Internet of Things security positioning method and device based on risk weight assessment

Abstract

The invention provides an Internet of Things security positioning method and device based on risk weight assessment. The method comprises the following steps: each anchor node measures an angle and a distance between the anchor node and a target node so as to estimate position information of the target node, and initial position information of the target node is calculated according to the position information estimated by all the anchor nodes; subtracting the estimated position information of each anchor node from the initial position information of the target node to obtain an initial positioning error of each anchor node, and performing hypothesis testing on each anchor node according to the initial positioning error to divide all the anchor nodes into malicious nodes and normal nodes; setting the risk weight of the malicious node as 0; performing fuzzy comprehensive evaluation on the normal nodes subjected to the hypothesis test to obtain a risk weight of each normal node; and performing weighted positioning according to the estimated position information of each anchor node and the corresponding risk weight to obtain final position information of the target node.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Things security positioning, and particularly relates to an Internet of Things security positioning method and device based on risk weight evaluation. BACKGROUND

[0002] In the Internet of Things environment, positioning technology is the basis for realizing device state monitoring, production process control and intelligent decision-making. The core of the service security problem in positioning is the security hidden danger brought by the maliciousness and heterogeneity of nodes, which directly affects the reliability and accuracy of the positioning system. In the open and distributed Internet of Things environment, some nodes may be attacked, maliciously tampered with, forged or interfered with positioning data, causing the positioning system to fail or mislead decision-making. In addition, due to the differences in physical characteristics of sensors, environmental interference and equipment aging factors, different sensors may have deviations or inconsistencies in measuring the position of the same target node. This will lead to inaccurate positioning results, and further affect the overall performance of the network. The above reasons cause the positioning in the Internet of Things to face serious security problems, and a method is urgently needed to solve the security positioning problem to ensure the stable operation of the Internet of Things system and support the positioning demand in complex environments.

[0003] Current research has made some progress in addressing security risks caused by node malice and heterogeneity in the Internet of Things (IoT). Regarding node malice, many studies currently focus on detecting malicious nodes using machine learning and trust value assessment. Osama et al., in their paper "Dual-stage machine learning approach for advanced malicious node detection in WSNs. (Ad Hoc Networksvol. vol. 166, pp. 103672, 2025)," utilized machine learning to enhance malicious node identification in wireless sensor networks, proposing a novel two-stage malicious node detection scheme to improve detection accuracy. Zhao et al., in their paper "A cooperative detection scheme for malicious nodes based on DS trustevidence reasoning in mobile crowdsensing networks. (Computers and Electrical Engineering. vol.118, pp. 109456, 2024)," proposed a cooperative malicious node detection scheme based on trust evidence reasoning. This scheme employs trust decay and control mechanisms to dynamically update the trust evidence chain, achieving malicious node detection through consensus among trusted nodes. While these methods can detect malicious nodes, most of them require additional storage space to store the node's historical behavior information, increasing computational and storage overhead.

[0004] Furthermore, existing research primarily employs machine learning, trust value evaluation, and outlier detection to improve positioning accuracy in IoT applications, addressing the issues of malicious node detection and node heterogeneity. However, these methods still face challenges in computationally limited and complex environments, requiring further in-depth research to enhance both positioning accuracy and security. Summary of the Invention

[0005] To address the security issues arising from the maliciousness and heterogeneity of positioning nodes in existing IoT secure positioning methods, this invention proposes an IoT secure positioning method and device based on Risk Weight Assessment (RWA-SL). This invention reduces the impact of malicious nodes while optimizing positioning accuracy and system robustness, ensuring the security and precision of the positioning system.

[0006] In a first aspect, the present invention provides an IoT security positioning method based on risk weight assessment, comprising: Step 1: Each anchor node measures the angle and distance between itself and the target node to estimate the position information of the target node. The initial position information of the target node is calculated based on the position information estimated by all anchor nodes. Step 2: Subtract the estimated position information of each anchor node from the initial position information of the target node to obtain the initial positioning error of each anchor node. Perform hypothesis testing on each anchor node based on the initial positioning error to classify all anchor nodes into malicious nodes and normal nodes. Step 3: Set the risk weight of malicious nodes to 0; and perform fuzzy comprehensive evaluation on normal nodes after hypothesis testing to obtain the risk weight of each normal node; Step 4: Perform weighted positioning based on the estimated location information of each anchor node and the corresponding risk weight to obtain the final location information of the target node.

