CN104581750B - A kind of opportunistic network can the construction method of computation model - Google Patents

Abstract

The present invention relates to a kind of opportunistic network can the construction method of computation model, described opportunistic network model comprises n node and foundation-free facility, described method is by quantizing the basic parameter feature of opportunistic network, utilize walk random model to portray the movement law of described node, and then calculate communication time and frequency, wherein said walk random model has following scene: each node moves in the region of area for S with random direction and fixing speed, when air line distance between any two nodes is less than or equal to communication distance, communication is started between node, and according to certain tactful forwarding data.

Description

A kind of opportunistic network can the construction method of computation model

Technical field

The present invention relates to field of wireless, particularly a kind of construction method of opportunistic network model.

Background technology

The video file that user terminal produces is not of uniform size, and in severe network environment, transmission has very large uncertainty.If video file small volume, under known node average communication duration and internodal communication bandwidth condition, effective transmission can be completed as a rule.But the size of video is often determined by multiple factor, as definition or resolution, the recording duration of video, the coding, compress mode etc. of video of video.When video file is larger, limited communication time and communication bandwidth are difficult to ensure that video completes in an internodal communication process, internodal connection may disconnect at any time, if video data is failed to transmit completely complete before disconnecting the connection, the data passed probably are caused to use, make inter-node communication failure, passed data and abandoned by node.In order to address this problem, video data can be divided into multiple deblockings that volume is identical, size is suitable, at the inter-node transmission producing opportunistic communication in units of piecemeal, when destination node is through a fixed response time, complete video can be merged into by correlation method after receiving all piecemeals, reach effective transmission.

In order to realize effective large transfer of data in the same area between social networks node, improve efficiency of transmission and Consumer's Experience, first the basic parameter feature of the opportunistic network quantizing to know is needed, as moving area area, number of nodes, node motion rule, node communication radius, node connect the duration and be connected interval duration, inter-node communication bandwidth etc.Subnetwork parameter easily obtains, as moving area area, number of nodes, communication bandwidth etc., and some parameters there is no fixing value, constantly change in time, Water demand its long time statistics and the regularity of distribution determine, as internodal communication time, node communication frequency.Affect the multiple because have of node communication duration, as node communication radius increases, node motion rate reduction can extend internodal communication time, even if these two features of all nodes are identical, the angle that node enters another one node still cannot ensure that all nodes have fixing communication time when meeting, because also can bring impact to communication time.Therefore it may be necessary the probabilistic model of defined node movement at random to portray the movement law of node, and then the core feature such as computing node communication time and frequency.

Summary of the invention

For above-mentioned subproblem, the invention provides a kind of construction method of opportunistic network model, the model built by described method can describe the mode of operation of opportunistic network more accurately, and the quantitative model that it provides describes the methods and strategies research contributing to improving opportunistic communication.

A construction method for opportunistic network model, described opportunistic network model comprises n node { N ₀, N ₁..., N _n-1and foundation-free facility, described method, by quantizing the basic parameter feature of opportunistic network, utilizes walk random model to portray the movement law of described node, and then calculates communication time and frequency, wherein:

Described basic parameter feature comprises: joint movements region area S, total simulation time T, node communication radius R, joint movements speed V ₀, inter-node communication bandwidth B, average nodal call duration time t _a, size of data M, deblocking size m;

Described walk random model has following scene:

Each node moves in the region of area for S with random direction and fixing speed, when the air line distance between any two nodes is less than or equal to communication distance, starts communication between node, and according to certain tactful forwarding data.

Can be verified by emulation, emulate the communication time of node and number of communications that obtain and carry out by this model the communication time that calculates and number of communications close, the model that the inventive method provides can portray the movement law of opportunistic network interior joint effectively; Adopt the method for transmitting video data of piecemeal effectively can improve the utilance of opportunistic communication in Internet Transmission in described model, reduce propagation delay time, also for the large transfer of data research of other types provides basis.

Accompanying drawing explanation

The scenario simulation of Fig. 1 joint movements;

Fig. 2 communication time calculates;

Between two nodes that Fig. 3 speed is identical, relative velocity calculates;

Fig. 4 node communication Annual distribution;

Fig. 5 average communication duration block diagram;

The total number of communications block diagram of Fig. 6.

