CN107786660A - A kind of radio sensing network code distribution method based on umbrella multipath - Google Patents

Abstract

The present invention provides a kind of radio sensing network code distribution method based on umbrella multipath, and it has advantages below：1. building umbrella multirouting path, spatial multiplex ratio is improved, is distributed beneficial to the quick mirror image of follow-up phase, improves distribution efficiency；2. " receive single-turn hair " pseudo- route transmission mode equipped with retransmission mechanism expands mirror image range of receiving more, and ensure that the reliability in path；3. being explored based on neighbours, the local transmission relation that competition subscription thought formation is not influenceed by concealed terminal, the generation of message transition collision is avoided, effectively shortens the code distribution time.

Description

A Code Distribution Method for Wireless Sensor Networks Based on Umbrella Multipath

technical field

The present invention relates to the technical field of wireless sensor networks, and more specifically, to a code distribution method for wireless sensor networks based on umbrella multipath.

Background technique

In a wireless sensor network, the deployment of nodes has the characteristics of a large number of nodes and widely dispersed spatial locations. Some application scenarios are relatively harsh. When the nodes are deployed and the user proposes new network function requirements, the software code image needs to be updated. The traditional The method of manual wired programming has the disadvantages of high labor costs and difficulty in meeting the needs. In this context, using wireless channels as a medium to distribute and upgrade code mirroring through single-hop and multi-hop methods, that is, a code distribution protocol that uses mirroring to update, can effectively solve the above problems.

Traditional code distribution protocols are divided into pure broadcast routing protocols and protocols with routing plus broadcast framework; pure broadcast routing protocols mainly perform mirror updates on all nodes in the WSN in a multi-hop manner, and the source node can complete the mirroring of surrounding nodes through one broadcast Distribution does not require the source node to interact with the surrounding nodes in order to save time, but before the node forwards, it must perform listening management, which will cause a large amount of time delay; the routing plus broadcast framework protocol, on the one hand, uses the routing The characteristics of fast transmission greatly shorten the time for code distribution. On the other hand, broadcasting can cover all corners of the entire network, but because the routing method uses a single path, the space coverage speed is relatively slow, and the transmission method uses "single receiving single Forwarding" mode, high energy consumption, relatively low efficiency.

Regardless of the pure broadcast routing protocol or the routing plus broadcast framework protocol, there is a situation where control messages and mirrored data messages are aliased in time and space, resulting in message collisions and affecting the reliability of mirrored transmission. The traditional way to suppress this kind of problem is to add a certain delay before sending messages to avoid message collisions. However, for the entire code distribution, the cumulative delay is large, which affects the efficiency of image transmission.

Contents of the invention

In order to solve the technical defect of low image transmission efficiency in the prior art, the present invention provides a wireless sensor network code distribution method based on umbrella multipath.

For realizing above-mentioned purpose of the invention, the technical scheme that adopts is:

A wireless sensor network code distribution method based on umbrella multipath, comprising the following steps:

S1. Determining the node at the opposite corner of the source node in the edge of the network as the first-level seed node, and determining the two nodes at the two adjacent corners of the source node in the edge of the network as the second-level seed node and the third-level seed node respectively;

S2. The source node broadcasts the RREQ message across the network. After receiving the RREQ message, the first-level seed node replies back to the source node with the RREP message. After the source node receives the RREP message, the main path is formed;

S3. Let the node at the midpoint of the main path between the source node and the first-level seed node be a super node. After the main path is constructed, the source node broadcasts a notifyTarget message to the main path. After receiving the notifyTarget message, the super node confirms its identity as The super node broadcasts the RREQ message and initiates a path search for the secondary seed node. After receiving the RREQ message, the secondary node delays the reverse path to reply to the RREP message. After the super node receives the RREP message, the distance between the super node and the secondary seed node The path is successfully established; follow the above method to establish the path between the super node and the third-level seed node; after the path between the super node and the second-level seed node and the third-level seed node is established, the super node sends a notifyTarget reversely along the main path message, notifying the source node that the establishment of umbrella multipath is completed;

