Improved node positioning method based on DV-Hop

Abstract

The invention discloses an improved node positioning method based on DV-Hop, and belongs to the technical field of remote sensing. According to the method, on the basis of a traditional DV-Hop algorithm, the minimum hop value and the average hop distance of the DV-Hop algorithm are improved respectively. The method comprises the following steps: firstly, classifying hop count values according to the distance between a transmitting node and a receiving node, and refining the hop count values into three parts; a minimum hop value R in a traditional DV-Hop algorithm is canceled, but an anchor node broadcasts own coordinates for three times according to communication radiuses of 1 / 3R, 2 / 3R and R, so that the whole network is better covered. And secondly, the average distance of anchor nodes between hops is corrected by applying a least square criterion and a weighting coefficient, the weighting coefficient is as follows, the positions of the nodes can be obtained by obtaining an improved distance calculation formula, and the node positioning error can be reduced to a certain extent. According to the method, the positioning performance of a node positioning algorithm is improved, the positioning precision between the nodes is improved, and the problems of a hop value error, an average hop distance error, a node position calculation error and the like existing in a traditional DV-Hop algorithm are solved.

Description

Technical Field

[0001] The present invention belongs to the field of remote sensing technology, and relates to an improved algorithm based on distance vector hop localization algorithm (DV-Hop). Specifically, it is a distributed hop-by-hop positioning technology, which first adopts the traditional distance vector exchange mechanism, which allows all nodes in the network to calculate the distance to the anchor node by the number of hops (calculated by the number of hops). Then, it estimates the average size of each hop and corrects the entire network. Finally, the position of the unknown node is calculated by the maximum likelihood positioning algorithm. On this basis, this method innovatively introduces the idea of ​​multiple communication radius, corrects the calculation error of the minimum hop value, and corrects the error of the average hop distance calculation by the least squares criterion and weighting coefficient, thereby improving the positioning performance of the positioning algorithm and improving the positioning accuracy between nodes. Background Art

[0002] At present, global forest resources are suffering from severe damage and reduction. The reduction of forest coverage not only threatens the stability of ecosystems, but also weakens the ecosystem services provided by forests, such as soil conservation, hydrological cycle and biodiversity maintenance. Forest fires are one of the main factors causing the destruction of global forest resources. The number of forest fires that occur worldwide each year is as high as hundreds of thousands, most of which are caused by natural factors, although human influence also plays a certain role. Forest fires can cause a large number of forests to be burned, destroy wildlife habitats, cause serious environmental pollution, and may also trigger natural disasters such as mudslides, threatening people's lives and property safety. In addition, forest fires can have a serious impact on the climate, leading to an increase in carbon dioxide emissions, further exacerbating the problem of climate change. Although the overall area of ​​my country's forest resources has increased year by year, due to excessive logging and forest fires, my country's forest coverage is still lower than that of many developed countries, and forest resources are in short supply. The number of forest fires that occur in my country each year is also rising, which has become a serious social problem. With the increasing scarcity of forest resources and the increasingly serious damage to the ecological environment caused by forest fires, how to effectively monitor forest fires has become an urgent problem to be solved.

[0003] In recent years, with the continuous development and maturity of wireless sensor technology, it has been widely used in forest fire monitoring and has become an efficient and practical monitoring method. Compared with traditional monitoring methods, the use of wireless sensor network technology has many advantages. First, its large coverage area, flexible networking, and low price allow it to be flexibly deployed at any location in the forest to form a seamless coverage network and achieve long-term real-time monitoring. Secondly, wireless sensor network technology can collect a variety of environmental parameters of forest fires, such as temperature, humidity, smoke concentration, etc., and transmit them to the monitoring department through wireless networks, helping monitoring personnel to more accurately understand changes in the forest environment, discover signs of fire in advance, and make prevention and rescue preparations. In addition, it can achieve all-weather uninterrupted monitoring and is not easily affected by environmental factors such as weather. Finally, through data analysis, monitoring personnel can identify abnormal data, improve monitoring efficiency, and reduce work burden. Therefore, using wireless sensor network technology for forest fire monitoring is an efficient and practical way, with advantages including low price, flexible networking, accuracy and timeliness, etc. It can warn of fires, improve rescue efficiency, and reduce the losses caused by fires to people.

