A blockchain-based anti-tampering system for marine environmental data

Abstract

This invention relates to the field of information security technology, specifically to a blockchain-based anti-tampering system for marine environmental data. The system includes a wave crest localization module, a waveform feature extraction module, a fingerprint generation module, a tag mapping module, and an on-chain storage module. In this invention, deep analysis of seawater multispectral light intensity waveform data is performed to extract wave crest curvature geometric indices and wavelength differences, constructing a set of morphological parameters. Component coupling weights are used to generate an on-chain fingerprint sequence reflecting the constraints of water quality components. Dynamic state tags are generated by mapping instantaneous flow velocity to tidal cycle flow velocity ranges. The fingerprint sequence and dynamic tags are encapsulated into blocks and subjected to fluid dynamic continuity verification. A distributed ledger consensus mechanism is used to lock the physical temporal relationship of the data, realizing a transformation from simple numerical recording to multi-dimensional logical verification of environmental characteristics, effectively avoiding the tampering risks associated with centralized storage.

Description

Technical Field

[0001] This invention relates to the field of information security technology, and in particular to a blockchain-based system for preventing tampering with marine environmental data. Background Technology

[0002] The field of information security technology involves the integrity protection, access control, and trusted verification of various types of digital information during the collection, transmission, storage, and sharing processes. This includes confirming the authenticity of data at its source, protecting against tampering in network links, maintaining consistency in distributed environments, and building data traceability under conditions involving multiple stakeholders.

[0003] Among them, the marine environmental data anti-tampering system refers to a system that collects data on marine water quality parameters, air-sea exchange indicators, ocean currents and tides by periodically reading values ​​through sensors at the acquisition end, encapsulating them into time series records, and then sending them to the central server via a wireless transmission link. The server uses checksum comparison, timestamp sorting, and log file recording to perform consistency verification and historical retention of the received data.

[0004] Current marine environmental monitoring mainly relies on sensors to collect data and then aggregate it to a central server via wireless links. This centralized storage architecture makes the data highly susceptible to single-point failure risks and malicious tampering threats under the control of administrators. Verification codes and log records can only verify the integrity of data transmission, but cannot deeply identify anomalies between the logical content of the data and the physical environmental characteristics. There is a lack of a correlation verification mechanism for water quality spectral morphology and hydrodynamic state, which makes it difficult to guarantee the original authenticity of monitoring records when facing complex network attacks or internal data anomalies. Once the storage medium is compromised or key logs are targeted for deletion, historical environmental data will lose its credibility and immutability, seriously weakening the evidentiary value of marine ecological tracing. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a blockchain-based anti-tampering system for marine environmental data.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a blockchain-based marine environmental data anti-tampering system includes:

[0007] The peak positioning module acquires multispectral light intensity waveform data of seawater light intensity as wavelength changes during the monitoring period, compares the light intensity waveforms of adjacent sampling points during the monitoring period, and marks the peak wavelength position.

[0008] The waveform feature extraction module locates the peak wavelength position in the entire multispectral light intensity waveform data, including the peak coordinates and the coordinates of the corresponding adjacent sampling points. It calculates the peak curvature geometric index and the peak wavelength difference and performs standardization processing on them respectively to construct a morphological parameter group.

[0009] The fingerprint generation module sets component coupling weights based on the peak curvature geometric index and peak wavelength difference in the morphological parameter group, and constructs an on-chain fingerprint sequence.

[0010] The label mapping module collects and compares the current seawater current velocity reading as the instantaneous ocean current velocity, and the tidal current amplitude range in the current velocity data of a specified tidal period to construct a dynamic state label.

[0011] The on-chain storage module uses the power status tag as header information to write data packets containing the on-chain fingerprint sequence to construct data blocks to be uploaded to the chain. The on-chain fingerprint sequence, which passes continuity verification and meets the consensus benchmark, is written into the distributed ledger to establish a tamper-proof marine environment data chain.

[0012] As a further aspect of the present invention, the wave crest wavelength position is specifically the horizontal axis scale corresponding to the local maximum value marked by comparing the multispectral light intensity waveform amplitude values ​​of adjacent sampling points within the monitoring period. The morphological parameter group includes the wave crest curvature geometric index calculated based on the wave crest point coordinates and the coordinates of adjacent sampling points, and the wave crest wavelength difference between adjacent wave crest wavelength positions. The on-chain fingerprint sequence is specifically a numerical sequence generated by performing a weighted operation on the morphological parameter group using component coupling weights. The dynamic state tag is specifically an identification number assigned according to the interval position of the instantaneous ocean current velocity within the tidal current amplitude range of the specified tidal cycle current velocity data. The marine environment tamper-proof data chain includes a data block to be uploaded to the chain encapsulated with the dynamic state tag and the on-chain fingerprint sequence, and a distributed ledger after continuity verification and consensus benchmark verification.

[0013] As a further aspect of the present invention, the peak positioning module includes:

[0014] The waveform data acquisition submodule monitors the seawater body within the current monitoring period, drives the multispectral sensor to acquire spectral signals that characterize the light intensity of water quality components as a function of wavelength, analyzes the light intensity amplitude and wavelength corresponding to each sampling point in the spectral signal, and establishes multispectral light intensity waveform data.

[0015] The adjacent amplitude comparison submodule traverses the multispectral light intensity waveform data corresponding to each sampling point, compares the light intensity amplitude of the current sampling point with the light intensity amplitudes of the adjacent sampling points to the left and right of the current sampling point, and filters out sampling points whose light intensity amplitude is greater than both the values ​​of the adjacent sampling points to the left and the adjacent sampling points to the right, generating local maximum points.

[0016] The wavelength position marking submodule locates the horizontal axis coordinates corresponding to each local maximum point in the waveform coordinate system, extracts the scale values ​​on the horizontal axis coordinates as feature representations of the identified wave peak location, and summarizes the horizontal axis coordinates corresponding to all local maximum points within the monitoring period to obtain the wave peak wavelength position.

[0017] As a further aspect of the present invention, the waveform feature extraction module includes:

[0018] The peak coordinate positioning submodule retrieves the sampling point index corresponding to the peak wavelength position value in the multispectral light intensity waveform data, locates the peak point coordinates and the coordinates of the adjacent sampling points on the left and right sides of the peak point at the sampling point index position, and constructs a left vector starting from the coordinates of the adjacent sampling point on the left and ending at the coordinates of the peak point, and a right vector starting from the coordinates of the peak point and ending at the coordinates of the adjacent sampling point on the right, respectively, to generate peak point position vector data.

