A multi-dimensional physical feature-based key generation method, system, device and medium

Abstract

The application discloses a kind of based on multi-dimensional physical feature key generation method, system, device and medium, belong to electric power system information security and wireless communication technical field.Method includes based on original feature sequence and obtains comprehensive difference feature sequence;Based on comprehensive difference feature sequence determines quantization interval and discard interval, according to the quantization interval of effective sample assignment gray code bit, in sensing terminal obtains sensing terminal initial key bit stream, in distribution terminal obtains distribution terminal initial key bit stream;In sensing terminal, sensing terminal initial key bit stream is encoded to obtain parity check symbol and convolution check bit;In distribution terminal, distribution terminal initial key bit stream is decoded, and in decoding process, according to parity check symbol and convolution check bit, distribution terminal initial key bit stream is corrected, to obtain negotiation key;Hash operation is carried out to negotiation key, to obtain working key.The application can fuse multi-dimensional physical feature, adapt to low mobility strong interference electric power scene.

Description

Technical Field

[0001] This invention belongs to the field of power system information security and wireless communication technology, specifically relating to a key generation method, system, device and medium based on multi-dimensional physical characteristics. Background Technology

[0002] With the deepening of smart grid construction, the number of low-power wireless sensing terminals such as smart meters, fault indicators, and distribution switch status sensors is rapidly increasing. These terminals communicate with the master station via low-power wide-area network protocols such as LoRa and NB-IoT to transmit sensitive data such as electricity consumption, equipment status, and control commands. The inherent openness of wireless channels exposes these terminals to threats such as eavesdropping, replay attacks, and man-in-the-middle attacks. Encryption is a core protection measure, while the secure distribution of keys is a key bottleneck restricting the overall security of the system.

[0003] Current key management schemes for power wireless sensor networks have the following key problems:

[0004] 1. Existing physical layer key generation schemes rely on the single wireless received signal strength (RSSI) as a source of channel features. In power scenarios, electromagnetic interference from high-power equipment leads to significantly higher noise in RSSI measurements. The randomness of keys extracted from a single feature is unstable, resulting in a low pass rate in the NIST randomness test.

[0005] 2. Power sensing nodes are usually fixed in locations such as poles and switchgear, with extremely low node mobility and insufficient channel time-varying properties. Existing quantization schemes based on continuous RSSI differential have a large number of bits in the generated key exhibiting long runs due to the high correlation between adjacent samples, which cannot meet the key randomness requirements of encryption algorithms such as AES-128.

[0006] 3. The unique harmonic pollution (2nd to 25th harmonics) and the rich physical random information carried by the transient load switching of the power system have not been utilized by any existing key generation scheme. Furthermore, the existing key negotiation mechanism is not compatible with the frame structure of the distribution automation communication protocol IEC 60870-5-101 / 104, which introduces additional communication overhead and affects the real-time performance of distribution automation. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide a key generation method, system, device and medium based on multi-dimensional physical characteristics, which can integrate the multi-dimensional physical characteristics of power channels, adapt to low mobility and strong interference power scenarios, and seamlessly integrate with distribution automation communication protocols.

[0008] This invention provides the following technical solution:

[0009] Firstly, a key generation method based on multi-dimensional physical characteristics is provided, which is executed separately in the sensing terminal and the power distribution terminal;

[0010] The key generation method includes:

[0011] Generate an original feature sequence including multidimensional features based on multiple pre-acquired physical features;

[0012] The multidimensional features in the original feature sequence are subjected to discontinuous differencing and normalization respectively to obtain the normalized difference sequence of each feature. The Shannon information entropy of the normalized difference sequence of each feature is calculated. The normalized difference sequence of the multidimensional features is weighted and summed according to the Shannon information entropy to obtain a comprehensive difference feature sequence including several samples.

[0013] The empirical cumulative distribution function is calculated based on the comprehensive differential feature sequence. The quantization interval and the discard interval are determined according to the empirical cumulative distribution function and the principle of equal probability. Gray code bits are allocated according to the quantization interval to which the valid samples belong. The initial key bit stream of the sensing terminal is obtained at the sensing terminal and the initial key bit stream of the power distribution terminal is obtained at the power distribution terminal. Among them, the samples in the union of the discard set of the sensing terminal and the discard set of the power distribution terminal are discarded, and the remaining samples are regarded as valid samples. The discard set includes the samples located within the discard interval.

[0014] At the sensing terminal, RS encoding and convolutional encoding are performed on the initial key bit stream to obtain parity check symbols and convolutional check bits. At the distribution terminal, Viterbi decoding and RS decoding are performed on the initial key bit stream, and during the decoding process, the initial key bit stream is corrected based on the obtained parity check symbols and convolutional check bits to obtain the negotiation key. The negotiation key at the sensing terminal is the same as the negotiation key at the distribution terminal.

[0015] Perform a hash operation on the negotiated key to obtain the working key.

