Internet of Things-based Information Resource Management System

Abstract

This invention relates to the field of Internet of Things (IoT) information management technology, and more particularly to an IoT-based information resource management system. It acquires data information transmitted by data acquisition devices, identifies the data properties of target data, determines the data characteristic values ​​of the target data based on these properties, performs frequency analysis on usage information to determine the inertial utilization rate of the target data, combines the data characteristic values ​​and inertial utilization rate of each target device for analysis to obtain the priority of the target device, determines the storage level of the current target device based on the priority, and finally, the space adjustment module can dynamically adjust the storage location according to the current storage level of each device's data, ensuring that data storage always matches actual usage needs. When the usage frequency or importance of device data changes, the system can automatically migrate the data to a suitable storage level, ensuring fast access to frequently used data and improving the overall response speed of the system.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) information management technology, and more particularly to an IoT-based information resource management system. Background Technology

[0002] With the rapid development of IoT technology, the number of various sensing devices and smart terminals has exploded, forming massive, multi-source, and heterogeneous information resources.

[0003] The prior art CN115373832A discloses an IoT smart cloud management platform, including a user login layer, a cloud management portal, and adapters. The cloud management portal includes a resource management platform, a cost management platform, a monitoring platform, an organizational structure platform, a self-service management platform, and a data display platform. The adapters include public cloud, virtual machine, private cloud, and container. It dynamically and timely interconnects the data of various application systems within an enterprise. Through server virtualization technology, it virtualizes various hardware and software resources into one or more resource pools, and intelligently and automatically manages and allocates these virtual resources through the system management platform.

[0004] However, the frequency of use of data generated by different devices varies significantly in actual applications. At the same time, traditional information resource management systems lack dynamism in storage space adjustment and cannot flexibly adjust storage locations according to real-time data conditions. As the scale of data continues to expand, this static storage management model is difficult to meet the needs of the Internet of Things (IoT) for efficient data processing, rapid resource scheduling, and low-cost operation and maintenance, thus restricting the in-depth application of IoT technology in various industries. Summary of the Invention

[0005] The purpose of this invention is to solve the problems in the background art by proposing an information resource management system based on the Internet of Things.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] Internet of Things (IoT) based information resource management systems include:

[0008] The information acquisition module is used to collect data information from the Internet of Things;

[0009] The data analysis module is used to acquire data information transmitted by the data acquisition device, set the data acquisition device as the target device, mark the data transmitted by the target device as target data, identify the data properties of the target data, including steady-state data and fluctuating data, and determine the data characteristic values ​​of the target data based on the data properties of the target data, including steady-state data and fluctuating data.

[0010] The frequency analysis module is used to acquire usage information of the target data, perform frequency analysis on the usage information, and determine the inertial utilization rate of the target data.

[0011] The structure setting module is used to combine and analyze the data characteristic values ​​and inertial utilization rate of each target device to obtain the priority of the target device. Based on the priority, the storage level of the current target device is determined, where the storage level includes the first level, the second level and the third level.

[0012] The spatial adjustment module is used to dynamically adjust the storage location of the data acquisition devices based on the storage level of each data acquisition device at the current time.

[0013] As a further aspect of the present invention, the method for identifying the data properties of the target data includes:

[0014] Acquire historically received data information and identify the device source of the data information transmission, where the device source refers to the data acquisition device corresponding to the data information;

[0015] Arbitrarily select a data acquisition device and mark it as the target device. Based on the device source of each data information transmission, extract the historical data information transmitted by the target device and mark this data information as the target data. Then arrange the target data in chronological order to obtain the data sequence.

[0016] Di, where i represents different target data and i∈[1,n], indicates that there are n target data in the data sequence.

[0017] Using formula The data fluctuation value B is obtained, and Da is the mean of all target data in the data sequence;

[0018] The data fluctuation value B is compared with the fluctuation threshold By. When B < By, the target data is marked as steady-state data; otherwise, when B ≥ By, the target data is marked as fluctuating data.

[0019] As a further aspect of the present invention, the method for determining data feature values ​​includes:

[0020] When the target data is steady-state, the mean Da is used as the characteristic value of the target data.

