Distributed power line communication data management system based on block chain

Abstract

The invention relates to the technical field of power line communication, in particular to a distributed power line communication data management system based on a block chain, in the invention, through innovative application of a convolutional neural network in an error repair module, accurate coding correction and rapid repair of error bits are realized, the integrity and effectiveness of data transmission are ensured, and the data transmission efficiency is improved. The method provides higher accuracy and adaptive capacity when reading and repairing bit streams bit by bit, deals with various error types in a complex communication environment, efficiently repairs error data packets in communication, and dynamically monitors and analyzes the signal-to-noise ratio and the error rate of a channel through the use of a finite-state machine, so that the error rate of the channel is effectively reduced, and the error rate of the channel is effectively reduced. The transmission strategy is adjusted in time when the channel state changes, so that the communication process is kept stable and efficient, the efficient transmission strategy is automatically generated according to the change of the channel state, the waste of channel resources is reduced, the bandwidth utilization rate is improved, and the data transmission failure caused by the unstable channel is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of power line communication, and in particular to a distributed power line communication data management system based on blockchain. Background Art

[0002] The purpose of the field of power line communication technology is to transmit data through the existing power line network, realize the dual transmission functions of data and power, avoid the laying of additional communication lines, reduce the cost of infrastructure construction, use power lines as communication media, cover a wide geographical area, improve the communication efficiency of the power system, realize interconnection between equipment, and at the same time ensure the stability and reliability of data transmission to meet the high requirements of smart grids for real-time communication.

[0003] The distributed power line communication data management system based on blockchain aims to solve the security, integrity and transparency issues of traditional power line communication systems in data management. It aims to ensure that the data transmitted in power line communication maintains high security in the entire communication network, prevent data from being tampered with or lost, and record all communication behaviors to achieve data traceability and credibility, thereby improving the efficiency and reliability of data management in the power system.

[0004] Traditional power line communication systems have insufficient accuracy in error detection and repair. Conventional error detection methods have poor ability to identify multiple types of errors in complex communication environments, resulting in some erroneous data not being discovered and repaired in a timely manner, affecting the integrity of data transmission. They also rely on fixed transmission strategies, lack the ability to dynamically monitor and adjust channel status, lack distributed architecture and reliable evidence storage methods, and are prone to data loss or tampering during data transmission. Data tracing is also difficult, posing hidden dangers to data management in power systems. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and propose a distributed power line communication data management system based on blockchain.

[0006] In order to achieve the above purpose, the present invention adopts the following technical solution: A distributed power line communication data management system based on blockchain includes:

[0007] Node authentication module: based on the physical address and logical address of the power line transformer and the communication node, perform address resolution, match the physical address with the logical address one by one, detect abnormal conditions in the matching state, filter invalid node data, confirm the node identity, and generate an authentication node table through bidirectional mapping;

[0008] Channel quality monitoring module: based on the node connection in the authentication node table, a finite state machine is used to sample the signal-to-noise ratio and bit error rate of the channel, and the channel parameters are compared with the set thresholds one by one. According to the thresholds, the channel state changes are judged, and high, medium and low channel states are marked. The state change trend is analyzed, and the changes in the remaining channel bandwidth and bit error rate are correlated to generate a channel transmission strategy;

[0009] Error location module: based on the channel status marked in the channel transmission strategy, performs bit-by-bit scanning on the bit stream of the communication data packet, detects data anomalies bit by bit, locates the index where the error bit is located, summarizes the error position to generate an error mask, and parses the error bit of the corresponding data packet in the mask to obtain error position information;

[0010] Error repair module: Based on the error position information, a convolutional neural network is used to read the bit stream bit by bit in sequence, encode and correct the error bits, rewrite each repaired bit into the original data packet, and complete the overall repair of the data packet by looping through all error positions to generate a repaired data packet set;

[0011] Channel scheduling module: based on the repaired data packet set, checks the matching of the data packet size with the remaining channel bandwidth one by one, allocates each data packet to an available channel, adjusts the data packet transmission order according to the channel status, records each transmission path and time, and establishes a channel scheduling record;

[0012] Data consistency check module: based on the channel scheduling record, read the transmitted bit stream one by one, compare it with the original bit stream bit by bit, mark the matching and non-matching data packets, compare the non-matching data packets to trigger the retransmission process, summarize the comparison results after processing, and generate the consistency check result;

[0013] Blockchain evidence module: Based on the consistency check result, read the path and status of the successfully transmitted data packet, write the path and status information into the hash table of the blockchain network, record the evidence time, store the transmission status of all data packets on the chain, and establish blockchain evidence information.

[0014] As a further solution of the present invention, the node authentication module includes an address resolution submodule, a node verification submodule, and a mapping generation submodule, wherein:

[0015] Address parsing submodule: Based on the physical address and logical address of the power line transformer and the communication node, each address is read separately, and the physical address is compared with the logical address in byte order, the unmatched byte segments are marked, and all the marks are summarized to generate the parsed address result;

[0016] Node verification submodule: Based on the parsed address result, check the marks in the comparison result one by one, compare the matching relationship between the physical and logical addresses, identify abnormal nodes in all addresses, update the address table by deleting abnormal nodes, and filter out all normal nodes to generate a valid node set;

[0017] Mapping generation submodule: Based on the valid node set, read the physical address and logical address of the normal node one by one, and store the addresses in pairs into the mapping table, establish a node mapping structure through a bidirectionally associated address set, enter the mapping structure into the node table, and generate an authentication node table.

