CN116132024A - Quantum communication method based on decision tree - Google Patents

Quantum communication method based on decision tree Download PDF

Info

Publication number
CN116132024A
CN116132024A CN202211455561.1A CN202211455561A CN116132024A CN 116132024 A CN116132024 A CN 116132024A CN 202211455561 A CN202211455561 A CN 202211455561A CN 116132024 A CN116132024 A CN 116132024A
Authority
CN
China
Prior art keywords
decision tree
quantum communication
prediction model
characteristic information
training
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202211455561.1A
Other languages
Chinese (zh)
Inventor
李冬芬
郑云丹
刘晓芳
刘明哲
唐小川
周让
周杰
杨小龙
谭玉乔
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chengdu Univeristy of Technology
Original Assignee
Chengdu Univeristy of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chengdu Univeristy of Technology filed Critical Chengdu Univeristy of Technology
Priority to CN202211455561.1A priority Critical patent/CN116132024A/en
Publication of CN116132024A publication Critical patent/CN116132024A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L9/00—Cryptographic mechanisms or cryptographic arrangements for secret or secure communications; Network security protocols
    • H04L9/08—Key distribution or management, e.g. generation, sharing or updating, of cryptographic keys or passwords
    • H04L9/0816—Key establishment, i.e. cryptographic processes or cryptographic protocols whereby a shared secret becomes available to two or more parties, for subsequent use
    • H04L9/0852—Quantum cryptography
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Security & Cryptography (AREA)
  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Databases & Information Systems (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention discloses a quantum communication method based on a decision tree, which comprises the steps of collecting information of a communicator, extracting required characteristic information from the information, converting the characteristic information into data, training a decision tree prediction model, and outputting the grade of the communicator; evaluating the trained decision tree prediction model by adopting cross validation and resampling, and optimizing the model; when in communication, corresponding channel particles are configured for the communicator class output by the decision tree prediction model, and communication is started; configuring corresponding measurement operations for receivers of different grades, and informing the receivers of the result through a classical channel after the measurement operations are executed; after receiving the measurement result, the receiver performs a unitary operation on the collapsed state to reconstruct the information state. The invention utilizes the decision tree algorithm in machine learning to grade the communicators, realizes the grade judgment of the communicators, so as to allocate matched channel particles for the communicators with corresponding grades, solve the problem of allocation of channel particles, save quantum resources and improve the communication efficiency.

Description

Quantum communication method based on decision tree
Technical Field
The invention relates to the technical field of channel particle distribution in quantum communication, in particular to a quantum communication method based on a decision tree.
Background
Quantum communication is a novel communication mode for information transmission by utilizing quantum superposition states and entanglement effects, and is mainly based on the theory of quantum entanglement states, and information transmission is realized by using modes such as quantum invisible transmission states (transmission), quantum information separation (transmission), quantum key distribution (transmission) and the like. The quantum communication process is as follows: a pair of particles with entangled states (entangled particles have long-distance physical connection), two particles are respectively put on two communication parties, the particles with unknown quantum states and the particles of a sender are combined and measured (an operation), then the particles of the receiver instantaneously collapse (change: based on the change generated by the physical connection between the particles) to a state which is symmetrical with the state after the particles of the sender collapse (change), then the combined measured information is transmitted to the receiver through a classical channel, and the receiver performs unitary transformation (equivalent to inverse transformation) on the collapsed particles according to the received information, so that the unknown quantum states which are identical with the unknown quantum states of the sender can be obtained.