Embodiment

Provide a kind of construction method of opportunistic network model in one embodiment, described opportunistic network model comprises n node { N ₀, N ₁..., N _n-1and foundation-free facility, described method is by quantizing the basic parameter feature of described opportunistic network, utilize walk random model to portray the movement law of described node, and then calculate communication time and frequency, wherein said basic parameter feature comprises: joint movements region area S, total simulation time T, node communication radius R, joint movements speed V ₀, inter-node communication bandwidth B, average nodal call duration time t _a, size of data M, deblocking size m;

Described walk random model has following scene:

Each node moves in the region of area for S with random direction and fixing speed, when the air line distance between any two nodes is less than or equal to communication distance, starts communication between node, and according to certain tactful forwarding data.

In this basic embodiment, by carrying out abstract to the movement of opportunistic network interior joint, for meeting of node has the feature calculation model of general significance with connection setup.Described opportunistic network model for describing the movement law of comprised node, and then calculates communication time and communication frequency; And definite network essential characteristic could provide basis for efficient transmission of messages and video block transmission method, described opportunistic network model also can be applied to other the large transfer of data research except video.

In opportunistic network, the status of all nodes is equality, and network configuration is non-hierarchical, is complanation.Node in chance net, except possessing basic communication capacity, also possesses certain storage capacity, computing capability and certain energy reserve.In addition, opportunistic network interior joint moves according to certain mode, facilitate storing-carrying-forward carrying out, realize opportunistic communication.Some special nodes are defined in some opportunistic networks, these special joints generally can not move, there is stable supply of electric power, be furnished with jumbo storage and stronger communication capacity, exist as " infrastructure " that strengthens inter-node communication, this method only relates to the internodal opportunistic communication relevant issues of foundation-free facility.

In one embodiment, the movement of the node in described opportunity model is walk random, with the model sport of model molecule Brownian movement, communication.

In another embodiment, described node is a random selected impact point in the region of S at area, then with constant speed, moves to described impact point, waits after arriving target, then select next impact point to move.

No matter which kind of move mode the node in described opportunity model adopts, in order to describe the mode of operation of opportunistic network more accurately, and research improves the methods and strategies of opportunistic communication under quantitative pattern description, needs to do necessarily abstract to opportunistic network model, and provide definition, definition N={N ₀, N ₁..., N _n-1it is the opportunistic network with n node, each node all can move in given scenario, for the ease of modeling and calculating, the structure of arranging each node is identical, there is identical rate travel, identical energy reserve, identical spatial cache and identical communication range, and fix tentatively as energy is all enough large, the communication activity in the fixed time can be maintained.Definition opportunistic network correlated variables, parameter are as shown in the table.

Table 1:

Joint movements scene as shown in Figure 1.

Under walk random mobility model, each node moves in the scene of area for S with random direction and fixing speed, when the air line distance between any two nodes is less than communication distance, starts communication between node, and according to certain tactful forwarding data.

In large transmission of video, need video data to be divided into the identical deblocking of multiple volume in advance at source node, the scope of the value m of point block size can be reduced into:

M=α t _ab, wherein 0 < α≤1 (1).

In order to quantitatively communication time can be calculated, in one embodiment, certain specific node can be selected, as N ₀analyze.In order to simplify problem, facilitate modeling, can by N ₀regard geo-stationary as, then, in S region, other nodes are all with between 0 ~ 2V ₀between relative speed move according to random direction.

Further, described node N ₀communication range arrive greatly and can see moving linearly by approximate for the motion of other node in described communication range.At node N ₀communication range in, constantly have other nodes to enter and leave, when Nodes is in a larger scene, can by other nodes, such as N ₃at N ₀motion in communication range is approximate sees moving linearly, as shown in Figure 2.