[0018] As a further solution of the present invention, the channel quality monitoring module includes a signal-to-noise ratio sampling submodule, a bit error rate detection submodule, and a state evaluation submodule, wherein:

[0019] Signal-to-noise ratio sampling submodule: based on the node connections in the authentication node table, reads the signal-to-noise ratio parameters from each communication channel, and records all parameters by time period, processes the signal-to-noise ratio curve in each time period in sections, and screens out abnormal fluctuation points, and generates sampling signal-to-noise ratio data for all records;

[0020] Bit error rate detection submodule: Based on the sampled signal-to-noise ratio data, the bit stream in the data packet is read packet by packet, the error bits between the actual transmitted data and the original data are compared, the error bits of all data packets are counted, and the total amount of bit errors is calculated. After comparing each packet of data with the threshold, the result is marked to generate detection bit error rate data;

[0021] State evaluation submodule: Based on the detected bit error rate data, a finite state machine is used to analyze the bit error rate and signal-to-noise ratio changes of each channel one by one, determine the channel distribution in high, medium and low states, summarize the state change trend, select the appropriate transmission mode according to the bit error and signal-to-noise ratio correlation information, and obtain the channel transmission strategy.

[0022] As a further solution of the present invention, the finite state machine is according to the formula:

[0023]

[0024] Where: P e represents the bit error rate, erfc(x) represents the complementary error function, E b represents the energy per bit, N0 represents the noise power spectral density, W represents the channel bandwidth, M represents the modulation order, α represents the noise interference coefficient, I represents the external interference power, β represents the power control coefficient, P t Indicates the transmit power.

[0025] As a further solution of the present invention, the error code location module includes a bit stream scanning submodule, an error bit detection submodule, and an error mask parsing submodule, wherein:

[0026] Bit stream scanning submodule: based on the channel transmission strategy, reads the bit stream from the communication data packet, traverses and checks the voltage fluctuation in the data stream bit by bit, records the bit positions that do not match the reference range, stores all abnormal positions in a table, and generates a bit abnormal position table;

[0027] Error bit detection submodule: based on the bit abnormal position table, read the bit value of each marked position, compare the recorded transmission data with the reference data, detect the difference bit by bit and confirm the error bit position, store the data packet index associated with each error bit in the structure table, and generate an error bit index table;

[0028] Error mask parsing submodule: Based on the error bit index table, read the index information of all error bits, generate a mask structure according to the index, parse the mask structure to associate the corresponding data packet with the error bit index, establish a mapping table between the error bit and the data packet, and obtain the error location information.

[0029] As a further solution of the present invention, the error repair module includes an error bit reading submodule, an error bit correction submodule, and a data packet reassembly submodule, wherein:

[0030] Error bit reading submodule: Based on the error position information, a convolutional neural network is used to locate the error bits from the data packet, read and extract the error bits and their positions bit by bit, summarize the positions and original values ​​into an error bit record table, and generate a read error bit list;

[0031] Error bit correction submodule: based on the read error bit list, read the error bits in order and correct them, perform correction operations on the bit stream error bits, overwrite the error bits in the original data packet with the corrected bits, update the bit stream structure of each data packet, and generate a corrected bit data packet;

[0032] Data packet reassembly submodule: based on the corrected bit data packet, extract the data packet content from all corrected bit streams, reassemble the data packet according to the original structure, traverse all error bits and check the repair status to ensure the integrity of the data packet structure, and generate a repaired data packet set.

[0033] As a further solution of the present invention, the convolutional neural network is according to the formula:

[0034]

[0035] in: is the confidence score that the i-th bit is an error bit, x i is the bit value of the i-th bit in the input data packet, w iis the weight parameter of the convolution kernel for the i-th bit during the convolution process, d i is the density information of the local position of the ith bit in the data packet, r is the transmission rate of the data packet, α is the weight coefficient of the density information, β is the weight coefficient of the transmission rate, b is the bias term, and σ(·) is the activation function.

[0036] As a further solution of the present invention, the channel scheduling module includes a bandwidth matching submodule, a transmission sequence adjustment submodule, and a transmission path recording submodule, wherein:

[0037] Bandwidth matching submodule: Based on the repaired data packet set, read the size of each data packet, and read the current available bandwidth by checking the remaining bandwidth of the channel, calculate the ratio of the data packet size to the remaining bandwidth of the channel, mark the data packets that meet the ratio as matching, and generate bandwidth matching records;

[0038] Transmission order adjustment submodule: based on the bandwidth matching record, checks the size and priority of the matching data packets one by one, sorts all the matching data packets from high to low according to the remaining bandwidth of the channel, and puts the high priority data packets in the front according to the priority, and generates the adjusted transmission order;

[0039] Transmission path recording submodule: Based on the adjusted transmission order, read the transmission path information of each data packet, record the start and end time of the transmission in the path table, store the path data in the order of the data packets, establish a complete path structure, and generate a channel scheduling record.

[0040] As a further solution of the present invention, the data consistency check module includes a bit stream reading submodule, a data packet comparison submodule, and a retransmission triggering submodule, wherein:

[0041] The bit stream reading submodule reads the transmitted data packets one by one based on the channel scheduling record, expands the bit stream in the data packet bit by bit, reads the original bit stream and the transmission bit stream synchronously, stores them in a comparison table, and generates the bit stream data after transmission;

[0042] Data packet comparison submodule: based on the transmitted bit stream data, compare the original and transmitted bit streams bit by bit, mark the matching and non-matching bits in the comparison table respectively, and extract the data packet indexes of all non-matching bits to generate a comparison mark table;

[0043] Retransmission trigger submodule: Based on the comparison mark table, read the index position of the mismatched data packet, and trigger retransmission according to the difference, compare the retransmitted data with the original data packet bit by bit for confirmation, summarize the status of all data packets after the retransmission is completed, and generate a consistency check result.

[0044] As a further solution of the present invention, the blockchain evidence storage module includes a path reading submodule, a state storage submodule, and a hash writing submodule, wherein:

[0045] Path reading submodule: based on the consistency check result, read the path of the successfully transmitted data packet, extract the start and end time of each data packet, organize the path and time into a structure table, and generate a path status table;

[0046] The state storage submodule reads the path and time of each data packet based on the path state table, arranges the paths of all data packets in chronological order, organizes them into a writable state table, and generates storage state data;

[0047] Hash writing submodule: Based on the storage status data, the path and status data of each data packet are written one by one and stored in the hash table of the blockchain, and the writing time is stored as a block record, so that all data packet statuses are stored on the chain and blockchain evidence information is established.