[0021] When the target data is fluctuating, the corresponding data sequence is obtained. Starting from the first target data in the data sequence, a sliding window and its corresponding starting byte 'a' are set. The sliding window is moved sequentially through the data sequence according to its size, and the data within the window is extracted. All data within each window are marked as window sequences. All window sequences are obtained, and the data similarity of each window is calculated sequentially to obtain m data similarity values. The number of similarity values ​​greater than Xy is counted and marked as Ls. Then, the similarity ratio BZ is obtained using the formula BZ = Ls ÷ m. The similarity ratio BZ is compared with the ratio threshold Zy. If BZ ≥ Zy, a similarity signal is generated. Otherwise, if BZ < Zy, the number of bytes in the sliding window is increased by one byte, and the resulting byte is used as the current sliding window size 'a', i.e., a = a + 1. The above method is then used to process the data sequence until BZ ≥ Zy and a similarity signal is generated, at which point the sliding window processing ends.

[0022] When similar signals are identified, all window sequences under similar signals are obtained. At the same time, a linear fitting method is used to fit each window sequence into a curve, and this curve is marked as a window curve. All window curves are aligned, and the curve fitting algorithm is used again to fit multiple window curves into a comprehensive curve, and this comprehensive curve is marked as the data feature value of the target data.

[0023] As a further embodiment of the present invention, the starting byte of the sliding window is set to 2, and the step size of the sliding window is equal to the size of the sliding window.

[0024] As a further aspect of the present invention, after the target data transmitted by the data acquisition device is received by the system, the system will first perform data cleaning on the received data. Data cleaning includes missing value processing, outlier processing, and noise data smoothing.

[0025] As a further aspect of the present invention, the method for determining the inertial utilization rate of target data includes:

[0026] Set a unit time and arbitrarily select a target data. Take the time when the target data is received as the time start, count the number of times the target data is used within the unit time, and mark it as the number of times it is used Uc.

[0027] Divide the number of uses Uc by the unit time to get the data usage frequency;

[0028] Obtain the data usage frequency of all target data under the target device, and perform average processing on the data usage frequency. Mark the obtained average result as the inertial usage rate f of the target data.

[0029] As a further aspect of the present invention, the method for setting up the information storage structure includes:

[0030] Set up storage tiers, which include a first tier, a second tier, and a third tier, with the storage priority as follows: first tier > second tier > third tier;

[0031] Identify the data characteristic value and inertial utilization rate of the target device at the current time, and then use the formula Y=T×α1+f×α2 to obtain the priority value Y of the target device at the current time, where T represents the data characteristic value of the target device at the current time, α1 and α2 are both proportional coefficients, and α1+α2=1;

[0032] The priority value Y of all data acquisition devices at the current time is obtained using the method described above. Then, the priority value Y is compared with the node thresholds k1 and k2. When Y≥k1, the storage location of the data transmitted by this data acquisition device is set as the first level. When k1>Y≥k2, the storage location of the data transmitted by this data acquisition device is set as the second level. When Y<k2, the storage location of the data transmitted by this data acquisition device is set as the third level, where k1>k2.

[0033] As a further embodiment of the present invention, the first layer is local storage space, the second layer is edge storage space, and the third layer is cloud storage.

[0034] As a further aspect of the present invention, when the target data is fluctuating data, the data feature value of the target data is a comprehensive curve. Then, the position of the target data transmitted in the current window sequence is identified in this comprehensive curve, the specific value T of the data feature value is determined in the comprehensive curve, and the value T is input into the priority value calculation formula to determine the priority value Y of the corresponding target device at this time.