S4. After receiving the notifyTarget message, the source node sends all the image pages of the code image file to the super node through the main path, and the super node then transmits the received image pages to the first-level seed node and the second-level seed node through the established path and third-level seed nodes; the path nodes between the source node and the super node, the path nodes between the super node and the first-level seed node, the second-level seed node and the third-level seed node receive the mirror image during the transmission of the mirror page Store the page first, and then transmit the mirrored page to the next hop node; after receiving and storing all the mirrored pages, the node performs the code distribution update of steps S5-S9 on the nodes to be updated around it;

S5. Let the nodes that store all mirror pages and have nodes to be updated around them be GRAY nodes, and let the nodes to be updated be WHITE nodes, then the GRAY nodes will enter the MAINTAIN state after receiving all the mirror pages, and start a random timer TM, broadcast the adv message and transition to the HALFLINKING state as soon as the timing is up; when the WHITE node receives the adv message in the SILENT state, it transitions to the HALFLINKING state, and records the ID number of the GRAY node that made it enter the HALFLINKING state;

S6. After the WHITE node enters the HALFLINKING state, start the semi-connection timer T _Half2 , and reply the advReply message to the GREY node at a random time before the timer T _Half2 expires. The duration of the timer T _Half2 is λ*neighborsCount; the GREY node enters After the HALFLINKING state, the semi-connection timer T _Half1 will be turned on. The duration of T _Half1 is λ*N _{WhiteNeighbours} , where λ is a constant. The GREY node counts the ID number of the WHITE node that replies to the advReply message before the timer T _Half1 expires, and will They are recorded in the local receive collection;

S7. After the timer T _Half1 expires, the GREY node in the HALFLINKING state judges whether the local neighbor set and the local receiving set are the same, and if they are the same, immediately broadcast the sub message; if not, the GREY node returns to the MAINTAIN state and executes again S5 step;

S8. After receiving the sub message, the WHITE node will send a subReply message to the GRAY node after a short random time, and enter the RX stable state at the same time; the GREY node that receives the subReply message will determine whether the message is sent to itself , if so, transition to the TX stable state; if not, transition to the SILENT stable state;

S9. After the WHITE node and the GRAY node enter the TX stable state, the code is distributed.

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

① Construct an umbrella-shaped multi-routing path, which improves the space multiplexing rate, facilitates the rapid image distribution in the subsequent stage, and improves the distribution efficiency; ② The "multi-reception and single forwarding" pseudo-routing transmission method equipped with a retransmission mechanism expands the image receiving range, And ensure the reliability of the path; ③Based on the idea of neighbor exploration and subscription, a local transmission relationship that is not affected by hidden terminals is formed, which avoids the occurrence of message collisions and effectively shortens the code distribution time.

Description of drawings

Figure 1 Umbrella protocol code distribution framework

Figure 2 Example of Umbrella protocol selecting seed nodes under different network structures

Figure 3 The message interaction diagram of Umbrella protocol establishing umbrella multipath (routing stage)

Figure 4 Umbrella protocol umbrella multipath effect diagram

Figure 5 Comparison of traditional pure routing transmission mode and Umbrella protocol pseudo-routing transmission mode

Figure 6 The message interaction process of the continuous multi-page transmission mechanism of the Umbrella protocol

Figure 7 Example of retransmission timing of node packet loss in the Umbrella protocol routing phase

Figure 8 Umbrella protocol's special handling of the last mirror page

Figure 9 Flowchart of timing control of Umbrella protocol source node to the whole network (broadcast stage)

Figure 10 The state table corresponding to different node types

Figure 11 The state machine of the GRAY node during the competitive subscription period

Figure 12 The state machine of the WHITE node during the competition subscription period

specific

Detailed ways

The accompanying drawings are for illustrative purposes only and cannot be construed as limiting the patent;

The present invention will be further elaborated below in conjunction with the accompanying drawings and embodiments.