[0004] At present, the node positioning algorithms in sensor networks are mainly divided into two types: distance-based and distance-free. Among them, the distance-based positioning algorithms mainly include received signal strength indication, arrival time, arrival time difference and arrival angle. The distance-free positioning algorithms include centroid, weighted centroid, DV-Hop, amorphous and approximate triangle interior point test positioning algorithms. Compared with the distance-based positioning algorithms, the positioning accuracy of the distance-free positioning algorithms can meet most of the requirements, so they are very popular among users. Among them, the DV-Hop positioning algorithm is one of the most famous distance-free positioning algorithms in WSN because of its strong robustness and simplicity. However, in complex situations, the positioning accuracy of the DV-Hop algorithm is limited, and there are problems such as jump value error, average jump distance error and node position calculation error. Therefore, it is necessary to improve the problems existing in DV-Hop to improve the positioning accuracy. Summary of the invention

[0005] In view of the problems of hop value error, average hop distance error and node position calculation error in the traditional DV-Hop algorithm, the present invention aims to provide an improved node positioning method based on distance vector hop number positioning (DV-Hop) to improve the positioning accuracy of wireless sensors. This method improves the minimum hop value and average hop distance in the traditional DV-Hop algorithm. The main idea is to refine the unknown node radius through multiple communication radii to minimize the error of the corrected minimum hop value; then the average hop value is corrected according to the least squares criterion.

[0006] The technical solution of the present invention is: an improved node positioning method based on DV-Hop, the method comprising:

[0007] Step 1: Network initialization;

[0008] Step 2: Multiple communication radius correction adjustment;

[0009] In the network, the anchor node broadcasts its coordinates three times according to the communication radius of 1 / 3R, 2 / 3R and R. The size of R is determined according to the actual situation of the communication network. In the first broadcast, the anchor node broadcasts sensor information to the surrounding nodes. Its communication radius is 1 / 3R, covering the nodes within the range of 1 / 3R from the anchor node. After receiving the broadcast, the neighboring nodes record the number of hops from them to the anchor node, which is 1 / 3R. This hop number is 1 / 3. In the second broadcast, the anchor node transmits sensor information to nearby nodes with a communication radius of 2 / 3R. This range is 1 / 3. The surrounding nodes that have been recorded in the first broadcast are not included in the neighboring nodes; when the neighboring nodes receive the broadcast information, they will record the number of hops from them to the anchor node, which is 2 / 3R. This number of hops is 2 / 3; in the third broadcast, the anchor node will send sensor information to the neighboring nodes, and its communication radius is R, and the anchor node will exclude the neighboring nodes that have been recorded in the second broadcast to ensure that the transmission of information will not be repeated; when the neighboring nodes receive the broadcast information, they will record the number of hops from them to the anchor node, which is 1; at this time, each node obtains the corrected minimum number of hops;

[0010] Step 3: The beacon node calculates the average hop distance and broadcasts it;

[0011] Step 4: Correct the average jump distance;

[0012] Step 5: Estimated distance of unknown nodes;

[0013] Step 6: Determine whether the unknown node has four or more beacon node information. If yes, proceed to step 7. If not, the unknown node cannot be located and proceed to step 8.

[0014] Step 7: Calculate the position according to the maximum likelihood positioning method;

[0015] Step 8: Calculation ends.

[0016] Furthermore, the specific method of step 2 is:

[0017] Step 2.1: Initialize i=1;

[0018] Step 2.2: The anchor node broadcasts its own information with a communication radius of i / 3R;

[0019] Step 2.3: The receiving node queries whether the on-duty node has received the broadcast information of the anchor node. If not, it receives and records the number of information hops; if yes, it proceeds to step 2.4;

[0020] Step 2.4: Update i=i+1;

[0021] Step 2.5: Determine whether i is less than 3. If so, return to step 2.2. If not, proceed to step 2.6.