[0007] Further, in step 2, hypothesis testing is performed on each anchor node based on the initial positioning error to classify all anchor nodes into malicious nodes and candidate nodes, specifically including: Define the null hypothesis The anchor node is a normal node; alternative assumption The anchor node is a malicious node; Define the impact of the presence or absence of malicious nodes on the distance measurement results as follows: in, This represents the distance between anchor node i and the target node in the k-th measurement. Indicates the actual location of the target node. This represents the location information of the target node measured by anchor node i. This represents the positioning error of anchor node i in the k-th measurement, which follows a Gaussian distribution with zero mean. This indicates the impact on distance measurement when anchor node i is a malicious node; This allows us to derive the likelihood functions for normal and malicious nodes, as well as the likelihood ratio between the two likelihood functions, and ultimately, the rejection region. for: in, Both c and c are constants. This represents the positioning error variable. For likelihood ratio, It is the initial positioning error about the target node measured by anchor node i; Given that the positioning error variable is when the anchor node is a normal node. It follows a normal distribution η~N(0,1 / K), thus we obtain: Where α is the probability that the anchor node is judged as a malicious node when it is not a malicious node, and K is the number of times the anchor node measures the target node; let The test level is then obtained as test function as follows: in, This indicates that anchor node i is a malicious node, and vice versa. This indicates that anchor node i is a normal node.

[0008] Furthermore, step 3 specifically includes: A factor set is constructed based on the factors that normal nodes themselves can influence the positioning results; The evaluation set is obtained by normalizing the performance of normal nodes on different factors; The evaluation matrix is ​​obtained by integrating the evaluation sets of all normal nodes; Based on the evaluation matrix, the weights of all factors are determined according to the entropy weight method, and the weights of all factors are integrated to obtain the factor weight matrix. The risk weight of each normal node is obtained by weighting the evaluation matrix using the factor weight matrix.

[0009] Furthermore, step 4 specifically includes: in, Let i be the risk weight of anchor node i. The location information of the target node measured for anchor node i, where M is the total number of anchor nodes. This is the estimated final location information of the target node.

[0010] Secondly, the present invention provides an IoT security positioning device based on risk weight assessment, comprising: The position estimation module is used by each anchor node to measure the angle and distance between itself and the target node to estimate the position information of the target node, and calculates the initial position information of the target node based on the position information estimated by all anchor nodes. The malicious node screening module is used to subtract the estimated position information of each anchor node from the initial position information of the target node to obtain the initial positioning error of each anchor node. Based on the initial positioning error, hypothesis testing is performed on each anchor node to classify all anchor nodes into malicious nodes and normal nodes. The weight configuration module is used to set the risk weight of malicious nodes to 0; and to perform fuzzy comprehensive evaluation on normal nodes after hypothesis testing to obtain the risk weight of each normal node. The position correction module is used to perform weighted positioning based on the estimated position information of each anchor node and the corresponding risk weight, so as to obtain the final position information of the target node.

[0011] Thirdly, the present invention provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method as described in the first aspect.

[0012] Fourthly, the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method described in the first aspect.

[0013] The beneficial effects of this invention are as follows: This invention provides an IoT secure positioning method and device based on risk weight assessment, effectively addressing the issues of malicious and heterogeneous positioning nodes in IoT positioning systems. First, it employs hypothesis testing to identify malicious nodes, improving system security. Second, it uses fuzzy comprehensive evaluation to uniformly assess node heterogeneity, enhancing positioning accuracy. Finally, it designs a secure positioning algorithm based on service security assessment results, reducing the impact of malicious nodes while optimizing positioning accuracy and system robustness. In summary, this invention can ensure both security and accuracy in positioning while estimating the location of target nodes. Simulation results show that, in the presence of malicious nodes, the positioning accuracy of this invention is improved by an average of 46.18%. Attached Figure Description