[0016] As an optional technical solution of the present invention, the step of generating an original feature sequence including multi-dimensional features based on multiple pre-acquired physical features includes:

[0017] The multidimensional physical characteristics include wireless received signal strength, harmonic noise index, and electromagnetic disturbance index;

[0018] The harmonic noise index is expressed as:

[0019] ;

[0020] in, Indicates the sensor terminal and the power distribution terminal. Harmonic noise figure of the second exchange Indicates the current transformer number During the second sampling The power of the subharmonics Indicates fundamental frequency power. This indicates the total number of exchanges between the sensing terminal and the power distribution terminal.

[0021] The electromagnetic disturbance index is expressed as:

[0022] ;

[0023] in, Indicates the sensor terminal and the power distribution terminal. Electromagnetic disturbance index of the second exchange. This indicates that the RF front-end, during the frame interval, transmits the first signal within a specified bandwidth near the center frequency. Energy received at each scan point This represents the average received energy during the scan. Indicates the number of scan points;

[0024] The original feature sequence is represented as follows:

[0025] ;

[0026] in, Indicates the sensor terminal and the power distribution terminal. The original feature sequence after the second exchange. Sensing terminals and power distribution terminals The strength of the wireless received signal during the second exchange. Indicates transpose;

[0027] Linear interpolation compensation is applied to the original feature sequence of any lost frame, as follows:

[0028] ;

[0029] in, Indicates the sensor terminal and the power distribution terminal. The original feature sequence after the second exchange. Indicates the sensor terminal and the power distribution terminal. The original feature sequence after the second exchange.

[0030] As an optional technical solution of the present invention, the multidimensional features in the original feature sequence are subjected to discontinuous differencing and normalization respectively to obtain normalized difference sequences of each dimension feature. The Shannon information entropy of the normalized difference sequences of each dimension feature is calculated. Based on the Shannon information entropy, the normalized difference sequences of the multidimensional features are weighted and summed to obtain a comprehensive difference feature sequence including several samples, including:

[0031] Discontinuous differencing is performed on the multidimensional features in the original feature sequence, as follows:

[0032] ;

[0033] ;

[0034] ;

[0035] in, Indicates the sensor terminal and the power distribution terminal. The difference in the strength of the wireless received signal after each exchange. Indicates the number of differential execution steps. Sensing terminals and power distribution terminals The strength of the wireless received signal during the second exchange. Indicates the sensor terminal and the power distribution terminal. The difference in the harmonic noise figure of the second exchange. Indicates the sensor terminal and the power distribution terminal. Harmonic noise figure of the second exchange Indicates the sensor terminal and the power distribution terminal. The difference in the electromagnetic disturbance index of the second exchange. Indicates the sensor terminal and the power distribution terminal. Electromagnetic disturbance index of the second exchange;

[0036] The difference values ​​of each feature dimension are normalized and expressed as follows:

[0037] ;

[0038] ;

[0039] ;

[0040] in, Indicates the sensor terminal and the power distribution terminal. The normalized difference value of the wireless received signal strength after the second exchange The mean of the differences in the strength of the wireless received signal. The standard deviation of the difference values ​​representing the strength of the wireless received signal. Indicates the sensor terminal and the power distribution terminal. The normalized difference of the harmonic noise figure of the second exchange. The mean of the differences in the harmonic noise figure. The standard deviation of the difference values ​​of the harmonic noise figure. Indicates the sensor terminal and the power distribution terminal. The normalized difference of the electromagnetic disturbance index of the second exchange. The mean of the differences in the electromagnetic disturbance index. The standard deviation of the difference values ​​in the electromagnetic disturbance index;

[0041] Based on the pre-acquired probability density histogram, the Shannon information entropy of the normalized difference sequence of each feature dimension is calculated, and expressed as:

[0042] ;

[0043] in, Representation of features Shannon information entropy of normalized difference sequences Representation of features The normalized difference sequence falls into the first The probability of each interval. Indicates the total number of intervals. Indicates the strength of the wireless received signal. Indicates the harmonic noise figure. Indicates the electromagnetic disturbance index;

[0044] The weights of the normalized difference sequences for each feature dimension are calculated and expressed as follows:

[0045] ;

[0046] ;

[0047] in, Representation of features The weights of the normalized difference sequence, The Shannon information entropy represents the normalized difference sequence of the wireless received signal strength. The Shannon information entropy representing the normalized difference sequence of the harmonic noise index. The Shannon information entropy represents the normalized difference sequence of the electromagnetic disturbance index. The weights represent the normalized difference sequence of the wireless received signal strength. The weights of the normalized difference sequence representing the harmonic noise index. The weights represent the normalized difference sequence of the electromagnetic disturbance index;

[0048] The samples in the comprehensive differential feature sequence are represented as follows:

[0049] ;

[0050] in, Indicates the sensor terminal and the power distribution terminal. The sample corresponding to each exchange.

[0051] As an optional technical solution of the present invention, the step of calculating the empirical cumulative distribution function based on the comprehensive differential feature sequence, and determining the quantization interval and the discard interval according to the empirical cumulative distribution function and the principle of equal probability includes:

[0052] The empirical cumulative distribution function is calculated as follows:

[0053] ;

[0054] in, Represents the empirical cumulative distribution function. This represents an arbitrary point in the comprehensive difference feature sequence. express The value is 1 if the condition is true, and 0 otherwise.

