CN103532667A - Reliable wireless sensor network data transmission method based on Chinese remainder theorem - Google Patents

Abstract

The invention discloses a reliable wireless sensor network data transmission method based on a Chinese remainder theorem. The method comprises the following steps: dispersing nodes randomly, performing topology control on a network in combination with a Gabriel graph on the basis of a UDG (Unit Disk Graph), adjusting the node transmitting power, and updating a neighbor node set to form a network data communication topology; performing linear encoding on a data packet to be transmitted, encoding data into a plurality of linearly-irrelevant data, partitioning the data into small sub-data packets by adopting the Chinese remainder theorem, transmitting the sub-data packets to a sink node by adopting a minimum hop routing algorithm along multiple paths, merging the partitioned data through the sink node, and decoding the encoded data. The data communication method disclosed by the invention is easy to implement, the service efficiency of network energy can be increased effectively without any complex route protocol, the occurrence of network congestion is reduced, the encoding effect is brought into full play, and data can be reliably transmitted to the sink node; the method can be applied to large-scale wireless sensor networks.

Description

Wireless sensor network data method for reliable transmission based on Chinese remainder theorem

Technical field

The invention belongs to technical field of wireless, relate to the data transmission method in a kind of wireless sensor network, specifically, relate to a kind of wireless sensor network data method for reliable transmission based on Chinese remainder theorem.

Background technology

Radio sensing network WSNs be take the multi-hop wireless communication network that a large amount of sensing nodes are the basic element of character.These node deployments are behind given application region, and node is information the generated data in this region of perception separately, then data is passed to sink node and be further processed by sink node by the mode of wireless multi-hop.

In the application of radio sensing network, sensor node needs to form network by Ad hoc mode after being dispersed at random certain surveyed area, and in the region of oneself data acquisition.Node conventionally will reported the positional information of node when destination node sends data simultaneously.Destination node can judge accurately the region that this event occurs after receiving data, and takes further measures.For example, in forest fire protection, when sensor node monitors fire generation, fire fighter needs the position that accurately location fire occurs.Generally, sensor node can utilize GPS (global positioning system) positioner or located in connection agreement to determine geographical position, and by broadcast data packet and neighbor node exchange position information.

Sensor node is conventionally by the powered battery of finite energy, to the service efficiency team node of electric weight and most important concerning whole sensor network.Wherein research shows, the energy that node communication stage transmission data consume accounts for the major part that node energy consumes.Method, mechanism or the agreement of WSNs energy efficiency have been proposed to improve according to These characteristics researchers.

Aspect reducing network energy consumption, extending network lifetime, the method effect of controlling based on topology is remarkable, gets the attention and studies.Topology is controlled the two kinds of methods that generally mainly contain: a kind of based on power control, another is to adopt layering measure.The method of controlling based on power is to meet under the prerequisite that is communicated with covering basic demand, reduces as far as possible node transmitting power and improves energy efficiency.The thought of layering be according to node in abutting connection with situation, manage to change cellular logic topology, common way is divided into bunch shape structure one by one by whole network.

That at power, controls realizes in approach, and people often use and close on figure.Gabriel figure is a kind of conventional figure that closes on.Generally on the Gabriel figure of structure, can find and the suitable path of former figure lowest energy consumption.In document AnEnergy-saving Algorithm of WSN based on Gabriel Graph, author proposes to using the GG limit of each node maximum as the corresponding communication radius of transmitting power adjustment, allows the process of transmitting of data carry out on Gabriel figure simultaneously.The alternative down hop of each node only limits to and its node that has GG limit to be connected like this, and is no longer all neighbours within the scope of maximum communication, has reduced communication amount of calculation and Communication Jamming odds.

In order to fall low-energy-consumption and to consider the complexity of algorithm.In document ANovel Reliable and Energy-Saving Forwarding Technique for Wireless Sensor Networks.The applicant proposes, at source node, data are carried out to data compression, and the initial data that needs are sent utilizes Chinese remainder theorem to cut apart, and larger data are divided into a series of less data, makes each node only need to send each little subdata.Sink node needs only correct these subdatas of receiving, just can utilize them to be reduced into former data.Here only need to suppose that the computing capability of sink node and energy are greater than traditional sensor node and just go for WSN and can keep lower transfer of data complexity.But whether the structure of not considering network topology in article can bring Communication Jamming, do not consider the impact of insecure communication link on reliable data transmission simultaneously yet.

