invention proposes Node perceived radius is carried out isomery process, in conjunction with actual environment and hardware condition, if minimum perception radius is R_si, maximum perception Radius is R_sa, node s can be determined as follows_iPerception radius R_s(s_i)。

$\left\{\begin{array}{c}{R}_s\left(s_i\right)=\frac{D(s_i,Sink)}{D_2}\×(R_{sa}-R_{si})+R_{si}\\ D(s_i,Sink)=\sqrt{{\left(x_i-\frac{x_m}{2}\right)}^2+{\left(y_i-\frac{y_m}{2}\right)}^2+{h_i}^2}\end{array}\right.---\left(4\right)$

3, after each node determines self perception radius, self communication radius should be further determined that.For guaranteeing that network obtains relatively Reducing inter-node communication interference while high connected ratio, the present invention is different from the node communication radius isomorphism processing mode of routine, Propose node communication radius is carried out equally isomery process.With node s_iAs a example by, if its initial communication radius is R_cin, communication half It is B that footpath increases step-length_c, initial communication rank is L_in=1, node increases its communication rank L (s step by step according to equation below_i), with Increase its communication radius R_c(s_i), until searching out the degree of depth neighbor node less than own depth.

R_c(s_i)=R_cin+(L(s_i)-L_in)×B_c (5)

Then, periphery monitoring region is detected by each node, determines the impact point number self covered, simultaneously to neighbour Occupy node and send notification M_t(content such as impact point number including self ID, position, covering).Subsequently, each node is from deeply Spend and less neighbor node selects coverage goal count out those at least as self bunch head.Turn owing to a bunch head undertakes bigger information Sending out pressure, communication energy consumption is relatively large, advantageously reduces its perception energy consumption using coverage goal those at least of counting out as a bunch head, carries High network energy consumption equalizes.As it is shown in figure 5, node s_jAnd s_kIt is node s_iNeighbor node, the two degree of depth is respectively less than node s_i's The degree of depth, due to node s_jThe impact point number covered is less than node s_k, therefore node s_jWill be as node s_iBunch head, be responsible for By node s_iDuring the network operation obtain monitoring information be forwarded to upper layer node.When each node all establishes self bunch head After, each bunch of interior nodes regulation self communication radius, until can communicate with a bunch head；Each leader cluster node adjusts self communication radius, directly To communicating with all bunches of interior nodes and the upper strata bunch head of self.Now, network has reached full-mesh state.

4, self bunch interior nodes is checked by each bunch of head further, if finding to exist bunch interior nodes of invalid covering, bunch Head node may call for bunch interior nodes and temporarily enters resting state.As shown in Figure 6, with node s₁As leader cluster node, node s₂、 s₃、s₄As bunch interior nodes bunch in, impact point t₃By node s₃And s₂Cover, impact point t simultaneously₄By node s₃And s₄Cover simultaneously Lid, it can thus be assumed that node s₃Impact point in its monitoring range is covered as invalid covering, leader cluster node s₁May call for node s₃ Temporarily enter resting state.Subsequently, network start by wheel run, each take turns in all nodes should monitor its perception radius Interior target information, sends acquired monitoring information to water surface Sink node with single-hop or multi-hop mode again.When some Bunch interior nodes because of the reasons such as depleted of energy will become feeble and die time, dormancy waking up nodes can be delayed the network coverage by leader cluster node Decay.Dormancy owing to being proposed exists only in localized network with the mechanism of waking up up, and leader cluster node can directly require that a bunch internal segment clicks on Enter dormancy or wake-up states.Compared to the network topology reconstructing method of remaining routine, the feature of its distributed implementation contributes to subtracting Few communication energy consumption, extends network life cycle.

Claims (1)

1. a new underwater wireless sensor network point overlapping control method, it is characterised in that the method passes for underwater wireless Sensor nexus Coverage Control demand, is adjusted to basis with the practicable node degree of depth first, it is possible to realize the network coverage The complex optimum of rate, connected ratio, energy consumption size and the indexs such as equilibrium, life cycle；The method comprises the following steps: (1) is to prison Surveying waters to model with cube, Sink node is fixed on water surface center, and remaining sensor node is deployed in the water surface the most equably On, the target to be monitored of some is distributed in three-dimension monitor waters uniformly at random；Each sensor node according to self with The distance of Sink node, and combine other two reference distance (the i.e. distance on Sink node and water surface summit and monitor waters The degree of depth), determine self submerged depth；The node submerged depth the nearest apart from Sink node is the least, otherwise the biggest；With this guarantee with Sink node is the biggest at a distance of the nearest Area Node density, otherwise the least；Owing to each sensor node is both needed to by single-hop or many Acquired monitoring information is sent to Sink node by jumping mode, and said method can effectively alleviate joint in Sink node near zone The information of point forwards pressure, improves network energy consumption equilibrium；

(2) after the ID that completes in step (1) adjusts, each node according to self distance with Sink node, in conjunction with Sink node and the distance on water-bed summit, with reference to the node minimum and maximum perception radius under actual environment and hardware condition, if Put self perception radius；Being different from the node perceived radius isomorphism processing mode of routine, the method is substantially to node perceived half Footpath has carried out isomery process, on the one hand can improve the network coverage further；On the other hand the perception of nodes can be balanced With traffic load so that the sensor node perception closer to Sink node loads the least, retain more multi-energy participation information and turn Send out, improve network energy consumption equilibrium；

(3) after determining perception radius by step (2), it is different from the node communication radius isomorphism processing mode of routine, the party Method proposes node should inquire, to increase communication radius mode step by step, the neighbor node that the degree of depth is less, and from the less neighbours of the degree of depth Node set select coverage goal number those at least as self leader cluster node (next-hop node that i.e. information forwards)；The method Substantially node communication radius is also carried out isomery process, on the one hand can guarantee net on the premise of reducing Communication Jamming as far as possible Network full-mesh；On the other hand can reduce the perception energy consumption of leader cluster node as far as possible, improve network energy consumption equilibrium；

(4) after completing network cluster dividing by step (3), each leader cluster node calculates in a distributed manner and dispatches, it is first determined Administered bunch in whether there is the node of invalid covering；If existing, then leader cluster node requires that this category node temporarily enters dormancy State；Along with network runs by wheel, if some bunch of interior nodes will be dead because of depleted of energy, leader cluster node can directly wake up dormancy up Node；Compared with remaining general networks reconstructing method, the method is (to refer to bunch head and bunch interior nodes between the node in subrange Between) directly communicating to complete network topology reconstruct, it is possible to reduce network energy consumption, delay the decay of network covering property, extend Network life cycle.