In practical communication, a communication scheme is generally designed directly according to transmitted information, wherein the communication grades of communication participants need to be manually judged, and then corresponding particle positions are allocated to the communication participants according to the grades, however, because the number of people participating in the communication is quite huge, the condition of judging the grades is too complicated, and the artificial workload is quite easy to judge errors, the grades of the communication participants are difficult to judge, even some communication participants cannot possess the particle positions and cannot participate in the communication, and therefore, the problem of allocation of communication channel particles exists in the conventional quantum communication process.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a quantum communication method based on a decision tree, which solves the grade judgment problem of a communicator by introducing the method of the decision tree in machine learning into quantum communication, thereby solving the distribution problem of communication channel particles.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
a quantum communication method based on decision tree, comprising the steps of:
s10, collecting the communicator information of the quantum communication participants, extracting a plurality of characteristic information required by a decision tree prediction model from the communicator information, training the decision tree prediction model after carrying out data preprocessing on the characteristic information, and outputting the high-level and low-level classification of the quantum communication participants;
s20, evaluating the decision tree prediction model obtained through training by adopting a cross-validation and resampling method, and optimizing the decision tree prediction model according to an evaluation result;
s30, determining the high-level and low-level classification of the quantum communication participants according to the decision tree prediction model during communication, distributing corresponding channel particles for the quantum communication participants of different grades according to preset information, and starting communication;
s40, the receiver respectively executes corresponding measurement operations according to the specified rules for quantum communication participants of different grades, and after the measurement operations are executed, the receiver is informed of the measurement results through classical channels;
and S50, after receiving all the measurement results, the receiver executes a unitary operation on the collapsed state according to the corresponding results, and recovers the information sent by the sender.
Specifically, the process of step S10 includes:
s11, presetting characteristic information required by a decision tree prediction model according to the target requirement of quantum communication;
s12, extracting corresponding characteristic information from the acquired communicator information according to preset requirements;
s13, digitizing the extracted characteristic information;
s14, taking the digitized characteristic information as a data set, and dividing the data set into a training set and a testing set according to a specified proportion;
s15, inputting data of the training set into a set decision tree prediction model for model training, and outputting results according to the set high-level and low-level classification of the quantum communication participants, thereby determining each parameter of the decision tree prediction model;
s16, checking the obtained decision tree prediction model by adopting a test set, carrying out statistical analysis and judgment on the error rate according to the output result of the test set, and carrying out pruning optimization on the decision tree prediction model obtained by training by adopting a loss function.
Specifically, the characteristic information required by the preset decision tree prediction model is the ratio of the times of integrity in all the communication, the times of integrity in the last three times of communication, the dishonest possibility judged by a third party notarization platform and whether the communication is marketed or not.
Specifically, the set decision tree prediction model adopts a base index model, which is expressed as:
D 1 ＝{(x i )∈DA j (x)＝a},0≤a≤x i ，D 2 ＝D-D 1
Figure BDA0003952845480000031
Figure BDA0003952845480000032
wherein D is a data set, A j Represents the j-th characteristic information, x i An ith value representing a certain characteristic information, a represents a branching condition value of the corresponding characteristic information, and a is calculated according to the valueThe value of a divides the data set D into D 1 And D 2 Two sub-sets of branches; k represents the number of classes of the characteristic information in D, C k Representing a subset of samples belonging to the kth class in D; gini (D) represents the base index of the data set D, gini (D, a) represents the base index of the data set D when the characteristic information a is taken;
and selecting the characteristic information corresponding to the minimum Basni index as an optimal segmentation point, distributing samples in a training set into two branch subsets according to the characteristic information during model training, and performing repeated calculation and selection in a circulating way until the number of samples in the branch subsets is smaller than a preset threshold value or no characteristic is selected, wherein each set of the two similar branch subsets corresponds to a high-level quantum communication participant and a low-level quantum communication participant in an output result respectively.
Specifically, the loss function is expressed as:
C α (T)＝C(T)+αT
wherein T represents any subtree, C (T) represents the prediction error of training data, namely the corresponding radix index, T represents the number of leaf nodes of the subtree, and C α And (T) represents the overall loss of the subtree T when the parameter alpha is the same, and the set parameter alpha is used for balancing the fitting degree of training data and the complexity of the model.
Specifically, the process of step S20 includes:
s21, according to a cross-validation error kfodLoss function and a resampling error resubLoss function, reassigning a data set D, training and testing a set decision tree prediction model, and calculating a cross-validation error;
s22, repeating the process of the step S21 for a plurality of times, and calculating the average value of the cross verification errors after a plurality of times of distribution, wherein the average value is used for evaluating a decision tree prediction model for training optimization;
s23, pruning the training optimized decision tree prediction model under the minimum cross validation error is selected according to the evaluation result of the step S22, and the decision tree prediction model is adjusted to achieve re-optimization of the decision tree prediction model.
