CN105704754A - Wireless sensor network routing method - Google Patents

Abstract

The invention discloses a wireless sensor network routing method, which relates to the technical field of wireless sensor network communication. The wireless sensor network routing method comprises the steps of: establishing a pressure gradient function according to relative distance among a current node, neighbor nodes and a convergent node; establishing a specific resistance function by utilizing the length of a current buffer queue of the nodes and predicted traffic; establishing a inter-node link traffic model according to the established pressure gradient function and the established specific resistance function; and selecting the neighbor node with maximum link traffic as a next-hop node, thereby completing the establishment of route. The wireless sensor network routing method utilizes the concepts and mechanisms of hydraulics, effectively controls network congestion, reduces total energy consumption of the network, improves overall throughput of the network, and has low packet loss rate and time delay.

Description

A kind of wireless sensor network routing method

Technical field

The invention belongs to network communication of wireless sensor technical field, particularly relate to a kind of wireless sensor network routing method。

Background technology

Wireless sensor network be by being deployed in monitored area a large amount of low costs, low-power consumption, possess perception, data process, the sensor node of storage and wireless communication ability is formed by Ad hoc mode network, its objective is to gather collaboratively, process and the information of perceived object in transmission network region, can be widely applied to the fields such as environmental monitoring, Military Application, road traffic, health care。In sensor network, node is generally adopted the battery that energy is extremely limited, cannot recharge typically upon dispose and change power supply。It is directed to this, how to extend network life, it is achieved the effectiveness of network energy consumption and the harmonious study hotspot becoming wireless sensor network routing method。

But in some specific occasion, such as power information collection, transmission line of electricity monitoring or node adopt the situations such as solar powered, the sustainable acquisition energy supply of sensor node, will not be dead because node energy exhausts。Meanwhile, the features such as collection inter-area traffic interarea generally has periodically, flow is big, frequently resulting in offered load has exceeded the transmittability of network, and then causes network congestion。Existing method for routing is mostly for the situation that node energy is limited, and to extend network lifecycle for primary goal。Meanwhile, only minority method for routing simply according to the loading level of node queue length decision node, it does not have consider the trend of changes in flow rate, network congestion can not be efficiently controlled。

Summary of the invention

It is an object of the invention to, it is provided that a kind of wireless sensor network routing method, be used for solving the defects such as consideration network congestion limited just for node energy in prior art, not yet in effect。

To achieve these goals, the technical scheme is that a kind of wireless sensor network routing method is characterized in that the method comprises the following steps:

Step 1: according to the relative distance build-up pressure gradient function between present node, neighbor node and aggregation node；

Step 2: utilize node current buffer queue length and predicted flow rate, sets up and compares damping function；

Step 3: build inter-node link discharge model according to the pressure gradient function having built up with than damping function；

Step 4: choose the maximum neighbor node of link flow as next-hop node, complete the foundation of route。

In described step 1, the computing formula of pressure gradient function is:

Wherein, (i, a) for the pressure gradient functional value of node i Yu neighbor node a for J；(i, a) for the distance between node i to neighbor node a for d；The neighbor node collection that nbr (i) is node i；P_iAnd P_aThe respectively pressure of node i and node a。

Described P_iAnd P_aComputing formula be:

Wherein, l is the data volume that node sends every time；(x, s) for the distance of node x to aggregation node for d；E_elec、E_fsAnd E_mpEnergy consumption parameter for single order radio communication energy consumption model；D₀Distance threshold for single order radio communication energy consumption model。

Described step 2 than the computing formula of damping function is:

Wherein, (i, a) for the resistivity functional value of node i Yu neighbor node a for A；β is the degree of load threshold value of node, 0 < β < 1；C_aLoad angle value for neighbor node a。

Described C_aComputing formula be:

Wherein, Q_aQueue length for neighbor node a relief area；Q_maxHeap(ed) capacity for node relief area；Tr_aPredicted flow rate for neighbor node next cycle of a。

Described Tr_aComputing formula be:

Wherein, T is the execution cycle of method for routing of the present invention；C_aAnd P_aThe respectively child node collection of current period neighbor node a and father node collection；R_a、r_iaAnd r_ajRespectively next cycle of node a self produces the predictive value of flow rate, the predictive value of the predictive value of the average input flow rate speed from node i to node a and the average output flow speed from node a to node j；。

Described flow rate predictive value computing formula be:

Wherein, r_y' for the volume forecasting value of node a current period；N_a、n_iaAnd n_ajRespectively current period node a self produces number of data packets, enters the number of data packets of node a and leaves the number of data packets of node a；ω is the adjustment parameter of flow rate prediction, 0 < ω < 1。