[0055] The boundary of the quantization interval is determined according to the principle of equal probability, and is expressed as:

[0056] ;

[0057] in, Indicates the first The boundaries of a quantization interval, This represents the pseudo-inverse of the empirical distribution function. Indicates the total number of quantization intervals;

[0058] The quantization interval is represented as The half-interval width of the quantization interval is calculated and expressed as:

[0059] ;

[0060] in, Indicates the first The width of half interval of each quantization interval. Indicates the first The boundaries of a quantization interval, Indicates the first The boundaries of each quantization interval;

[0061] For each boundary, the discard interval is represented as:

[0062] ;

[0063] in, Indicates the first The discard interval at the boundary of each quantization interval. Indicates that the parameter is discarded. Indicates the first The width of half interval of each quantization interval. Indicates the first The width of half of each quantization interval.

[0064] As an optional technical solution of the present invention, the process of obtaining the discard set of the sensing terminal and the discard set of the power distribution terminal includes:

[0065] At the sensing terminal or power distribution terminal, samples located within the discard interval are added to the discard set of the sensing terminal or power distribution terminal.

[0066] As an optional technical solution of the present invention, the step of performing RS encoding and convolutional encoding on the initial key bit stream of the sensing terminal to obtain parity check symbols and convolutional check bits includes:

[0067] RS encoding is performed on the initial key bitstream of the sensing terminal to obtain the first codeword and parity check symbol, which are represented as follows:

[0068] ;

[0069] ;

[0070] ;

[0071] ;

[0072] ;

[0073] in, Indicates the codeword length. Indicates message length. Indicates the number of bits per symbol. Indicates the number of correctable sign errors. Indicates the target key length. Indicates the first character. Indicates the parity check symbol;

[0074] The first codeword is convolutionally encoded to obtain the convolutional parity bit.

[0075] As an optional technical solution of the present invention, the step of performing Viterbi decoding and RS decoding on the initial key bit stream of the power distribution terminal at the power distribution terminal, and correcting errors in the initial key bit stream of the power distribution terminal based on the obtained parity check symbol and convolution check bit during the decoding process to obtain the negotiation key, includes:

[0076] The second codeword is obtained by performing RS encoding on the initial key bitstream of the power distribution terminal;

[0077] Add convolutional parity bits to the end of the second codeword to obtain the concatenated codeword, and perform Viterbi decoding on the concatenated codeword to obtain the error-corrected RS codeword;

[0078] After replacing the parity part of the error correction RS codeword with the parity check symbol, RS decoding is performed to obtain the negotiation key.

[0079] In a second aspect, a key generation system based on multidimensional physical features is provided, for executing the steps of the key generation method based on multidimensional physical features described in the first aspect at a sensing terminal and a power distribution terminal, respectively, including:

[0080] The multidimensional feature acquisition and alignment module is used to generate an original feature sequence including multidimensional features based on multiple pre-acquired physical features;

[0081] The multidimensional weighted difference module is used to perform discontinuous difference and normalization on the multidimensional features in the original feature sequence to obtain the normalized difference sequence of each dimension feature. The Shannon information entropy of the normalized difference sequence of each dimension feature is calculated. The normalized difference sequence of the multidimensional features is weighted and summed according to the Shannon information entropy to obtain a comprehensive difference feature sequence including several samples.

[0082] The MD-LCDF quantization module is used to calculate the empirical cumulative distribution function based on the comprehensive differential feature sequence, determine the quantization interval and the discard interval according to the empirical cumulative distribution function and the principle of equal probability, allocate Gray code bits according to the quantization interval to which the valid samples belong, obtain the initial key bit stream of the sensing terminal at the sensing terminal, and obtain the initial key bit stream of the power distribution terminal at the power distribution terminal; wherein, the samples in the union of the discard set of the sensing terminal and the discard set of the power distribution terminal are discarded, and the remaining samples are regarded as valid samples, and the discard set includes the samples located within the discard interval.

[0083] The frame-aware RSC negotiation module is used to perform RS encoding and convolutional encoding on the initial key bit stream of the sensing terminal at the sensing terminal to obtain parity check symbols and convolutional check bits; and to perform Viterbi decoding and RS decoding on the initial key bit stream of the distribution terminal at the distribution terminal. During the decoding process, the initial key bit stream of the distribution terminal is corrected based on the obtained parity check symbols and convolutional check bits to obtain the negotiation key; the negotiation key at the sensing terminal is the same as the negotiation key at the distribution terminal.

[0084] The privacy amplification module is used to perform a hash operation on the negotiated key to obtain the working key.

[0085] Thirdly, a key generation apparatus based on multidimensional physical characteristics is provided, comprising a processor and a storage medium; the storage medium is used to store instructions; the processor is used to perform operations according to the instructions to execute the steps of the method described in the first aspect.

[0086] Fourthly, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps of the method described in the first aspect.

