Discrete variable quantum key distribution data coordination method

Abstract

The invention belongs to the technical field of quantum communication, and particularly relates to a discrete variable quantum key distribution data coordination method and system based on a code rate adaptive multi-system LDPC code. Aiming at the defect of performance bottleneck of an existing discrete variable quantum key distribution system adopting a binary low-density parity check code in a long-distance and high-noise environment, the invention provides a discrete variable quantum key distribution data coordination method. The method comprises the following steps: preparing and transmitting a quantum state; screening data; error rate estimation and eavesdropping detection; data coordination based on symbol-level code rate adaptation and / or bit-level code rate adaptation; and confidentiality enhancement. According to the method disclosed by the invention, the construction and storage pressure of the system on error correction codes with different code rates is obviously reduced, the robustness of the system under a complex channel condition is enhanced, and efficient and unconditionally safe key distribution is ensured; the eavesdropping attack can be effectively resisted, and the communication security is ensured.

Description

[0001] Methodology

[0002] The present invention belongs to the field of quantum communication methods, and in particular relates to a discrete variable quantum key distribution data coordination method based on rate-adaptive multi-base LDPC codes. Background Art

[0003] Quantum Key Distribution (QKD) is a key distribution protocol implemented using the principles of quantum mechanics. Its core concept is to establish a shared key through the transmission of quantum states. Discrete variable quantum key distribution (DV-QKD), an important form of QKD, uses discrete variables (such as the polarization or phase of photons) to encode information, offering advantages such as relatively simple implementation and high security. Compared to continuous variable quantum key distribution (CV-QKD), DV-QKD has a higher degree of practical maturity in long-distance transmission and performs particularly well in backbone network environments. Its representative protocol, the BB84 protocol, has become the mainstream solution for QKD.

[0004] Despite significant progress in DV-QKD, its practical performance is still limited by the efficiency of data coordination in the post-processing stage. Data coordination requires efficient error correction within a limited number of interactions while also limiting information leakage to ensure security. Traditional coordination methods employ the multi-round Cascade protocol, whose interaction complexity is positively correlated with the quantum bit error rate (QBER), severely limiting the key generation rate over long distances and in high-loss scenarios.

[0005] The Forward Error Correction (FEC) method achieves error correction through a single round of interaction, making it the key to improving data coordination efficiency. Low-density parity-check codes (LDPC codes) have become the mainstream error correction scheme for QKD systems using FEC methods due to their error correction performance approaching the Shannon limit. However, traditional binary LDPC codes have two major bottlenecks: first, in actual environments where the channel QBER changes dynamically, the fixed code rate characteristic leads to a decrease in error correction efficiency, and the need to pre-store multiple sets of check matrices creates a storage burden; second, in high QBER scenarios caused by long-distance transmission, the symbol efficiency of binary coding is insufficient, and the error correction performance is significantly degraded. Summary of the Invention

[0006] In response to the performance bottleneck of existing discrete variable quantum key distribution systems using binary low-density parity-check codes over long distances and in high-noise environments, the present invention provides a discrete variable quantum key distribution data coordination method based on rate-adaptive multi-base LDPC codes. By introducing multi-base LDPC codes and their sum-product decoding algorithm, combined with a rate-adaptive mechanism, the error correction capability and system robustness are significantly improved.

[0007] To achieve the above objectives, the present invention adopts the following method scheme: a discrete variable quantum key distribution data coordination method, comprising:

[0008] Under the premise of reverse negotiation, the sender Alice prepares the original key data into a quantum state and transmits it to the receiver Bob through the quantum channel;

[0009] Alice purifies the original key data and obtains the associated data sequence shared with Bob after filtering;

[0010] Alice compares the statistical characteristics of the measurement results of the partial associated data sequence with Bob through the authentication channel and calculates the QBER of quantum key distribution;

[0011] Based on the QBER calculation results, Alice selects a multi-bit LDPC code with a code rate that matches Bob's. She then uses symbol-level and / or bit-level code rate adaptation methods to dynamically optimize the code rate, balancing efficiency and system overhead. She then uses syndrome decoding for error correction.

[0012] Alice selects the same hash function as Bob from the collision-resistant hash function family, and converts the original key of the coordinated error-corrected original key into the information-theoretic secure final key through a compression mapping.