[0019] The feature index calculation submodule performs a dot product operation on the left and right vectors corresponding to the same peak point in the peak point position vector data. Based on the dot product operation result, it calculates the cosine value of the angle between the two vectors as a peak curvature geometric index characterizing the sharpness of the peak. At the same time, it extracts the wavelength scale of two adjacent peak points on the horizontal axis of the waveform coordinate system, calculates the difference between the wavelength positions of adjacent peaks, obtains the peak wavelength difference value characterizing the peak interval feature, and summarizes it into a peak geometric difference index.

[0020] The morphological parameter group construction submodule obtains the maximum curvature and maximum wavelength interval within a specified observation period based on the peak geometric difference index. It calculates the standardized curvature parameter based on the peak curvature geometric index and the maximum curvature, calculates the normalized wavelength interval parameter based on the peak wavelength difference and the maximum wavelength interval, and integrates the standardized curvature parameter and the standardized wavelength interval parameter to construct the morphological parameter group.

[0021] As a further aspect of the present invention, the fingerprint generation module includes:

[0022] The difference comparison and statistics submodule calculates the absolute difference between any standardized curvature parameter or standardized wavelength interval parameter in the morphological parameter group and all other standardized curvature parameters and standardized wavelength interval parameters in the group, filters parameter pairs with an absolute difference less than a preset coupling threshold, and counts the number of coupling pairs.

[0023] The coupling weight setting submodule extracts the total number of all parameters within the morphological parameter group as the benchmark value for weight calculation based on the number of coupling pair counts. It calculates the ratio between the number of pair counts corresponding to each parameter and the total number of parameters, and establishes it as a quantization coefficient that characterizes the degree of correlation between parameters. Based on the quantization coefficient, it sets the weight value of each parameter in the morphological parameter group to obtain the component coupling weight.

[0024] The fingerprint sequence construction submodule combines the standardized curvature parameters and standardized wavelength interval parameters in the morphological parameter group with the corresponding component coupling weights, calculates the weighted value of each parameter, and accumulates the weighted values ​​of all parameters to reflect the multi-component constraint relationship of the water body and establishes an on-chain fingerprint sequence.

[0025] As a further aspect of the present invention, the tag mapping module includes:

[0026] The current velocity data acquisition submodule monitors the current seawater flow status and drives the current meter to acquire seawater velocity readings, representing the instantaneous current velocity that characterizes the current water flow intensity. At the same time, it acquires velocity data covering the entire specified tidal cycle, extracts the maximum and minimum velocity values ​​from the velocity data, calculates the numerical difference between the maximum and minimum velocity values, and obtains the tidal amplitude range.

[0027] The interval division coding submodule divides the tidal current amplitude range according to a preset step size interval value, constructs multiple continuously distributed velocity sub-intervals, assigns a unique numerical number to each velocity sub-interval in ascending order of numerical value, and establishes a velocity sub-interval index.

[0028] The dynamic label mapping submodule traverses the numerical range of each velocity sub-interval in the velocity sub-interval index, determines the velocity sub-interval into which the instantaneous ocean current velocity falls, extracts the numerical code corresponding to the target velocity sub-interval, maps the numerical code to an identifier that characterizes the dynamic state of the water body, and generates a dynamic state label.

[0029] As a further aspect of the present invention, the on-chain storage module includes:

[0030] The data block construction submodule writes the power status tag as header information into a data packet containing the on-chain fingerprint sequence, performs encapsulation processing on the data packet, and generates a data block to be uploaded to the chain.

[0031] The continuity verification submodule broadcasts the data block to be uploaded to the distributed node network, extracts the dynamic state tag in the current data block to be uploaded to the blockchain and the dynamic state tag recorded in the latest block in the blockchain ledger, calculates the difference between the two dynamic state tags, and removes blocks whose difference exceeds the preset fluid dynamics continuity threshold to obtain blocks that meet the continuity constraints.

[0032] The ledger is written to the storage submodule, which counts the frequency of the blocks that meet the continuity constraints in the network broadcast, establishes the dynamic state tag with the highest frequency as the consensus benchmark, selects data blocks that pass the continuity verification and whose dynamic state tags are consistent with the consensus benchmark, writes the target data blocks to be added to the chain into the distributed ledger to lock the physical time sequence constraint relationship, and establishes a marine environment tamper-proof data chain.

[0033] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0034] In this invention, by performing in-depth analysis of seawater multispectral light intensity waveform data, the geometric indices of wave crest curvature and wavelength differences are extracted and a set of morphological parameters is constructed. Composition coupling weights are used to generate an on-chain fingerprint sequence reflecting the constraint relationship of water quality components. Dynamic state tags are generated by mapping instantaneous flow velocity with tidal cycle flow velocity ranges. The fingerprint sequence and dynamic tags are encapsulated into blocks and fluid dynamic continuity verification is performed. A distributed ledger consensus mechanism is used to lock the physical temporal relationship of the data, realizing the transformation from simple numerical recording to multidimensional logical verification of environmental characteristics. This effectively avoids the risk of tampering caused by centralized storage and ensures the uniqueness of marine environmental data throughout the entire acquisition and transmission chain. Attached Figure Description

[0035] Figure 1 This is a system flowchart of the present invention;

[0036] Figure 2 This is a flowchart of the peak positioning module of the present invention;

[0037] Figure 3 This is a flowchart of the waveform feature extraction module of the present invention;

[0038] Figure 4 This is a flowchart of the fingerprint generation module of the present invention;

[0039] Figure 5 This is a flowchart of the label mapping module of the present invention;

[0040] Figure 6 This is a flowchart of the on-chain storage module of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0042] Please see Figure 1 A blockchain-based marine environmental data tamper-proof system includes:

[0043] The peak positioning module acquires multispectral light intensity waveform data of seawater light intensity as wavelength changes during the monitoring period, compares the light intensity waveforms of adjacent sampling points during the monitoring period, and marks the peak wavelength position.

[0044] The waveform feature extraction module locates the peak wavelength position in the entire multispectral light intensity waveform data, including the coordinates of the peak point and the corresponding adjacent sampling points. It calculates the peak curvature geometric index and the peak wavelength difference and performs standardization processing on them respectively to construct a set of morphological parameters.

[0045] The fingerprint generation module sets component coupling weights based on the peak curvature geometric index and peak wavelength difference in the morphological parameter group, and constructs an on-chain fingerprint sequence.

[0046] The label mapping module collects and compares the current seawater current velocity reading as the instantaneous ocean current velocity, and the tidal current amplitude range in the current velocity data of a specified tidal period to construct a dynamic state label.

[0047] The on-chain storage module writes the power status tag as header information into a data packet with an on-chain fingerprint sequence to construct a data block to be uploaded to the chain. The on-chain fingerprint sequence, which is verified for continuity and conforms to the consensus benchmark, is written into the distributed ledger to establish a tamper-proof data chain for the marine environment.