Summary of the invention

The object of the invention is to overcome the defect that above-mentioned technology exists, a kind of wireless sensor network data method for reliable transmission based on Chinese remainder theorem is provided, consider topology control and the impact of unreliable communication link on transfer of data simultaneously, by Adoption Network, encode and increase the reliability that transmitting redundancy improves transfer of data, effectively reduce the dependence to individual data, reduced the impact that unreliable communication link brings, can improve the energy utilization rate of whole network simultaneously, reduce the Communication Jamming of network, meet the transmission requirement of large-scale wireless sensing network.

Its concrete technical scheme is:

A wireless sensor network data method for reliable transmission for Chinese remainder theorem, for its network energy consumption balance of wireless sensing and transmit reliable problem, adopts linear network encoding and jete routing algorithm to realize data communication, comprises the steps:

(1) in the plane domain of L * L, shed at random N wireless sensor node, Sink node is positioned at edge, monitored area, is responsible for accepting image data and data are carried out to relevant treatment;

(2) in wireless sensor network WSNs, sensor node, with maximum transmission power broadcast position information, and records the information of neighbor node, forms the unit circle figure UDG of network;

(3) according to unit circle figure UDG, utilize Gabriel figure construction algorithm, form the topological structure of network based on Gabriel figure;

(4), according to the topological structure of Gabriel figure, form data communication topological structure;

(4a) node i is calculated respectively the distance with each neighbor node under Gabriel figure, finds out ultimate range wherein, and adjusts the transmitting power of self, adjusts its communication radius with consistent to large distance, reduction Communication Jamming;

(4b) node i sends query messages metadata information after adjusting radius under new communication radius, finds out the neighbor node of asymmetric link, thereby obtains the neighbor node collection N (i) of symmetric links.;

(4c) repeat (4a), (4b), until all node i have all obtained neighbor node collection N (i) in whole network, form the data communication topology of network;

(5), in the data communication topology building, after completing respective handling by packet, the node that need to send data is delivered to sink node with minimum hop count routing algorithm;

(5a) first the node i that need to send data searches the number of neighbor node in self neighbor node collection N (i) | N (i) |;

If (5b) number of neighbor node collection is 1 o'clock, node i is left intact to data and data is passed to unique neighbor node j, and node j searches the number of the neighbor node collection N (j) of self | N (j) |;

If (5c) number of neighbor node collection | N (j) | when being more than or equal to 2, first packet is encoded the data to the data of many parts of linear independences by uniform enconding, and then each data acquisition is carried out to Data Segmentation by Chinese remainder theorem, the number of Data Segmentation is identical with the number of the neighbor node collection of node i;

(5d) data after cutting apart are delivered to sink node with minimum hop count routing algorithm;

(5e) data of sink node after receiving and cutting apart, first merge the data after cutting apart, then by network code, carry out data decode and recover initial data, the data that source node will send;

(5f), in other times section, if there is other sensor nodes u need to send the packet of its image data in network, repeats (5a)-(5e) just make sink node and receive the data that source node sends.

Further preferably, described in step (3) according to unit circle figure UDG, utilize Gabriel figure building method, form the Gabriel figure topology of network, carry out as follows:

(3a) node i is according to the positional information of own and neighbor node j, calculates to take home position and the radius of the circle that i and the two line of j be diameter;

(3b) node i is calculated the above-mentioned center of circle to the distance of other neighbor nodes, and judges whether this distance is less than above-mentioned radius, if be less than, this node is concentrated and is deleted from neighbor node;

(3c) repeating step (2a)-(2b), until node i all operates all neighbor node j, thereby completes the renewal of this node neighbor node;

(3d) repeating step (2c), knows that all node k in whole wireless sensor network have carried out above-mentioned operation, and this has just completed the Gabriel topology constructing of whole network;

Further preferably, the number of the described neighbor node collection as node j of step (5c) | N (j) | when being more than or equal to 2, first packet is encoded the data to the data of many parts of linear independences by uniform enconding, and then each data acquisition is carried out to Data Segmentation by Chinese remainder theorem, carry out in accordance with the following steps;

(5c1) when the number of the neighbor node collection of node j | N (j) | while being more than or equal to 2, search in its neighbor node collection N (j) and whether comprise sink node, if sink node belongs to its neighbor node collection, directly selecting sink node is next-hop node, and to wanting packet to be left intact; Otherwise, execution step (5c2);