[0035] Compared with existing technologies, the advantages of this invention are:

[0036] This invention identifies the properties of target data, distinguishes between steady-state and fluctuating data, and determines corresponding data characteristic values. Then, based on the number of times the data is used per unit time, it determines the inertial utilization rate. Based on the inertial utilization rate of the target data and the device data characteristic values, it determines device priority and further divides different storage levels, achieving on-demand allocation of storage resources. Simultaneously, the space adjustment module can dynamically adjust the storage location according to the current storage level of each device's data, ensuring that data storage always matches actual usage needs. When the usage frequency or importance of device data changes, the system can automatically migrate the data to a suitable storage level, ensuring fast access to frequently used data and improving the overall system response speed. Attached Figure Description

[0037] Figure 1 This is a schematic diagram of the system structure of the present invention. Detailed Implementation

[0038] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0039] Reference Figure 1 The Internet of Things-based information resource management system includes an information acquisition module, a data analysis module, a frequency analysis module, a structure setting module, and a space adjustment module.

[0040] The information acquisition module is used to collect data information from the Internet of Things and transmit it to the data analysis module and the frequency analysis module respectively. The data information includes the amount of data transmitted and the number of times the data is used.

[0041] The data analysis module is used to analyze the received data and retrieve historical data to determine the data characteristic values ​​of each data acquisition device. Specific methods for determining these data characteristic values ​​include:

[0042] S1: Obtain historically received data information and identify the device source of the data information transmission, where the device source refers to the data acquisition device corresponding to the data information;

[0043] S2: Randomly select a data acquisition device and mark it as the target device. Based on the device source of each data information transmission, extract the historical data information transmitted by the target device and mark this data information as the target data. Then arrange the target data in chronological order to obtain the data sequence.

[0044] Identify the data properties of the target data and determine its data characteristic values ​​based on these properties. The data properties include steady-state data and fluctuating data. Specific methods for determining the data properties of the target data include:

[0045] Di, where i represents different target data and i∈[1,n], indicates that there are n target data in the data sequence.

[0046] Using formula The data fluctuation value B is obtained, and Da is the mean of all target data in the data sequence;

[0047] The data fluctuation value B is compared with the fluctuation threshold By. When B < By, the target data is marked as steady-state data. Conversely, when B ≥ By, the target data is marked as fluctuating data. The specific value of the fluctuation threshold By is obtained by those skilled in the art based on big data calculations.

[0048] When the target data is steady-state, the mean Da is used as the characteristic value of the target data.

[0049] When the target data is fluctuating, the corresponding data sequence is obtained. Starting from the first target data in the data sequence, a sliding window and its corresponding starting byte 'a' are set. The sliding window is moved sequentially through the data sequence according to its size, extracting data within each window. All data within each window are marked as window sequences. Data similarity is calculated for each window, resulting in m data similarity values. The number of similarity values ​​greater than Xy is counted and marked as Ls. Then, the similarity ratio BZ is obtained using the formula BZ = Ls ÷ m. The similarity ratio BZ is compared with the ratio threshold Zy. If BZ ≥ Zy, the similarity ratio is calculated. If BZ < Zy, then a similar signal is generated. Conversely, if BZ < Zy, then the number of bytes in the sliding window is increased by one byte, and the resulting byte is used as the current number of bytes in the sliding window, a, i.e., a = a + 1. Then the above method is used for processing until BZ ≥ Zy and a similar signal is generated, at which point the sliding window processing of the data sequence ends. The specific value of the proportional threshold Zy is obtained by those skilled in the art based on big data calculations. In this embodiment, the data similarity algorithm is the cosine similarity algorithm. The specific processing process belongs to the prior art and will not be described in detail here. Xy is the similarity threshold, and the specific value is obtained by those skilled in the art based on big data calculations.

[0050] Furthermore, the starting byte of the sliding window is set to 2, and the step size of the sliding window is equal to the size of the sliding window, meaning that there is no overlap between data when the sliding window moves;

[0051] When similar signals are identified, all window sequences under similar signals are obtained. At the same time, a linear fitting method is used to fit each window sequence into a curve, and this curve is marked as a window curve. All window curves are aligned, and the curve fitting algorithm is used again to fit multiple window curves into a comprehensive curve. This comprehensive curve is marked as the data feature value of the target data. In this embodiment, the curve fitting algorithm uses the polynomial fitting method in linear fitting. The specific processing process is existing technology and will not be described in detail here.

[0052] It should be further explained that after the target data transmitted by the data acquisition device is received by the system, the system will first perform data cleaning on the received data. Data cleaning includes missing value processing, outlier processing, and noise data smoothing. The specific data cleaning methods are existing technologies and will not be elaborated here.