[0048] The peak wavelength position is specifically the horizontal axis scale corresponding to the local maximum value marked by comparing the multispectral light intensity waveform amplitude values ​​of adjacent sampling points within the monitoring period. The morphological parameter group includes the peak curvature geometric index calculated based on the peak point coordinates and the coordinates of adjacent sampling points, and the peak wavelength difference between adjacent peak wavelength positions. The on-chain fingerprint sequence is specifically a numerical sequence generated by performing a weighted operation on the morphological parameter group using component coupling weights. The dynamic state tag is specifically an identification number assigned based on the interval position of the instantaneous ocean current velocity within the tidal current amplitude range of the specified tidal cycle current data. The marine environmental tamper-proof data chain includes a data block to be uploaded to the chain encapsulated with the dynamic state tag and the on-chain fingerprint sequence, and a distributed ledger after continuity verification and consensus benchmark verification.

[0049] Please see Figure 2 The peak positioning module includes:

[0050] The waveform data acquisition submodule monitors the seawater body within the current monitoring period, drives the multispectral sensor to acquire spectral signals that characterize the light intensity of water quality components as a function of wavelength, analyzes the light intensity amplitude and wavelength corresponding to each sampling point in the spectral signal, and establishes multispectral light intensity waveform data.

[0051] The monitoring task for the seawater body within the current monitoring period is initiated by sending a data acquisition command to a multispectral sensor deployed underwater. Upon receiving the command, the photoelectric conversion element inside the multispectral sensor activates the exposure program. The sensor scans a preset visible to near-infrared wavelength range, with the scanning range set to 400 nm to 900 nm and the sampling interval set to 1 nm, acquiring a total of 501 independent wavelength sampling channels. At the wavelength position corresponding to each sampling channel, the sensor measures the radiant energy of the incident light and transmits the analog electrical signal to an analog-to-digital converter (ADC). The ADC uses a 12-bit precision quantization standard to convert the analog electrical signal into a digital quantization value between 0 and 4095, which represents the light intensity amplitude at the current wavelength point. The waveform data acquisition submodule sequentially reads the digital quantization values ​​corresponding to the 501 wavelength channels, using the wavelength values ​​as horizontal indices and the light intensity amplitude digital quantization values ​​as vertical data, constructing a raw spectral data sequence composed of pairs of wavelengths and light intensity amplitudes. For example, the light intensity amplitude read at a wavelength of 450 nm is 1200, the light intensity amplitude read at a wavelength of 451 nm is 1215, and so on, to complete the establishment of multispectral light intensity waveform data throughout the entire monitoring period.

[0052] The adjacent amplitude comparison submodule iterates through the multispectral light intensity waveform data corresponding to each sampling point, compares the light intensity amplitude of the current sampling point with the light intensity amplitudes of the adjacent sampling points to the left and right of the current sampling point, and filters out the sampling points whose light intensity amplitude is greater than both the values ​​of the adjacent sampling points to the left and the adjacent sampling points to the right, generating local maximum points.

[0053] The system reads the established multispectral light intensity waveform data and initializes a data pointer. This pointer starts from the second sampling point in the data sequence and ends at the second-to-last sampling point, traversing each sampling point sequentially. During the traversal, the light intensity amplitude value of the sampling point currently pointed to by the pointer is extracted. Simultaneously, the light intensity amplitude values ​​of the adjacent sampling point to the left of the previous index position and the adjacent sampling point to the right of the next index position are retrieved. The numerical comparison logic is executed: first, it is determined whether the light intensity amplitude of the current sampling point is strictly greater than the light intensity amplitude of the adjacent sampling point to the left. If the result is yes, then it continues to determine whether the light intensity amplitude of the current sampling point is strictly greater than the light intensity amplitude of the adjacent sampling point to the right. There are four possible outcomes: First, if the current light intensity is greater than both the left and right light intensity, the point is identified as a local maximum. Second, if the current light intensity is less than or equal to the left light intensity but greater than the right light intensity, the point is not a local maximum. Third, if the current light intensity is greater than the left light intensity but less than or equal to the right light intensity, the point is not a local maximum. Fourth, if the current light intensity is less than or equal to both the left and right light intensity, the point is not a local maximum. Only when the first condition is met is the sampling point identified as a local maximum. By traversing the entire sequence, all sampling point indices that satisfy the bilateral descent condition are selected, generating a list containing the indices of all local maximum points. For example, if the light intensity amplitude at index 100 is 1500, at index 99 it is 1400, and at index 101 it is 1450, then we determine whether 1500 is greater than 1400, and the result is yes; we determine whether 1500 is greater than 1450, and the result is yes, then index 100 is confirmed as a local maximum point.

[0054] The wavelength position marking submodule locates the horizontal axis coordinates corresponding to each local maximum point in the waveform coordinate system, extracts the scale values ​​on the horizontal axis coordinates as the feature representation of the identified wave peak location, and summarizes the horizontal axis coordinates corresponding to all local maximum points within the monitoring period to obtain the wave peak wavelength position.

[0055] The specific process of constructing the waveform coordinate system is as follows:

[0056] Analyze the spectral signals collected by the multispectral sensor and extract the wavelength values ​​and light intensity amplitude values ​​corresponding to all sampling points in the spectral signals;

[0057] The minimum and maximum values ​​of the selected wavelengths are set as the start and end limits of the horizontal axis, respectively, defining the horizontal axis as the dimension representing the wavelength distribution.

[0058] The minimum and maximum values ​​of the light intensity amplitude values ​​are filtered out and set as the bottom and top limits of the vertical axis, respectively, to define the vertical axis as the dimension representing the light intensity response.

[0059] Establish the perpendicular orthogonal relationship between the horizontal axis and the vertical axis at the initial boundary to generate a two-dimensional planar projection space;

[0060] Based on the sampling frequency parameters of the multispectral sensor, the wavelength difference between adjacent sampling points is calculated, and the wavelength difference is set as the scale step of the horizontal axis. Wavelength scale is generated within the range from the start limit to the end limit of the horizontal axis.

[0061] Based on the analog-to-digital conversion bit depth parameter of the multispectral sensor, the minimum resolvable light intensity amplitude increment is calculated, and the light intensity amplitude increment is set as the scale step of the vertical axis. Light intensity scale is generated within the range from the bottom limit to the top limit of the vertical axis.