[0013] As an improvement, Alice purifies the original key data through a dual filtering mechanism of bit filtering and base filtering, including:

[0014] Alice performs bit filtering on the raw data: invalid events where the detector did not respond are eliminated, and only the bit positions where both parties successfully detected are retained, forming a data sequence after preliminary filtering;

[0015] Alice compares its own encoding basis sequence with the measurement basis sequence disclosed by Bob through the authentication channel bit by bit, discarding bits with inconsistent bases. After basis screening, Alice and Bob share a set of basis-matched associated data sequences as input for subsequent processing.

[0016] As an improvement, the process of calculating the QBER of quantum key distribution includes:

[0017] Alice randomly selects some bits from the associated data sequence after base filtering as test samples and publishes the corresponding bit values. Alice's public values are compared with the bit values at the corresponding positions of Bob, and the number of inconsistent bits is counted to calculate the QBER. If the QBER exceeds the preset security threshold, it is determined that eavesdropping has occurred or the channel noise is too high, and the protocol is terminated and restarted. If the QBER is lower than the threshold, the next step is entered.

[0018] After completing the QBER comparison, Alice removes the published test sample data, and the remaining unpublished basis matching data is used for subsequent key negotiation and confidentiality enhancement.

[0019] As an improvement, the process of the symbol-level rate adaptation method includes:

[0020] The symbols of the multi-ary LDPC code are punctured and / or shortened as a whole. The formula for calculating the adjusted code rate is:

[0021]

[0022] Where R = K / N, which is the original code rate, K is the length of the information bit when the code rate adaptation method is not used, N is the code length of the multi-ary LDPC code, s is the number of shortened symbols, and p is the number of punctured symbols. The process of the bit-level code rate adaptation method includes:

[0023] Let the finite field GF(Q), Q=2 q The multi-ary symbol is split into q bits, and fine bit rate adjustment is achieved by puncturing and / or shortening the bit-level data. The formula for calculating the adjusted bit rate is:

[0024]

[0025] Where s is the number of shortened symbols, p is the number of punctured symbols, b is the number of shortened bits per symbol, c is the number of punctured bits per symbol, and R = K / N is the original bit rate before using the rate adaptation method.

[0026] Alice selects a symbol-level and / or bit-level solution based on system overhead requirements and determines related parameters.

[0027] As an improvement, Bob adjusts the data at the symbol level and / or bit level according to the parameters determined by the code rate adaptation method to optimize the code rate to achieve the best coordination efficiency; specifically,

[0028] In symbol-level processing, Bob determines the number of shortened symbols, s, and the number of punctured symbols, p, and randomly generates s shortened symbols and p punctured symbols, inserting them into designated positions in the original data sequence. Bob then sends the insertion positions and related parameters to Alice over an authenticated channel. Upon receiving this information, Alice inserts a shortened symbol with the same value as Bob's at the same position and fills the punctured symbol position with a random value. After these operations, Alice and Bob generate new data sequences, X' and Y', respectively, both with lengths N' = N + s + p.

[0029] In bit-level processing, Bob further determines the number of shortened bits, b, and the number of punctured bits, c, randomly generates b shortened bits, and inserts them into specified positions in the original bit stream. Simultaneously, Bob selects c punctured bit positions and fills these positions with random bit values. Subsequently, Bob sends the bit-level insertion positions, b and c, and the shortened bit values to Alice via an authenticated channel. Upon receiving this information, Alice inserts the same shortened bits as Bob at the same bit positions and fills the punctured positions with random bits. Finally, Alice and Bob generate new bit sequences, X' and Y', respectively, both of length N' = N + (s·b + p·c) / log2 Q.

[0030] As an improvement, the syndrome decoding method is as follows: Alice receives the syndrome of the data calculated by Bob through an authenticated channel, and Alice locates and corrects the error bits through iterative verification and side information exchange.

[0031] As an improvement, the process of using the syndrome decoding method to correct data errors includes:

[0032] Bob uses the parity check matrix of the multi-ary LDPC code and the data Y' to calculate the syndrome S. The formula of syndrome S is expressed as:

[0033] S=Y'H T

[0034] Where H T is the transpose of the parity check matrix of the multi-ary LDPC code;

[0035] Then the syndrome S is sent to Alice as side information to assist Alice in correcting the data X'.

[0036] Alice uses the multi-ary belief propagation decoding algorithm to correct the data X' based on the syndrome S.

[0037] As an improvement, Alice selects the same hash function from the collision-resistant hash function family and converts the original key of the coordinated error-corrected original key into the final key with information-theoretic security through a compression mapping, including:

[0038] Alice chooses a suitable hash function G to compress the partial information L obtained by Eve into a very small security parameter, making it impossible for Eve to extract useful information from it.