[0048] Compared with the prior art, the advantages and positive effects of the present invention are:

[0049] 1. In the present invention, through the innovative application of convolutional neural networks in the error repair module, accurate coding correction and rapid repair of erroneous bits are achieved, ensuring the integrity and effectiveness of data transmission, providing higher accuracy and adaptability when reading and repairing bit streams bit by bit, coping with various error types in complex communication environments, and efficiently repairing erroneous data packets in communications.

[0050] 2. In the present invention, the signal-to-noise ratio and bit error rate of the channel are dynamically monitored and analyzed through the use of a finite state machine, and the transmission strategy is adjusted in time when the channel state changes, so that the communication process remains stable and efficient. An efficient transmission strategy is automatically generated according to the change of the channel state, thereby reducing the waste of channel resources and improving bandwidth utilization, and avoiding data transmission failure caused by channel instability. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a system flow chart of the present invention;

[0052] Figure 2 It is a schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0054] See also Figure 1The present invention provides a technical solution: a distributed power line communication data management system based on blockchain includes:

[0055] Node authentication module: Based on the physical address and logical address of the power line transformer and the communication node, address resolution is performed, the physical address is matched with the logical address one by one, and abnormal conditions in the matching status are detected, invalid node data is filtered, node identity is confirmed, and an authentication node table is generated through bidirectional mapping;

[0056] Channel quality monitoring module: Based on the node connections in the authentication node table, a finite state machine is used to sample the signal-to-noise ratio and bit error rate of the channel, and the channel parameters are compared with the set thresholds one by one. The channel state changes are judged according to the thresholds, and high, medium and low channel states are marked. The state change trend is analyzed, and the changes in the remaining channel bandwidth and bit error rate are correlated to generate a channel transmission strategy.

[0057] Error location module: Based on the channel status marked in the channel transmission strategy, the bit stream of the communication data packet is scanned bit by bit, data anomalies are detected bit by bit, the index of the error bit is located, the error position is summarized to generate an error mask, and the error bit of the corresponding data packet in the mask is parsed to obtain the error position information;

[0058] Error repair module: Based on the error position information, a convolutional neural network is used to read the bit stream bit by bit in sequence, encode and correct the error bits, rewrite each repaired bit into the original data packet, and complete the overall repair of the data packet by looping through all error positions to generate a set of repaired data packets;

[0059] Channel scheduling module: Based on the repaired data packet set, check the matching of the data packet size and the remaining channel bandwidth one by one, assign each data packet to an available channel, adjust the data packet transmission order according to the channel status, record each transmission path and time, and establish a channel scheduling record;

[0060] Data consistency check module: Based on the channel scheduling record, read the transmitted bit stream one by one, compare it with the original bit stream bit by bit, mark the matching and non-matching data packets, compare the non-matching data packets to trigger the retransmission process, summarize the comparison results after processing, and generate the consistency check result;

[0061] Blockchain evidence module: Based on the consistency check results, read the path and status of the successfully transmitted data packet, write the path and status information into the hash table of the blockchain network, record the evidence time, store the transmission status of all data packets on the chain, and establish blockchain evidence information.

[0062] See also Figure 2 , the node authentication module includes an address resolution submodule, a node verification submodule, and a mapping generation submodule, wherein:

[0063] Address parsing submodule: Based on the physical address and logical address of the power line transformer and the communication node, each address is read separately, and the physical address is compared with the logical address in byte order, the unmatched byte segments are marked, and all the marks are summarized to generate the parsed address result;

[0064] Node verification submodule: Based on the parsed address results, check the marks in the comparison results one by one, compare the matching relationship between physical and logical addresses, identify abnormal nodes in all addresses, update the address table by deleting abnormal nodes, and filter out all normal nodes to generate a valid node set;

[0065] Mapping generation submodule: Based on the valid node set, read the physical address and logical address of the normal node one by one, and store the addresses in pairs into the mapping table, establish the node mapping structure through the bidirectional association address set, enter the mapping structure into the node table, and generate the authentication node table;

[0066] Address parsing submodule: Based on the physical address and logical address of the power line transformer and the communication node, a byte-by-byte comparison method is used to read each address separately. First, the low address byte is used to start, and then the high address byte is gradually moved to the physical address and the logical address are compared one by one in byte order. If a byte does not match, the mark instruction is used to mark the byte segment. The mark content is summarized by setting the flag bit. Specifically, each flag bit corresponds to the mismatch state of a byte segment. The bitmap of the marked address segment is generated by reading the flag bit information, and all the marks are summarized to generate the parsed address result.

[0067] Node verification submodule: Based on the address resolution result, the byte-by-byte search method is used to compare the marks in the comparison result. The abnormal nodes are identified one by one by checking the matching relationship between the physical address and the logical address. The mark bit in the resolution result is compared with each byte of the physical address and the logical address in turn. If there is a byte segment where the mark bit is "1" in the comparison between the physical address and the logical address, it is determined to be an abnormal node. The delete instruction is called to perform a delete operation on the abnormal node and update the address table. The specific deletion method is to remove the physical address and logical address corresponding to the abnormal node from the address table, perform table compression, filter all nodes with "0" marks, and store them in the valid node set to generate a valid node set;

[0068] Mapping generation submodule: Based on the valid node set, a bidirectional linked list structure storage algorithm is used to read the physical address and logical address of the normal node one by one, and the linked list insertion function is called to store the mapping of the physical address and the logical address in pairs into the mapping table. During the insertion process, the linked list is sorted, and a node mapping structure is established by bidirectionally associating the front and back pointers of each address set. The structure is entered into the node table by setting the linked list pointer to generate an authentication node table.