In Fig. 2, circle N ₀represent node N ₀communication range, line segment AB represents node N ₃at node N ₀movement locus in communication range, line segment AB is with random angle and position and justify N ₀intersect, always there is line segment x and be positioned on the radius vertical with AB, AB locality is random, so the length of x is between 0 ~ R, and is evenly distributed.Two node N ₀and N ₃speed identical are all V ₀, two nodes are when starting to communicate, and the direction of motion is random, as shown in Figure 3.

The two relative velocity is:

$\begin{array}{c}{V}_{03}=\sqrt{{V_0}^2+{V_0}^2-{{2V}_0}^2\mathrm{Cos\&theta;}}\\ =\sqrt{2(1-\mathrm{Cos\&theta;})}\&CenterDot;V_0\end{array}---\left(2\right)$

Wherein 0≤θ≤π

Calculate to obtain two node relative velocities, can N be obtained ₀with N ₃communication time be:

$t_{03}=\frac{\stackrel{\&OverBar;}{\mathrm{AB}}}{V_{03}}=\frac{2\sqrt{R^2-x^2}}{\sqrt{2(1-\mathrm{Cos\&theta;})\&CenterDot;V_0}},---\left(3\right)$

Wherein 0≤x≤R; 0≤θ≤π

θ is N ₃with N ₀direction of motion angle, when θ goes to zero, represents N ₀and N ₃node geo-stationary, its communication time levels off to+∞, levels off to scene running time T in this example.Two node communication duration distribution situations are drawn as shown in Figure 4 after the span (0.2 ~ π) of restriction θ.

After described scene runs certain hour, the communication time of node presents stable statistical law, known, x ~ U (0, R).

$\mathrm{AB}=2\sqrt{R^2-x^2}$

Calculate line segment AB length to expect:

$\begin{array}{c}E\left(\mathrm{AB}\right)\\ =\underset{-\&infin;}{\overset{+\&infin;}{\&Integral;}}2\sqrt{R^2{-x}^2}\&CenterDot;f\left(x\right)\mathrm{dx}\\ =\underset{0}{\overset{R}{\&Integral;}}\frac{1}{R}\&CenterDot;2\sqrt{R^2-x^2}\mathrm{dx}\\ =\frac{2}{R}\underset{0}{\overset{R}{\&Integral;}}\sqrt{R^2-x^2\mathrm{dx}}\\ =\frac{2}{R}(\frac{R^2}{2}\arcsin \frac{x}{R}+\frac{x}{2}\sqrt{R^2-x^2}){|}_{x=0}^R\\ =\frac{\mathrm{\&pi;R}}{2}\end{array}---\left(4\right)$

Further, for convenience of calculating the expectation of relative speed, distribute assuming that the random moving direction of n node in described opportunistic network model obeys U (0 ~ 2 π).Assuming that N ₀the random direction of movement is θ ₁, θ ₁~ U (0,2 π), N ₃movable random direction θ ₂, θ ₂~ U (0,2 π), the two moving direction angle theta=θ ₁-θ ₂, make θ ₂'=-θ ₂, then θ ₂' ~ U (-2 π, 0), density function is:

And θ=θ ₁+ θ ₂', θ again ₁density function be

Then the probability density function of θ is:

$f\left(\mathrm{\&theta;}\right)=\underset{-\&infin;}{\overset{\&infin;}{\&Integral;}}{f}_1\left({\mathrm{\&theta;}}_1\right){f_2}^{\&prime;}(\mathrm{\&theta;}-{\mathrm{\&theta;}}_1)d{\mathrm{\&theta;}}_1$

Due to f ₁(θ ₁) f ₂' (θ-θ ₁) non-zero region is:

$\left\{\begin{array}{c}{\mathrm{\&theta;}}_1\&Element;\left(\mathrm{0,2}\mathrm{\&pi;}\right)\\ \mathrm{\&theta;}-{\mathrm{\&theta;}}_1\&Element;(-2\mathrm{\&pi;},0)\end{array}\right.$

Therefore,

And then calculate two node relative velocities expectations:

$\begin{array}{c}E\left(v_{03}\right)=\underset{-\&infin;}{\overset{\&infin;}{\&Integral;}}{v}_0\sqrt{2(1-\mathrm{Cos\&theta;})}\&CenterDot;f\left(\mathrm{\&theta;}\right)\mathrm{d\&theta;}\\ =\underset{-2\mathrm{\&pi;}}{\overset{0}{\&Integral;}}{v}_0\sqrt{2(1-\mathrm{Cos\&theta;})}\&CenterDot;\frac{1}{4{\mathrm{\&pi;}}^2}(\mathrm{\&theta;}+2\mathrm{\&pi;})\mathrm{d\&theta;}+\\ \underset{0}{\overset{2\mathrm{\&pi;}}{\&Integral;}}{v}_0\sqrt{2(1-\mathrm{Cos\&theta;})}\&CenterDot;\frac{1}{4{\mathrm{\&pi;}}^2}(2\mathrm{\&pi;}-\mathrm{\&theta;})\mathrm{d\&theta;}\\ =\frac{4}{\mathrm{\&pi;}}\&CenterDot;v_0\end{array}$

Therefore, average communication duration should be:

$t_{03}=\frac{E\left(\mathrm{AB}\right)}{E\left(V_{03}\right)}=\frac{\frac{\mathrm{\&pi;R}}{2}}{\frac{4}{\mathrm{\&pi;}}\&CenterDot;v_0}=\frac{{\mathrm{\&pi;}}^{2}R}{8v_0}---\left(5\right)$

Can find out that average communication duration affects by node communication radius and translational speed by formula (5).

In order to calculate communication frequency, when applying described opportunity model and calculating, the described node N of supposition further ₀geo-stationary, the position of other node in the region of area S is even random distribution.

Long data block is divided into some piecemeals and is propagated by opportunistic network, consider the size of piecemeal, also will analyze its propagation delay time simultaneously.In opportunistic network, the transmission of data relies on internodal opportunistic communication, increases the chance communicated between node and effectively can shorten propagation delay time.Qualitative analysis, node motion speed is higher, and number of nodes is more, and node communication radius is larger, and place scene area is less, and the chance of communication is more.

Particularly, during all-network running time T, still suppose specified node N ₀be in relative static conditions, then for any specified node, such as N ₃, or be in N ₀communication context in, or be in N ₀communication context outside.

If by N ₃movement locus be depicted as circuit, then some circuit is in the communication radius of N0, and all the other circuits are in N ₀communication radius outside, and N ₃velocity attitude be random, speed, between 0 ~ 2V0, obtains thus, when scene running time is longer, N ₃the position being positioned at zone of action S is even random distribution.Namely in T time, N ₃with N ₀accumulative communication time mean value be

${t_{03}}^{\&prime;}=T\&CenterDot;\frac{\mathrm{\&pi;}\&CenterDot;R^2}{S}$

Obtain the total average communication duration of two nodes and single average communication duration thus, can N be calculated ₃with N ₀total number of communications expect, total number of communications formula is as follows:

${i_{03}}^{\&prime;}=\frac{{t_{03}}^{\&prime;}}{t_{03}}=\frac{T\&CenterDot;\frac{\mathrm{\&pi;}\&CenterDot;R^2}{S}}{\frac{{\mathrm{\&pi;}}^{2}R}{8v_0}}=\frac{8v_0\&CenterDot;T\&CenterDot;R}{\mathrm{\&pi;}\&CenterDot;S}$

N ₃with N ₀for arbitrary node, if total n node in scene, then global communication number of times is:

$C=\frac{n\&CenterDot;(n-1)}{2}\&CenterDot;\frac{8v_0\&CenterDot;T\&CenterDot;R}{\mathrm{\&pi;}\&CenterDot;S}$

In another embodiment, by the calculating utilizing model to carry out communication time and communication frequency, and carry out emulation experiment, the validity of verification model simultaneously.In l-G simulation test, the checking of communication time and number of communications can use opportunistic network environment (OpportunisticNetworkEnvironment, ONE) simulator to build simulated scenario and complete, and the design parameter of ONE simulator arranges as shown in table 2.

Table 2:

In ONE simulator, build corresponding scene by above-mentioned parameter, node diagnostic and quantity are set, obtain concrete node communication daily record through emulation, can add up and obtain communication time and number of communications, simultaneously according to above-mentioned formulae discovery analog value, as shown in the table.