Specifically, the step S30 determines that the quantum communication participants are ranked as an advanced quantum communication participant and a low quantum communication participant.
Specifically, the step S40 specifies that the corresponding measurement operation is performed as follows:
when the receiver is an advanced quantum communication participant, executing single particle measurement by any other advanced quantum communication participant, and informing the receiver of the measurement result through a classical channel;
when the receiver is a low-level quantum communication participant, single-event measurement is performed by all other quantum communication participants, and the measurement result is told to the receiver through a classical channel.
Compared with the prior art, the invention has the following beneficial effects:
(1) The invention provides a method for designing a quantum communication scheme according to quantum communication participants, which skillfully utilizes a decision tree algorithm in machine learning to divide and grade the quantum communication participants, realizes grade judgment of the communicants, so as to distribute matched channel particles for the communicants of the corresponding grade, solve the distribution problem of the channel particles, save quantum resources and improve the quantum communication efficiency. The invention has the advantages of ingenious design, relatively simple process, convenient and reliable realization and suitability for application in quantum communication.
(2) The invention collects the information of the quantum communication participants, extracts the needed characteristic information, and more accurately realizes the high-level and low-level classification judgment of the communicators by processing the characteristic information, thereby providing a reliable basis for the quantum communication process.
(3) On the basis of designing the decision tree prediction model based on the characteristic information, the model is evaluated by the method of cross verification and resampling, so that the decision tree prediction model is modified and pruned, and the decision tree prediction model has better performance.
Drawings
FIG. 1 is a schematic flow chart of an embodiment of the present invention.
FIG. 2 is a schematic diagram of a decision tree prediction model obtained by training in an embodiment of the present invention.
FIG. 3 is a schematic diagram of a decision tree predictive model optimized by test set verification in an embodiment of the invention.
FIG. 4 is a schematic diagram of a decision tree prediction model optimized according to the evaluation result in the embodiment of the invention.
Detailed Description
The invention will now be further described with reference to the accompanying drawings and examples, embodiments of which include, but are not limited to, the following examples.
The existing quantum communication method is designed directly according to the transmitted information and is mainly based on the quantum entanglement theory. The method uses methods such as quantum invisible transmission state (a transmission mode), quantum information separation (a transmission mode) and the like to realize information transmission. The process is generally as follows: first, a pair of entangled particles a (entangled particles have long-distance physical connection), two particles are placed on both sides of a communication, and particles with unknown quantum states and particles of a sender are measured together (one operation), then the particles of a receiver will collapse instantaneously (change means change based on physical connection between particles), and the collapse (change) will be a specific state symmetrical to the state of the collapsed (changed) particles of the sender, and then the information of the joint measurement is transmitted to the receiver through classical channels. The receiver performs a unitary operation (equivalent to a reverse operation) on the collapsed particles based on the received information to obtain the same unknown quantum state as the transmitter. For example: with 2 bits of information transmitted, then 1 bit quantum state is required
Figure BDA0003952845480000051
To carry information as information particles, in order to transmit this quantum state, quantum information separation is used as a transmission mode, 4-bit quantum entangled state +.>
Figure BDA0003952845480000052
As quantum channels, the sender owns particles 1 and 2 of 1-bit quantum state and 4-bit channel particle quantum state, the receiver owns particles 3, and the third party quantum communication participant owns particles 4. The sender performs joint measurement on the particle information particles and the channel particles 1 and 2, if the sender wants the receiver to obtain the information, the sender informs the third party of the measurement result to the quantum communication parametersAnd the others. The third party quantum communication measures the owned particles 4, and if the receiver wants to obtain the original information, the receiver needs the measurement results of the sender and the third party quantum communication participants, and performs the unitary operation according to the measurement results, so that the receiver can obtain the original information.