[0062] By mapping the wavelength scale and light intensity scale onto a two-dimensional planar projection space, a waveform coordinate system is obtained;

[0063] First, a waveform coordinate system is constructed by analyzing the multispectral light intensity waveform data and extracting the wavelength and intensity amplitude sets of all sampling points. The wavelength set is iterated through to identify the minimum wavelength value of 400 nm and the maximum wavelength value of 900 nm. 400 nm is set as the starting boundary of the horizontal axis, and 900 nm as the ending boundary, with the horizontal axis representing the wavelength distribution dimension. Simultaneously, the intensity amplitude set is iterated through to identify the minimum intensity value of 0 and the maximum intensity value of 4095, which are set as the bottom and top boundaries of the vertical axis, respectively, representing the intensity response dimension. A perpendicular and orthogonal two-dimensional planar projection space is established with the intersection of the horizontal axis's starting boundary and the vertical axis's bottom boundary as the origin. Based on the 1 nm sampling frequency parameter of the multispectral sensor, the wavelength difference between adjacent sampling points is calculated to be 1 nm, and this is set as the scale step of the horizontal axis. Wavelength scales are generated at 1 nm intervals within the horizontal axis's starting boundary of 400 nm to its ending boundary of 900 nm. Based on the 12-bit analog-to-digital conversion bit depth parameter of the multispectral sensor, the minimum resolvable light intensity amplitude increment is calculated to be 1 unit, which is set as the scale step size of the vertical axis. Light intensity scales are generated at 1-unit intervals within the vertical axis, from the bottom boundary 0 to the top boundary 4095. The wavelength scale and light intensity scale are mapped onto a two-dimensional planar projection space to obtain a waveform coordinate system. In this waveform coordinate system, the local maximum point index output by the adjacent amplitude comparison submodule is read, and the wavelength scale value corresponding to each index is located on the horizontal axis. For example, for the local maximum point corresponding to index 100, its horizontal axis coordinate is read as 500 nanometers. This scale value is extracted as a feature representation of the peak location, and the same horizontal axis coordinate extraction operation is performed on all identified local maximum points within the monitoring period. Finally, a set of peak wavelength position sequences containing multiple wavelength values ​​is obtained.

[0064] Please see Figure 3 The waveform feature extraction module includes:

[0065] The peak coordinate positioning submodule retrieves the sampling point index corresponding to the peak wavelength position value in the multispectral light intensity waveform data, locates the peak point coordinates at the sampling point index position and the coordinates of the adjacent sampling points on the left and right sides of the peak point, and constructs a left vector starting from the coordinates of the adjacent sampling point on the left and ending at the coordinates of the peak point, and a right vector starting from the coordinates of the peak point and ending at the coordinates of the adjacent sampling point on the right, respectively, to generate peak point position vector data.

[0066] Based on the numerical values ​​in the wavelength position sequence of the wave crest, such as 500 nm, a reverse search is performed in the multispectral intensity waveform data to locate the sampling point index value that precisely matches this wavelength value, such as index 100. After locating the target wave crest sampling point index, the complete coordinate data of this index in the waveform coordinate system is read, including the wavelength horizontal coordinate and the intensity vertical coordinate, such as horizontal coordinate 500 and vertical coordinate 1500, which are recorded as the wave crest coordinates. Subsequently, the index value is decremented by one to obtain the complete coordinate data of the adjacent sampling point on the left, such as the horizontal coordinate 499 and vertical coordinate 1400 corresponding to index 99; the index value is incremented by one to obtain the complete coordinate data of the adjacent sampling point on the right, such as the horizontal coordinate 501 and vertical coordinate 1450 corresponding to index 101. Based on the definition of planar geometric vectors, a left vector is constructed with the coordinates of the adjacent sampling point on the left (499, 1400) as the starting point and the coordinates of the wave crest (500, 1500) as the ending point. The horizontal component of the left vector is calculated as the horizontal coordinate of the ending point minus the horizontal coordinate of the starting point, i.e. The ordinate component of the left vector is calculated by subtracting the ordinate of the starting point from the ordinate of the ending point. The left vector is represented as Similarly, starting from the peak coordinates (500, 1500) and ending at the adjacent sampling point coordinates (501, 1450), a right vector is constructed. The x-axis component of the right vector is calculated by subtracting the starting x-coordinate from the ending x-coordinate. The ordinate component of the right vector is calculated by subtracting the ordinate of the starting point from the ordinate of the ending point. The right vector is represented as Perform this vector construction process on all peak points to generate a set of peak point position vector data containing paired vector data.

[0067] The feature index calculation submodule performs a dot product operation on the left and right vectors corresponding to the same peak point in the peak point position vector data. Based on the dot product operation result, it calculates the cosine value of the angle between the two vectors, which serves as the peak curvature geometric index characterizing the sharpness of the peak. At the same time, it extracts the wavelength scale of two adjacent peak points on the horizontal axis of the waveform coordinate system, calculates the difference between the wavelength positions of adjacent peaks, obtains the peak wavelength difference value characterizing the peak interval feature, and summarizes it into a peak geometric difference index.

[0068] For each peak point, its corresponding left and right vector data are extracted, and the peak curvature geometric index is calculated using vector cosine symmetry correction. The algorithm formula is as follows: ,in, The geometric index of crest curvature; Let x and y be the x-axis and y-axis components of the left vector, respectively. In this example, ; Let x and y be the x-axis components and y-axis components of the right vector, respectively. In this example, ; To prevent singularity robustness, this coefficient is set based on the square of the sensor's minimum resolvable signal amplitude. This prevents the denominator from becoming singular due to a vector magnitude of 0. For example, if the sensor's minimum amplitude is 0.001, then this coefficient is set to... ; This is an asymmetric penalty factor, set based on the reciprocal of the average modulus of historical standard peak shapes. It is used to balance the influence of the modulus magnitude on exponential decay. For example, if the statistically obtained average modulus is 200, then this factor is set... .

[0069] The specific numerical calculation process is as follows: First, calculate the dot product. Next, calculate the square of the magnitude of the left vector. The left vector has a magnitude of ; Calculate the square of the magnitude of the right vector. The right vector has a magnitude of Next, calculate the product of the moduli in the denominator. Plus a robustness coefficient The denominator is . The value of the cosine term is . Then calculate the exponential correction term; the absolute value of the modulus difference is... The index part is The value of the exponential function is Final calculation The closer the result is to -1 and the higher the symmetry, the sharper and more regular the peak shape. Simultaneously, the wavelength difference between adjacent peaks is calculated by subtracting the current peak wavelength of 500 nm from the next peak wavelength of 530 nm, yielding 30 nm. The results are then compiled into a dataset containing both peak curvature geometric indices and peak geometric difference indices.