[0039] By using the hashed compressed information, Alice generates a final key sequence that is completely consistent with Bob's and is unconditionally secure in information theory.

[0040] A discrete variable quantum key distribution data coordination method, comprising:

[0041] Under the premise of reverse negotiation, the receiver Bob receives the quantum state key data prepared by the sender Alice and transmitted through the quantum channel, and obtains the corresponding original key data through measurement;

[0042] Bob purifies the original key data and obtains the associated data sequence that is shared with Alice after screening;

[0043] Bob compares the statistical characteristics of the measurement results of part of the associated data sequence with Alice through the authentication channel to calculate the QBER of quantum key distribution;

[0044] Based on the QBER calculation results, Bob selects a rate-adaptive multi-ary LDPC code and uses symbol-level and / or bit-level rate adaptation methods for dynamic optimization, balancing efficiency and system overhead. He then uses syndrome decoding for error correction:

[0045] Bob selects the same hash function as Alice from the collision-resistant hash function family, and converts the original key of the coordinated error-corrected original key into the final key that is information-theoretically secure through a compression mapping.

[0046] As an improvement, the process by which Bob obtains the corresponding original key data through measurement includes:

[0047] After receiving the quantum state key data, Bob randomly selects a measurement basis to measure each quantum state and records the measurement results through a single-photon detector. Due to the randomness of the measurement basis, Bob only obtains valid bits when it is consistent with Alice's coding basis.

[0048] The present invention provides a discrete variable quantum key distribution data coordination method based on rate-adaptive multi-level LDPC codes. By introducing multi-level LDPC codes and their sum-product decoding algorithms, and combining them with a rate-adaptive mechanism, the rate-adaptive mechanism includes symbol-level and / or bit-level rate-adaptive methods. Coarse-grained rate adjustment is achieved at the symbol level, while fine-grained adjustment is achieved at the bit level. Coordination efficiency is balanced with system overhead, significantly improving error correction capability and system robustness. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 is a flow chart of a data coordination method according to an embodiment of the present invention.

[0050] Figure 2 This is a schematic diagram of the specific implementation process of the data coordination method according to an embodiment of the present invention.

[0051] Figure 3 This is a schematic diagram of the process of constructing a new sequence using symbol-level and bit-level code rate adaptation methods in the data coordination method of an embodiment of the present invention.

[0052] Figure 4Schematic diagram of the symbol-level and bit-level code rate adaptation method of the data coordination method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The following is an explanation and description of the method scheme of the present invention, but the following embodiments are only preferred embodiments of the present invention, not all. Based on the embodiments in the embodiments, other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0054] See also Figures 1 to 4 The present invention provides a data coordination method for quantum key distribution based on rate-adaptive multi-ary LDPC codes. This method is applicable to both forward coordination and reverse coordination. Taking reverse coordination as an example, the specific process of the data coordination method for quantum key distribution based on rate-adaptive multi-ary LDPC codes in the present invention is as follows:

[0055] Step S1: During quantum key distribution, Alice first encodes a true random number sequence into a quantum state using a quantum state preparation device. The prepared quantum state is then transmitted to Bob via a quantum channel (e.g., an optical fiber or free-space channel). Bob then randomly selects a measurement basis set according to the protocol and measures the received quantum state using a single-photon detector or other device, ultimately obtaining the corresponding original key data.

[0056] In step S2, the data screening phase purifies the original key data through a dual filtering mechanism of bit screening and basis screening. Bit screening is responsible for identifying and eliminating abnormal data units caused by transmission errors or device noise, ensuring the physical integrity of the data units. Basis screening, based on the quantum measurement basis comparison results pre-negotiated by both parties, screens out data fragments with consistent measurement basis and removes invalid data caused by basis mismatch. The synergistic effect of the two can significantly improve data consistency, build a low-noise, high-reliability data foundation for subsequent steps, and effectively reduce the system's quantum bit error rate (QBER).

[0057] In step S3, Alice and Bob compare the statistical properties of the measurement results of the partially correlated data sequence over the public channel and calculate the QBER of the quantum key distribution. This QBER is caused by quantum channel noise, bias in Bob's random basis selection, and interference from Eve, a potential eavesdropper. If the QBER exceeds a preset security threshold, the channel is deemed eavesdropping-risky and the communication process is terminated. If it falls below the threshold, the channel is deemed secure and subsequent data reconciliation and confidentiality enhancement operations proceed. This process provides a dynamic, quantitative assessment of system security.