[0022] Step 2.6: The anchor node broadcasts its own information with R as the communication radius;

[0023] Step 2.7: The receiving node compares and saves the minimum hop count, then increases the hop count by 1 and forwards the information.

[0024] Furthermore, the specific method of step 4 is:

[0025] Step 4.1: Assume that anchor node i receives the location information of other anchor nodes, and anchor node i calculates the average hop number HopSize i The following criteria should be met:d ij =HopSize i *hops ij +ε ij , hops ij represents the number of hops between anchor node i and anchor node j, d ij is the actual length between anchor nodes i and j, ε ij It indicates the error caused by replacing the actual length with the product of the average distance per hop and the number of hops;

[0026] Step 4.2: Calculate the average hop distance:

[0027] Step 4.3: Add weighting coefficients hop ij represents the minimum hop value between anchor node i and unknown node j, n represents the number of anchor nodes, and the improved distance calculation formula is obtained:

[0028] Furthermore, the average jump distance formula of step 4.2 is obtained in the following way;

[0029] As the total error, the formula is as follows:

[0030]

[0031] use The least squares corrected average jump distance formula is as follows:

[0032]

[0033] Beneficial effects of the present invention: The present invention proposes an improved node positioning algorithm based on DV-Hop. This is a distributed hop-by-hop positioning method, which improves the minimum hop value and the average hop distance on the basis of the traditional DV-Hop algorithm, estimates through the maximum likelihood positioning algorithm, and finally obtains a more accurate node location method. The present invention avoids the error problem caused by the linear distance being smaller than the communication radius by improving the minimum hop value, and corrects the average hop distance. The average hop distance between nodes is no longer a fixed value, but is dynamically corrected to a suitable average hop distance value according to the changes in the surrounding anchor nodes of the unknown node, which can reduce the node positioning error caused by the average hop distance calculation error to a certain extent. The positioning performance of the node positioning algorithm is improved, and the positioning accuracy between nodes is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic diagram of the overall method flow of the present invention.

[0035] Figure 2 This is a schematic diagram of multiple communication radius hops.

[0036] Figure 3 The figure is a comparison diagram of node positioning errors between the method of the present invention and the traditional DV-Hop algorithm under Matlab simulation.

[0037] Figure 4 This is a comparison chart of the node positioning results of the two algorithms under Matlab simulation. DETAILED DESCRIPTION

[0038] The present invention will be further described below in conjunction with the accompanying drawings.

[0039] The algorithm improvement process of this paper is as follows Figure 1 As shown in the figure, the first and second steps of the traditional DV-hop algorithm are mainly improved. First, the idea of ​​multiple communication radius is introduced to correct the calculation error of the minimum number of hops, and then the error of the average hop distance calculation is corrected by the least squares criterion and weighted coefficient. The following will explain the improvement of the first and second steps of the traditional DV-hop algorithm respectively.

[0040] (1) Optimization of minimum hop value

[0041] In traditional DV-Hop, as long as the linear distance between two nodes is smaller than the communication radius, it will be considered as one hop. This practice will significantly increase the positioning error of the algorithm because the actual distance may be very different. In order to solve this problem, the algorithm needs to be optimized. In this chapter, we will use multiple communication radius concepts to optimize the hop value. First, the communication radius can be refined, and then the number of hops can be refined by decimals to minimize the error. This can avoid the error problem caused by the linear distance being smaller than the communication radius.