[0014] Figure 1 A flowchart illustrating an IoT security positioning method based on risk weight assessment, provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of malicious node detection provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of risk weight evaluation provided in an embodiment of the present invention; Figure 4 The diagram illustrates the analysis results of the relationship between positioning error and the number of anchor nodes provided in the embodiments of the present invention. Figure 5 The graph shows the analysis results of the relationship between positioning error and the number of measurements provided in the embodiments of the present invention; Figure 6 The diagram illustrates the analysis results of the relationship between positioning error and the proportion of malicious nodes, provided by an embodiment of the present invention. Figure 7 A schematic diagram of the structure of an IoT security positioning device based on risk weight assessment provided in an embodiment of the present invention; Figure 8 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the embodiments of this invention will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0016] like Figure 1 As shown, this embodiment of the invention provides an IoT security positioning method based on risk weight assessment, including the following steps: S101: Each anchor node measures the angle and distance between itself and the target node to estimate the position information of the target node. The initial position information of the target node is calculated based on the position information estimated by all anchor nodes. Specifically, existing measurement methods, such as ranging based on signal reception strength, can be used by the anchor nodes to measure their distance and angle to the target node. After obtaining the target node's position information measured by all anchor nodes, the initial position information of the target node is obtained by averaging the measurement information of all anchor nodes. .

[0017] S102: Subtract the estimated position information of each anchor node from the initial position information of the target node to obtain the initial positioning error of each anchor node. Perform hypothesis testing on each anchor node based on the initial positioning error to classify all anchor nodes into malicious nodes and normal nodes. Set the risk weight of the malicious node to 0. If not, assign the initial risk weight of the anchor node to 1. Thus, the malicious node and normal node can be distinguished by the initial weight value.

[0018] Specifically, such as Figure 2 As shown, to address the issue of malicious nodes in IoT positioning systems, a malicious node detection method is designed to detect malicious nodes in the system, including the following sub-steps: S1021: Define the null hypothesis The anchor node is a normal node; alternative assumption The anchor node is a malicious node.

[0019] Specifically, from a mathematical perspective, the distance-based localization problem can be formulated as a system of linear equations, describing the observed distance between the anchor node and the target node as a composite result of additive noise contamination and potential malicious attack interference on their true geometric distance. Therefore, malicious nodes in the system can be modeled from the perspective of statistical hypothesis testing.

[0020] Considering that each observation distance may come from two distributions: the presence or absence of malicious nodes, two hypotheses are set, namely the null hypothesis. The anchor node is a normal node; alternative assumption The anchor node is a malicious node.

[0021] S1022: Based on the above assumptions, the impact of the presence or absence of malicious nodes on distance measurement results is defined as follows: in, This represents the distance between anchor node i and the target node in the k-th measurement. Indicates the actual location of the target node. This represents the location information of the target node measured by anchor node i. This represents the positioning error of anchor node i in the k-th measurement, which follows a Gaussian distribution with zero mean. This indicates the impact on distance measurement when anchor node i is a malicious node.

[0022] S1023: Based on the above, derive the likelihood functions for normal nodes respectively. Likelihood function with malicious nodes : Where K represents the number of measurements taken by a single anchor node from the target node. This represents the noise standard deviation based on the maximum likelihood estimation. , This indicates the measured distance between anchor node i and the target node.

[0023] Furthermore, the likelihood ratio between two likelihood functions can be expressed as follows: in, It is the initial positioning error of anchor node i with respect to the target node; Obviously, when hour, for Since it is a strictly increasing function, the rejection region can be obtained as: in, Both c and are constants, i.e., the likelihood ratio Equivalent to It is greater than another constant c.

[0024] S1024: Given that when the anchor node is a normal node, the measured positioning error conforms to a normal distribution η~N(0,1 / K), based on this, we can obtain: Where α is the probability of identifying an anchor node as malicious when the anchor node is not malicious. At this point, there will be... .make , Under standard normal distribution quantiles.