In described step 3, the computing formula of discharge model is:

Wherein, (i, a) for the link flow value of node i to neighbor node a for Q；(i, a) for J, (its computing formula is J ' for i, normalized value a)Wherein, J_min(i, a) and J_max(i, a) for the node i minima to neighbor node a pressure gradient and maximum；λ is the adjustment parameter of discharge model。

The beneficial effects of the present invention is:

(1) method for routing of the present invention have effectively achieved the effectiveness of node energy consumption by build-up pressure gradient function, the total energy consumption making network is lower, namely it is applicable to the unrestricted sensor network of node energy, overcomes in traditional method just for the effective problem of node energy。Meanwhile, the present invention flow load degree by volume forecasting more precisely decision node, and then network congestion can be effectively prevented from。

(2) method for routing of the present invention considers the energy consumption of network and congested when being routed, and utilize link flow pressure gradient and two functions of resistivity to be merged, characterize node and become the probability of next-hop node, so can make wireless sensor network while energy consumption is relatively low, there is more excellent network performance。

Accompanying drawing explanation

Fig. 1 is the wireless sensor network node scattergram that the embodiment of the present invention uses；

Fig. 2 is the comparing result of the method for routing described in the embodiment of the present invention and MICRO, TADR network total energy consumption；

Fig. 3 is the comparing result of the method for routing described in the embodiment of the present invention and MICRO, TADR packet loss；

Fig. 4 is the comparing result of the method for routing described in the embodiment of the present invention and MICRO, TADR route average number of hops；

Fig. 5 is the comparing result of the method for routing described in the embodiment of the present invention and MICRO, TADR network throughput；

Detailed description of the invention

Below in conjunction with accompanying drawing, preferred embodiment is elaborated。It is emphasized that the description below is merely exemplary, rather than in order to limit the scope of the present invention and application thereof。

Instant invention overcomes the deficiency of existing wireless sensor network routing method, a kind of new wireless sensor network routing method is proposed, flow load degree according to the relative distance between present node, neighbor node and aggregation node and node build-up pressure gradient function respectively and compare damping function, and utilize link flow pressure gradient and two factors of resistivity to be merged, characterize node and become the probability of next-hop node, thus reducing the energy consumption of network, improve network throughput, and ensure that network has relatively low packet loss and time delay。

The present invention comprises the following steps:

Step 1: according to the relative distance build-up pressure gradient function between present node, neighbor node and aggregation node；

Step 2: utilize node current buffer queue length and predicted flow rate, sets up and compares damping function；

Step 3: build inter-node link discharge model according to the pressure gradient function having built up with than damping function；

Step 4: choose the maximum neighbor node of link flow as next-hop node, complete the foundation of route。

Fig. 1 is the network node scattergram that the embodiment of the present invention uses。

Initialize network environment: 100 sensor nodes with same communication ability and perception are random, be evenly distributed in the region of 100 meters × 100 meters in, sensor node is known the geographical position of oneself and has unique ID；Aggregation node is deployed in network central, and coordinate is (50,50)；The maximum communication radius of each sensor node is 30 meters。

When being routed, node can become next-hop node be transferred to this node institute consumed energy with the flow load degree of node relevant。We preferentially choose the transmission node that energy consumption is low and flow load is light as next-hop node, but owing to this relation is difficult to state definitely, the present invention adopts pressure gradient and resistivity function to characterize node becomes the probability of next-hop node, and it is fused into link flow based on hydraulics theory, it is routed according to the maximum principle of link flow。

Step 1: the relative distance build-up pressure gradient function according between present node, neighbor node and aggregation node:

Wherein, (i, a) for the pressure gradient functional value of node i Yu neighbor node a for J；(i, a) for the distance between node i to neighbor node a for d；The neighbor node collection that nbr (i) is node i；P_iAnd P_aThe respectively pressure of node i and node a, its computing formula is:

Wherein, l is the data volume that node sends every time；(x, s) for the distance of node x to aggregation node for d_；E_elec、E_fsAnd E_mpEnergy consumption parameter for single order radio communication energy consumption model；D₀Distance threshold for single order radio communication energy consumption model。

Step 2: utilize node current buffer queue length and predicted flow rate, sets up than damping function:

Wherein, (i, a) for the resistivity functional value of node i Yu neighbor node a for A；β is the degree of load threshold value of node, 0 < β < 1；C_aFor the load angle value of neighbor node a, the computing formula of node load degree is:

Wherein, Q_aQueue length for neighbor node a relief area；Q_maxHeap(ed) capacity for node relief area；Tr_aFor the predicted flow rate in neighbor node next cycle of a, its computing formula is:

Wherein, T is the execution cycle of method for routing of the present invention；C_aAnd P_aThe respectively child node collection of current period neighbor node a and father node collection；R_a、r_iaAnd r_ajRespectively next cycle of node a self produces the predictive value of flow rate, the predictive value of the predictive value of the average input flow rate speed from node i to node a and the average output flow speed from node a to node j, the computing formula of the flow rate predictive value adopted is:

Wherein, r_y' for the volume forecasting value of node a current period；N_a、n_iaAnd n_ajRespectively current period node a self produces number of data packets, enters the number of data packets of node a and leaves the number of data packets of node a；ω is the adjustment parameter of flow rate prediction, 0 < ω < 1。

Step 3: build inter-node link discharge model according to the pressure gradient function having built up with than damping function:

Wherein, (i, a) for the link flow value of node i to neighbor node a for Q；(i, a) for J, (its computing formula is J ' for i, normalized value a)Wherein, J_min(i, a) and J_max(i, a) for the node i minima to neighbor node a pressure gradient and maximum；λ is the adjustment parameter of discharge model。

Step 4: choose the maximum neighbor node of link flow as next-hop node, complete the foundation of route。

In order to check the performance of wireless sensor network routing method that the present invention proposes, carry out emulating and comparing under identical network environment with LiuguoYin et al. MICRO, FengyuanRen proposed et al. TADR proposed by it。

Adopting Matlab as emulation tool, system emulation ambient parameter is provided that

1) 100 nodes are randomly dispersed in the region of 100 meters × 100 meters, abscissa scope (0,100), vertical coordinate scope (0,100), and sensor node does not possess mobility；

2) aggregation node is static, and position is (50,50)；

3) launch and receiver circuit to process the energy that 1 Bit data consumes be 50 joules^-9, i.e. E_elec=50 joules^-9/ bit；

4) adopt free space model to launch and receiver circuit to send, to unit are, the energy that 1 Bit data consumes be 100 joules^-12, i.e. E_fs=100 joules^-12/ bit/rice²；

5) adopt Multipath Transmission model to launch and receiver circuit to send, to unit are, the energy that 1 Bit data consumes be 0.0013 joule^-12, i.e. E_mp=0.0013 joule^-12/ bit/rice⁴；

6) distance threshold of single order radio communication energy consumption model is 87 meters, i.e. d₀=87 meters；

7) the adjustment parameter of flow rate prediction is 0.1, i.e. ω=0.1；

8) the degree of load threshold value of node is 0.7, i.e. β=0.7；

9) the adjustment parameter of discharge model is 2.5, i.e. λ=2.5；

10) data volume that node sends every time is 2048 bits, i.e. l=2048 bit；

11) in TADR, weight value is α=0.71；

Above parameter is also non-constant, can change some parameter as required for different emulation contents。

Fig. 2 be equivalent environment lower node quantity from 100 to 300 change time corresponding network total energy consumption simulation result。

It can be seen that the network total energy consumption of the method for routing of present invention proposition is significantly lower than MICRO and TADR, and along with the increase of number of nodes, its advantage is more and more obvious。

Fig. 3 be equivalent environment lower node buffer size from 10 to 100 change time corresponding network packet loss rate simulation result。

Visible along with the change of buffer size, the packet loss of the method for routing that the present invention proposes is all significantly lower than other two kinds of methods。When node relief area is wrapped lower than 50, it is with the obvious advantage that the method for routing in the present invention compares other two kinds of methods；When relief area continues to increase, nodes is congested all to be compared gently, therefore three kinds of method divergences are less, but the packet loss of the method for routing in the present invention is still minimum。

Fig. 4 be equivalent environment lower node quantity from 100 to 300 change time corresponding network route average number of hops simulation result。

Can be seen that, the present invention proposes the route average number of hops of method significantly lower than MICRO and TADR, and along with the increase of node number changes little, illustrate that network congestion is had stronger regulating power by the method for routing that the present invention proposes, and then there is more stable route average number of hops。

Fig. 5 be under equivalent environment data rate from 1 to 7 change time corresponding network throughput simulation result。

It can be seen that handling capacity relatively MICRO and the TADR of the method for present invention proposition is respectively increased nearly 60% and 20%, and along with the increase of number of nodes, its advantage is more and more obvious。

Simulation result shows, method for routing in the present invention optimizes choosing of next-hop node, efficiently controls network congestion, reduces the total energy consumption of transmitted data on network, and ensure that network has relatively low packet loss and route average number of hops, significantly improve the handling capacity of network simultaneously。