[0042] like Figure 2 As shown in the figure, nodes B, C, and D are all within the communication radius of node A. If the traditional DV-Hop protocol is used, the number of hops from node A to them will be recorded as one hop. The optimization ideas of this paper are as follows: In order to more accurately express the actual distance between nodes, this paper will classify the hop value according to the distance between the transmitting node and the receiving node, and refine it into three parts. In this way, the real distance between nodes can be better reflected and the error caused by the inaccurate hop value can be reduced. In the network, the anchor node broadcasts its coordinates three times according to the communication radius of 1 / 3R, 2 / 3R, and R to better cover the entire network. This design can increase the applicability of the algorithm, so that the DV-Hop algorithm can also obtain higher positioning accuracy when the distance difference is large. In the first broadcast, the anchor node broadcasts sensor information to the surrounding nodes. Its communication radius is 1 / 3R, which can cover nodes within 1 / 3R of the anchor node. After receiving the broadcast, the neighboring nodes will record the number of hops from them to the anchor node, which is 1 / 3. In the second broadcast, the anchor node will transmit sensor information to nearby nodes with a communication radius of 2 / 3R. This range does not include the surrounding nodes that have been recorded in the first broadcast. When the neighboring nodes receive the broadcast information, they will record the number of hops from them to the anchor node as 2 / 3R, which is 2 / 3. In the third broadcast, the anchor node will send sensor information to the neighboring nodes with a communication radius of R. In the third broadcast, the anchor node will exclude the neighboring nodes that have been recorded in the previous second broadcast to ensure that the transmission of information will not be repeated. When the neighboring nodes receive the broadcast information, they will record the number of hops from them to the anchor node as 1. At this time, each node can obtain the corrected minimum number of hops. Therefore, Figure 2 The hop count is corrected to 1 / 3, and the hop count between the anchor node and its neighbors is no longer an integer, but a more precise decimal. This decimal value has a closer positive relationship with the distance between nodes, so it can greatly enhance the accuracy of the data and reduce the positioning error.

[0043]

[0044] In the formula, the straight-line distance between the anchor node and the adjacent node is d, and the number of hops between the anchor node and the adjacent node is h. ij .

[0045] (2) Average jump distance correction

[0046] This method applies the least squares criterion and weighted coefficients to correct the average distance between anchor nodes in each hop. In the first part of the DV-hop algorithm, it is assumed that anchor node i receives the location information of other anchor nodes, and the average HopSize calculated by anchor node i is i The following criteria should be met:

[0047]

[0048] Where j represents other anchor nodes in the data table, hops ij represents the number of hops between anchor node i and anchor node j, d ij is the actual length between anchor nodes i and j, ε ij Indicates the error caused by using the product of the average distance per hop and the number of hops instead of the actual length. A reasonable HopSize i The total error should be minimized. According to the principle of least squares method, As the total error, the formula is as follows:

[0049]

[0050] The least squares corrected average jump distance formula can be obtained as follows:

[0051]

[0052] The number of anchor nodes does not necessarily improve the positioning performance of unknown nodes. The closer the anchor nodes are to the unknown nodes, the more valuable the information they provide. The farther away they are, the less valuable the information they provide, and they may even be misleading. Therefore, this paper also introduces a weighting coefficient to increase the importance of close anchor nodes and weaken the importance of distant anchor nodes. The weighting coefficient is as follows:

[0053]

[0054] where w i is the average hop distance weighted coefficient of anchor node i, hop i,j is the minimum hop value between anchor node i and unknown node j, and n is the number of anchor nodes. Therefore, after introducing the weighted coefficient, the average hop distance formula of the unknown node is expressed as:

[0055]

[0056] After improvement, the distance calculation formula between nodes is:

[0057] d i =hop i,j Hopsize j (1-7)

[0058] It can be seen that in the improved distance calculation formula, the average hop distance between nodes is no longer a fixed value Hopsize i Instead, it dynamically corrects the average hop distance to a suitable value according to the changes of the surrounding anchor nodes of the unknown node, which can reduce the node positioning error caused by the average hop distance calculation error to a certain extent.

[0059] (3) Comparison of simulation results

[0060] Figure 3 (a) shows the relative positioning error distribution of the traditional DV-Hop algorithm among 80 unknown nodes, which is between 20% and 50%, and the relative positioning error of a very small number of nodes can reach about 70%. In contrast, Figure 3 (b) shows the performance of the improved DV-Hop algorithm under the same number of nodes. Its relative positioning error is distributed between 10% and 25%, and the relative positioning error of a very small number of nodes can reach about 45%. The above results show that the improved DV-Hop algorithm performs better than the traditional DV-Hop algorithm in relative positioning error performance in the same scenario, and can more accurately estimate the positions between nodes.