(5c2) need to send m the packet that hypothesis will send be B to hypothesis source node S ₁, B ₂, L, B _m, source node S is from finite field F _qin choose at random coding vector

by the m of source node packet B ₁, B ₂, L, B _mbe encoded into n packet E ₁, E ₂, L, E _n(n>=m):

$E_i=\underset{j=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\α}}_{\mathrm{ij}}{B}_j;i=\mathrm{1,2},L,n$

Like this by adopting uniform enconding, the m that just source node S will be sent originally packet B ₁, B ₂, L, B _mbe converted to E ₁, E ₂, L, E _n, the benefit of this coding is different data to blend into one, and has reduced the dependence to individual data, has increased the transmission reliability of data;

(5c3) the data E after transfer encoding _ktime, at intra-node, define intrinsic prime number p _ki, wherein the number of prime number is identical with the number of its neighbor node, known according to Chinese remainder theorem, to any given integer set { E _k1, E _k2..., E _kN, there is unique integer

make E _k=E _ki(modp _ki), therefore, at a data E of transmission _ktime, only need on each paths of its neighbor node, transmit E _ki.

Further preferably, in step (5e), sink node is received the data after cutting apart, and first the data after cutting apart is merged, then by network code, decoding data is recovered to initial data, carries out in accordance with the following steps:

(5e1) when sink node is received the data E after cutting apart _kiafter, utilize Chinese remainder theorem by the data E of cutting apart _kimerge, obtain the data E before cutting apart _k, E _kcomputational methods be:

coefficient wherein $c_{\mathrm{ki}}=Q_{\mathrm{ki}}{q}_{\mathrm{ki}},M_k=\underset{i}{\overset{N}{\mathrm{\Π}}}{p}_{\mathrm{ki}},Q_{\mathrm{ki}}=\frac{M_k}{p_{\mathrm{ki}}},$ Q _kimake c _ki=1 (modp _ki);

(5e2) the data E before being cut apart _kafter, m the data that tentative segment point is received are respectively E ₁, E ₂... E _m, further judge corresponding m coefficient vector α ₁, α ₂... α _klinear independence whether, if so, data are decoded by coding:

$\left[\begin{array}{c}{B}_1\\ B_2\\ M\\ B_m\end{array}\right]={\left(\begin{array}{ccc}{\mathrm{\α}}_{11}& K& {\mathrm{\α}}_{1m}\\ M& O& M\\ {\mathrm{\α}}_{m1}& L& {\mathrm{\α}}_{\mathrm{mm}}\end{array}\right)}^{-1}\left[\begin{array}{c}{E}_1\\ E_2\\ M\\ E_m\end{array}\right]$

By above-mentioned decoding, just data are reduced into initial data B ₁, B ₂..., B _m.

Compared with prior art, beneficial effect of the present invention is:

The present invention carries out topology in conjunction with Gabriel figure to WSNs and controls, and limits the communication range of each node in conjunction with Gabriel figure, has reduced the energy consumption of node data transmitting-receiving, and has reduced internodal Communication Jamming.

The present invention utilizes the broadcast characteristic of wireless transmission and network coding technique to propose the reliable transmission plan of data, has avoided the impact of the unreliability of wireless link on transfer of data.By guaranteeing the transmitting of local node, give full play to coded actions, guarantee that data are transferred to sink node reliably.

The present invention utilizes Chinese remainder theorem that original larger Data Segmentation is become to a series of little subdata bags in Adoption Network coding, and each node only need transmit subdata bag, has avoided occurring network congestion in the time of balanced energy consumption.

The present invention is applicable to large-scale wireless sensor network, has good extensibility.

Accompanying drawing explanation

Fig. 1 is the general flow chart that the present invention is based on the wireless sensor network data method for reliable transmission of Chinese remainder theorem;

Fig. 2 needs the node S that transmits data to sending the procedure chart of data data processing in the present invention;

Fig. 3 is the present invention unit circle figure UDG that emulation generates in given network and Gabriel figure topology, wherein,

Fig. 3 (a) is the present invention unit circle figure UDG that emulation generates in given network topology;

Fig. 3 (b) is the present invention Gabriel that emulation generates in given network figure topology;

Fig. 4 is that the present invention adopts 100 resulting ceiling capacities of Chinese remainder theorem post-simulation to reduce the ordering chart of the factor (MERF);

Fig. 5 is that the present invention adopts the front comparison diagram with cutting apart the large smallest number of rear different pieces of information of Chinese remainder theorem Data Segmentation, wherein:

Fig. 5 (a) is the quantity that the present invention does not adopt the front data package size of Chinese remainder theorem Data Segmentation;

Fig. 5 (b) is the quantity that the present invention adopts data package size after Chinese remainder theorem Data Segmentation;

Fig. 6 is that the present invention adopts uniform enconding and the transmission reliability comparison diagram that does not adopt uniform enconding.