[0070] The morphological parameter group construction submodule obtains the maximum curvature and maximum wavelength interval within a specified observation period based on the peak geometric difference index. It calculates the standardized curvature parameter based on the peak curvature geometric index and the maximum curvature, and calculates the normalized wavelength interval parameter based on the peak wavelength difference and the maximum wavelength interval. Finally, it integrates the standardized curvature parameter and the standardized wavelength interval parameter to construct the morphological parameter group.

[0071] First, all accumulated peak curvature geometric indices within the specified observation period are scanned, and the largest value is selected as the maximum curvature reference value. Simultaneously, all peak wavelength differences are scanned, and the largest value is selected as the maximum wavelength interval reference value. For example, if the peak curvature geometric index sequence obtained within the observation period is... The maximum curvature reference value is -0.78 (the logic of taking the largest value is to select the smoothest peak closest to 0 as the normalization benchmark to highlight the relative characteristics of sharp peaks). To eliminate the influence of different dimensions, a standardization process is performed: read the curvature geometric index of each peak, for example, the first peak curvature geometric index is -0.99, take the absolute value of 0.99 and divide it by the absolute value of the maximum curvature reference value of 0.78 to calculate... This refers to the standardized curvature parameter. The wavelength difference between each peak is read; for example, the wavelength difference of the first peak is 30 nanometers. This is divided by the maximum wavelength interval reference value of 60 nanometers to calculate the value. This refers to the standardized wavelength interval parameter. The values ​​of both the standardized curvature parameter and the standardized wavelength interval parameter range from 0 to 1 (Note: For ease of explanation of the normalization principle, in actual calculations, if the numerator is greater than the denominator, the value can be greater than 1, depending on the selection logic of the normalization benchmark; in this embodiment, the relative proportion is retained). All calculated standardized curvature parameters and standardized wavelength interval parameters are arranged and integrated according to their order of appearance in the spectrum to construct a morphological parameter set describing the current spectral morphological characteristics of the water body.

[0072] Please see Figure 4 The fingerprint generation module includes:

[0073] The difference comparison and statistics submodule calculates the absolute difference between any standardized curvature parameter or standardized wavelength interval parameter in the morphological parameter group and all other standardized curvature parameters and standardized wavelength interval parameters in the group, filters parameter pairs with an absolute difference less than a preset coupling threshold, and counts the number of coupling pairs.

[0074] The system receives a set of morphological parameters and sets a coupling threshold. This coupling threshold is based on historical spectral characteristic data of the water composition of this specific water body. It collects 10,000 sets of spectral parameter difference distribution data for similar water bodies, calculates the average absolute difference between all parameter pairs, and sets 10% of this average as the preset coupling threshold. For example, if the average absolute difference calculated through historical statistics is 0.5, then the coupling threshold is set to... The module initiates a double-loop traversal procedure. For any target parameter within the parameter group, such as the first standardized curvature parameter 1.269, it compares its value with all the remaining parameters in the group. In each comparison, the difference between the target parameter and the comparison parameter is calculated, and the absolute value of this difference is taken. Then, it is determined whether this absolute value is less than a preset coupling threshold of 0.05. If the absolute value is less than 0.05, a strong coupling is determined between the two parameters, and the coupling pair count is incremented by one. If the absolute value is greater than or equal to 0.05, it is not counted as a pair. After the module completes the comparison of the current target parameter with all other parameters in the group, the final count value is recorded as the number of coupling pairs corresponding to that parameter. The above statistical process is performed for each parameter in the parameter group, generating a pair count list corresponding to the parameter.

[0075] The coupling weight setting submodule extracts the total number of all parameters within the morphological parameter group based on the number of coupling pair counts as the benchmark value for weight calculation. It calculates the ratio between the number of pair counts for each parameter and the total number of parameters, and establishes it as a quantization coefficient that characterizes the degree of correlation between parameters. Based on the quantization coefficient, it sets the weight value of each parameter in the morphological parameter group to obtain the component coupling weight.

[0076] First, count the total number of parameters in the morphological parameter group. For example, if there are 10 parameters in the group, define this as the baseline denominator for weight calculation, with a value of 10. Then, iterate through the number of coupling pairs for each parameter. For example, if a parameter has 4 coupling pairs, perform a division operation, dividing the number of coupling pairs for that parameter by the total number of parameters. The quantization coefficient is obtained. The value of the quantization coefficient ranges from 0 to 1. This quantization coefficient is directly set as the weight value of the corresponding morphological parameter. For example, the weight value of this parameter is set to 0.4. By performing this calculation on each parameter, a set of component coupling weights corresponding to the morphological parameter is obtained.

[0077] The fingerprint sequence construction submodule combines the standardized curvature parameters and standardized wavelength interval parameters in the morphological parameter group with the corresponding component coupling weights, calculates the weighted correction value of each parameter, and accumulates the weighted values ​​of all parameters to reflect the multi-component constraint relationship of the water body and establishes an on-chain fingerprint sequence.

[0078] Read each normalized curvature parameter or normalized wavelength interval parameter from the morphological parameter group and match its corresponding component coupling weight. For example, a normalized curvature parameter of 0.8 corresponds to a component coupling weight of 0.4. Multiply the parameter value by its corresponding weight value, i.e. This yields the weighted adjusted values. After completing the weighted calculation for all parameters, all weighted adjusted values ​​are summed. For example, if the parameter set contains 10 weighted adjusted values... Then the summation is calculated as follows: This accumulated sum, incorporating spectral morphological characteristics and their internal correlation structure, comprehensively reflects the constraint relationships among multiple components of the water body. This accumulated sum, 4.16, is converted into a binary data stream and combined with timestamp information to establish a unique on-chain fingerprint sequence.

[0079] Please see Figure 5 The tag mapping module includes:

[0080] The current velocity data acquisition submodule monitors the current seawater flow status and drives the current meter to acquire seawater velocity readings, representing the instantaneous current velocity that characterizes the current water flow intensity. At the same time, it acquires velocity data covering the entire specified tidal cycle, extracts the maximum and minimum velocity values ​​from the velocity data, calculates the numerical difference between the maximum and minimum velocity values, and obtains the tidal amplitude range.

[0081] The acoustic Doppler current meter deployed at the monitoring point is activated to emit acoustic pulses in real time and receive the echoes reflected from scattering objects in the water. Using the Doppler frequency shift principle, the current seawater velocity reading is calculated to two decimal places in meters per second (m / s), serving as the instantaneous current velocity; for example, the current instantaneous current velocity is 1.25 m / s. Simultaneously, local memory is accessed to retrieve a historical current velocity data sequence covering the most recent complete specified tidal cycle (12 hours and 25 minutes). This historical sequence is iterated through, and a numerical comparison algorithm is used to filter out the maximum and minimum current velocity values ​​within the cycle. For example, the maximum current velocity is 2.00 m / s, and the minimum current velocity is 0.50 m / s. A subtraction operation is performed, subtracting the minimum current velocity value from the maximum current velocity value. Meters per second, calculate the amplitude of the velocity change within the tidal cycle, i.e., the range of tidal amplitude.