[0061] like Figure 4 As shown in (b), the actual positioning effect of the improved node has been effectively demonstrated. In the figure, the blue points represent the real coordinates of the node, while the red points represent the coordinates estimated by the algorithm. It can be clearly seen that the coordinates estimated by the algorithm are almost all within 10 meters of the real position of the node. Compared with the positioning results of traditional DV-hop, the positioning effect of the improved algorithm has been significantly improved.

Claims

1. An improved node positioning method based on DV-Hop, the method comprising: Step 1: Network initialization; Step 2: Multiple communication radius correction adjustment; In the network, the anchor node broadcasts its coordinates three times according to the communication radius of 1 / 3R, 2 / 3R and R. The size of R is determined according to the actual situation of the communication network. In the first broadcast, the anchor node broadcasts sensor information to the surrounding nodes. Its communication radius is 1 / 3R, covering the nodes within 1 / 3R of the anchor node. After receiving the broadcast, neighbor nodes record the distance from them to the anchor node as 1 / 3R, which is 1 / 3 of the hop count; In the second broadcast, the anchor node transmits sensor information to nearby nodes with a communication radius of 2 / 3R, which does not include the surrounding nodes recorded in the first broadcast; when the neighboring nodes receive the broadcast information, they will record the number of hops from them to the anchor node, which is 2 / 3R. This number of hops is 2 / 3; in the third broadcast, the anchor node will send sensor information to the neighboring nodes, whose communication radius is R, and the anchor node will exclude the neighboring nodes that have been recorded in the second broadcast to ensure that the transmission of information will not be repeated; when the neighboring nodes receive the broadcast information, they will record the number of hops from them to the anchor node, which is 1; at this time, each node obtains the corrected minimum number of hops; Step 3: The beacon node calculates the average hop distance and broadcasts it; Step 4: Correct the average jump distance; Step 5: Estimated distance of unknown nodes; Step 6: Determine whether the unknown node has four or more beacon node information. If yes, proceed to step 7. If not, the unknown node cannot be located and proceed to step 8. Step 7: Calculate the position according to the maximum likelihood positioning method; Step 8: Calculation ends.

2. The improved node positioning method based on DV-Hop as claimed in claim 1, characterized in that: The specific method of step 2 is: Step 2.1: Initialize i=1; Step 2.2: The anchor node broadcasts its own information with a communication radius of i / 3R; Step 2.3: The receiving node queries whether the on-duty node has received the broadcast information of the anchor node. If not, it receives and records the number of information hops; if yes, it proceeds to step 2.4; Step 2.4: Update i=i+1; Step 2.5: Determine whether i is less than 3. If so, return to step 2.

2. If not, proceed to step 2.

6. Step 2.6: The anchor node broadcasts its own information with R as the communication radius; Step 2.7: The receiving node compares and saves the minimum hop count, then increases the hop count by 1 and forwards the information.

3. The improved node positioning method based on DV-Hop as claimed in claim 1, characterized in that: The specific method of step 4 is: Step 4.1: Assume that anchor node i receives the location information of other anchor nodes, and anchor node i calculates the average hop number HopSize i The following criteria should be met:d ij =HopSize i *hops ij +ε ij , hops ij represents the number of hops between anchor node i and anchor node j, d ij is the actual length between anchor nodes i and j, ε ij It indicates the error caused by replacing the actual length with the product of the average distance per hop and the number of hops; Step 4.2: Calculate the average hop distance: Step 4.3: Add weighting coefficients hop ij represents the minimum hop value between anchor node i and unknown node j, n represents the number of anchor nodes, and the improved distance calculation formula is obtained:

4. The improved node positioning method based on DV-Hop as claimed in claim 1, characterized in that: The average jump distance formula in step 4.2 is obtained as follows; As the total error, the formula is as follows: use The least squares corrected average jump distance formula is as follows:

Images