Embodiment

Below in conjunction with the drawings and specific embodiments, technical scheme of the present invention is described in more detail.

With reference to Fig. 1, the present invention is based on the wireless sensor network data method for reliable transmission of Chinese remainder theorem, concrete steps are as follows:

Step 1., in the plane domain of L * L, is shed N wireless sensor node at random, and Sink node is positioned at edge, monitored area, is responsible for accepting image data and data are carried out to relevant treatment.

In step 2. wireless sensor network WSNs, sensor node is with maximum transmission power broadcast position information, and records the information of neighbor node, forms the unit circle figure UDG of network.

(2.1) in network, each node is broadcasted the positional information of self with maximum transmission power;

(2.2) node i is received after the positional information of neighbor node j, records this positional information and j is added to adjacency list, forms the initial adjacency list of node i;

(2.3) repeat (2.1)-(2.2), until node i all adds adjacency list by all nodes of received positional information, form initial adjacency list;

(2.4) repeat (2.3), until the initial adjacency list that all nodes of network all obtain, thereby the unit circle figure UDG of formation network.

Step 3., according to unit circle figure UDG, is utilized Gabriel figure construction algorithm, forms the topological structure of network based on Gabriel figure.

(3.1) node i is according to the positional information of own and neighbor node j, calculates to take home position and the radius of the circle that i and the two line of j be diameter;

(3.2) node i is calculated the above-mentioned center of circle to the distance of other neighbor nodes, and judges whether this distance is less than above-mentioned radius, if be less than, this node is concentrated and is deleted from neighbor node;

(3.3) repeating step (3.1)-(3.2), until node i all operates all neighbor node j, thereby complete the renewal of this node neighbor node;

(3.4) repeating step (3.3), knows that all node k in whole wireless sensor network have carried out above-mentioned operation, and this has just completed the Gabriel topology constructing of whole network.

Step 4., according to the topological structure of Gabriel figure, forms data communication topological structure:

(4.1) node i is calculated respectively the distance with each neighbor node under Gabriel figure, finds out ultimate range wherein, and adjusts the transmitting power of self, adjusts its communication radius with consistent to large distance, reduction Communication Jamming;

(4.2) node i sends query messages metadata information after adjusting radius under new communication radius, finds out the neighbor node of asymmetric link, thereby obtains the neighbor node collection N (i) of symmetric links;

(4.3) repeat (4.1)-(4.2) until in whole network all node i all obtained neighbor node collection N (i), form the data communication topology of network.

Step 5. is in the data communication topology building, and the node that need to send data carries out packet with minimum hop count routing algorithm, along multipath, to be delivered to sink node after compressed encoding processing.

(5.1) first the node i that need to send data searches the number of neighbor node in self neighbor node collection N (i);

(5.2) if the number of neighbor node collection is 1 o'clock, node i is left intact to data and data is passed to unique neighbor node j, and node j searches the number of the neighbor node collection N (j) of self;

(5.3) if first packet is encoded the data to the data of many parts of linear independences when the number of neighbor node collection is more than or equal to 2 by uniform enconding, and then each data acquisition is carried out to Data Segmentation by Chinese remainder theorem, the number of Data Segmentation is identical with the number of the neighbor node collection of node i;

(5.3.1) when the number of the neighbor node collection of node j | N (j) | while being more than or equal to 2, search in its neighbor node collection N (j) and whether comprise sink node, if sink node belongs to its neighbor node collection, directly selecting sink node is next-hop node, and to wanting packet to be left intact; Otherwise, execution step (5c2);

(5.3.2) need to send m the packet that hypothesis will send be B to hypothesis source node S ₁, B ₂, L, B _m, source node S is from finite field F _qin choose at random coding vector

by the m of source node packet B ₁, B ₂, L, B _mbe encoded into n packet E ₁, E ₂, L, E _n(n>=m):