[0053] Then, a one-way communication connection is established between the data analysis module and the structure setting module;

[0054] The frequency analysis module is used to acquire usage information of the target data and perform frequency analysis on the usage information to determine the inertial utilization rate of the target data. Specific methods for determining the inertial utilization rate include:

[0055] Set a unit time and arbitrarily select a target data. Take the time when the target data is received as the time starting point, count the number of times the target data is used within the unit time, and mark it as the number of uses Uc. The unit time is a threshold, and the specific value of the unit time is set by those skilled in the art based on big data experience.

[0056] Divide the number of uses Uc by the unit time to get the data usage frequency;

[0057] Obtain the data usage frequency of all target data under the target device, and perform average processing on the data usage frequency. Mark the obtained average result as the inertial usage rate f of the target data.

[0058] A one-way communication connection is established between the frequency analysis module and the structure setting module, and the inertia utilization rate is transmitted to the structure setting module;

[0059] The structure setting module is used to combine and analyze the data characteristic values ​​and inertial utilization rate of each target device to determine the information storage structure. The specific methods for setting the information storage structure include:

[0060] Set up storage tiers, which include a first tier, a second tier, and a third tier, with the storage priority as follows: first tier > second tier > third tier;

[0061] In this embodiment, the first level is local storage space, the second level is edge storage space, and the third level is cloud storage;

[0062] The data characteristic value and inertial utilization rate of the target device at the current time are identified. Then, the priority value Y of the target device at the current time is obtained by using the formula Y=T×α1+f×α2, where T represents the data characteristic value of the target device at the current time, α1 and α2 are both proportional coefficients, and α1+α2=1. In this embodiment, the specific values ​​of α1 and α2 are obtained by those skilled in the art based on big data calculations.

[0063] It should be further explained that when the target data is fluctuating data, the data feature value of the target data is the composite curve. Then, the position of the target data transmitted in the current window series in this composite curve is identified, the specific value T of the data feature value is determined in the composite curve, and the value T is input into the priority value calculation formula to determine the priority value Y of the corresponding target device at this time.

[0064] The priority value Y of all data acquisition devices at the current time is obtained using the above method. Then, the priority value Y is compared with the node thresholds k1 and k2. When Y≥k1, the storage location of the data transmitted by this data acquisition device is set to the first level. When k1>Y≥k2, the storage location of the data transmitted by this data acquisition device is set to the second level. When Y<k2, the storage location of the data transmitted by this data acquisition device is set to the third level.

[0065] Wherein, k1 > K2, and the specific values ​​of k1 and k2 are obtained by those skilled in the art based on big data calculations;

[0066] Then, a one-way communication connection is established between the structure setting module and the space adjustment module;

[0067] The spatial adjustment module is used to dynamically adjust the storage location of the data acquisition devices based on the storage level of each data acquisition device at the current time.