Example 1

The technical solution of the present invention is to adopt a wireless sensor network code distribution method based on umbrella-shaped multi-path non-interference multi-local transmission, including two phase strategies, which are umbrella multi-path establishment phase strategy and non-interference multi-local transmission Mechanism strategy, as shown in Figure 1, will be explained separately below.

1. Umbrella multipath establishment phase strategy

1. Umbrella multi-path establishment phase strategy, including: seed node selection, information interaction establishment path.

At this stage, the selection of seed nodes is carried out first, based on the principle: select a network edge node located at the opposite corner of the source node in the network as the first-level seed node, and the second-level and third-level seed nodes should be located at two adjacent corners of the source node. Figure 2(a)(b)(c) are examples of seed node selection for circular network, lattice network and irregular network respectively.

After the node is selected, the path is established through the interaction of the RREQ message, RREP message, and notifyTarget three types of messages, as shown in Figure 3. The specific interaction process is described as follows:

1) The source node broadcasts the RREQ message throughout the network, and initiates a path search for the first-level seed node.

2) After receiving the RREQ message, the first-level seed node will reply the RREP message to the source node along the reverse path.

3) Once the RREP message sent by the first-level seed node is forwarded to the source node, the main path is established.

Subsequently, the source node sends a notifyTarget message to the super node at the midpoint of the main path, in order to notify the super node to initiate a path search to the secondary seed node.

4) The super node broadcasts the RREQ message and initiates the path search for the secondary seed node.

5) After receiving the RREQ message trying to find itself, the secondary seed node immediately replies to the RREP message along the reverse path leading to the super node.

6) When the super node receives the RREP message replied by the secondary seed node, the second bidirectional path between the source node and the secondary seed node is established successfully. Then the super node initiates the third RREQ path search to the third-level seed node.

7) After receiving the third RREQ message, the third-level seed node will reply the RREP message to the super node along the reverse path to form a forward path from the super node to the third-level seed node.

8) At the moment when the super node receives the RREP message sent by the third-level seed node, the three bidirectional paths of the network are all established. Finally, the super node sends a notifyTarget message reversely along the main path to notify the source node that the umbrella multipath is established. As shown in Figure 4.

After receiving the notifyTarget message, the source node sends all the image pages of the code image file to the super node through the main path, and the super node then transmits the received image pages to the first-level seed node, the second-level seed node and the third-level seed node through the established path. level seed node; the path node between the source node and the super node, the path node between the super node and the first-level seed node, the second-level seed node and the third-level seed node are in the transmission process of the mirror page, after receiving the mirror page Store first, and then transmit the mirrored page to the next hop node; after the node receives and stores all the mirrored pages, it distributes and updates the code to the surrounding nodes to be updated. The second part of the specific distribution process is more non-interfering The local transmission mechanism strategy will be described in detail, as shown in Figure 5.

2. The information interaction of continuous multi-page transmission is shown in Figure 6, and its pseudocode can be described as follows:

while(currentPage<PAGE_SUM){

if(currentPage still has mirror blocks not finished sending){

Turn on the timer, the timing duration is T_Block, when the timing is up, the next mirror will be sent

block;

}else{

currentPage++;

Turn on the timer, the timing duration is T_Page, and the next mirror page will be sent when the timing is up

//The next mirror page is the mirror page of currentPage;

}

}

Among them, when the super node constructs the second and third transmission paths, it will respectively record the next-hop node ID number of the forward path to the second and third seed nodes, and the super node itself already knows that it is in the The ID number of the next hop node on the forward main path, so the super node will have three next hop nodes on the forward path. After the common path node receives the image data, it will first query the ID number of the next-hop node in the forward routing table, and then forward the image data to the next-hop node. Before forwarding the data message, the super node will set three next-hop nodes in the forward path in the message. After receiving the data message, the three next-hop nodes will continue to forward the message along their respective forward paths until to the seed nodes of the respective paths.