[0087] Beneficial effects

[0088] This invention provides a key generation method based on multidimensional physical features. The method adaptively weights and fuses multidimensional physical features into a comprehensive differential feature sequence based on normalized Shannon information entropy. The comprehensive differential feature sequence is then quantized using an empirical cumulative distribution function to generate an initial key bit stream. Key negotiation is performed on the initial key bit stream, and privacy amplification is applied to the negotiated key to obtain the working key. This method can fuse multidimensional physical features, adapt to low-mobility and high-interference power scenarios, and seamlessly integrate with distribution automation communication protocols. Attached Figure Description

[0089] Figure 1 This is a flowchart illustrating a key generation method based on multidimensional physical features in an embodiment of the present invention. Detailed Implementation

[0090] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0091] Example 1

[0092] This embodiment provides a key generation method based on multi-dimensional physical features. For example... Figure 1 As shown, the process is executed separately in the sensing terminal (ED) and the distribution terminal (DTU). The sensing terminal (ED) integrates a LoRa RF chip and a current transformer sampling circuit; the aggregation master station (GW) is integrated into the distribution terminal (DTU) and communicates with the master station server via the IEC 60870-5-104 protocol. The steps include:

[0093] Step 1: Generate an original feature sequence including multidimensional features based on multiple pre-acquired physical features.

[0094] The sensing terminal initiates an uplink LoRa frame every sampling period, and the aggregation master station responds with a downlink frame within the receiving window, completing one bidirectional frame exchange. Therefore, the sensing terminal and the power distribution terminal acquire the same physical characteristics.

[0095] The multidimensional physical characteristics include wireless received signal strength, harmonic noise index, and electromagnetic disturbance index. The harmonic noise index is obtained by performing a Fast Fourier Transform on the signal acquired by the current transformer, calculating the base-10 logarithm of the ratio of the sum of the 2nd to 25th harmonic powers to the fundamental power. The electromagnetic disturbance index is obtained by the wireless RF chip switching to energy detection mode during frame intervals, scanning the background electromagnetic energy within a specified bandwidth near the operating carrier frequency, and calculating the standard deviation of the resulting energy sequence.

[0096] The harmonic noise index is expressed as:

[0097] ;

[0098] in, Indicates the sensor terminal and the power distribution terminal. Harmonic noise figure of the second exchange Indicates the current transformer number During the second sampling The power of the subharmonics Indicates fundamental frequency power. This indicates the total number of exchanges between the sensor terminal and the power distribution terminal.

[0099] The electromagnetic disturbance index is expressed as:

[0100] ;

[0101] in, Indicates the sensor terminal and the power distribution terminal. Electromagnetic disturbance index of the second exchange. This indicates that the RF front-end, during the frame interval, transmits the first signal within a specified bandwidth near the center frequency. Energy received at each scan point This represents the average received energy during the scan. Indicates the number of scan points.

[0102] The original feature sequence is represented as follows:

[0103] ;

[0104] in, Indicates the sensor terminal and the power distribution terminal. The original feature sequence after the second exchange. Sensing terminals and power distribution terminals The strength of the wireless received signal during the second exchange. This indicates transpose.

[0105] If the sensing terminal does not receive a downlink frame from the aggregation master station in either the receiving window RX1 or RX2, it sends a frame carrying the sequence number to the aggregation master station. The flag frame. After the main station extracts the sequence number carried in the flag frame, it sets the location... Marked as a lost frame. Linear interpolation compensation is applied to the original feature sequence of any lost frame, expressed as:

[0106] ;

[0107] in, Indicates the sensor terminal and the power distribution terminal. The original feature sequence after the second exchange. Indicates the sensor terminal and the power distribution terminal. The original feature sequence after the second exchange.

[0108] Step 2: Perform discontinuous differencing and normalization on the multidimensional features in the original feature sequence to obtain the normalized difference sequence of each feature dimension. Calculate the Shannon information entropy of the normalized difference sequence of each feature dimension. Based on the Shannon information entropy, perform weighted summation on the normalized difference sequence of the multidimensional features to obtain a comprehensive difference feature sequence including several samples.

[0109] Discontinuous differencing is performed on the multidimensional features in the original feature sequence, as follows:

[0110] ;

[0111] ;

[0112] ;

[0113] in, Indicates the sensor terminal and the power distribution terminal. The difference in the strength of the wireless received signal after each exchange. Indicates the number of differential execution steps. , Sensing terminals and power distribution terminals The strength of the wireless received signal during the second exchange. Indicates the sensor terminal and the power distribution terminal. The difference in the harmonic noise figure of the second exchange. Indicates the sensor terminal and the power distribution terminal. Harmonic noise figure of the second exchange Indicates the sensor terminal and the power distribution terminal. The difference in the electromagnetic disturbance index of the second exchange. Indicates the sensor terminal and the power distribution terminal. Electromagnetic disturbance index of the second exchange.

[0114] The number of effective samples after differencing is To unify the dimensions, the difference values ​​of each feature dimension are normalized, and expressed as:

[0115] ;

[0116] ;

[0117] ;

[0118] in, Indicates the sensor terminal and the power distribution terminal. The normalized difference value of the wireless received signal strength after the second exchange The mean of the differences in the strength of the wireless received signal. The standard deviation of the difference values ​​representing the strength of the wireless received signal. Indicates the sensor terminal and the power distribution terminal. The normalized difference of the harmonic noise figure of the second exchange. The mean of the differences in the harmonic noise figure. The standard deviation of the difference values ​​of the harmonic noise figure. Indicates the sensor terminal and the power distribution terminal. The normalized difference of the electromagnetic disturbance index of the second exchange. The mean of the differences in the electromagnetic disturbance index. The standard deviation of the difference value of the electromagnetic disturbance index.