[0058] Step S4: Based on the QBER calculation results, both parties select a rate-adaptive multi-ary LDPC code and utilize symbol-level or bit-level rate adaptation methods to dynamically optimize the balance between coordination efficiency and system overhead. Syndrome decoding is then used for error correction: Bob calculates the syndrome of the data and sends it to Alice via an authenticated channel. Both parties locate and correct erroneous bits through iterative verification and side information exchange. This process achieves efficient error correction with limited communication overhead, ensuring complete consistency of the data shared by both parties.

[0059] Step S5: To eliminate information that could be obtained by a potential eavesdropper, Eve, via the quantum or classical channel, Alice and Bob perform a confidentiality enhancement operation. Both parties select the same hash function from a family of collision-resistant hash functions and transform the original key, after coordinated error correction, into an information-theoretically secure final key using a compression mapping. By controlling the compression ratio using a preset security parameter, the amount of Eve's information can be compressed to a negligible level, ultimately generating an unconditionally secure key sequence that meets the confidentiality requirements of quantum communication.

[0060] In this embodiment, step S1 includes:

[0061] Step S11: Alice randomly selects a coding basis, encodes the key bits generated by the true random number into the quantum state (such as the polarization state or phase state) of the photon through the quantum modulator, and sends the modulated quantum state sequence X0 to Bob through the quantum channel;

[0062] Step S12: After receiving the quantum state, Bob randomly selects a measurement basis to measure each quantum state and records the measurement result Y0 through a single-photon detector. Due to the randomness of the measurement basis, Bob obtains valid bits only when it is consistent with Alice's coding basis.

[0063] In this embodiment, in step S2, Alice and Bob purify the original key data X0 and Y0 using a dual filtering mechanism of bit filtering and base filtering, including:

[0064] Step S21: Both parties first perform bit screening on the original data: invalid events where the detectors did not respond (such as the case where no photons were detected due to channel loss) are eliminated, and only the bit positions that were successfully detected by both parties are retained, forming the data sequences X0′ and Y0′ after preliminary screening;

[0065] In step S22, Bob discloses his measurement basis selection sequence through the authentication channel. Alice compares its own encoding basis sequence with Bob's measurement basis sequence bit by bit, discarding bits where the bases are inconsistent. After basis selection, Alice and Bob share a set of basis-matched associated data sequences (denoted as X and Y) as input for subsequent processing.

[0066] In this embodiment, step S3 is bit error rate estimation and eavesdropping detection, including:

[0067] Step S31: Alice randomly selects some bits from the associated data sequences X and Y after base filtering as test samples and publishes their corresponding bit values. Bob compares the bit values at the corresponding positions with Alice's public values, counts the number of inconsistent bits, and calculates the QBER. If the QBER exceeds the preset security threshold, it is determined that eavesdropping has occurred or the channel noise is too high, and the protocol is terminated and restarted. If the QBER is lower than the threshold, the protocol proceeds to the subsequent steps.

[0068] Step S32: After the QBER comparison (ie, eavesdropping detection) is completed, both parties remove the published test sample data, and the remaining unpublished basis matching data (denoted as X and Y) is used for subsequent key negotiation and confidentiality enhancement.

[0069] In this embodiment, step S4 includes: both parties select a multi-ary LDPC code with a suitable code rate based on the QBER estimation result, and dynamically adjust the code rate through a symbol-level or bit-level code rate adaptation method to balance coordination efficiency and system overhead.

[0070] The process of the symbol-level rate adaptation method includes:

[0071] The symbols of the multi-ary LDPC code are punctured (p symbols) and / or shortened (s symbols). The formula for calculating the adjusted code rate is:

[0072]

[0073] Here, R = K / N is the original code rate, K is the length of the information bit when the code rate adaptation method is not used, N is the code length of the multi-binary LDPC code, s is the number of shortened symbols, and p is the number of punctured symbols. The symbol-level code rate adaptation method is simple to implement, but the code rate adjustment granularity is coarse.

[0074] The process of the bit-level rate adaptation method includes:

[0075] Let the finite field GF(Q), Q=2 q The multi-ary symbol is split into q bits, and fine bit rate adjustment is achieved by puncturing and / or shortening the bit-level data. The formula for calculating the adjusted bit rate is:

[0076]

[0077] Where s is the number of shortened symbols, p is the number of punctured symbols, b is the number of shortened bits per symbol, c is the number of punctured bits per symbol, R = K / N is the original bit rate before using the rate adaptation method; the bit-level rate adaptation method is slightly more complex, but has better error correction performance. Both parties choose the symbol-level (coarse-grained) or bit-level (fine-grained) solution based on the system overhead requirements and determine parameters such as s and p. The puncturing and / or shortening process can be found in [1]. Figure 3 .