[0069] See also Figure 2 The channel quality monitoring module includes a signal-to-noise ratio sampling submodule, a bit error rate detection submodule, and a state evaluation submodule, wherein:

[0070] SNR sampling submodule: Based on the node connections in the authentication node table, read the SNR parameters from each communication channel, record all parameters by time period, segment the SNR curve in each time period, filter out abnormal fluctuation points, and generate sampling SNR data from all records;

[0071] Bit error rate detection submodule: Based on the sampled signal-to-noise ratio data, the bit stream in the data packet is read packet by packet, the error bits between the actual transmitted data and the original data are compared, the error bits of all data packets are counted, and the total amount of bit errors is calculated. After comparing each packet of data with the threshold, the result is marked to generate the detection bit error rate data;

[0072] State evaluation submodule: Based on the bit error rate data, a finite state machine is used to analyze the bit error rate and signal-to-noise ratio changes of each channel one by one, determine the channel distribution of high, medium and low states, summarize the state change trend, select the appropriate transmission mode according to the bit error and signal-to-noise ratio correlation information, and obtain the channel transmission strategy;

[0073] Signal-to-noise ratio sampling submodule: Based on the node connection in the authentication node table, the time-period sampling algorithm is adopted to read the signal-to-noise ratio parameters from each communication channel, call the reading function of the communication interface, read the signal-to-noise ratio in each period of time in the channel by byte, and store it in the timing cache queue. The time period is set to 100 milliseconds, record all signal-to-noise ratio parameters, and use the piecewise linear fitting algorithm to segment the signal-to-noise ratio curve in each period of time. During the segment processing, the fluctuation points exceeding the set threshold are screened out by setting the threshold difference, and the position information of the abnormal fluctuation points is recorded in the abnormal point map, and the abnormal fluctuation report is generated by accumulating the abnormal point information, and all signal-to-noise ratio data and abnormal information are integrated to generate sampling signal-to-noise ratio data;

[0074] Bit error rate detection submodule: Based on the sampled signal-to-noise ratio data, the packet-by-packet comparison method is adopted to read the bit stream in the data packet packet by packet. First, the decoding function is called for each data packet to read the actual transmission data bit by bit, and the original data comparison function is called to compare the bit stream difference between the actual transmission data and the original data bit by bit, and the error information of each bit is recorded. The error bits of each data packet are accumulated and counted through the counter, and the error statistics function is called to calculate the total bit error of each data packet. The bit error rate threshold is set to 0.01, and the total bit error of each data packet is compared with the threshold. If the total error exceeds the threshold, the bit error rate of the data packet is marked using the marking function, and all marking information is summarized to generate the detection bit error rate data;

[0075] State evaluation submodule: Based on the bit error rate detection data, a finite state machine is used for state judgment. The bit error rate and signal-to-noise ratio changes of each channel are analyzed one by one. First, the bit error rate and signal-to-noise ratio of the channel are binary combined. The state judgment standard is set as the bit error rate threshold of 0.01 and the signal-to-noise ratio threshold of 20dB. The state distribution of the finite state machine is divided into three categories: high state, medium state and low state. The bit error rate and signal-to-noise ratio combination is compared. If the signal-to-noise ratio is higher than the threshold and the bit error rate is lower than the threshold, it is judged as a high state, and the medium state and low state are deduced accordingly. The state judgment function is called to generate the state of each channel, and a state change trend report is generated through the summary function. Then, according to the correlation information between the bit error and the signal-to-noise ratio, the appropriate transmission mode is selected through the algorithm, and the transmission strategy generation function is called to generate the channel transmission strategy.

[0076] Finite state machine, according to the formula:

[0077]

[0078] Where: P e represents the bit error rate, erfc(x) represents the complementary error function, E b represents the energy per bit, N0 represents the noise power spectral density, W represents the channel bandwidth, M represents the modulation order, α represents the noise interference coefficient, I represents the external interference power, β represents the power control coefficient, P t Indicates the transmit power;

[0079] Execution process: First, the bandwidth value W is obtained based on the detected channel parameters. The higher the bandwidth, the greater the data transmission capacity, and the more significant the noise and interference. At the same time, the modulation order M is determined according to the currently used modulation scheme. The bandwidth W is multiplied by the modulation order M to obtain a quantitative index of transmission complexity, and the energy per bit E per unit time is detected. b Combined with the product of bandwidth and modulation order, it ensures a more accurate reflection of the effective power level of the signal in the channel. At the same time, the noise power spectral density N0 in the measurement environment is used to describe the basic impact of noise on signal quality. The noise interference coefficient represented by the parameter α is used to weigh the interference factors in the complex environment. The external interference power I during operation is detected to reflect the interference intensity caused by the surrounding equipment or system, ensuring that the superposition effect of environmental noise is fully considered when calculating the bit error rate. In terms of power control, the power control coefficient β is calculated by the weighted average method based on the collected historical transmission data and bit error rate trend to adjust the signal transmission power. Subsequently, the coefficient β is added to the transmission power P t Multiply them to get the adjusted power value, which is used to compensate for the signal attenuation in the channel, and calculate the bit error rate P of each channel through the complementary error function erfc(x) e, generate accurate bit error rate analysis results, complete channel status evaluation, and provide a basis for further transmission strategy optimization.