[0119] According to pre-acquired The probability density histograms of equal-width intervals are used to calculate the Shannon information entropy of the normalized difference sequences of each feature dimension, expressed as:

[0120] ;

[0121] in, Representation of features Shannon information entropy of normalized difference sequences Representation of features The normalized difference sequence falls into the first The probability of each interval. Indicates the total number of intervals. Indicates the strength of the wireless received signal. Indicates the harmonic noise figure. This indicates the electromagnetic disturbance index.

[0122] The weights of the normalized difference sequences for each feature dimension are calculated and expressed as follows:

[0123] ;

[0124] ;

[0125] in, Representation of features The weights of the normalized difference sequence, The Shannon information entropy represents the normalized difference sequence of the wireless received signal strength. The Shannon information entropy representing the normalized difference sequence of the harmonic noise index. The Shannon information entropy represents the normalized difference sequence of the electromagnetic disturbance index. The weights represent the normalized difference sequence of the wireless received signal strength. The weights of the normalized difference sequence representing the harmonic noise index. The weights represent the normalized difference sequence of the electromagnetic disturbance index.

[0126] The samples in the comprehensive differential feature sequence are represented as follows:

[0127] ;

[0128] in, Indicates the sensor terminal and the power distribution terminal. The sample corresponding to each exchange.

[0129] Step 3: Calculate the empirical cumulative distribution function based on the comprehensive differential feature sequence, determine the quantization interval and the discard interval according to the empirical cumulative distribution function and the principle of equal probability, allocate Gray code bits according to the quantization interval to which the effective sample belongs, and obtain the initial key bit stream of the sensing terminal at the sensing terminal and the initial key bit stream of the power distribution terminal at the power distribution terminal.

[0130] The empirical cumulative distribution function is calculated as follows:

[0131] ;

[0132] in, Represents the empirical cumulative distribution function. This represents an arbitrary point in the comprehensive difference feature sequence. express The value is 1 if it is true, otherwise it is 0.

[0133] The boundary of the quantization interval is determined according to the principle of equal probability, and is expressed as:

[0134] ;

[0135] in, Indicates the first The boundaries of a quantization interval, This represents the pseudo-inverse of the empirical distribution function. This indicates the total number of quantization intervals.

[0136] The quantization interval is represented as The half-interval width of the quantization interval is calculated and expressed as:

[0137] ;

[0138] in, Indicates the first The width of half interval of each quantization interval. Indicates the first The boundaries of a quantization interval, Indicates the first The boundaries of each quantization interval.

[0139] For each boundary, the discard interval is represented as:

[0140] ;

[0141] in, Indicates the first The discard interval at the boundary of each quantization interval. Indicates that the parameter is discarded. To discard parameters, Do not discard any samples. Discard all samples at that time. Indicates the first The width of half interval of each quantization interval. Indicates the first The width of half of each quantization interval.

[0142] The samples in the union of the discard sets of the sensor terminals and the power distribution terminals are discarded, and the remaining samples are considered valid samples, ensuring that the quantized sample sets of the sensor terminals and the power distribution terminals are completely consistent. The discard set includes samples located within the discard interval. The process of obtaining the discard sets of the sensor terminals and the power distribution terminals includes: adding samples located within the discard interval to the discard set of the sensor terminal or the power distribution terminal.

[0143] Gray code bits are allocated according to the quantization interval to which the valid samples belong. The lengths of the initial key bit stream of the sensing terminal and the initial key bit stream of the power distribution terminal are expressed as follows:

[0144] ;

[0145] in, Indicates the length of the initial key bit stream of the sensing terminal and the initial key bit stream of the power distribution terminal. This represents the union of the discard sets of sensor terminals and the discard sets of power distribution terminals.

[0146] Step 4: Generate negotiation keys in both the sensing terminal and the power distribution terminal.

[0147] (1) Perform RS encoding and convolution encoding on the initial key bit stream of the sensing terminal at the sensing terminal to obtain the parity check symbol and convolution check bit.

[0148] Due to the non-complete reciprocity of the channel, there are a small number of mismatched bits between the initial key bit stream of the sensing terminal and the initial key bit stream of the distribution terminal. This mismatch is measured by the key inconsistency rate and expressed as:

[0149] ;

[0150] in, Indicates the key inconsistency rate. The first bit of the initial key bit stream of the sensing terminal bits, The first bit of the initial key bit stream of the power distribution terminal 1 bit.

[0151] Zero padding is performed when the initial key bit stream of the sensing terminal is less than 385 bits.

[0152] RS encoding is performed on the initial key bitstream of the sensing terminal to obtain the first codeword and parity check symbol, which are represented as follows:

[0153] ;

[0154] ;

[0155] ;

[0156] ;

[0157] ;

[0158] in, Indicates the codeword length. Indicates message length. Indicates the number of bits per symbol. Indicates the number of correctable sign errors. Indicates the target key length. Indicates the first character. This represents the parity check symbol.