Table 3:

By Plotting data block diagram in table 3 as illustrated in Figures 5 and 6, in legend, " ONE " represents simulation result, and in legend, " formula " represents that above use, corresponding formula carries out the result calculated.As can be seen from Fig. 5 and 3,6 experiments of 6 associative lists, the formulae discovery result of average communication duration and total number of communications is close with corresponding simulation result, and meanwhile, due to node motion and the randomness communicated, experimental result and formulae results also exist difference among a small circle.In addition, it should be noted that, what total number of communications represented is the number of times that node enters communication range mutually, in actual applications, there is node with the situation of single-channel communication, as Imote2, CSIR O, the communication pattern of the nodes such as Mica2 is single-channel communication, namely when two nodes are all within the scope of the 3rd node communication, 3rd node can only communicate with in the first two node, at this moment, enter in multiple nodes of same node communication scope, a node can only be had to communicate with, be limited to this situation, actual node communication number of times can be less than notional result.

In this specification, each embodiment adopts the mode of going forward one by one to describe, and what stress is all the difference with other embodiments, part identical, similar between each embodiment mutually see.For system embodiment, due to itself and embodiment of the method basic simlarity, so description is fairly simple, relevant part illustrates see the part of embodiment of the method.

Above to the capable detailed introduction of the construction method of a kind of opportunistic network model that the disclosure provides, apply specific case herein to set forth principle of the present disclosure and execution mode, the explanation of above embodiment just understands method of the present invention and core concept thereof for helping; Meanwhile, for those skilled in the art, according to thought of the present disclosure, all will change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention.

Claims (7)

1. opportunistic network can the construction method of computation model, and it is characterized in that, described opportunistic network model comprises n node { N ₀, N ₁..., N _n-1, and foundation-free facility, described method, by quantizing the basic parameter feature of opportunistic network, utilizes walk random model to portray the movement law of described node, and then calculates communication time and frequency, wherein:

Described basic parameter feature comprises: joint movements region area S, total simulation time T, node communication radius R, joint movements speed V ₀, inter-node communication bandwidth B, average nodal call duration time t _a, size of data M, deblocking size m;

Described walk random model has following scene:

Each node moves in the region of area for S with random direction and fixing speed, when the air line distance between any two nodes is less than or equal to communication distance, starts communication between node, and according to certain tactful forwarding data;

Described method hypothesis n node structure is identical, and have identical energy deposit, spatial cache and communication range, described energy can maintain the communication activity in the fixed time;

Described node N ₀communication range arrive greatly and can see moving linearly by approximate for the motion of other node in described communication range; By node N ₀regard geo-stationary as, then communication range is with node N ₀for center of circle communication radius is the circle of R, and then the communication time of two nodes can be calculated by the relation between movement angle, Distance geometry speed three;

Except described node N ₀the outer position of other node in the S of zone of action is even random distribution, then in T time, and other individual node and described node N ₀accumulative communication time mean value be two internodal total number of communications can be tried to achieve in conjunction with described average communication duration, and then can in the hope of the global communication number of times of n node.

2. method according to claim 1, is characterized in that, the direction of motion of a described n node is random, and speed is identical.

3. method according to claim 1, is characterized in that, the random moving direction of n node in described opportunistic network model obeys the random distribution in U (0 ~ 2 π) scope.

4. method according to claim 3, it is characterized in that, by the desired value of the desired value of calculating communication distance length, the relative speed of two nodes, can be that the desired value of communication distance length solves average communication duration with the ratio of the desired value of the relative speed of two nodes by average communication duration.

5. method according to claim 1, is characterized in that, by data in advance piecemeal when carrying out transfer of data between two nodes communicated, the scope of point block size meets: m=α t _ab, wherein 0 < α≤1.

6. method according to claim 1, is characterized in that, the movement of described node is walk random, and the mobility model of described node is model molecule Brownian movement.

7. method according to claim 1, is characterized in that, described node is a random selected impact point in the region of S at area, then with constant speed, move to described impact point, wait after arriving target, wait for a period of time, and then select next impact point to move.