[0080] See also Figure 2 The error location module includes a bit stream scanning submodule, an error bit detection submodule, and an error mask analysis submodule, wherein:

[0081] Bit stream scanning submodule: Based on the channel transmission strategy, it reads the bit stream from the communication data packet, traverses and checks the voltage fluctuations in the data stream bit by bit, records the bit positions that do not match the reference range, stores all abnormal positions in a table, and generates a bit abnormal position table;

[0082] Error bit detection submodule: Based on the bit abnormal position table, read the bit value of each marked position, compare the recorded transmission data with the reference data, detect the difference bit by bit and confirm the error bit position, store the data packet index associated with each error bit in the structure table, and generate the error bit index table;

[0083] Error mask parsing submodule: Based on the error bit index table, read the index information of all error bits, generate a mask structure according to the index, parse the mask structure to associate the corresponding data packet with the error bit index, establish a mapping table between the error bit and the data packet, and obtain the error location information;

[0084] Bit stream scanning submodule: Based on the channel transmission strategy, the bit stream is read from the communication data packet using the bit-by-bit traversal method, and the data packet reading function is called to scan each bit stream bit by bit. First, the communication interface is called to read the bit stream information in the communication data packet bit by bit, and the corresponding voltage fluctuation data in each bit stream is stored in the cache area. By setting the reference voltage range, the voltage fluctuation of each bit is compared and checked. The reference range is set to 0.8V to 1.2V. If the bit voltage exceeds the range, the marking function is used to record the position of the bit, and the position information of each abnormal bit is written into the abnormal position table, and the bit abnormal position table is generated by accumulating records;

[0085] Error bit detection submodule: Based on the bit abnormal position table, the bit value of the abnormal position is read bit by bit by using the bit-by-bit comparison method. First, the bit reading function is called to read the bit value at the marked position, and the recorded transmission bit is compared with the reference bit by calling the bit value comparison function. If the two are inconsistent, the bit is confirmed as an error bit, and the position information of the error bit is recorded in the structure table by calling the error bit recording function. The data packet index reading function is called to store the data packet index associated with the error bit in the structure table, and each error bit and the corresponding data packet index are recorded to generate an error bit index table;

[0086] Error mask parsing submodule: Based on the error bit index table, the mask generation algorithm is used to gradually read the index information of all error bits, call the index reading function, read the index value of each error bit, and generate a mask structure corresponding to the error bit through the mask generation function. The mask structure generates a mask byte for every 8 bits, and marks the specific location of the error bit with binary "1" and "0". The mask parsing function is called to parse the mask structure, and each parsed mask bit is associated with its corresponding data packet. A mapping table between error bits and data packets is established, and the mapping relationship between error bits and data packets is stored in the data table to generate a mapping table between error bits and data packets.

[0087] See also Figure 2 The error repair module includes an error bit reading submodule, an error bit correction submodule, and a data packet reassembly submodule, wherein:

[0088] Error bit reading submodule: Based on the error position information, a convolutional neural network is used to locate the error bits from the data packet, read and extract the error bits and their positions bit by bit, summarize the positions and original values ​​into an error bit record table, and generate a read error bit list;

[0089] Error bit correction submodule: Based on the error bit list, the error bits are read in order and corrected, the correction operation is performed on the bit stream error bits, the corrected bits are used to overwrite the error bits in the original data packet, the bit stream structure of each data packet is updated, and a corrected bit data packet is generated;

[0090] Packet reassembly submodule: Based on the corrected bit packets, extract the packet content from all corrected bit streams, reassemble the packets according to the original structure, traverse all error bits and check the repair status to ensure the integrity of the packet structure, and generate a set of repaired packets;

[0091] Error bit reading submodule: Based on the error position information, a convolutional neural network is used to locate the error bit from the data packet. First, the convolution layer of the convolutional neural network is called, and a 3×3 convolution kernel is used to extract local features of the bit stream data. Each convolution kernel extracts the error features in the bit stream in turn during the sliding process. The convolution output is passed to the ReLU activation function, and the negative values ​​in the convolution result are set to zero. After multiple layers of convolution and pooling operations, the position with significant error features is extracted, and the position information is passed to the fully connected layer. The specific index position of the error bit is calculated through the fully connected layer, and the error bit and its position are read and extracted bit by bit. Each position and the original value are summarized and stored in the error bit record table to generate a read error bit list;

[0092] Error bit correction submodule: Based on the reading error bit list, a sequential correction algorithm is adopted to read the error bits in sequence and perform correction operations. First, the reading function is called to read the error bits in sequence, and the correction function is called to compare the error bits in the bit stream with the correct bit values ​​in the reference data bit by bit. If the error bit values ​​are inconsistent, the overwrite write instruction is called to overwrite the error bits in the original data packet with the correct bit values ​​in the reference data to complete the correction operation. Each corrected bit is overwritten and written into the original data packet. The bitstream update function is called to update the bitstream structure of each data packet, and the corrected bitstream is saved in the new data packet structure to generate a corrected bit data packet.

[0093] Packet reassembly submodule: Based on the corrected bit packets, a reassembly traversal algorithm is used to extract the packet content from all corrected bit streams. First, the extraction function is called to extract the corrected bit stream bit by bit, and the new packets are stored one by one according to the structure of the original packet. The traversal check function is called to traverse and check the error bits in each packet to confirm that all error bits have been corrected. The structural integrity of each packet is checked. If an abnormal data structure is found, the correction function is called for a secondary correction operation. After the traversal is completed, all repaired packets are reassembled to generate a set of repaired packets.

[0094] Convolutional neural network, according to the formula:

[0095]

[0096] in: is the confidence score that the i-th bit is an error bit, x i is the bit value of the i-th bit in the input data packet, w i is the weight parameter of the convolution kernel for the i-th bit during the convolution process, d i is the density information of the local position of the ith bit in the data packet, r is the transmission rate of the data packet, α is the weight coefficient of the density information, β is the weight coefficient of the transmission rate, b is the bias term, and σ(·) is the activation function;

[0097] Execution process: First, the bit streams in the input data packets are sent to the convolutional neural network for processing one by one. The convolution kernel is processed in the sliding window according to the weight parameter w i For each bit x i Weighted calculation is performed. During the sliding process of the convolution window, all bits are covered one by one to ensure that each bit participates in the judgment. The density information d of the bit position is introduced during the convolution process. i If the density is high, the model will determine that the probability of bit error here is higher, and control d through the density information weight coefficient α iThe influence of α on the results is optimized through the training data set, and the grid search is used to find the optimal value. At the same time, the transmission rate r of the data packet is considered. The higher the rate, the more likely the data packet bits are to be interfered and errors occur. The influence of this parameter is adjusted by the weight coefficient β of the transmission rate. The optimization of β is also determined by the performance of the validation set. Then the weighted sum in the convolution window is combined with the density information d i The weighted results of the transmission rate r are accumulated, and the bias term b is added as the basic adjustment. The accumulated result is mapped to a probability value y between 0 and 1 through the activation function σ(·). When the probability exceeds the set threshold, the corresponding bit is marked as an error bit, and its position and original value are stored in the error bit record table. All error bits are summarized to generate a read error bit list.