[0025] Depend on The significance level can be obtained as follows: The test function is shown in the following formula: in, This indicates that anchor node i is a malicious node, and vice versa. This indicates that anchor node i is a normal node.

[0026] After the above process, when the initial positioning error of the anchor node measurement... Greater than the threshold At this time, the anchor node is marked as a malicious node, thereby filtering out malicious nodes. Simultaneously, the weight of the malicious node is adjusted. Setting it to 0 ensures that it will not affect the positioning results during subsequent positioning processes.

[0027] Specifically, in determining whether an anchor node is malicious, two types of errors may occur: (1) judging it as malicious when it is not malicious; and (2) judging it as normal when it is malicious. Generally, when attempting to decrease either the probability of committing a Type (1) error or the probability of committing a Type (2) error, the probability of committing the other type of error increases. The Neyman-Pearson lemma fixes one of the two types of errors and then minimizes the other.

[0028] In this invention, the probability of a normal node being identified as a malicious node is fixed within an acceptable range to minimize the probability of a malicious node being identified as a normal node. This is because: if an anchor node is a normal node but is identified as a malicious node, it will not participate in the localization task, only causing some additional time overhead; while if an anchor node is a malicious node but is identified as a normal node, it may lead to receiving incorrect information, or even cause system paralysis.

[0029] S103: Perform fuzzy comprehensive evaluation on normal nodes after hypothesis testing to obtain the risk weight of each normal node; Specifically, such as Figure 3 As shown, a risk weighting assessment method is designed to address the node heterogeneity problem in IoT positioning systems. The evaluation of anchor nodes typically involves multiple factors, such as the distance and angle between the anchor node and the target node, social relationships, and the anchor node's battery level. However, the impact of these factors on the positioning results is fuzzy, imprecise, and uncertain. This necessitates a comprehensive evaluation based on multiple factors, rather than relying solely on a single factor.

[0030] Fuzzy mathematics theory is a mathematical theory and method that uses precise mathematical means to study and handle fuzzy phenomena. Fuzzy comprehensive evaluation and decision-making is an effective multi-factor decision-making method for making a comprehensive evaluation of things influenced by multiple factors. Therefore, based on the characteristics of fuzzy mathematics theory, the fuzzy comprehensive evaluation method is used to calculate the comprehensive capability of each anchor node. Specifically, it includes the following sub-steps: S1031: Construction of Factor Set Preparation. In this invention, the factors that can affect the positioning results of the anchor node itself are used as the factor set for fuzzy comprehensive evaluation. During the positioning process, the factors affecting the positioning results mainly include the following: the distance and angle between the anchor node and the target node, the maximum measurable distance of the anchor node, the stability of the anchor node in different environments, the equipment resource occupancy of the anchor node, and the social relationships between anchor nodes, etc.

[0031] S1032: Constructing the evaluation set. In this invention, the evaluation set is obtained by normalizing the performance of all anchor nodes on different factors. The anchor node is then evaluated based on its performance on these different factors.

[0032] Taking distance as an example, the process of evaluating anchor nodes is as follows: Assume the distance range between the target node and the anchor node is... , The anchor node i is assigned a value based on its distance from the target node, and then normalized. Similarly, the evaluation results for other factors can be obtained. Finally, the evaluation set for anchor node i is obtained: in, This refers to the number of factors considered that influence the positioning results. Indicates that anchor node i is at the 1st epoch. Evaluation values ​​for each factor.

[0033] Assuming there are M anchor nodes, there are corresponding M evaluation sets, thus the evaluation matrix can be obtained: S1033: Calculate the factor weight matrix. The entropy weight method is used to calculate the weight of each factor.

[0034] First, the data in the evaluation matrix needs to be normalized according to the formula to obtain the normalized set: in, This represents the value of the element in the i-th row and j-th column of the evaluation matrix before normalization. It is the value of the element in the i-th row and j-th column of the evaluation matrix after normalization.

[0035] Then, calculate the normalized entropy for each column of data, using the following formula: in, This represents the normalized entropy of the j-th column (i.e., the j-th factor).