As shown in Figure 7, Tc in the figure is the duration of the retransmission timer. When the path node and the first-hop node of the path receive discontinuous mirror pages, the retransmission timer is started. The duration of Tc can ensure that the retransmission is always on Initiated between page-to-page intervals, thereby avoiding conflicts with mirrored page messages. Since nodes receive mirrored pages in an orderly manner, a mirrored page can only be used after all mirrored blocks in a mirrored page have been received. Path nodes and path one-hop nodes can only receive the data of the next adjacent mirror page that is 1 greater than the current mirror page number of the node, and cannot receive across page numbers. When a path node or path one-hop node receives a cross-page mirror block, it will start a retransmission timer. After the timer expires, the path node or path one-hop node will follow the reverse path to the source node Send a retransmission request message, that is, a specialReq message. After the source node receives the retransmission request message, it will sequentially send each mirror block of the requested mirror page at the next sending time.

Since packet loss detection is based on continuous mirror pages, it is impossible to ensure that the last page is received correctly. Therefore, it is necessary to introduce special processing for the last mirror page to ensure correct transmission. This method is verified by introducing the srcDataCompleted message. As shown in Figure 8, the protocol makes the source node forward the srcDataCompleted message once along the umbrella multipath after completing continuous multi-page transmission. The source node starts a timer immediately after sending the srcDataCompleted message. The timing duration is the size of the mirror page transmission time interval Tpage. If no retransmission request message is received before the timer expires, the path node and path one-hop node are considered The last page is also received correctly. At this point, the task of the routing transmission phase is completed, and the source node can notify the entire network to enter the broadcast transmission phase. If a retransmission application message requesting the last page is received before the timer expires, the source node resends the last mirrored page, and then repeats the above process.

2. Non-interference multi-local transmission mechanism strategy

In order to implement the strategy of non-interference multi-local transmission mechanism, we mainly adopt the mechanism of exploring neighbors and competing for subscription, and the specific process is shown in Figure 9. Here, the three states of the nodes are called GREY, WHITE, and BLACK, respectively, where the WHITE node represents the node to be updated, and the BLACK node and the GRAY node are the nodes with the new image. The difference is that BLACK nodes do not have WHITE nodes, and there are WHITE nodes around GRAY nodes, which are potential sending nodes of GRAY nodes.

In the neighbor discovery phase, the GREY node detects whether there are WHITE nodes around. If there is no WHITE node, it becomes a BLACK node. If there is a WHITE node, it records the total number of WHITE nodes, and then enters the competitive subscription phase. During the competitive subscription period, the BLACK node does not operate, and the GRAY node communicates with the WHITE node. During this stage, the WHITE node may receive subscription signals from multiple GRAY nodes. Here, a node optimization selection strategy is used to select a GRAY node for subscription response. After the GRAY node sends the subscription request, it will continue to collect the subscription response signal and count the total number. If the total number is equal to the total number of WHITEs explored by the neighbors, that is, all the surrounding WHITE nodes subscribe to themselves, that is, in the non-collision transmission situation, the subscription is successful; if the total number is not the same , that is, the collision-free transmission relationship is not satisfied, the subscription fails, and the neighbor discovery phase needs to be re-entered.

In order to ensure that messages in the neighbor discovery and competitive subscription phases will not cause subscription message packet loss due to message collisions, this method uses random delay message sending technology, where the order of magnitude of random delay time is very small. Sending information with a random time delay can greatly reduce the phenomenon of collision and packet loss caused by nodes receiving messages at the same time.

At the same time, each node will start a fixed-time timer in the competition phase. The use of the timer can ensure that the node can receive the message in this phase, and at the same time, it can supplement and solve the small probability of message collision and packet loss that still occurs in the random delay message sending technology. Problem, if the timer expires, if the subscription fails, it will no longer wait for the subscription response, and re-enter the neighbor discovery and subscription subscription phases.