[0098] See also Figure 2 The channel scheduling module includes a bandwidth matching submodule, a transmission order adjustment submodule, and a transmission path recording submodule, wherein:

[0099] Bandwidth matching submodule: Based on the repaired data packet set, read the size of each data packet, and read the current available bandwidth by checking the remaining bandwidth of the channel, calculate the ratio of the data packet size to the remaining bandwidth of the channel, mark the data packets that meet the ratio as matching, and generate bandwidth matching records;

[0100] Transmission order adjustment submodule: Based on the bandwidth matching records, check the size and priority of the matching data packets one by one, sort all the matching data packets from high to low according to the remaining bandwidth of the channel, and put the high-priority data packets in the front according to the priority, and generate the adjusted transmission order;

[0101] Transmission path recording submodule: Based on the adjusted transmission order, read the transmission path information of each data packet, record the start and end time of the transmission in the path table, store the path data in the order of the data packets, establish a complete path structure, and generate a channel scheduling record;

[0102] Bandwidth matching submodule: Based on the repaired data packet set, the bandwidth share calculation algorithm is used to read the size of each data packet, call the data packet reading function to obtain the number of bytes of each data packet, and call the bandwidth check function when reading the remaining bandwidth of the channel. The function detects the remaining bandwidth value of the channel in real time and stores it in the bandwidth cache queue, and then calls the bandwidth share calculation function. The calculation formula is "data packet size ÷ channel remaining bandwidth". The calculated share value is compared with the set bandwidth threshold. The bandwidth threshold is set to 80%. If the data packet size share is within the threshold range, the marking function is used to mark the data packet as a match, and the marking information is written into the matching record table to generate a bandwidth matching record;

[0103] Transmission order adjustment submodule: Based on the bandwidth matching record, the priority sorting algorithm is used to check the size and priority of each matching data packet one by one. First, the size check function is called to obtain the number of bytes of each data packet, and the priority reading function is called to obtain the priority value of each data packet. The priority value range is set to 1 to 10, and then the sorting function is called to sort all matching data packets by size according to the sorting standard of "channel remaining bandwidth from high to low". After the sorting is completed, the priority adjustment function is called to adjust the data packet with a higher priority value to the priority position. If two data packets have the same bandwidth share, the one with a higher priority value will be ranked first to generate the adjusted transmission order;

[0104] Transmission path recording submodule: Based on the adjusted transmission order, the path tracking algorithm is used to gradually read the transmission path information of each data packet. First, the path reading function is called to obtain the transmission path information of each data packet. The path information includes the timestamp data of the starting node and the ending node. The timestamp recording function is called to store the start and end time of the transmission into the path table. The path table records the path information of each data packet in the transmission order. The sequence storage function is called to gradually rearrange the transmission data in the path table according to the adjusted transmission order, and the complete path structure is established by calling the path structure establishment function to generate a channel scheduling record.

[0105] See also Figure 2 The data consistency check module includes a bit stream reading submodule, a data packet comparison submodule, and a retransmission trigger submodule, wherein:

[0106] Bit stream reading submodule: Based on the channel scheduling record, read the transmitted data packets one by one, expand the bit stream in the data packet bit by bit, read the original bit stream and the transmitted bit stream synchronously, store them in the comparison table, and generate the bit stream data after transmission;

[0107] Packet comparison submodule: Based on the bit stream data after transmission, compare the original and transmitted bit streams bit by bit, mark the matching and non-matching bits in the comparison table, extract the packet index of all non-matching bits, and generate a comparison mark table;

[0108] Retransmission trigger submodule: Based on the comparison mark table, read the index position of the mismatched data packet, trigger retransmission based on the difference, compare the retransmitted data with the original data packet bit by bit, summarize the status of all data packets after retransmission is completed, and generate consistency check results;

[0109] Bitstream reading submodule: Based on the channel scheduling record, the bit-by-bit expansion algorithm is adopted to read the transmitted data packets one by one, and the data packet reading function is called to expand the bit stream in the data packet bit by bit. The expansion method is to read byte by byte and disassemble it into single bits. The bitstream synchronous reading function is called to synchronously read the original bit stream and the transmission bit stream. The synchronous reading window size is set to 8 bits. After each synchronous reading, the synchronized bit stream is written into the comparison table, and the offset and bit position of the bit stream are recorded by storing it bit by bit in the comparison table to generate the bit stream data after transmission;

[0110] Data packet comparison submodule: Based on the bit stream data after transmission, a bit-by-bit comparison algorithm is used to compare the original and transmitted bit streams bit by bit. First, the comparison function is called to read each bit in the original bit stream and the transmitted bit stream bit by bit. By bit-by-bit comparison, the window size of each comparison is set to 8 bits. The matching mark function is called to mark the matching and non-matching bits in the comparison table respectively. The mark value of the matching bit is set to "1", and the mark value of the non-matching bit is set to "0". The index of each non-matching bit of the data packet is extracted by calling the index extraction function, and the index information is stored in the non-matching bit index table to generate a comparison mark table;

[0111] Retransmission trigger submodule: Based on the comparison mark table, the difference retransmission trigger algorithm is adopted to read the index position of the unmatched data packet. First, the index reading function is called to read the index position of the unmatched bit in the comparison mark table one by one, and the retransmission trigger function is called to trigger the retransmission operation according to the difference in the bit stream. The retransmitted data packet is read bit by bit and transmitted to the comparison module, and the bit stream comparison function is called to compare the retransmitted data with the bit stream in the original data packet bit by bit, and confirm that the retransmitted bit value is consistent with the original bit stream bit by bit. After the retransmission is completed, the status summary function is called to summarize the transmission status of all data packets and store them in the status table to generate a consistency check result.