[0159] In this embodiment , , , It can correct up to 36 symbol errors, each group requires Bit input. To ensure sufficient information-theoretic security of the 128-bit working key output after privacy amplification, the target key length is chosen to be... Bits, in conjunction with the parameters of the embodiment The constraints are satisfied.

[0160] The first codeword is convolutionally encoded to obtain the convolutional parity bit.

[0161] Parity symbols and convolutional parity bits are encapsulated as parity information in an application service data unit conforming to the IEC 60870-5-104 specification and transmitted to the distribution terminal along with the telemetry message. The total additional bytes are approximately 175 bytes, which are aligned with the maximum frame length of 255 bytes of ASDU and do not require fragmentation.

[0162] (2) Viterbi decoding and RS decoding are performed on the initial key bit stream of the power distribution terminal at the power distribution terminal. During the decoding process, the initial key bit stream of the power distribution terminal is corrected according to the obtained parity check symbol and convolution check bit to obtain the negotiated key.

[0163] The second codeword is obtained by performing RS encoding on the initial key bitstream of the power distribution terminal.

[0164] Add convolutional parity bits to the end of the second codeword to obtain a concatenated codeword, and perform Viterbi decoding on the concatenated codeword to obtain the error-corrected RS codeword.

[0165] After replacing the parity part of the error correction RS codeword with the parity check symbol, RS decoding is performed to obtain the negotiation key.

[0166] The negotiation key of the sensing terminal is the same as the negotiation key of the power distribution terminal. .

[0167] Step 5: Perform a hash operation on the negotiated key to obtain the working key.

[0168] In this embodiment, SHA3-256 hashing is performed, and the first 128 bits are extracted as the AES-128 working key. This eliminates some prior information introduced by the leakage of verification information during the key negotiation process, and is represented as follows:

[0169] ;

[0170] in, Indicates the working key. This represents the negotiation key.

[0171] Example 2

[0172] This embodiment provides a key generation system based on multidimensional physical features, used to execute the steps of the key generation method based on multidimensional physical features described in Embodiment 1 at both a sensing terminal and a power distribution terminal, including:

[0173] The multidimensional feature acquisition and alignment module is used to generate an original feature sequence including multidimensional features based on multiple pre-acquired physical features;

[0174] The multidimensional weighted difference module is used to perform discontinuous difference and normalization on the multidimensional features in the original feature sequence to obtain the normalized difference sequence of each dimension feature. The Shannon information entropy of the normalized difference sequence of each dimension feature is calculated. The normalized difference sequence of the multidimensional features is weighted and summed according to the Shannon information entropy to obtain a comprehensive difference feature sequence including several samples.

[0175] The MD-LCDF quantization module is used to calculate the empirical cumulative distribution function based on the comprehensive differential feature sequence, determine the quantization interval and the discard interval according to the empirical cumulative distribution function and the principle of equal probability, allocate Gray code bits according to the quantization interval to which the valid samples belong, obtain the initial key bit stream of the sensing terminal at the sensing terminal, and obtain the initial key bit stream of the power distribution terminal at the power distribution terminal; wherein, the samples in the union of the discard set of the sensing terminal and the discard set of the power distribution terminal are discarded, and the remaining samples are regarded as valid samples, and the discard set includes the samples located within the discard interval.

[0176] The frame-aware RSC negotiation module is used to perform RS encoding and convolutional encoding on the initial key bit stream of the sensing terminal at the sensing terminal to obtain parity check symbols and convolutional check bits; and to perform Viterbi decoding and RS decoding on the initial key bit stream of the distribution terminal at the distribution terminal. During the decoding process, the initial key bit stream of the distribution terminal is corrected based on the obtained parity check symbols and convolutional check bits to obtain the negotiation key; the negotiation key at the sensing terminal is the same as the negotiation key at the distribution terminal.

[0177] The privacy amplification module is used to perform a hash operation on the negotiated key to obtain the working key.

[0178] Example 3

[0179] This embodiment provides a key generation device based on multi-dimensional physical features, including a processor and a storage medium. The storage medium is used to store instructions. The processor is used to perform operations according to the instructions to execute the steps of the key generation method based on multi-dimensional physical features described in Embodiment 1.

[0180] Example 4

[0181] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the key generation method based on multidimensional physical features described in Embodiment 1.

[0182] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0183] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0184] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0185] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0186] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A key generation method based on multidimensional physical features, characterized in that, Executed separately at the sensing terminal and the power distribution terminal; The key generation method includes: Generate an original feature sequence including multidimensional features based on multiple pre-acquired physical features; The multidimensional features in the original feature sequence are subjected to discontinuous differencing and normalization respectively to obtain the normalized difference sequence of each feature. The Shannon information entropy of the normalized difference sequence of each feature is calculated. The normalized difference sequence of the multidimensional features is weighted and summed according to the Shannon information entropy to obtain a comprehensive difference feature sequence including several samples. The empirical cumulative distribution function is calculated based on the comprehensive differential feature sequence. The quantization interval and the discard interval are determined according to the empirical cumulative distribution function and the principle of equal probability. Gray code bits are allocated according to the quantization interval to which the valid samples belong. The initial key bit stream of the sensing terminal is obtained at the sensing terminal and the initial key bit stream of the power distribution terminal is obtained at the power distribution terminal. Among them, the samples in the union of the discard set of the sensing terminal and the discard set of the power distribution terminal are discarded, and the remaining samples are regarded as valid samples. The discard set includes the samples located within the discard interval. At the sensing terminal, RS encoding and convolutional encoding are performed on the initial key bit stream to obtain parity check symbols and convolutional check bits. At the distribution terminal, Viterbi decoding and RS decoding are performed on the initial key bit stream, and during the decoding process, the initial key bit stream is corrected based on the obtained parity check symbols and convolutional check bits to obtain the negotiation key. The negotiation key at the sensing terminal is the same as the negotiation key at the distribution terminal. Perform a hash operation on the negotiated key to obtain the working key.