According to a communication transmission efficiency calculation formula:
Figure BDA0003952845480000061
c is the number of quantum bits transmitted, q is the number of bits used by the quantum channel, t is the classical number of bits, which is the number of bits transmitted through the classical channel. The ciphertext to be transmitted can meet all the requirements of transmission by two particle states, so that fewer channel particles can be used for transmitting information, such as the channel particles are +.>
Figure BDA0003952845480000062
Figure BDA0003952845480000063
Etc. How does there be so many ways to communicate, how is it known what is what is most appropriate we need to communicate? And the channel particles are the same, the states may also be different, such as |0000>+|0101>+|1010>+|1111>Sum |0000>+|0101>+|1010>-|1111>How does we change per increment of one correspondent? How does it determine that the communication scheme is optimal? In order to solve these problems, the judgment is conventionally performed manually based on experience, which is quite complex and prone to error.
Under the background, the inventor proposes the quantum communication method based on the decision tree, firstly, a quantum communication scheme is designed according to communicators needing to participate, then, a decision tree algorithm in machine learning is introduced to solve the grade judgment problem of the communicators, so that the distribution problem of channel particles is solved, and finally, the accurate, efficient and low-quantum resource consumption quantum communication process is realized.
As shown in fig. 1, the quantum communication method based on the decision tree comprises the following steps:
s10, collecting the communicator information of the quantum communication participants, extracting a plurality of characteristic information required by a decision tree prediction model from the communicator information, training the decision tree prediction model after carrying out data preprocessing on the characteristic information, and outputting the high-level and low-level classification of the quantum communication participants; the specific process is as follows:
s11, presetting characteristic information required by a decision tree prediction model according to the target requirement of quantum communication; for example, the target needs to more stand the integrity of the communicants, the preset required characteristic information is: four characteristics of the ratio of the times of integrity in all communication, the times of integrity in the last three times of communication, the possibility of dishonest judgment by a third party notarization platform, whether to market or not and the like; in another example, if the target requirement is more qualified for the correspondent, the preset required characteristic information may be: corporate personnel scale, corporate asset valuations, whether it is a world-wide encyclopedia, etc.; if the target needs to pay more attention to the credit of the communicator, the preset required characteristic information can be: default conditions, loan conditions, whether to repay the loan, etc.;
s12, extracting corresponding characteristic information from the acquired communicator information according to preset requirements; the collected information of the communicators is generally comprehensive various information of the quantum communication participants, but if comprehensive information processing is used, the algorithm complexity is overlarge, so that partial characteristic operation processing is extracted according to the target demand trend, the algorithm processing complexity can be well simplified while pertinence of the result is ensured, and the data processing efficiency is improved; in practical application, the types of the communication information during collection can also be set based on the preset target requirement and the basic information, so that the collection data volume and the requirement on data storage can be reduced to a certain extent;
s13, digitizing the extracted characteristic information, wherein the directly extracted characteristic information is not necessarily the data required by the algorithm, so that the digital processing of the extracted characteristic information realizes the standardization matched with the algorithm, the characteristic information of the digital result is processed according to the appointed standard, and the characteristic information of the non-digital result is processed according to the algorithm requirement; for example: taking part in any decimal result with the percentage of the honest times between 0 and 1 in all communication, and taking a value according to one decimal during digitalization; the last three times of communication have 0 times, 1 time, 2 times and 3 times of results, and the digitization is respectively carried out according to the corresponding values of 0, 1, 2 and 3; the dishonest possibility judged by the third party notarization platform has dishonest and honest results, the dishonest results are assigned to be 1 when the dishonest results are digitized, and the dishonest results are assigned to be 0; if yes and no results are available in the market, the yes results are assigned to 0 when the digital result is digitalized, and if no results are assigned to 1;
s14, taking the digitized characteristic information as a data set, dividing the data set into a training set and a test set according to a specified proportion, wherein the training set is used for model training, the test set is used for model checking, and the training set and the test set can be used according to 8:2, other proportions can be set according to the requirements;
s15, inputting data of a training set into a set decision tree prediction model for model training, and determining each parameter of the decision tree prediction model according to the high-level and low-level classification output results of the set quantum communication participants, wherein the model parameters mainly comprise the number of branch nodes of the decision tree, the ordering of each branch node and the branching condition of each branch node; the set decision tree prediction model adopts a base index model, and is expressed as:
D 1 ＝{(x i )∈D|A j (x)＝a},0≤a≤x i ，D 2 ＝D-D 1
Figure BDA0003952845480000071
Figure BDA0003952845480000081
wherein D is a data set, and a training set is used during training; a is that j Represents the j-th characteristic information, x i An ith value representing a certain characteristic information, a represents a branching condition value of the corresponding characteristic information, and the data set D is divided into D according to the value of a 1 And D 2 Two branchesSet of two branches corresponding to a branch node, e.g. j E [1,4 in the previous example]And the value of i corresponding to different j is 2, 3 or more, and the different j corresponds to the value of the branch condition matched with the value of the i; k represents the number of types of characteristic information in the data set D, and is matched with the maximum value of j, C k Representing a subset of samples belonging to the kth class in the data set D, gini (D) representing the base index of the data set D, gini (D, a) representing the base index of the data set D when the characteristic information a is taken; the formula corresponds to one branch node of the decision tree, and the other branch nodes are similar to the branch node, but branch subsets divided by each branch node can be different, and all branch subsets of the same class are summarized, namely, the high-level and low-level classes of the quantum communication participants are correspondingly set;
the smaller the base index is, the smaller the uncertainty of the data set is, the corresponding characteristic information A is the optimal characteristic, the characteristic information corresponding to the minimum base index is selected as the optimal dividing point, the samples in the training set are distributed into two branch subsets according to the characteristic information during model training, and the samples in the branch subsets are calculated and selected repeatedly in a circulating way until the number of the samples in the branch subsets is smaller than a preset threshold value or the characteristics are not selected, and the respective sets of the two similar branch subsets correspond to a high-level quantum communication participant and a low-level quantum communication participant in an output result respectively;
s16, checking the obtained decision tree prediction model by adopting a test set, carrying out statistical analysis and judgment on the error rate according to the output result of the test set, and carrying out pruning optimization on the decision tree prediction model obtained by training by adopting a loss function. And repeatedly calling leaf nodes of the generated original decision tree prediction model, and performing pruning optimization by calculating and comparing prediction errors of the decision tree models before and after limiting the leaf nodes. The loss function is expressed as:
C α (T)＝C(T)+α|T|
wherein T represents any subtree, C (T) represents the prediction error of training data, namely the corresponding radix index, T represents the number of leaf nodes of the subtree, and C α (T) represents the overall loss of the subtree T when the parameter alpha is set to balance the fitting degree of training data with the modulusType complexity. The decision tree prediction model can be simplified through pruning, so that the unknown data can be predicted better, and the generalization capability of the model is improved.
S20, evaluating a decision tree prediction model obtained by training and optimizing by adopting a cross-validation and resampling method, and optimizing the decision tree prediction model according to an evaluation result;
s21, redistributing a data set D according to a cross-validation error kfodLoss function and a resampling error resubLoss function to obtain a new training set and a new testing set, then re-acquiring a decision tree prediction model by adopting the processes of the steps S15 and S16, and comparing and calculating a cross-validation error E with the preset high-level and low-level grades of quantum communication participants;
s22, repeating the process of the step S21 for a plurality of times, and calculating the average value of the cross verification errors after a plurality of times of distribution, wherein the average value is used for evaluating a decision tree prediction model for training optimization; expressed as:
Figure BDA0003952845480000091
in the method, in the process of the invention,
Figure BDA0003952845480000092
mean value of cross-validation errors, E i Indicating the i-th calculated cross-validation error, n being the total number of times step S21 is performed;
s23, pruning the training optimized decision tree prediction model under the minimum cross validation error is selected according to the evaluation result of the step S22, and the decision tree prediction model is adjusted to achieve re-optimization of the decision tree prediction model.
S30, determining the high-level and low-level classification of the quantum communication participants according to the decision tree prediction model during communication, distributing corresponding channel particles for the quantum communication participants of different grades according to preset information, and starting communication.
S40, the receiver respectively executes corresponding measurement operations according to the specified rules for quantum communication participants of different grades, and after the measurement operations are executed, the receiver is informed of the measurement results through classical channels; wherein the specified rule performs the corresponding measurement operation as:
when the receiver is an advanced quantum communication participant, executing single particle measurement by any other advanced quantum communication participant, and informing the receiver of the measurement result through a classical channel;
when the receiver is a low-level quantum communication participant, single-event measurement is performed by all other quantum communication participants, and the measurement result is told to the receiver through a classical channel.