[0082] The interval division coding submodule divides the tidal current amplitude range according to the preset step size interval value, constructs multiple continuously distributed velocity sub-intervals, assigns a unique number to each velocity sub-interval in ascending order of value, and establishes a velocity sub-interval index.

[0083] The calculated tidal current amplitude range is discretized based on a preset step interval. The preset step interval is set based on historical current velocity statistics for the sea area. The standard deviation of all current velocity monitoring data for the past year is collected, and 10% of this standard deviation is set as the step interval. For example, if the historical current velocity standard deviation is 1.00 m / s, then the step interval is set to... Meters per second. Taking the calculated tidal current amplitude range of 0.50 m / s to 2.00 m / s as an example, starting from the minimum velocity value of 0.50 m / s, a node is divided every 0.10 m / s, cutting the entire tidal current amplitude range into multiple continuously distributed velocity sub-intervals. The first sub-interval is... The second sub-interval is until the last sub-interval Following the increasing order of flow velocity values, starting from number 1, each velocity sub-interval is assigned a unique integer number. The first sub-interval... Number 1, the second Numbered from 2, until the last sub-interval Numbered 15. Store these interval ranges and their corresponding numerical numbers in a lookup table to create a speed sub-interval index.

[0084] The dynamic label mapping submodule iterates through the numerical range of each velocity sub-interval in the velocity sub-interval index, determines the velocity sub-interval into which the instantaneous ocean current velocity falls, extracts the numerical code corresponding to the target velocity sub-interval, maps the numerical code to an identifier that characterizes the dynamic state of the water body, and generates a dynamic state label.

[0085] Get the instantaneous ocean current speed at the current moment, for example, 1.25 meters per second. Iterate through the velocity sub-interval indices, checking the value range of each sub-interval. Determine if the instantaneous ocean current speed of 1.25 meters per second falls within each velocity sub-interval. For example, check the sub-interval... ,judge and The result is yes, meaning the speed is determined to fall within the target speed sub-interval. Extract the target speed sub-interval. The corresponding numerical code is 8. This code is converted to a fixed-length hexadecimal string "08" and used as an identifier representing the current dynamic state of the water body, generating a dynamic state label. This label indicates that the water flow velocity is at a moderate to high level.

[0086] Please see Figure 6 The on-chain storage module includes:

[0087] The data block construction submodule writes the power status tag as header information into the data packet with the on-chain fingerprint sequence, performs encapsulation processing on the data packet, and generates the data block to be uploaded to the chain.

[0088] A new data buffer is allocated in memory. First, the power status tag "08" generated by the power tag mapping submodule is written to the header field of the buffer. Then, the on-chain fingerprint sequence generated by the fingerprint sequence construction submodule (e.g., the binary data stream obtained by summing and converting to 4.16) is written to the data payload portion of the buffer. The data in the buffer is encapsulated by adding standard start, stop, and length check bits. The cyclic redundancy check (CRC) code for the entire data packet is calculated and appended to the end of the packet. The encapsulated data structure then becomes the data block to be uploaded to the blockchain.

[0089] The continuity verification submodule broadcasts the data blocks to be uploaded to the distributed node network, extracts the dynamic state tag from the current data block to be uploaded to the blockchain and the dynamic state tag recorded in the latest block in the blockchain ledger, calculates the difference between the two dynamic state tags, and removes blocks whose difference exceeds the preset fluid dynamics continuity threshold to obtain blocks that meet the continuity constraints.

[0090] The process of broadcasting data blocks to be uploaded to the blockchain to the distributed node network is as follows:

[0091] The system retrieves the identifiers of neighboring nodes that are currently online and maintain an active connection with the sending node by calling the P2P network routing table stored locally.

[0092] Perform binary serialization encoding on the data block to be uploaded to the blockchain to construct a network transmission data packet with complete block content and cyclic redundancy check code;

[0093] Write a time-to-live parameter to limit the number of times the data packet is forwarded and a list of node signatures to record the data propagation path into the header field of the network transmission data packet;

[0094] Establish a network communication socket corresponding to the neighbor node identification information, and send network transmission data packets to each directly adjacent node in the distributed node network through an asynchronous parallel transmission channel;

[0095] Start the data packet transmission monitoring mechanism and wait to receive the data acknowledgment frame returned by the directly adjacent node within the preset response time window;

[0096] For directly adjacent nodes that do not return a data acknowledgment frame after the response time window has expired, perform packet retransmission operations until the number of retransmissions reaches the preset maximum retransmission threshold;

[0097] Configure data flooding and forwarding logic based on the Gossip protocol to drive the directly adjacent nodes that receive network transmission data packets to perform a decrement operation on the time-to-live parameter, and forward the network transmission data packets to the secondary adjacent nodes when the operation result is greater than zero;

[0098] First, the broadcast procedure is executed, calling the locally stored P2P network routing table to retrieve the identifiers of currently online and actively connected neighboring nodes. The data block to be uploaded to the blockchain is binary serialized and encoded, with a time-to-live (TTL) parameter (set to 10) and a list of node signatures written in the header. A network communication socket is established, and the data packet is sent to all directly adjacent nodes via an asynchronous parallel transmission channel. Upon receiving the data packet, the adjacent node decrements the TTL parameter by one based on the Gossip protocol logic; if the result is greater than 0, it forwards the packet to the next next-level neighboring node. Simultaneously with the broadcast, the dynamic state tag "08" is extracted from the current block to be uploaded to the blockchain, and the dynamic state tag recorded in the latest block of the blockchain ledger is queried. For example, the dynamic state tag of the latest block is "07," corresponding to a flow velocity of 1.15 meters per second. The preset fluid dynamic continuity threshold at the current moment is calculated using the "fluid dynamics turbulence boundary constraint algorithm," with the following formula: ,in, The preset fluid dynamics continuity threshold is used; The confidence interval safety factor is set based on the statistical law of normal distribution. For example, a value of 2.0 is chosen to correspond to a 95% confidence level, which means that it covers the vast majority of natural fluctuations. The historical average maximum acceleration amplitude is obtained by statistically analyzing the absolute average of the rate of change of current velocity per minute in this sea area over the past year; for example, the statistical value is 0.002 meters per square second. The time interval between the current block and the previous block, for example, 300 seconds; The turbulence diffusion coefficient is set based on the variance growth rate of historical current velocity data in this sea area, reflecting the uncertainty of the random walk of current velocity. For example, a value of 0.01 m / s to the power of 1.5 is obtained through linear regression (here, it is only used as a coefficient value and is taken as 0.01).