2. The key generation method based on multidimensional physical features according to claim 1, characterized in that, The step of generating an original feature sequence including multidimensional features based on multiple pre-acquired physical features includes: The multidimensional physical characteristics include wireless received signal strength, harmonic noise index, and electromagnetic disturbance index; The harmonic noise index is expressed as: ； in, Indicates the sensor terminal and the power distribution terminal. Harmonic noise figure of the second exchange Indicates the current transformer number During the second sampling The power of the subharmonics Indicates fundamental frequency power. This indicates the total number of exchanges between the sensing terminal and the power distribution terminal. The electromagnetic disturbance index is expressed as: ； in, Indicates the sensor terminal and the power distribution terminal. Electromagnetic disturbance index of the second exchange. This indicates that the RF front-end, during the frame interval, transmits the first signal within a specified bandwidth near the center frequency. Energy received at each scan point This represents the average received energy during the scan. Indicates the number of scan points; The original feature sequence is represented as follows: ； in, Indicates the sensor terminal and the power distribution terminal. The original feature sequence after the second exchange. Sensing terminals and power distribution terminals The strength of the wireless received signal during the second exchange. Indicates transpose; Linear interpolation compensation is applied to the original feature sequence of any lost frame, as follows: ； in, Indicates the sensor terminal and the power distribution terminal. The original feature sequence after the second exchange. Indicates the sensor terminal and the power distribution terminal. The original feature sequence after the second exchange.

3. The key generation method based on multidimensional physical features according to claim 2, characterized in that, The process involves performing discontinuous differencing and normalization on the multidimensional features in the original feature sequence to obtain normalized difference sequences for each dimension. The Shannon information entropy of each dimension's normalized difference sequences is calculated. Based on the Shannon information entropy, a weighted sum of the normalized difference sequences of the multidimensional features is obtained, resulting in a comprehensive difference feature sequence comprising several samples, including: Discontinuous differencing is performed on the multidimensional features in the original feature sequence, as follows: ； ； ； in, Indicates the sensor terminal and the power distribution terminal. The difference in the strength of the wireless received signal after each exchange. Indicates the number of differential execution steps. Sensing terminals and power distribution terminals The strength of the wireless received signal during the second exchange. Indicates the sensor terminal and the power distribution terminal. The difference in the harmonic noise figure of the second exchange. Indicates the sensor terminal and the power distribution terminal. Harmonic noise figure of the second exchange Indicates the sensor terminal and the power distribution terminal. The difference in the electromagnetic disturbance index of the second exchange. Indicates the sensor terminal and the power distribution terminal. Electromagnetic disturbance index of the second exchange; The difference values ​​of each feature dimension are normalized and expressed as follows: ； ； ； in, Indicates the sensor terminal and the power distribution terminal. The normalized difference value of the wireless received signal strength after the second exchange The mean of the differences in the strength of the wireless received signal. The standard deviation of the difference values ​​representing the strength of the wireless received signal. Indicates the sensor terminal and the power distribution terminal. The normalized difference of the harmonic noise figure of the second exchange. The mean of the differences in the harmonic noise figure. The standard deviation of the difference values ​​of the harmonic noise figure. Indicates the sensor terminal and the power distribution terminal. The normalized difference of the electromagnetic disturbance index of the second exchange. The mean of the differences in the electromagnetic disturbance index. The standard deviation of the difference values ​​in the electromagnetic disturbance index; Based on the pre-acquired probability density histogram, the Shannon information entropy of the normalized difference sequence of each feature dimension is calculated, and expressed as: ； in, Representation of features Shannon information entropy of normalized difference sequences Representation of features The normalized difference sequence falls into the first The probability of each interval. Indicates the total number of intervals. Indicates the strength of the wireless received signal. Indicates the harmonic noise figure. Indicates the electromagnetic disturbance index; The weights of the normalized difference sequences for each feature dimension are calculated and expressed as follows: ； ； in, Representation of features The weights of the normalized difference sequence, The Shannon information entropy represents the normalized difference sequence of the wireless received signal strength. The Shannon information entropy representing the normalized difference sequence of the harmonic noise index. The Shannon information entropy represents the normalized difference sequence of the electromagnetic disturbance index. The weights represent the normalized difference sequence of the wireless received signal strength. The weights of the normalized difference sequence representing the harmonic noise index. The weights represent the normalized difference sequence of the electromagnetic disturbance index; The samples in the comprehensive differential feature sequence are represented as follows: ； in, Indicates the sensor terminal and the power distribution terminal. The sample corresponding to each exchange.