In the competitive subscription stage, the three types of nodes will go through various intermediate states from their respective default states and finally transform into a stable state. The status is shown in Figure 10.

1. The behavior status of GREY nodes during the competitive subscription period can be described as the following points, as shown in Figure 11:

When the G.1GREY node enters the MAINTAIN state, it starts a random timer TM, broadcasts an adv message and transitions to the HALFLINKING state as soon as the timer expires.

G.2 If the GRAY node in the MAINTAIN state monitors that there are adv messages sent around, it indicates that there are Gray neighbors trying to initiate a relationship establishment request. Then postpone the sending of its own adv message by μ*N _{WhiteNeighbours} +τ _rand , where μ is a constant and τ _rand is a short random delay time. The purpose is to provide a quieter environment for GRAY neighbor nodes trying to initiate relationship establishment.

After the G.3GREY node enters the HALFLINKING state, it will start the half-connection timer T _Half1 , and the duration of T _Half1 is λ*N _{WhiteNeighbours} , where λ is a constant. Before the T _Half1 timer expires, it will count the WHTIE node ID numbers that reply the advReply message of the current GRAY node, and record them in the local receiving set.

G.4 The GREY node in the HALFLINKING state will judge whether the local neighbor set and the local receiving set are the same after T _Half1 expires. ① If they are the same, it means that all its WHITE neighbors have restored to it. Then immediately broadcast a subscription message (sub message), the purpose is to immediately initiate its subscription request to all WHITE nodes. ② If they are not the same, it means that some nodes among the WHITE neighbor nodes of the current node have preferentially participated in the transmission relationship of other GRAY nodes, and the current node itself has given up being the sending node of the transmission relationship. The GRAY node returns to the MAINTAIN state, and re-executes step G.1.

G.5 Regardless of whether it is in the MAINTAIN state or the HALFLINKING state, as long as the gray node listens to the sub message, it indicates that the gray neighbors around it have established a transmission relationship, and they will no longer have the opportunity to participate in this transmission relationship. Immediately close all timers and enter the SILENT steady state.

G.6 The GREY node that receives the subReply message will judge whether the message is sent to itself. ①If yes, then transition to TX stable state; ②If not, then transition to SILENT stable state.

2. The behavior status of the WHITE node during the competitive subscription period can be described as the following points, as shown in Figure 12:

W.1 When the WHITE node receives the adv message in the SILENT state, it transitions to the HALFLINKING state, and records the ID number of the GREY node that makes the current WHITE node enter the HALFLINKING state, that is, the ID number of the sender of the adv message. When entering the HALFLINKING state, start the half-connection timer T _Half2 at the same time, and the duration of T _Half2 is λ*neighborsCount. Because the neighboursCount field comes from the received adv message (the neighboursCount field is the number of WHITE neighbors of the GREY node), so the duration of T _Half1 and T _Half2 are actually equal.

W.2 After the WHITE node enters the HALFLINKING state, it will reply the advReply message at a random time before the timer T _Half2 expires. The WHITE node does not reply to the adv message received after entering the HALFLINKING state.

W.3 After the timer T _Half2 expires, the WHITE node will clear the previously recorded semi-connected GRAY node ID number and return to the SILENT state, so that the WHITE node has the opportunity to participate in the establishment of other transmission relationships.

W.4 After receiving the sub message, the WHITE node will send a subReply message after a short random time and enter the RX stable state at the same time.

3. The optimal selection strategy of the sending node is to let the GRAY node start a random timer TM in the MAINTAIN state, following the following formula:

Among them, T _low and T _high are the lower limit and upper limit of the expected value range of TM, N _{AverageNeighbours} is the average number of neighbors of network nodes, N _{WhiteNeighbours} is the number of neighbor nodes to be updated, τ _rand is a short delay. The purpose is to make more gray nodes with WHITE neighbors more likely to become the sending node of the transmission relationship.