[0036] Next, calculate the weight of the j-th column (i.e., the j-th factor): This allows us to obtain the factor weight matrix, which consists of the weights of all factors. S1034: Calculate the final evaluation result.

[0037] Multiplying the evaluation matrix by the factor weight matrix yields the final risk weight for each anchor node, the process of which is expressed by the formula: in, Let B represent the evaluation result matrix, with a size of ; This represents the factor weight matrix, with a size of ; Representing the evaluation matrix Transpose of; The content of the calculated evaluation result matrix B is the risk weight of each anchor node. .

[0038] S104: Perform weighted positioning based on the location information calculated from the anchor node and the final risk weight to obtain the final positioning result.

[0039] Specifically, using the malicious node detection results from step S102, the risk weight of the malicious node is set to 0; the evaluation results obtained from the fuzzy comprehensive evaluation of each anchor node in step S103 are used as the risk weight of the anchor node. The final location information of the target node is obtained by weighting the measured position of the target node by the risk weight of the anchor node, as expressed by the formula: in, The location information of the target node measured for the i-th anchor node, where M is the total number of anchor nodes. This is the estimated final location information of the target node.

[0040] To verify the effectiveness of the technical solution of this invention, the following experiment was conducted: The simulation parameters were configured as follows: 100 nodes were uniformly distributed within an area centered at (100, 100). During the simulation, zero-mean random variables were used to simulate distance and angle measurement errors under a Gaussian distribution. To account for the heterogeneity of the device hardware, 10% of the anchor nodes were configured with weaker positioning capabilities, leading to larger measurement errors. Considering the possibility of malicious activity, some anchor nodes in the system were configured as malicious nodes to provide false positioning information. To reduce the impact of random variables on the results, all simulations were run independently for an average of over 1000 times.

[0041] The comparison schemes include Accurate Privacy Protection Location (APPL), Homomorphic encryption based Privacy-Preserving Localization (HPPL), Gaussiannoise based Privacy Preserving Localization (GPPL), A Trust Evaluation Scheme for Federated Learning in Digital Twin (TFL), and Greedy Localization Algorithm (Greedy).

[0042] Figure 4The relationship between the number of anchor nodes, positioning range, and positioning error under different positioning algorithms is shown. As the number of anchor nodes increases, the positioning error of all algorithms gradually decreases. This is because the increased number of anchor nodes provides richer information, thus improving positioning accuracy. RWA-SL exhibits the smallest positioning error, reducing it by 66.4%, 57.3%, 52.5%, and 27.1% compared to GPPL, HPPL, APPL, and TFL algorithms, respectively. While the increased number of anchor nodes provides richer information, it also leads to heterogeneity issues among anchor nodes. When anchor nodes are heterogeneous, RWA-SL of this invention can mitigate the impact of node heterogeneity on positioning results through fuzzy comprehensive evaluation, resulting in better position estimation. In particular, as the number of anchor nodes increases, the positioning error of RWA-SL of this invention increasingly approaches that of the greedy algorithm, demonstrating the superiority of RWA-SL.

[0043] Figure 5 The impact of the number of measurements on positioning accuracy under different algorithms is shown. HPPL, GPPL, APPL, RWA-SL, the greedy algorithm, and TFL all exhibit stability. Compared to HPPL, GPPL, APPL, and TFL, RWA-SL has higher positioning accuracy, with positioning errors reduced by approximately 58.3%, 49.6%, 43.7%, and 35.2%, respectively. This is because the RWA-SL algorithm can detect malicious nodes in the system and considers multiple factors that can affect positioning accuracy. Although the greedy algorithm shows the best positioning performance because it can traverse all positioning nodes to find the optimal positioning effect, it also increases the time overhead. This indicates that RWA-SL has a better positioning effect when anchor nodes are heterogeneous and malicious nodes are present.