4. The key generation method based on multidimensional physical features according to claim 3, characterized in that, The calculation of the empirical cumulative distribution function based on the comprehensive differential feature sequence, and the determination of the quantization interval and the discard interval based on the empirical cumulative distribution function and the principle of equal probability, include: The empirical cumulative distribution function is calculated as follows: ； in, Represents the empirical cumulative distribution function. This represents an arbitrary point in the comprehensive difference feature sequence. express The value is 1 if the condition is true, and 0 otherwise. The boundary of the quantization interval is determined according to the principle of equal probability, and is expressed as: ； in, Indicates the first The boundaries of a quantization interval, This represents the pseudo-inverse of the empirical distribution function. Indicates the total number of quantization intervals; The quantization interval is represented as The half-interval width of the quantization interval is calculated and expressed as: ； in, Indicates the first The width of half interval of each quantization interval. Indicates the first The boundaries of a quantization interval, Indicates the first The boundaries of each quantization interval; For each boundary, the discard interval is represented as: ； in, Indicates the first The discard interval at the boundary of each quantization interval. Indicates that the parameter is discarded. Indicates the first The width of half interval of each quantization interval. Indicates the first The width of half of each quantization interval.

5. The key generation method based on multidimensional physical features according to claim 1, characterized in that, The process of obtaining the discard set of the sensing terminal and the discard set of the power distribution terminal includes: At the sensing terminal or power distribution terminal, samples located within the discard interval are added to the discard set of the sensing terminal or power distribution terminal.

6. The key generation method based on multidimensional physical features according to claim 1, characterized in that, The step of performing RS encoding and convolutional encoding on the initial key bit stream of the sensing terminal to obtain parity check symbols and convolutional check bits includes: RS encoding is performed on the initial key bitstream of the sensing terminal to obtain the first codeword and parity check symbol, which are represented as follows: ； ； ； ； ； in, Indicates the codeword length. Indicates message length. Indicates the number of bits per symbol. Indicates the number of correctable sign errors. Indicates the target key length. Indicates the first character. Indicates the parity check symbol; The first codeword is convolutionally encoded to obtain the convolutional parity bit.

7. The key generation method based on multidimensional physical features according to claim 6, characterized in that, The process involves performing Viterbi decoding and RS decoding on the initial key bitstream of the power distribution terminal, and correcting errors in the initial key bitstream based on the obtained parity check symbol and convolutional parity bits during the decoding process to obtain the negotiated key, including: The second codeword is obtained by performing RS encoding on the initial key bitstream of the power distribution terminal; Add convolutional parity bits to the end of the second codeword to obtain the concatenated codeword, and perform Viterbi decoding on the concatenated codeword to obtain the error-corrected RS codeword; After replacing the parity part of the error correction RS codeword with the parity check symbol, RS decoding is performed to obtain the negotiation key.

8. A key generation system based on multidimensional physical characteristics, characterized in that, The steps for executing the key generation method based on multidimensional physical features as described in any one of claims 1 to 7 at the sensing terminal and the power distribution terminal respectively include: The multidimensional feature acquisition and alignment module is used to generate an original feature sequence including multidimensional features based on multiple pre-acquired physical features; The multidimensional weighted difference module is used to perform discontinuous difference and normalization on the multidimensional features in the original feature sequence to obtain the normalized difference sequence of each dimension feature. The Shannon information entropy of the normalized difference sequence of each dimension feature is calculated. The normalized difference sequence of the multidimensional features is weighted and summed according to the Shannon information entropy to obtain a comprehensive difference feature sequence including several samples. The MD-LCDF quantization module is used to calculate the empirical cumulative distribution function based on the comprehensive differential feature sequence, determine the quantization interval and the discard interval according to the empirical cumulative distribution function and the principle of equal probability, allocate Gray code bits according to the quantization interval to which the valid samples belong, obtain the initial key bit stream of the sensing terminal at the sensing terminal, and obtain the initial key bit stream of the power distribution terminal at the power distribution terminal; wherein, the samples in the union of the discard set of the sensing terminal and the discard set of the power distribution terminal are discarded, and the remaining samples are regarded as valid samples, and the discard set includes the samples located within the discard interval. The frame-aware RSC negotiation module is used to perform RS encoding and convolutional encoding on the initial key bit stream of the sensing terminal at the sensing terminal to obtain parity check symbols and convolutional check bits; and to perform Viterbi decoding and RS decoding on the initial key bit stream of the distribution terminal at the distribution terminal. During the decoding process, the initial key bit stream of the distribution terminal is corrected based on the obtained parity check symbols and convolutional check bits to obtain the negotiation key; the negotiation key at the sensing terminal is the same as the negotiation key at the distribution terminal. The privacy amplification module is used to perform a hash operation on the negotiated key to obtain the working key.

9. A key generation device based on multidimensional physical characteristics, characterized in that, Including processor and storage media; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the key generation method based on multidimensional physical features as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the key generation method based on multidimensional physical features as described in any one of claims 1 to 7.

Images