[0099] The specific numerical calculation process is as follows: First, calculate the linear drift term, Next, to calculate the random diffusion term, first calculate the square root of the time interval. Then multiply by the diffusion coefficient Finally, the two items are added together to obtain the threshold. Meters per second. Then, the difference between the flow rate values ​​corresponding to the two labels is calculated, i.e. Meters per second. Determine if the calculated label difference of 0.10 meters per second exceeds the calculated threshold of 1.3732 meters per second. There are two possibilities: First, the difference is less than or equal to 1.3732 meters per second, indicating that the flow velocity change conforms to physical laws; second, the difference exceeds 1.3732 meters per second, indicating data anomaly. Since 0.10 ≤ 1.3732, it falls under the first possibility, and this block is retained and marked as a block that meets the continuity constraint.

[0100] The ledger is written to the storage submodule, which counts the frequency of blocks that meet the continuity constraints in the network broadcast, establishes the dynamic state tag with the highest frequency as the consensus benchmark, selects data blocks that pass the continuity verification and whose dynamic state tags are consistent with the consensus benchmark, writes the target data blocks to be added to the chain into the distributed ledger to lock the physical time sequence constraint relationship, and establishes a marine environment tamper-proof data chain.

[0101] The system monitors verification feedback for blocks that meet continuity constraints in the distributed network. A consensus time window is set, during which the power state tags broadcast by each node for the same time slice are counted. For example, within the consensus window, the tag "08" appears 70% of the time, and the tag "07" appears 30%. Comparing the frequencies, 70% is greater than 30%, so the tag "08," with the highest frequency, is established as the network-wide consensus benchmark. Subsequently, the locally selected blocks that meet continuity constraints are checked again to determine if their power state tags are completely consistent with the network-wide consensus benchmark "08." There are two possibilities: first, the power state tag is consistent with the consensus benchmark; second, the power state tag is inconsistent with the consensus benchmark. Only when the continuity check passes and the tag is consistent with the consensus benchmark—for example, if the tag of the current block meeting the continuity constraint is "08"—meeting the first possibility, is the block confirmed as the target data block to be added to the chain. The underlying blockchain interface is invoked, and a hash algorithm is used to connect the block with the previous blockchain, write it into the storage space of the distributed ledger, and broadcast a confirmation message to the entire network to complete the establishment of the marine environmental tamper-proof data chain.

[0102] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A blockchain-based tamper-proof system for marine environmental data, characterized in that, The system includes: The peak positioning module acquires multispectral light intensity waveform data of seawater light intensity as wavelength changes during the monitoring period, compares the light intensity waveforms of adjacent sampling points during the monitoring period, and marks the peak wavelength position. The waveform feature extraction module locates the peak wavelength position in the entire multispectral light intensity waveform data, including the peak coordinates and the coordinates of the corresponding adjacent sampling points. It calculates the peak curvature geometric index and the peak wavelength difference and performs standardization processing on them respectively to construct a morphological parameter group. For the geometric index of crest curvature, the formula is used: ； in, The geometric index of crest curvature; These are the horizontal and vertical components of the left vector, respectively. These are the x-axis and y-axis components of the right vector, respectively. To prevent singular robustness coefficients, It is an asymmetric penalty factor; The fingerprint generation module sets component coupling weights based on the peak curvature geometric index and peak wavelength difference in the morphological parameter group, and constructs an on-chain fingerprint sequence. The fingerprint generation module includes: The difference comparison and statistics submodule calculates the absolute difference between any standardized curvature parameter or standardized wavelength interval parameter in the morphological parameter group and all other standardized curvature parameters and standardized wavelength interval parameters in the group, filters parameter pairs with an absolute difference less than a preset coupling threshold, and counts the number of coupling pairs. The coupling weight setting submodule extracts the total number of all parameters within the morphological parameter group as the benchmark value for weight calculation based on the number of coupling pair counts. It calculates the ratio between the number of pair counts corresponding to each parameter and the total number of parameters, and establishes it as a quantization coefficient that characterizes the degree of correlation between parameters. Based on the quantization coefficient, it sets the weight value of each parameter in the morphological parameter group to obtain the component coupling weight. The fingerprint sequence construction submodule combines the standardized curvature parameters and standardized wavelength interval parameters in the morphological parameter group with the corresponding component coupling weights, calculates the weighted correction value of each parameter, and accumulates the weighted values ​​of all parameters to reflect the multi-component constraint relationship of the water body and establishes an on-chain fingerprint sequence. The label mapping module collects and compares the current seawater current velocity reading as the instantaneous ocean current velocity, and the tidal current amplitude range in the current velocity data of a specified tidal period to construct a dynamic state label. The on-chain storage module writes the dynamic state tag as header information into a data packet containing the on-chain fingerprint sequence, performs encapsulation processing on the data packet, constructs a data block to be uploaded to the chain, broadcasts the data block to be uploaded to the distributed node network, extracts the dynamic state tag in the current data block to be uploaded to the chain and the dynamic state tag recorded in the latest block in the blockchain ledger, calculates the difference between the two dynamic state tags, removes blocks whose difference exceeds a preset fluid dynamics continuity threshold, obtains blocks that meet the continuity constraint, counts the frequency of the blocks that meet the continuity constraint in the network broadcast, establishes the dynamic state tag with the highest frequency as the consensus benchmark, selects data blocks to be uploaded to the chain that simultaneously pass the continuity check and whose dynamic state tag is consistent with the consensus benchmark, writes the target data block to be uploaded to the distributed ledger, and establishes a marine environment tamper-proof data chain.

2. The blockchain-based marine environmental data tamper-proof system according to claim 1, characterized in that, The wave crest wavelength position is specifically the horizontal axis scale corresponding to the local maximum value marked by comparing the multispectral light intensity waveform amplitude values ​​of adjacent sampling points within the monitoring period. The morphological parameter group includes the wave crest curvature geometric index calculated based on the wave crest point coordinates and the coordinates of adjacent sampling points, and the wave crest wavelength difference between adjacent wave crest wavelength positions. The on-chain fingerprint sequence is specifically a numerical sequence generated by performing a weighted operation on the morphological parameter group using component coupling weights. The dynamic state tag is specifically an identification number assigned according to the interval position of the instantaneous ocean current velocity within the tidal current amplitude range of the specified tidal cycle current velocity data. The marine environmental tamper-proof data chain includes a data block to be uploaded to the chain encapsulated with the dynamic state tag and the on-chain fingerprint sequence, and a distributed ledger after continuity verification and consensus benchmark verification.