[0112] See also Figure 2 The blockchain evidence storage module includes a path reading submodule, a state storage submodule, and a hash writing submodule, among which:

[0113] Path reading submodule: Based on the consistency check result, read the path of the successfully transmitted data packet, extract the start and end time of each data packet, organize the path and time into a structure table, and generate a path status table;

[0114] State storage submodule: Based on the path state table, read the path and time of each data packet, arrange the paths of all data packets in chronological order, organize them into a writable state table, and generate storage state data;

[0115] Hash writing submodule: Based on the storage status data, the path and status data of each data packet are written one by one and stored in the hash table of the blockchain. The writing time is stored as a block record, and all data packet statuses are stored on the chain to establish blockchain evidence information;

[0116] Path reading submodule: Based on the consistency check result, the path extraction algorithm is used to read the path of the successfully transmitted data packet. First, the path reading function is called to read the transmission path of the data packet one by one. The path information includes the start node and the end node. The time extraction function is called to extract the start time and end time of each data packet. The time format is set to Unix timestamp. The extracted path and time information are organized into a structure table. By traversing all data packets, the path and corresponding time information of each data packet are stored in the structure table according to byte alignment, and finally a path status table is generated;

[0117] State storage submodule: Based on the path state table, the time sorting algorithm is used to read the path and time of each data packet. First, the path reading function is called to read the path and time information in the path state table one by one, and the time information of each data packet is arranged in ascending order. The time unit is milliseconds. The time sorting function is called to sort the timestamps of the data packets step by step. The sorted path information is rearranged in chronological order. The write buffer function is called to organize all sorted paths into a writable state table structure, and finally generate storage state data;

[0118] Hash writing submodule: Based on the storage status data, the SHA-256 hash algorithm is used to write the path and status data of each data packet one by one, call the hash generation function, read the path and status data of each data packet in sequence, call the SHA-256 hash generator, use the byte stream of the path and status data as input, generate a 256-bit hash value, write each hash value into the hash table, and then call the blockchain writing function to write the hash of each data packet into the blockchain, record the writing time as a Unix timestamp, call the block recording function to store the writing time and hash value as a block record, complete the on-chain evidence storage of the status of all data packets, and establish blockchain evidence information.

[0119] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A distributed power line communication data management system based on blockchain, characterized in that: The system comprises: Node authentication module: based on the physical address and logical address of the power line transformer and the communication node, perform address resolution, match the physical address with the logical address one by one, detect abnormal conditions in the matching state, filter invalid node data, confirm the node identity, and generate an authentication node table through bidirectional mapping; Channel quality monitoring module: based on the node connection in the authentication node table, a finite state machine is used to sample the signal-to-noise ratio and bit error rate of the channel, and the channel parameters are compared with the set thresholds one by one. According to the thresholds, the channel state changes are judged, and high, medium and low channel states are marked. The state change trend is analyzed, and the changes in the remaining channel bandwidth and bit error rate are correlated to generate a channel transmission strategy; Error location module: based on the channel status marked in the channel transmission strategy, performs bit-by-bit scanning on the bit stream of the communication data packet, detects data anomalies bit by bit, locates the index where the error bit is located, summarizes the error position to generate an error mask, and parses the error bit of the corresponding data packet in the mask to obtain error position information; Error repair module: Based on the error position information, a convolutional neural network is used to read the bit stream bit by bit in sequence, encode and correct the error bits, rewrite each repaired bit into the original data packet, and complete the overall repair of the data packet by looping through all error positions to generate a set of repaired data packets; Channel scheduling module: based on the repaired data packet set, checks the matching of the data packet size with the remaining channel bandwidth one by one, allocates each data packet to an available channel, adjusts the data packet transmission order according to the channel status, records each transmission path and time, and establishes a channel scheduling record; Data consistency check module: based on the channel scheduling record, read the transmitted bit stream one by one, compare it with the original bit stream bit by bit, mark the matching and non-matching data packets, compare the non-matching data packets to trigger the retransmission process, summarize the comparison results after processing, and generate the consistency check result; Blockchain evidence module: Based on the consistency check result, read the path and status of the successfully transmitted data packet, write the path and status information into the hash table of the blockchain network, record the evidence time, store the transmission status of all data packets on the chain, and establish blockchain evidence information.

2. The distributed power line communication data management system based on blockchain according to claim 1 is characterized in that: The node authentication module includes an address resolution submodule, a node verification submodule, and a mapping generation submodule, wherein: Address parsing submodule: Based on the physical address and logical address of the power line transformer and the communication node, each address is read separately, and the physical address is compared with the logical address in byte order, the unmatched byte segments are marked, and all the marks are summarized to generate the parsed address result; Node verification submodule: Based on the parsed address result, check the marks in the comparison result one by one, compare the matching relationship between the physical and logical addresses, identify abnormal nodes in all addresses, update the address table by deleting abnormal nodes, and filter out all normal nodes to generate a valid node set; Mapping generation submodule: Based on the valid node set, read the physical address and logical address of the normal node one by one, and store the addresses in pairs into the mapping table, establish a node mapping structure through a bidirectionally associated address set, enter the mapping structure into the node table, and generate an authentication node table.

3. The distributed power line communication data management system based on blockchain according to claim 1 is characterized in that: The channel quality monitoring module includes a signal-to-noise ratio sampling submodule, a bit error rate detection submodule, and a state evaluation submodule, wherein: Signal-to-noise ratio sampling submodule: based on the node connections in the authentication node table, reads the signal-to-noise ratio parameters from each communication channel, and records all parameters by time period, processes the signal-to-noise ratio curve in each time period in sections, and screens out abnormal fluctuation points, and generates sampling signal-to-noise ratio data for all records; Bit error rate detection submodule: Based on the sampled signal-to-noise ratio data, the bit stream in the data packet is read packet by packet, the error bits between the actual transmitted data and the original data are compared, the error bits of all data packets are counted, and the total amount of bit errors is calculated. After comparing each packet of data with the threshold, the result is marked to generate detection bit error rate data; State evaluation submodule: Based on the detected bit error rate data, a finite state machine is used to analyze the bit error rate and signal-to-noise ratio changes of each channel one by one, determine the channel distribution in high, medium and low states, summarize the state change trend, select the appropriate transmission mode according to the bit error and signal-to-noise ratio correlation information, and obtain the channel transmission strategy.