[0078] In this embodiment, in step S4, Bob adjusts the data at the symbol level and / or bit level according to the parameters determined by the code rate adaptation method to optimize the code rate to achieve the best coordination efficiency. Specifically,

[0079] First, during symbol-level processing, Bob determines the number of shortened symbols, s, and the number of punctured symbols, p. He then randomly generates s shortened symbols and p punctured symbols and inserts them into designated positions in the original data sequence. The insertion positions can be random or selected according to a predefined rule (e.g., uniform distribution). Bob then sends the insertion positions and related parameters (e.g., the shortened symbol value) to Alice over an authenticated channel. Upon receiving this information, Alice inserts a shortened symbol with the same value as Bob's at the same position and pads the punctured symbol positions with random values. After these operations, Alice and Bob generate new data sequences, X' and Y', respectively, each with a length of N' = N + s + p. The new sequences are identical at the shortened symbol positions, but may differ at other positions due to channel noise or eavesdropping. By synchronizing the shortened symbols, Alice and Bob can use these known symbols as "anchor points" in subsequent error correction, thereby improving error correction efficiency. The random padding of punctured symbols is used to adjust the code rate without affecting the consistency of the final key.

[0080] In bit-level processing, Bob further determines the number of shortened bits b and the number of punctured bits c, randomly generates b shortened bits and inserts them into the specified position of the original bit stream. The insertion position can be determined by uniform distribution or random selection. At the same time, Bob selects c punctured bit positions and fills them with random bit values. Subsequently, Bob sends the bit-level insertion position, b, c, and shortened bit value to Alice through the authentication channel. After receiving this information, Alice inserts the same shortened bit as Bob at the same bit position and fills the punctured position with random bits. Finally, Alice and Bob generate new bit sequences X' and Y', respectively, whose lengths are both

[0081] N' = N + (s·b + p·c) / log2 Q. (For ease of description, the new sequence after rate adaptation uses the same notation as the symbol-level adaptation method.) The new bit sequence is strictly consistent at the shortened bit positions, serving as "bit-level anchors" for error correction. Random padding at the puncture positions does not affect the consistency of the final key, while the remaining bits may differ due to noise or eavesdropping.

[0082] The introduction of bit-level rate adaptation technology enables more refined rate adjustment, enabling finer-grained adaptation to channel conditions. For example, in high-frequency noisy channels, error correction capabilities can be enhanced by increasing the ratio of shortened bits. Furthermore, bit-level and symbol-level technologies can be combined to form a hybrid strategy. For example, in higher-layer protocols, symbol-level adjustments can be prioritized, and then bit-level optimization can be used to fine-tune the rate, further improving system performance. By using both symbol-level and bit-level anchors, Alice and Bob can significantly improve error correction efficiency in multi-granular error correction.

[0083] In this embodiment, in step S4, the process of using the syndrome decoding method to correct data errors includes:

[0084] Bob uses the parity check matrix of the multi-ary LDPC code and the data Y' to calculate the syndrome S. The syndrome S formula is expressed as:

[0085] S=Y'H T

[0086] Where H T is the transpose of the parity check matrix of the multi-ary LDPC code;

[0087] Then the syndrome S is sent to Alice as side information to assist Alice in correcting the data X'.

[0088] Alice uses the multi-ary belief propagation decoding algorithm based on the syndrome S to correct the data X'. The specific process includes:

[0089] First, Alice performs channel initialization according to the channel type. For a binary symmetric channel with a crossover probability of ε, the channel initialization probability information is calculated using the fast Fourier transform sum-product decoding algorithm as follows:

[0090]

[0091] Among them, p(y i |x i ) means the sending symbol is x i , the received symbol is y i The probability of x i Send symbol for bit i, y iis the received symbol corresponding to the i-th transmitted symbol, both of which are binary bit units, i∈[0,q·N * ],N * This is the sequence length held by Alice and Bob at this time. It is N when the rate adaptive method is not used, and N' when the rate adaptive method is used;

[0092] When the channel information is transmitted in the form of a binary bit stream, for the definition of GF(Q), Q=2 q The receiver generates symbol information on the finite field GF(Q) by symbolically mapping every q consecutive bits.