3. The blockchain-based marine environmental data anti-tampering system according to claim 1, characterized in that, The peak location module includes: The waveform data acquisition submodule monitors the seawater body within the current monitoring period, drives the multispectral sensor to acquire spectral signals that characterize the light intensity of water quality components as a function of wavelength, analyzes the light intensity amplitude and wavelength corresponding to each sampling point in the spectral signal, and establishes multispectral light intensity waveform data. The adjacent amplitude comparison submodule traverses the multispectral light intensity waveform data corresponding to each sampling point, compares the light intensity amplitude of the current sampling point with the light intensity amplitudes of the adjacent sampling points to the left and right of the current sampling point, and filters out sampling points whose light intensity amplitude is greater than both the values ​​of the adjacent sampling points to the left and the adjacent sampling points to the right, generating local maximum points. The wavelength position marking submodule locates the horizontal axis coordinates corresponding to each local maximum point in the waveform coordinate system, extracts the scale values ​​on the horizontal axis coordinates as feature representations of the identified wave peak location, and summarizes the horizontal axis coordinates corresponding to all local maximum points within the monitoring period to obtain the wave peak wavelength position.

4. The blockchain-based marine environmental data anti-tampering system according to claim 3, characterized in that, The specific process of constructing the waveform coordinate system is as follows: Analyze the spectral signals collected by the multispectral sensor and extract the wavelength values ​​and light intensity amplitude values ​​corresponding to all sampling points in the spectral signals; The minimum and maximum values ​​of the selected wavelengths are set as the start and end limits of the horizontal axis, respectively, defining the horizontal axis as the dimension representing the wavelength distribution. The minimum and maximum values ​​of the light intensity amplitude values ​​are filtered out and set as the bottom and top limits of the vertical axis, respectively, to define the vertical axis as the dimension representing the light intensity response. Establish the perpendicular orthogonal relationship between the horizontal axis and the vertical axis at the initial boundary to generate a two-dimensional planar projection space; Based on the sampling frequency parameters of the multispectral sensor, the wavelength difference between adjacent sampling points is calculated, and the wavelength difference is set as the scale step of the horizontal axis. Wavelength scale is generated within the range from the start limit to the end limit of the horizontal axis. Based on the analog-to-digital conversion bit depth parameter of the multispectral sensor, the minimum resolvable light intensity amplitude increment is calculated, and the light intensity amplitude increment is set as the scale step of the vertical axis. Light intensity scale is generated within the range from the bottom limit to the top limit of the vertical axis. By mapping the wavelength scale and light intensity scale onto a two-dimensional plane projection space, a waveform coordinate system is obtained.

5. The blockchain-based marine environmental data tamper-proof system according to claim 3, characterized in that, The waveform feature extraction module includes: The peak coordinate positioning submodule retrieves the sampling point index corresponding to the peak wavelength position value in the multispectral light intensity waveform data, locates the peak point coordinates and the coordinates of the adjacent sampling points on the left and right sides of the peak point at the sampling point index position, and constructs a left vector starting from the coordinates of the adjacent sampling point on the left and ending at the coordinates of the peak point, and a right vector starting from the coordinates of the peak point and ending at the coordinates of the adjacent sampling point on the right, respectively, to generate peak point position vector data. The feature index calculation submodule performs a dot product operation on the left and right vectors corresponding to the same peak point in the peak point position vector data. Based on the dot product operation result, it calculates the cosine value of the angle between the two vectors as a peak curvature geometric index characterizing the sharpness of the peak. At the same time, it extracts the wavelength scale of two adjacent peak points on the horizontal axis of the waveform coordinate system, calculates the difference between the wavelength positions of adjacent peaks, obtains the peak wavelength difference value characterizing the peak interval feature, and summarizes it into a peak geometric difference index. The morphological parameter group construction submodule obtains the maximum curvature and maximum wavelength interval within a specified observation period based on the peak geometric difference index. It calculates the standardized curvature parameter based on the peak curvature geometric index and the maximum curvature, calculates the normalized wavelength interval parameter based on the peak wavelength difference and the maximum wavelength interval, and integrates the standardized curvature parameter and the standardized wavelength interval parameter to construct the morphological parameter group.

6. The blockchain-based marine environmental data tamper-proof system according to claim 1, characterized in that, The label mapping module includes: The current velocity data acquisition submodule monitors the current seawater flow status and drives the current meter to acquire seawater velocity readings, representing the instantaneous current velocity that characterizes the current water flow intensity. At the same time, it acquires velocity data covering the entire specified tidal cycle, extracts the maximum and minimum velocity values ​​from the velocity data, calculates the numerical difference between the maximum and minimum velocity values, and obtains the tidal amplitude range. The interval division coding submodule divides the tidal current amplitude range according to a preset step size interval value, constructs multiple continuously distributed velocity sub-intervals, assigns a unique numerical number to each velocity sub-interval in ascending order of numerical value, and establishes a velocity sub-interval index. The dynamic label mapping submodule traverses the numerical range of each velocity sub-interval in the velocity sub-interval index, determines the velocity sub-interval into which the instantaneous ocean current velocity falls, extracts the numerical code corresponding to the target velocity sub-interval, maps the numerical code to an identifier that characterizes the dynamic state of the water body, and generates a dynamic state label.

7. The blockchain-based marine environmental data tamper-proof system according to claim 1, characterized in that, The process of broadcasting data blocks to be uploaded to the blockchain to the distributed node network is as follows: The system retrieves the identifiers of neighboring nodes that are currently online and maintain an active connection with the sending node by calling the P2P network routing table stored locally. Perform binary serialization encoding on the data block to be uploaded to the blockchain to construct a network transmission data packet with complete block content and cyclic redundancy check code; Write a time-to-live parameter to limit the number of times the data packet is forwarded and a list of node signatures to record the data propagation path into the header field of the network transmission data packet; Establish a network communication socket corresponding to the neighbor node identification information, and send network transmission data packets to each directly adjacent node in the distributed node network through an asynchronous parallel transmission channel; Start the data packet transmission monitoring mechanism and wait to receive the data acknowledgment frame returned by the directly adjacent node within the preset response time window; For the directly adjacent nodes that have not returned a data acknowledgment frame within the response time window, perform data packet retransmission operations until the number of retransmissions reaches the preset maximum retransmission threshold; Configure data flooding and forwarding logic based on the Gossip protocol to drive the directly adjacent nodes that receive network transmission data packets to perform a decrement operation on the time-to-live parameter, and forward the network transmission data packets to the secondary adjacent nodes when the operation result is greater than zero.

Images