Apparently, the above-mentioned embodiments of the present invention are only examples for clearly illustrating the present invention, rather than limiting the implementation of the present invention. For those of ordinary skill in the art, other changes or changes in different forms can be made on the basis of the above description. It is not necessary and impossible to exhaustively list all the implementation manners here. All modifications, equivalent replacements and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the claims of the present invention.

Claims (7)

1. a wireless sensor network code distribution method based on umbrella multi-path, it is characterized in that: comprise the following steps: S1. Determining the node at the opposite corner of the source node in the edge of the network as the first-level seed node, and determining the two nodes at the two adjacent corners of the source node in the edge of the network as the second-level seed node and the third-level seed node respectively; S2. The source node broadcasts the RREQ message across the network. After receiving the RREQ message, the first-level seed node replies back to the source node with the RREP message. After the source node receives the RREP message, the main path is formed; S3. Let the node at the midpoint of the main path between the source node and the first-level seed node be a super node. After the main path is constructed, the source node broadcasts a notifyTarget message to the main path. After receiving the notifyTarget message, the super node confirms its identity as The super node broadcasts the RREQ message and initiates a path search for the secondary seed node. After receiving the RREQ message, the secondary node delays the reverse path to reply to the RREP message. After the super node receives the RREP message, the distance between the super node and the secondary seed node The path is successfully established; follow the above method to establish the path between the super node and the third-level seed node; after the path between the super node and the second-level seed node and the third-level seed node is established, the super node sends a notifyTarget reversely along the main path message, notifying the source node that the establishment of umbrella multipath is completed; S4. After receiving the notifyTarget message, the source node sends all the image pages of the code image file to the super node through the main path, and the super node then transmits the received image pages to the first-level seed node and the second-level seed node through the established path and third-level seed nodes; the path nodes between the source node and the super node, the path nodes between the super node and the first-level seed node, the second-level seed node and the third-level seed node receive the mirror image during the transmission of the mirror page Store the page first, and then transmit the mirrored page to the next hop node; after receiving and storing all the mirrored pages, the node performs the code distribution update of steps S5-S9 on the nodes to be updated around it; S5. Let the nodes that store all mirror pages and have nodes to be updated around them be GRAY nodes, and let the nodes to be updated be WHITE nodes, then the GRAY nodes will enter the MAINTAIN state after receiving all the mirror pages, and start a random timer TM, broadcast the adv message and transition to the HALFLINKING state as soon as the timing is up; when the WHITE node receives the adv message in the SILENT state, it transitions to the HALFLINKING state, and records the ID number of the GRAY node that made it enter the HALFLINKING state; S6. After the WHITE node enters the HALFLINKING state, start the semi-connection timer T _Half2 , and reply the advReply message to the GREY node at a random time before the timer T _Half2 expires. The duration of the timer T _Half2 is λ*neighborsCount; the GREY node enters After the HALFLINKING state, the semi-connection timer T _Half1 will be turned on. The duration of T _Half1 is λ*N _{WhiteNeighbours} , where λ is a constant. The GREY node counts the ID number of the WHITE node that replies to the advReply message before the timer T _Half1 expires, and will They are recorded in the local receive collection; S7. After the timer T _Half1 expires, the GREY node in the