[0093] For a Q-ary symmetric channel with a symbol-level crossing probability of μ, the symbol probability is calculated as:

[0094]

[0095] Among them, p(n j |m j ) indicates that the symbol to be sent is m i , the received symbol is n i The probability of m i Send symbol for bit j, n i The received symbol corresponding to the j-th transmitted symbol, both of which are multi-ary symbol units, j∈[0,N * ]; At this time p(n j |m j ) directly corresponds to the symbolic information on the finite field GF(Q);

[0096] The information at all symbols together constitutes the initial message vector L of the decoding process, with a length of N * ;

[0097] Second, Alice starts the decoding process. The variable node passes the message to the check node and updates the probability by multiplying the tensors term by term, which is expressed as:

[0098]

[0099] Where ×. and ∏. are defined as the term-by-term product of tensors, represents the message passed from the variable node v to the replacement node p at the i-th iteration, Indicates the information transferred from the replacement node p to the variable node v' at the i-1th iteration.

[0100] Third, the permutation step in the decoding process:

[0101] In the check matrix of the multi-ary LDPC code, non-zero elements are used as replacement nodes, and the corresponding coefficients are added to the variable nodes to participate in the corresponding check equation; after the replacement, the probability value remains unchanged, but the arrangement order is rearranged according to the replacement result; finally, after the replacement process, is converted into That is, the information transmitted from the replacement node p to the check node c during the i-th iteration.

[0102] Fourth, the message update process at the verification node:

[0103] In traditional belief propagation algorithms, the message update process at the check node involves convolution operations, which require a lot of computation. Therefore, a fast Fourier transform (FFT) is used to convert the convolution operation into a simple product in the frequency domain. In this way, the message update process at the check node can be expressed as follows:

[0104]

[0105] Where, represents the information transmitted by the check node c to the replacement node p during the i-th iteration;

[0106] During the decoding process of a 4-ary LDPC code The 4-point FFT used can be exemplified as:

[0107] F 0 =[f 0 +f 1 ]+[f 2 +f 3 ]

[0108] F 1 =[f 0 -f 1 ]+[f 2 -f 3 ]

[0109] F 2 =[f 0 +f 1 ]-[f 2 +f 3 ]

[0110] F 3 =[f 0 -f 1 ]-[f 2 -f 3 ]

[0111] Where, f 0 ,f 1 ,f 2 ,f 3 Respectively The first to fourth items in .

[0112] Fifth, reverse permutation process:

[0113] This is the permutation step that the message goes through when it passes from the check node to the variable node. It is the same as the operation in the third one, but in the reverse order. is converted to

[0114] Update results Make a judgment, if the judgment result X1 is consistent with the syndrome S' (=X1H T ) is exactly the same as the syndrome S sent by Bob, then the decoding is completed and the decoding is declared successful; if they are different, it returns to the second step and continues the decoding process until the decoding is successful or the set maximum number of iterations is reached.

[0115] Specifically, Figure 4 For a finite field GF(Q), Q=2 4 The message update process at the check node when a multi-ary LDPC code with a symbol-level symbol of 16 is split into 4 bits using symbol-level and bit-level rate adaptation techniques. Symbol-level rate adaptation operates on the entire symbol, i.e., on all 4 bits of information simultaneously; bit-level rate adaptation processes partial bits within a symbol to achieve more granular rate adjustment.

[0116] In this embodiment, step S5 includes:

[0117] Step S51: Alice and Bob select a suitable hash function G to compress the partial information L obtained by Eve into a very small security parameter, so that Eve cannot extract any useful information from it.

[0118] Step S52: Alice and Bob generate a final key sequence that is completely consistent and unconditionally secure based on information theory through the hashed compressed information.

[0119] The present invention proposes a data coordination method for discrete variable quantum key distribution based on rate-adaptive multi-level LDPC codes. This method introduces multi-level LDPC codes and a sum-product decoding algorithm, and designs symbol-level and bit-level rate adaptation methods based on the characteristics of multi-level LDPC codes. This significantly reduces the construction and storage pressure of the DV-QKD system for error-correcting codes with different rates, enhances the system's robustness under complex channel conditions, and ensures efficient and unconditionally secure key distribution. Furthermore, the method uses fast Fourier transforms to accelerate the decoding process, reducing computational complexity and improving processing efficiency. A reasonable QBER calculation and data screening mechanism effectively resists eavesdropping attacks and ensures communication security. Combined with the confidentiality enhancement process of hash functions, this method further enhances key security, preventing eavesdroppers from obtaining useful information, and greatly improves the feasibility and security of quantum key distribution systems in practical applications. The method requires only a single basic parity check matrix to cover a continuous code rate range, reducing storage requirements. By dynamically adjusting the code rate, the system maintains high coordination efficiency in real-world environments with dynamically changing quantum bit error rates, thereby ensuring efficient and unconditionally secure key distribution under complex channel conditions.