[0044] Figure 6The figure shows the relationship between the proportion of malicious nodes in the system and the positioning error. These malicious nodes may provide false positioning information, leading to significant positioning errors. As can be seen from the figure, the proposed RWA-SL algorithm performs better as the proportion of malicious nodes increases. This is because the RWA-SL algorithm can detect malicious nodes in the system and does not consider them when calculating the final positioning result, achieving improvements of 54.4%, 63.7%, 67.7%, and 31.4% compared to the APPL, HPPL, GPPL, and TFL algorithms, respectively. However, the positioning errors of the APPL, HPPL, and GPPL algorithms gradually increase because these algorithms do not consider the presence of malicious nodes in the system. While the TFL algorithm considers malicious nodes in positioning, it does not consider other factors affecting positioning accuracy. Furthermore, compared to the HPPL and GPPL algorithms, the APPL algorithm can select nodes to participate in positioning based on their own attributes; therefore, the APPL algorithm has a lower positioning error than the HPPL and GPPL algorithms.

[0045] In summary, this invention provides an IoT secure positioning method based on risk weight assessment, effectively addressing service security issues caused by the malice and heterogeneity of nodes in IoT positioning systems. Hypothesis testing techniques can detect malicious nodes in the system, improving system security. Furthermore, to improve positioning accuracy, this invention proposes a fuzzy comprehensive evaluation technique to comprehensively assess node performance, assigning different risk weights to nodes based on the evaluation results. The final positioning result is obtained through weighted averaging, thus improving positioning performance. Experimental results show that the proposed scheme can significantly reduce the interference of node malice and heterogeneity on the positioning system, effectively maintaining positioning accuracy.

[0046] like Figure 7 As shown, based on the same inventive concept, this embodiment of the invention also provides an IoT security positioning device based on risk weight assessment, including a location estimation module, a malicious node screening module, a weight configuration module, and a location correction module.

[0047] The system comprises the following modules: a position estimation module, where each anchor node measures the angle and distance between itself and the target node to estimate the target node's position information; an initial position information of the target node is calculated based on the estimated position information of all anchor nodes; a malicious node screening module, where the estimated position information of each anchor node is subtracted from the initial position information of the target node to obtain the initial positioning error of each anchor node; hypothesis testing is performed on each anchor node based on the initial positioning error to classify all anchor nodes into malicious nodes and normal nodes, and the risk weight of malicious nodes is set to 0; a weight configuration module, where fuzzy comprehensive evaluation is performed on the normal nodes after hypothesis testing to obtain the risk weight of each normal node; and a position correction module, where weighted positioning is performed based on the estimated position information of each anchor node and its corresponding risk weight to obtain the final position information of the target node.

[0048] It should be noted that the IoT security positioning device provided in this embodiment of the invention is for implementing the above method, and its specific functions can be found in the above method embodiments, which will not be repeated here. Figure 8 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 8 As shown, the electronic device may include: a processor 801, a communication interface 802, a memory 803, and a communication bus 804, wherein the processor 801, the communication interface 802, and the memory 803 communicate with each other through the communication bus 804. The processor 801 can call logic instructions in the memory 803 to execute an IoT security positioning method based on risk weight assessment. The method includes: Step 1: Each anchor node measures the angle and distance between itself and the target node to estimate the target node's position information; the initial position information of the target node is calculated based on the estimated position information of all anchor nodes; Step 2: The difference between the estimated position information of each anchor node and the initial position information of the target node is calculated to obtain the initial positioning error of each anchor node; hypothesis testing is performed on each anchor node based on the initial positioning error to classify all anchor nodes into malicious nodes and normal nodes; Step 3: The risk weight of malicious nodes is set to 0; fuzzy comprehensive evaluation is performed on the normal nodes after hypothesis testing to obtain the risk weight of each normal node; Step 4: Weighted positioning is performed based on the estimated position information of each anchor node and its corresponding risk weight to obtain the final position information of the target node.

[0049] Furthermore, when the logical instructions in the aforementioned memory 803 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0050] This invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, and when the program instructions are executed by a computer, the computer can execute an IoT security positioning method based on risk weight assessment provided in the above-described method embodiments.

[0051] This invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements an IoT security positioning method based on risk weight assessment provided in the above-described method embodiments.