4. The distributed power line communication data management system based on blockchain according to claim 1 is characterized in that: The finite state machine, according to the formula: Where: P e represents the bit error rate, erfc(x) represents the complementary error function, E b represents the energy per bit, N0 represents the noise power spectral density, W represents the channel bandwidth, M represents the modulation order, α represents the noise interference coefficient, I represents the external interference power, β represents the power control coefficient, P t Indicates the transmit power.

5. The distributed power line communication data management system based on blockchain according to claim 1 is characterized in that: The error code location module includes a bit stream scanning submodule, an error bit detection submodule, and an error mask analysis submodule, wherein: Bit stream scanning submodule: based on the channel transmission strategy, reads the bit stream from the communication data packet, traverses and checks the voltage fluctuation in the data stream bit by bit, records the bit positions that do not match the reference range, stores all abnormal positions in a table, and generates a bit abnormal position table; Error bit detection submodule: based on the bit abnormal position table, read the bit value of each marked position, compare the recorded transmission data with the reference data, detect the difference bit by bit and confirm the error bit position, store the data packet index associated with each error bit in the structure table, and generate an error bit index table; Error mask parsing submodule: Based on the error bit index table, read the index information of all error bits, generate a mask structure according to the index, parse the mask structure to associate the corresponding data packet with the error bit index, establish a mapping table between the error bit and the data packet, and obtain the error location information.

6. The distributed power line communication data management system based on blockchain according to claim 1 is characterized in that: The error repair module includes an error bit reading submodule, an error bit correction submodule, and a data packet reassembly submodule, wherein: Error bit reading submodule: Based on the error position information, a convolutional neural network is used to locate the error bits from the data packet, read and extract the error bits and their positions bit by bit, summarize the positions and original values ​​into an error bit record table, and generate a read error bit list; Error bit correction submodule: based on the read error bit list, read the error bits in order and correct them, perform correction operations on the bit stream error bits, overwrite the error bits in the original data packet with the corrected bits, update the bit stream structure of each data packet, and generate a corrected bit data packet; Data packet reassembly submodule: based on the corrected bit data packet, extract the data packet content from all corrected bit streams, reassemble the data packet according to the original structure, traverse all error bits and check the repair status to ensure the integrity of the data packet structure, and generate a repaired data packet set.

7. The distributed power line communication data management system based on blockchain according to claim 1 is characterized in that: The convolutional neural network, according to the formula: in: is the confidence score that the i-th bit is an error bit, x i is the bit value of the i-th bit in the input data packet, w i is the weight parameter of the convolution kernel for the i-th bit during the convolution process, d i is the density information of the local position of the ith bit in the data packet, r is the transmission rate of the data packet, α is the weight coefficient of the density information, β is the weight coefficient of the transmission rate, b is the bias term, and σ(·) is the activation function.

8. The distributed power line communication data management system based on blockchain according to claim 1, characterized in that: The channel scheduling module includes a bandwidth matching submodule, a transmission sequence adjustment submodule, and a transmission path recording submodule, wherein: Bandwidth matching submodule: Based on the repaired data packet set, read the size of each data packet, and read the current available bandwidth by checking the remaining bandwidth of the channel, calculate the ratio of the data packet size to the remaining bandwidth of the channel, mark the data packets that meet the ratio as matching, and generate bandwidth matching records; Transmission order adjustment submodule: based on the bandwidth matching record, checks the size and priority of the matching data packets one by one, sorts all the matching data packets from high to low according to the remaining bandwidth of the channel, and puts the high priority data packets in the front according to the priority, and generates the adjusted transmission order; Transmission path recording submodule: Based on the adjusted transmission order, read the transmission path information of each data packet, record the start and end time of the transmission in the path table, store the path data in the order of the data packets, establish a complete path structure, and generate a channel scheduling record.

9. The distributed power line communication data management system based on blockchain according to claim 1, characterized in that: The data consistency check module includes a bit stream reading submodule, a data packet comparison submodule, and a retransmission triggering submodule, wherein: The bit stream reading submodule reads the transmitted data packets one by one based on the channel scheduling record, expands the bit stream in the data packet bit by bit, reads the original bit stream and the transmission bit stream synchronously, stores them in a comparison table, and generates the bit stream data after transmission; Data packet comparison submodule: based on the transmitted bit stream data, compare the original and transmitted bit streams bit by bit, mark the matching and non-matching bits in the comparison table respectively, and extract the data packet indexes of all non-matching bits to generate a comparison mark table; Retransmission trigger submodule: Based on the comparison mark table, read the index position of the mismatched data packet, and trigger retransmission according to the difference, compare the retransmitted data with the original data packet bit by bit for confirmation, summarize the status of all data packets after the retransmission is completed, and generate a consistency check result.

10. The distributed power line communication data management system based on blockchain according to claim 1, characterized in that: The blockchain evidence storage module includes a path reading submodule, a state storage submodule, and a hash writing submodule, wherein: Path reading submodule: based on the consistency check result, read the path of the successfully transmitted data packet, extract the start and end time of each data packet, organize the path and time into a structure table, and generate a path status table; The state storage submodule reads the path and time of each data packet based on the path state table, arranges the paths of all data packets in chronological order, organizes them into a writable state table, and generates storage state data; Hash writing submodule: Based on the storage status data, the path and status data of each data packet are written one by one and stored in the hash table of the blockchain, and the writing time is stored as a block record, so that all data packet statuses are stored on the chain and blockchain evidence information is established.

Images