The above; being only the present invention preferably detailed description of the invention, but protection scope of the present invention is not limited thereto, any those familiar with the art is in the technical scope that the invention discloses; the change that can readily occur in or replacement, all should be encompassed within protection scope of the present invention。Therefore, protection scope of the present invention should be as the criterion with scope of the claims。

Claims (8)

1. a wireless sensor network routing method, is characterized in that the method includes:

Step 1: according to the relative distance build-up pressure gradient function between present node, neighbor node and aggregation node；

Step 2: utilize node current buffer queue length and predicted flow rate, sets up and compares damping function；

Step 3: build inter-node link discharge model according to the pressure gradient function having built up with than damping function；

Step 4: choose the maximum neighbor node of link flow as next-hop node, complete the foundation of route。

2. method according to claim 1, is characterized in that in described step 1, the computing formula of pressure gradient function is:

$J(i,a)=\frac{P_i-P_a}{d(i,a)},a\&Element;nbr\left(i\right),$

Wherein, (i, a) for the pressure gradient functional value of node i Yu neighbor node a for J；(i, a) for the distance between node i to neighbor node a for d；The neighbor node collection that nbr (i) is node i；P_iAnd P_aThe respectively pressure of node i and node a。

3. method according to claim 2, is characterized in that described P_iAnd P_aComputing formula be:

$P_x=\left\{\begin{array}{c}{\mathrm{lE}}_{elec}+{\mathrm{lE}}_{fs}{d}^2\left(x,s\right),d\left(x,s\right)<d_0\\ {\mathrm{lE}}_{elec}+{\mathrm{lE}}_{mp}{d}^4\left(x,s\right),d\left(x,s\right)\&GreaterEqual;d_0\end{array}\right.,x\&Element;\left\{i,a\right\},$

Wherein, l is the data volume that node sends every time；(x, s) for the distance of node x to aggregation node for d；E_elec、E_fsAnd E_mpEnergy consumption parameter for single order radio communication energy consumption model；D₀Distance threshold for single order radio communication energy consumption model。

4. method according to claim 1, it is characterized in that in described step 2 than the computing formula of damping function be:

$A(i,a)=\left\{\begin{array}{c}{C}_a,if(C_a\&le;\mathrm{\&beta;})\\ +\mathrm{\&infin;},if(C_a>\mathrm{\&beta;})\end{array}\right.,$

Wherein, (i, a) for the resistivity functional value of node i Yu neighbor node a for A；β is the degree of load threshold value of node, 0 < β < 1；C_aLoad angle value for neighbor node a。

5. method according to claim 4, is characterized in that described C_aComputing formula be:

$C_a=\frac{Q_a+{\mathrm{Tr}}_a}{Q_{max}},$

Wherein, Q_aQueue length for neighbor node a relief area；Q_maxHeap(ed) capacity for node relief area；Tr_aPredicted flow rate for neighbor node next cycle of a。

6. method according to claim 5, is characterized in that described Tr_aComputing formula be:

${\mathrm{Tr}}_a=\left(r_a+\underset{i\&Element;C_a}{\&Sigma;}{r}_{ia}-\underset{j\&Element;P_a}{\&Sigma;}{r}_{aj}\right)\&times;T,$

Wherein, T is the execution cycle of method for routing of the present invention；C_aAnd P_aThe respectively child node collection of current period neighbor node a and father node collection；R_a、r_iaAnd r_ajRespectively next cycle of node a self produces the predictive value of flow rate, the predictive value of the predictive value of the average input flow rate speed from node i to node a and the average output flow speed from node a to node j。

7. method according to claim 6, it is characterized in that described flow rate predictive value computing formula be:

$r_y=(1-\mathrm{\&omega;})\&times;{r_y}^{\&prime;}+\mathrm{\&omega;}\&times;\frac{n_y}{T},y\&Element;\{a,ia,aj\},$

Wherein, r '_yVolume forecasting value for node a current period；N_a、n_iaAnd n_ajRespectively current period node a self produces number of data packets, enters the number of data packets of node a and leaves the number of data packets of node a；ω is the adjustment parameter of flow rate prediction, 0 < ω < 1。

8. method according to claim 1, is characterized in that the computing formula of the discharge model of described step 3 is:

$Q(i,a)=\sqrt{J^{\&prime;}(i,a)/A(i,a)},$

Wherein, (i, a) for the link flow value of node i to neighbor node a for Q；(i, a) for J, (its computing formula is J ' for i, normalized value a)Wherein, J_min(i, a) and J_max(i, a) for the node i minima to neighbor node a pressure gradient and maximum。