[0052] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0053] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A secure positioning method for the Internet of Things (IoT) based on risk weight assessment, characterized in that, include: Step 1: Each anchor node measures the angle and distance between itself and the target node to estimate the position information of the target node. The initial position information of the target node is calculated based on the position information estimated by all anchor nodes. Step 2: Subtract the estimated position information of each anchor node from the initial position information of the target node to obtain the initial positioning error of each anchor node. Perform hypothesis testing on each anchor node based on the initial positioning error to classify all anchor nodes into malicious nodes and normal nodes. Step 3: Set the risk weight of malicious nodes to 0; Furthermore, a fuzzy comprehensive evaluation is performed on the normal nodes after hypothesis testing to obtain the risk weight of each normal node; Step 4: Perform weighted positioning based on the estimated location information of each anchor node and the corresponding risk weight to obtain the final location information of the target node.

2. The IoT security positioning method based on risk weight assessment according to claim 1, characterized in that, In step 2, hypothesis testing is performed on each anchor node based on the initial positioning error to classify all anchor nodes into malicious nodes and candidate nodes. Specifically, this includes: Define the null hypothesis The anchor node is a normal node; alternative assumption The anchor node is a malicious node; Define the impact of the presence or absence of malicious nodes on the distance measurement results as follows: in, This represents the distance between anchor node i and the target node in the k-th measurement. Indicates the actual location of the target node. This represents the location information of the target node measured by anchor node i. This represents the positioning error of anchor node i in the k-th measurement, which follows a Gaussian distribution with zero mean. This indicates the impact on distance measurement when anchor node i is a malicious node; This allows us to derive the likelihood functions for normal and malicious nodes, as well as the likelihood ratio between the two likelihood functions, and ultimately, the rejection region. for: in, Both c and c are constants. This represents the positioning error variable. For likelihood ratio, It is the initial positioning error about the target node measured by anchor node i; Given that the positioning error variable is when the anchor node is a normal node. It follows a normal distribution η~N(0,1 / K), thus we obtain: Where α is the probability that the anchor node is judged as a malicious node when it is not a malicious node, and K is the number of times the anchor node measures the target node; let The test level is then obtained as test function as follows: in, This indicates that anchor node i is a malicious node, and vice versa. This indicates that anchor node i is a normal node.

3. The IoT security positioning method based on risk weight assessment according to claim 1, characterized in that, Step 3 specifically includes: A factor set is constructed based on the factors that normal nodes themselves can influence the positioning results; The evaluation set is obtained by normalizing the performance of normal nodes on different factors; The evaluation matrix is ​​obtained by integrating the evaluation sets of all normal nodes; Based on the evaluation matrix, the weights of all factors are determined according to the entropy weight method, and the weights of all factors are integrated to obtain the factor weight matrix. The risk weight of each normal node is obtained by weighting the evaluation matrix using the factor weight matrix.

4. The IoT security positioning method based on risk weight assessment according to claim 1, step 4 specifically includes: in, Let i be the risk weight of anchor node i. The location information of the target node measured for anchor node i, where M is the total number of anchor nodes. This is the estimated final location information of the target node.

5. An IoT security positioning device based on risk weight assessment, characterized in that, include: The position estimation module is used by each anchor node to measure the angle and distance between itself and the target node to estimate the position information of the target node, and calculates the initial position information of the target node based on the position information estimated by all anchor nodes. The malicious node screening module is used to subtract the estimated position information of each anchor node from the initial position information of the target node to obtain the initial positioning error of each anchor node. Based on the initial positioning error, hypothesis testing is performed on each anchor node to classify all anchor nodes into malicious nodes and normal nodes. The weight configuration module is used to set the risk weight of malicious nodes to 0. Furthermore, a fuzzy comprehensive evaluation is performed on the normal nodes after hypothesis testing to obtain the risk weight of each normal node; The position correction module is used to perform weighted positioning based on the estimated position information of each anchor node and the corresponding risk weight, so as to obtain the final position information of the target node.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 4.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 4.

Images