[0120] The present invention also discloses a discrete variable quantum key distribution data coordination system, including a sender Alice and a receiver Bob. During key distribution, the system performs the following steps:

[0121] Step S1: Under the premise of reverse negotiation, the sender Alice prepares a quantum state and transmits it to the receiver Bob through a quantum channel. Bob obtains the corresponding original key data through measurement;

[0122] Step S2: Alice and Bob purify the original key data respectively to obtain a filtered associated data sequence;

[0123] Step S3: Alice and Bob compare the statistical characteristics of the measurement results of the partial correlation data sequence through the authentication channel and calculate the QBER of the quantum key distribution;

[0124] Step S4: Based on the QBER calculation result, both parties select a rate-adaptive multi-ary LDPC code and dynamically optimize it using symbol-level and / or bit-level rate adaptation methods to balance efficiency and system overhead; then, they use a syndrome decoding method for error correction.

[0125] Step S5: Both parties select the same hash function from the collision-resistant hash function family, and convert the original key of the coordinated error-corrected original key into the final key with information-theoretic security through compression mapping.

[0126] An embodiment of the present invention also discloses a computer device, including a processor and a storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by the processor, the aforementioned discrete variable quantum key distribution data coordination method is implemented.

[0127] An embodiment of the present invention also discloses a computer-readable storage medium having a computer program stored thereon. When the computer program is executed, the aforementioned discrete variable quantum key distribution data coordination method is implemented.

[0128] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Those familiar with the art should understand that the present invention includes, but is not limited to, the contents described in the above specific embodiments. Any modifications that do not deviate from the functional and structural principles of the present invention are intended to be included within the scope of the claims.

Claims

1. A discrete variable quantum key distribution data coordination method, characterized by: include: Under the premise of reverse negotiation, the sender Alice prepares the original key data into a quantum state and transmits it to the receiver Bob through the quantum channel; Alice purifies the original key data and obtains the associated data sequence shared with Bob after filtering; Alice compares the statistical characteristics of the measurement results of the partial associated data sequence with Bob through the authentication channel and calculates the QBER of quantum key distribution; Based on the QBER calculation results, Alice selects a multi-bit LDPC code with a code rate that matches Bob's. She then uses symbol-level and / or bit-level code rate adaptation methods to dynamically optimize the code rate, balancing efficiency and system overhead. She then uses syndrome decoding for error correction. Alice selects the same hash function as Bob from the collision-resistant hash function family, and converts the original key of the coordinated error-corrected original key into the information-theoretic secure final key through a compression mapping.

2. The discrete variable quantum key distribution data coordination method according to claim 1, characterized in that: Alice purifies the original key data through a dual filtering mechanism of bit filtering and base filtering, including: Alice performs bit filtering on the raw data: invalid events where the detector did not respond are eliminated, and only the bit positions where both parties successfully detected are retained, forming a data sequence after preliminary filtering; Alice compares its own encoding basis sequence with the measurement basis sequence disclosed by Bob through the authentication channel bit by bit, discarding bits with inconsistent bases. After basis screening, Alice and Bob share a set of basis-matched associated data sequences as input for subsequent processing.

3. The discrete variable quantum key distribution data coordination method according to claim 2, characterized in that: The process of calculating the QBER for quantum key distribution includes: Alice randomly selects some bits from the associated data sequence after base filtering as test samples and publishes the corresponding bit values. Alice's public values are compared with the bit values at the corresponding positions of Bob, and the number of inconsistent bits is counted to calculate the QBER. If the QBER exceeds the preset security threshold, it is determined that eavesdropping has occurred or the channel noise is too high, and the protocol is terminated and restarted. If the QBER is lower than the threshold, the next step is entered. After completing the QBER comparison, Alice removes the published test sample data, and the remaining unpublished basis matching data is used for subsequent key negotiation and confidentiality enhancement.