$E_i=\underset{j=1}{\overset{m}{\mathrm{\Σ}}}{\mathrm{\α}}_{\mathrm{ij}}{B}_j;i=\mathrm{1,2},L,n$

Like this by adopting uniform enconding, the m that just source node S will be sent originally packet B ₁, B ₂, L, B _mbe converted to E ₁, E ₂, L, E _n.The benefit of this coding is different data to blend into one, and has reduced the dependence to individual data, has increased the transmission reliability of data;

(5.3.3) the data E after transfer encoding _ktime, at intra-node, define intrinsic prime number p _ki, wherein the number of prime number is identical with the number of its neighbor node.Known according to Chinese remainder theorem, to any given integer set { E _k1, E _k2..., E _kN, there is unique integer

make E _k=E _ki(modp _ki).Therefore, at a data E of transmission _ktime, only need on each paths of its neighbor node, transmit E _ki.

(5.4) multi-path transmission of the data after cutting apart being pointed to neighbor node with minimum hop count routing algorithm along each is to sink node;

(5.4.1) nodes represents with h to the jumping figure of Sink node, and when initial, Sink node s arranges h _s=0, other Node configuration h=∞;

(5.4.2) Sink node broadcasts is with h _s=0 Route establishment information, neighbor node j receives after information, is arrived the jumping figure h of Sink node _jchange h into _u=h _j+ 1=1, and broadcast is with h _j=1 information;

(5.4.3) in network arbitrary node v receive node j broadcast with h<sub TranNum="227">j</sub>route establishment information after, do following processing: 1) if h<sub TranNum="228">v</sub><h<sub TranNum="229">j</sub>+ 1, do not do any processing; 2) if h<sub TranNum="230">v</sub>=h<sub TranNum="231">j</sub>+ 1, node j is added in routing table; 3) if h<sub TranNum="232">v</sub>>h<sub TranNum="233">j</sub>+ 1, empty existing routing table, node j is added in routing table, upgrade h<sub TranNum="234">v</sub>=h<sub TranNum="235">j</sub>+ 1, and broadcast is with h<sub TranNum="236">v</sub>route establishment information;

(5.4.4) along said process, upgrade Route establishment information, finally can set up the route based on minimum hop count.

(5.5) data of sink node after receiving and cutting apart, first merge the data after cutting apart.By network code, carry out data decode again and recover initial data, the data that source node will send.

(5.5.1) when sink node is received the data E after cutting apart _kiafter, utilize Chinese remainder theorem by the data E of cutting apart _kimerge, obtain the data E before cutting apart _k.E _kcomputational methods be:

coefficient wherein $c_{\mathrm{ki}}=Q_{\mathrm{ki}}{q}_{\mathrm{ki}},M_k=\underset{i}{\overset{N}{\mathrm{\&Pi;}}}{p}_{\mathrm{ki}},Q_{\mathrm{ki}}=\frac{M_k}{p_{\mathrm{ki}}},$ Q _kimake c _ki=1 (modp _ki);

(5.5.2) the data E before being cut apart _kafter, m the data that tentative segment point is received are respectively E ₁, E ₂... E _m, further judge corresponding m coefficient vector α ₁, α ₂... α _klinear independence whether, if so, data can be decoded by coding:

$\left[\begin{array}{c}{B}_1\\ B_2\\ M\\ B_m\end{array}\right]={\left(\begin{array}{ccc}{\mathrm{\&alpha;}}_{11}& K& {\mathrm{\&alpha;}}_{1m}\\ M& O& M\\ {\mathrm{\&alpha;}}_{m1}& L& {\mathrm{\&alpha;}}_{\mathrm{mm}}\end{array}\right)}^{-1}\left[\begin{array}{c}{E}_1\\ E_2\\ M\\ E_m\end{array}\right]$

By above-mentioned decoding, just data can be reduced into initial data B ₁, B ₂..., B _m.

Effect of the present invention can be further described by following emulation:

Simulated conditions

In the plane domain of L * L, shed at random N wireless sensor node, network node is evenly distributed in monitored area, Sink node is positioned at edge, monitored area, is responsible for accepting image data and data are carried out to relevant treatment.In tentation data transmitting procedure, do not carry out data fusion, suppose that whole network has the perfect medium access control MAC agreement that combines time division multiple access TDMA technology simultaneously.