[0068] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An information resource management system based on the Internet of Things, characterized in that, include: The information acquisition module is used to collect data information from the Internet of Things; The data analysis module is used to acquire data information transmitted by the data acquisition device, set the data acquisition device as the target device, mark the data transmitted by the target device as target data, identify the data properties of the target data, and determine the data characteristic values ​​of the target data based on the data properties, including steady-state data and fluctuating data. Methods for determining data feature values ​​include: When the target data is steady-state, the mean Da of all target data in the data sequence is taken as the data feature value of the target data. When the target data is fluctuating, obtain the corresponding data sequence. Starting from the first target data in the data sequence, set the sliding window and the corresponding starting byte 'a'. According to the window size, slide the sliding window sequentially in the data sequence to extract the data within the window. Mark all the data within each window as the window sequence and obtain all the window sequences. The starting byte of the sliding window is set to 2, and the step size of the sliding window is equal to the size of the sliding window. The data similarity of each window is calculated sequentially to obtain m data similarity values. The number of similarity values ​​greater than Xy is counted and marked as Ls, where Xy is the similarity threshold. Then, the similarity ratio BZ is obtained using the formula BZ=Ls÷m. The similarity ratio BZ is compared with the ratio threshold Zy. If BZ≥Zy, a similarity signal is generated. Otherwise, if BZ<Zy, the number of bytes in the sliding window is increased by one byte, and the resulting byte is used as the current number of bytes in the sliding window, i.e., a=a+1. The above method is then used to process the data sequence until BZ≥Zy and a similarity signal is generated, at which point the sliding window processing ends. When similar signals are identified, all window sequences under similar signals are obtained. At the same time, a linear fitting method is used to fit each window sequence into a curve and mark this curve as a window curve. All window curves are aligned and the curve fitting algorithm is used again to fit multiple window curves into a comprehensive curve and mark this comprehensive curve as the data feature value of the target data. The frequency analysis module is used to acquire usage information of the target data and perform frequency analysis on the usage information to determine the inertial utilization rate of the target data. The methods for determining the inertial utilization rate include: Set a unit time and arbitrarily select a target data. Take the time when the target data is received as the time start, count the number of times the target data is used within the unit time, and mark it as the number of times it is used Uc. Divide the number of uses Uc by the unit time to get the data usage frequency; Obtain the data usage frequency of all target data under the target device, and perform average processing on the data usage frequency. Mark the obtained average result as the inertial usage rate f of the target data. The structure setting module is used to combine and analyze the data characteristic values ​​and inertial utilization rate of each target device to obtain the priority of the target device. Based on the priority, the storage level of the current target device is determined, where the storage level includes the first level, the second level and the third level. The spatial adjustment module is used to dynamically adjust the storage location of the data acquisition devices based on the storage level of each data acquisition device at the current time.

2. The information resource management system based on the Internet of Things according to claim 1, characterized in that, Methods for identifying the data properties of target data include: Acquire historically received data information and identify the device source of the data information transmission, where the device source refers to the data acquisition device corresponding to the data information; Arbitrarily select a data acquisition device and mark it as the target device. Based on the device source of each data information transmission, extract the historical data information transmitted by the target device and mark this data information as the target data. Then arrange the target data in chronological order to obtain the data sequence. Di, where i represents different target data and i∈[1,n], indicates that there are n target data in the data sequence. Using formula The data fluctuation value B is obtained, and Da is the mean of all target data in the data sequence; The data fluctuation value B is compared with the fluctuation threshold By. When B < By, the target data is marked as steady-state data; otherwise, when B ≥ By, the target data is marked as fluctuating data.

3. The information resource management system based on the Internet of Things according to claim 2, characterized in that, Once the target data transmitted by the data acquisition device is received by the system, the system will first perform data cleaning on the received data. Data cleaning includes handling missing values, handling outliers, and smoothing noisy data.

4. The information resource management system based on the Internet of Things according to claim 1, characterized in that, The methods for setting up information storage structures include: Set up storage tiers, which include a first tier, a second tier, and a third tier, with the storage priority as follows: first tier > second tier > third tier; Identify the data characteristics and inertial utilization rate of the target device at the current time, and then use the formula... Obtain the priority value Y of the target device at the current time, where T represents the data characteristic value of the target device at the current time. and All are proportionality coefficients, and ; The priority value Y of all data acquisition devices at the current time is obtained using the method described above. Then, the priority value Y is compared with the node thresholds k1 and k2. When Y≥k1, the storage location of the data transmitted by this data acquisition device is set as the first level. When k1>Y≥k2, the storage location of the data transmitted by this data acquisition device is set as the second level. When Y<k2, the storage location of the data transmitted by this data acquisition device is set as the third level, where k1>k2.

5. The information resource management system based on the Internet of Things according to claim 4, characterized in that, The first tier is local storage space, the second tier is edge storage space, and the third tier is cloud storage.

6. The information resource management system based on the Internet of Things according to claim 4, characterized in that, When the target data is fluctuating data, the data characteristic value of the target data is the composite curve. Then, the position of the target data transmitted in the current window series in this composite curve is identified, the specific value T of the data characteristic value is determined in the composite curve, and the value T is input into the priority value calculation formula to determine the priority value Y of the corresponding target device at this time.

Images