4. The discrete variable quantum key distribution data coordination method according to claim 1, characterized in that: The process of the symbol-level rate adaptation method includes: The symbols of the multi-ary LDPC code are punctured and / or shortened as a whole. The formula for calculating the adjusted code rate is: Where R = K / N, which is the original code rate, K is the length of the information bit when the code rate adaptation method is not used, N is the code length of the multi-ary LDPC code, s is the number of shortened symbols, and p is the number of punctured symbols; The process of the bit-level rate adaptation method includes: Let the finite field GF(Q), Q=2 q The multi-ary symbol is split into q bits, and fine bit rate adjustment is achieved by puncturing and / or shortening the bit-level data. The formula for calculating the adjusted bit rate is: Where s is the number of shortened symbols, p is the number of punctured symbols, b is the number of shortened bits per symbol, c is the number of punctured bits per symbol, and R = K / N is the original bit rate before using the rate adaptation method. Alice selects a symbol-level and / or bit-level solution based on system overhead requirements and determines related parameters.

5. The discrete variable quantum key distribution data coordination method according to claim 4, characterized in that: Bob adjusts the data at the symbol level and / or bit level according to the parameters determined by the code rate adaptation method to optimize the code rate to achieve the best coordination efficiency; specifically, In symbol-level processing, Bob determines the number of shortened symbols, s, and the number of punctured symbols, p, and randomly generates s shortened symbols and p punctured symbols, inserting them into designated positions in the original data sequence. Bob then sends the insertion positions and related parameters to Alice over an authenticated channel. Upon receiving this information, Alice inserts a shortened symbol with the same value as Bob's at the same position and fills the punctured symbol position with a random value. After these operations, Alice and Bob generate new data sequences, X' and Y', respectively, both with lengths N' = N + s + p. In bit-level processing, Bob further determines the number of shortened bits, b, and the number of punctured bits, c, randomly generates b shortened bits, and inserts them into specified positions in the original bit stream. Simultaneously, Bob selects c punctured bit positions and fills these positions with random bit values. Subsequently, Bob sends the bit-level insertion positions, b and c, and the shortened bit values to Alice via an authenticated channel. Upon receiving this information, Alice inserts the same shortened bits as Bob at the same bit positions and fills the punctured positions with random bits. Finally, Alice and Bob generate new bit sequences, X' and Y', respectively, both of length N' = N + (s·b + p·c) / log2 Q.

6. The discrete variable quantum key distribution data coordination method according to claim 5, characterized in that: The syndrome decoding method is as follows: Alice receives the syndrome of the data calculated by Bob through an authenticated channel, and Alice locates and corrects the error bits through iterative verification and side information exchange.

7. The discrete variable quantum key distribution data coordination method according to claim 6, characterized in that: The process of using syndrome decoding method to correct data errors includes: Bob uses the parity check matrix of the multi-ary LDPC code and the data Y' to calculate the syndrome S. The formula of syndrome S is expressed as: S=Y'H T Where H T is the transpose of the parity check matrix of the multi-ary LDPC code; Then the syndrome S is sent to Alice as side information to assist Alice in correcting the data X'. Alice uses the multi-ary belief propagation decoding algorithm to correct the data X' based on the syndrome S.

8. The discrete variable quantum key distribution data coordination method according to claim 1, characterized in that: Alice selects the same hash function from the collision-resistant hash function family and converts the original key of the coordinated error-corrected original key into the final key with information-theoretic security through a compression mapping, including: Alice chooses a suitable hash function G to compress the partial information L obtained by Eve into a very small security parameter, making it impossible for Eve to extract useful information from it. By using the hashed compressed information, Alice generates a final key sequence that is completely consistent with Bob's and is unconditionally secure in information theory.

9. A discrete variable quantum key distribution data coordination method, characterized by: include: Under the premise of reverse negotiation, the receiver Bob receives the quantum state key data prepared by the sender Alice and transmitted through the quantum channel, and obtains the corresponding original key data through measurement; Bob purifies the original key data and obtains the associated data sequence that is shared with Alice after screening; Bob compares the statistical characteristics of the measurement results of part of the associated data sequence with Alice through the authentication channel to calculate the QBER of quantum key distribution; Based on the QBER calculation results, Bob selects a rate-adaptive multi-ary LDPC code and uses symbol-level and / or bit-level rate adaptation methods for dynamic optimization, balancing efficiency and system overhead. He then uses syndrome decoding for error correction: Bob selects the same hash function as Alice from the collision-resistant hash function family, and converts the original key of the coordinated error-corrected original key into the final key that is information-theoretically secure through a compression mapping.

10. The discrete variable quantum key distribution data coordination method according to claim 9, characterized in that: The process by which Bob obtains the corresponding original key data through measurement includes: After receiving the quantum state key data, Bob randomly selects a measurement basis to measure each quantum state and records the measurement results through a single-photon detector. Due to the randomness of the measurement basis, Bob only obtains valid bits when it is consistent with Alice's coding basis.

Images