Other simulation parameters are as shown in table 1:

Table 1

Region (m 2)	100×100
		Interstitial content	100
Sink position	(0，0)
		Primary power (J)	2
Maximum communication radius (m)	25
		Data package size (bits)	[10，80]

Emulation content

Emulation 1, in conjunction with above-mentioned parameter, generating network unit circle figure UDG and Gabriel figure topology, as shown in Figure 3, wherein Fig. 3 (a) is network unit circle figure UDG, Fig. 3 (b) is network G abriel figure topology.From relatively can finding out of Fig. 3 (a) and Fig. 3 (b), Gabriel figure topology has guaranteed the connectedness of whole network, compares with UDG simultaneously, and each node neighbor node decreased number, can reduce inter-node communication and disturb.

Emulation 2, in conjunction with above-mentioned simulated conditions, adopt respectively based on grandson be born in the transmission plan of theorem and network uniform enconding with only adopt minimum hop count reportedly defeated two kinds of communication plans compare, analyze consolidated network lower node Energy Expenditure Levels and data packet transmission situation, result as Figure 4-Figure 6.

In Fig. 4, we reduce factor M ERF with ceiling capacity consumption and assess two kinds of communication mode node ceiling capacity Expenditure Levels, wherein

w represents only to adopt reportedly defeated transmitted data package size of minimum hop count, w _cRTrepresent the size that adopts grandson to be born in packet that theorem is transmitted.Under random 100 different structures, send at random the packet of [10,80] bits size, and the result of 100 times is sorted.As can be seen from Figure 4, than only taking minimum jump set defeated, adopt Chinese remainder theorem and coded system to transmit the energy that each node can at least be saved 45% left and right.In Fig. 5, when having provided employing Chinese remainder theorem and not taked Chinese remainder theorem, send the contrast of data package size situation.As can be seen from Figure 5, adopt Chinese remainder theorem the data compression of [10,80] bits can be arrived to [10,20] bits.Reduced the energy consumption that individual node transmission larger data bag causes, balanced offered load, has avoided network congestion to a certain extent.In Fig. 6, provided while considering unreliable communication link, not the reliability contrast of Adoption Network coding and the transmission of Adoption Network coded data.Suppose that node acquiescence sends 3 packets, can send data encoding 5 packets while adopting uniform enconding.As can be seen from Figure 6, adopt the data transmission credibility of uniform enconding will be far longer than the data transmission credibility that does not adopt uniform enconding.

Symbol description: WSNs:Wireless Sensor Networks wireless sensor network

I: sensor node

J: the neighbor node of sensor node i

S: the source node that sends data

W: send data package size while not taking Chinese remainder theorem

W _cRT: while taking Chinese remainder theorem, send data package size

N (i): the neighbor node collection of node i

| N (i) |: node i neighbor node collection number

MERF:Max Energy Reduction Factor ceiling capacity reduces the factor

UDG:Unit Disk Graph unit circle figure

GG:Gabriel Graph Gabriel figure

MAC:Media Access Control medium access control

TDMA:Time Division Multiple Access time division multiple access

The above; it is only preferably embodiment of the present invention; protection scope of the present invention is not limited to this; anyly be familiar with those skilled in the art in the technical scope that the present invention discloses, the simple change of the technical scheme that can obtain apparently or equivalence are replaced and are all fallen within the scope of protection of the present invention.

Claims (4)

1. the wireless sensor network data method for reliable transmission based on Chinese remainder theorem, is characterized in that, comprises the steps:

(1) in the plane domain of L * L, shed at random N wireless sensor node, Sink node is positioned at edge, monitored area, is responsible for accepting image data and data are carried out to relevant treatment;

(2) in wireless sensor network WSNs, sensor node, with maximum transmission power broadcast position information, and records the information of neighbor node, forms the unit circle figure UDG of network;

(3) according to unit circle figure UDG, utilize Gabriel figure construction algorithm, form the topological structure of network based on Gabriel figure;

(4), according to the topological structure of Gabriel figure, form data communication topological structure;

(4a) node i is calculated respectively the distance with each neighbor node under Gabriel figure, finds out ultimate range wherein, and adjusts the transmitting power of self, adjusts its communication radius with consistent to large distance, reduction Communication Jamming;

(4b) node i sends query messages metadata information after adjusting radius under new communication radius, finds out the neighbor node of asymmetric link, thereby obtains the neighbor node collection N (i) of symmetric links.;

(4c) repeat (4a), (4b), until all node i have all obtained neighbor node collection N (i) in whole network, form the data communication topology of network;