HALFLINKING state judges whether the local neighbor set and the local receiving set are the same, and if they are the same, immediately broadcast the sub message; if not, the GREY node returns to the MAINTAIN state and executes again S5 step; S8. After receiving the sub message, the WHITE node will send a subReply message to the GRAY node after a short random time, and enter the RX stable state at the same time; the GREY node that receives the subReply message will determine whether the message is sent to itself , if so, transition to the TX stable state; if not, transition to the SILENT stable state; S9. After the WHITE node and the GRAY node enter the TX stable state, the code is distributed. 2. the wireless sensor network code distribution method based on umbrella multi-path according to claim 1, is characterized in that: in described step S5, if be in the GREY node of MAINTAIN state and monitor and have adv message to send around, postpone so The delay time for sending the own adv message is μ*N _{WhiteNeighbours} +τ _rand , where μ is a constant, and τ _rand is a short random delay time. 3. The wireless sensor network code distribution method based on umbrella multipath according to claim 1, characterized in that: in the step S5, if the GRAY node in the MAINTAIN state or the HALFLINKING state listens to the sub message, then the GRAY The node enters the SILENT stable state. 4. The wireless sensor network code distribution method based on umbrella multipath according to claim 1, characterized in that: in the step S6, after the timer T _Half2 expires, the WHITE node will clear the previously recorded half The ID number of the connected GRAY node, and return to the SILENT stable state. 5. the wireless sensor network code distribution method based on umbrella multi-path according to claim 1, is characterized in that: the regular duration of described random timer TM is: <mrow><msub><mi>T</mi><mi>M</mi></msub><mo>=</mo><msub><mi>T</mi><mrow><mi>h</mi><mi>i</mi><mi>g</mi><mi>h</mi></mrow></msub><mo>-</mo><mrow><mo>(</mo><msub><mi>T</mi><mrow><mi>h</mi><mi>i</mi><mi>g</mi><mi>h</mi></mrow></msub><mo>-</mo><msub><mi>T</mi><mrow><mi>l</mi><mi>o</mi><mi>w</mi></mrow></msub><mo>)</mo></mrow><mo>*</mo><msub><mi>log</mi><mn>10</mn></msub><mrow><mo>(</mo><mfrac><mrow><mn>10</mn><mo>*</mo><msub><mi>N</mi><mrow><mi>W</mi><mi>h</mi><mi>i</mi><mi>t</mi><mi>e</mi><mi>N</mi><mi>e</mi><mi>i</mi><mi>g</mi><mi>h</mi><mi>b</mi><mi>o</mi><mi>u</mi><mi>r</mi><mi>s</mi></mrow></msub></mrow><msub><mi>N</mi><mrow><mi>A</mi><mi>v</mi><mi>e</mi><mi>r</mi><mi>a</mi><mi>g</mi><mi>e</mi><mi>N</mi><mi>e</mi><mi>i</mi><mi>g</mi><mi>h</mi><mi>b</mi><mi>o</mi><mi>u</mi><mi>r</mi><mi>s</mi></mrow></msub></mfrac><mo>)</mo></mrow><mo>+</mo><msub><mi>&amp;tau;</mi><mrow><mi>r</mi><mi>a</mi><mi>n</mi><mi>d</mi></mrow></msub></mrow> Among them, T _low and T _high are the lower limit and upper limit of the expected value range of TM, N _{AverageNeighbours} is the average number of neighbors of network nodes, N _{WhiteNeighbours} is the number of WHITE nodes of GRAY nodes, and τ _rand is a short delay. 6. The wireless sensor network code distribution method based on umbrella multipath according to any one of claims 1 to 5, characterized in that: in the step S4, when the first-level seed node, the second-level seed node, the third-level seed node When the seed node, super node or node on the path receives discontinuous mirror pages, it starts the retransmission timer, and sends a retransmission request message to the source node after the retransmission timer expires, and the source node receives the retransmission After the application message is transmitted, each mirror block of the requested mirror page is sent sequentially at the next sending time. 7. The wireless sensor network code distribution method based on umbrella multi-path according to claim 6, characterized in that: in the step S4, after the source node has sent all the mirrored pages, it will then follow the path to The first-level seed node, second-level seed node, third-level seed node, and super node send the srcDataCompleted message, and the source node starts a timer immediately after sending the srcDataCompleted message. If no retransmission application message is received before it is full, it is considered that the last mirrored page has been received correctly.