(5), in the data communication topology building, after completing respective handling by packet, the node that need to send data is delivered to sink node with minimum hop count routing algorithm;

(5a) first the node i that need to send data searches the number of neighbor node in self neighbor node collection N (i) | N (i) |;

If (5b) number of neighbor node collection is 1 o'clock, node i is left intact to data and data is passed to unique neighbor node j, and node j searches the number of the neighbor node collection N (j) of self | N (j) |;

If (5c) number of neighbor node collection | N (j) | when being more than or equal to 2, first packet is encoded the data to the data of many parts of linear independences by uniform enconding, and then each data acquisition is carried out to Data Segmentation by Chinese remainder theorem, the number of Data Segmentation is identical with the number of the neighbor node collection of node i;

(5d) data after cutting apart are delivered to sink node with minimum hop count routing algorithm;

(5e) data of sink node after receiving and cutting apart, first merge the data after cutting apart, then by network code, carry out data decode and recover initial data, the data that source node will send;

(5f), in other times section, if there is other sensor nodes u need to send the packet of its image data in network, repeats (5a)-(5e) just make sink node and receive the data that source node sends.

2. the wireless sensor network data method for reliable transmission based on Chinese remainder theorem according to claim 1, described in step (3) according to unit circle figure UDG, utilize Gabriel figure building method, form the Gabriel figure topology of network, carry out as follows:

(3a) node i is according to the positional information of own and neighbor node j, calculates to take home position and the radius of the circle that i and the two line of j be diameter;

(3b) node i is calculated the above-mentioned center of circle to the distance of other neighbor nodes, and judges whether this distance is less than above-mentioned radius, if be less than, this node is concentrated and is deleted from neighbor node;

(3c) repeating step (2a)-(2b), until node i all operates all neighbor node j, thereby completes the renewal of this node neighbor node;

(3d) repeating step (2c), knows that all node k in whole wireless sensor network have carried out above-mentioned operation, and this has just completed the Gabriel topology constructing of whole network.

3. the wireless sensor network data method for reliable transmission based on Chinese remainder theorem according to claim 1, the number of the described neighbor node collection as node j of step (5c) | N (j) | when being more than or equal to 2, first packet is encoded the data to the data of many parts of linear independences by uniform enconding, and then each data acquisition is carried out to Data Segmentation by Chinese remainder theorem, carry out in accordance with the following steps;

(5c1) when the number of the neighbor node collection of node j | N (j) | while being more than or equal to 2, search in its neighbor node collection N (j) and whether comprise sink node, if sink node belongs to its neighbor node collection, directly selecting sink node is next-hop node, and to wanting packet to be left intact; Otherwise, execution step (5c2);

(5c2) need to send m the packet that hypothesis will send be B to hypothesis source node S ₁, B ₂, L, B _m, source node S is from finite field F _qin choose at random coding vector

by the m of source node packet B ₁, B ₂, L, B _mbe encoded into n packet E ₁, E ₂, L, E _n(n>=m):

Like this by adopting uniform enconding, the m that just source node S will be sent originally packet B ₁, B ₂, L, B _mbe converted to E ₁, E ₂, L, E _n;

(5c3) the data E after transfer encoding _ktime, at intra-node, define intrinsic prime number p _ki, wherein the number of prime number is identical with the number of its neighbor node, to any given integer set { E _k1, E _k2..., E _kN, there is unique integer

make E _k=E _ki(modp _ki), at a data E of transmission _ktime, only need on each paths of its neighbor node, transmit E _ki.

4. the wireless sensor network data method for reliable transmission based on Chinese remainder theorem according to claim 1, in step (5e), sink node is received the data after cutting apart, first the data after cutting apart are merged, by network code, decoding data is recovered to initial data again, carries out in accordance with the following steps:

(5e1) when sink node is received the data E after cutting apart _kiafter, utilize Chinese remainder theorem by the data E of cutting apart _kimerge, obtain the data E before cutting apart _k, E _kcomputational methods be:

coefficient wherein

q _kimake c _ki=1 (modp _ki);

(5e2) the data E before being cut apart _kafter, m the data that tentative segment point is received are respectively E ₁, E ₂... E _m, further judge corresponding m coefficient vector α ₁, α ₂... α _klinear independence whether, if so, data are decoded by coding:

By above-mentioned decoding, just data are reduced into initial data B ₁, B ₂..., B _m.

Images