And S50, after receiving all the measurement results, the receiver executes a unitary operation on the collapsed state according to the corresponding results, and recovers the information sent by the sender.
Verification and illustration of data by the following examples
Four characteristics for evaluating the grade of a communicator are set as follows: the ratio of the times of integrity in all communication, the times of integrity in the last three communication, the possibility of dishonest judgment by a third party notarization platform and whether to be marketed. The total sample size of all data collected was 1131.
The data set D is obtained by sorting 409 cases of correspondents which have a data rate of 0.5 or more and a honest possibility of 1 and are on the market as high-level correspondents and 722 cases as low-level correspondents. Data set D was run at 8: and after dividing the ratio of 2 into a training set and a test set, training a decision tree prediction model by using the training set to obtain a result shown in fig. 2, optimizing the result shown in fig. 2 by using the test set to obtain a result shown in fig. 3, and evaluating and optimizing the result shown in fig. 3 by adopting a cross-validation and resampling method to obtain the result shown in fig. 4. In the figure, x1 represents the ratio of the times of being honest in all the communication, x2 represents the times of being honest in the last three communication, x3 represents the possibility of dishonest judgment by a third party notarization platform, and x4 represents whether to be marketed.
After the allocation schemes of the high-level communicants and the low-level communicants are obtained according to the decision tree prediction model, channel particles are allocated to the high-level communicants and the low-level communicants according to a preset allocation strategy, for example, one of N communication participants is a sender (with particles 1 and 2) and one is a receiver (with particles 3 and 4), the decision high-level communicants Bob1, bob2 and …, bob x have particles 5 to 5+x bits, and the low-level communicants Charlie1, charlie 2 and …, charlie are particles with 6+x to 6+x+y bits; and then communication is carried out.
The comparison of the communication efficiency of the communication scheme OS of the present invention and the conventional communication schemes 1, 2, 3, and 4 is shown in the following table 1, wherein scheme 1 is a two-particle information three-particle channel transmission scheme, scheme 2 is a single-particle information 4-particle transmission scheme, scheme 3 is a single-particle information 4-particle separation transmission scheme, scheme 4 is a two-particle information eight-particle transmission scheme, and references are respectively:
1、Wen Z,Liu Y M,Zhang Z J,et al.Splitting a qudit state via Greenberger–Horne–Zeilinger states of qubits[J].Optics Communications,2010,283(4):628-632.
2、Jouguet P,Kunz-Jacques S,Leverrier A,et al.Experimental demonstration of long-distance continuous-variable quantum key distribution[J].Nature Photonics,2013,7(5):378-381.
3、Lu Y J.A Novel Practical Quantum Secure Direct Communication Protocol[J].International Journal of Theoretical Physics,2021(3).
4、Lucas L.Quantum Reinforcement Leaming with Quantum Photonics[J].Photonics,2021,8(2):33.
four aspects are compared: quantum resource consumption QS, transmitted quantum bit QT, transmission efficiency η, and communication participant population NC. Compared with the traditional communication schemes, the result of the invention saves quantum resources and improves transmission efficiency.
Table 1 comparison of communication efficiency
Schemes QS QT transmission efficiency(η％) NC
1 10 2 10％ 5
2 4 1 12.5％ 4
3 4 1 12.5％ 4
4 8 2 12.5％ 4
OS 8 2 14.3％ 5
Therefore, the invention can effectively reduce the complexity involved in the communication process and improve the overall working efficiency of the system by using the decision tree model in the quantum communication process, and simultaneously realize the resource optimization configuration.
The above embodiments are only preferred embodiments of the present invention, and not intended to limit the scope of the present invention, but all changes made by adopting the design principle of the present invention and performing non-creative work on the basis thereof shall fall within the scope of the present invention.

Claims (8)

1. The quantum communication method based on the decision tree is characterized by comprising the following steps of:
s10, collecting the communicator information of the quantum communication participants, extracting a plurality of characteristic information required by a decision tree prediction model from the communicator information, training the decision tree prediction model after carrying out data preprocessing on the characteristic information, and outputting the high-level and low-level classification of the quantum communication participants;
s20, evaluating the decision tree prediction model obtained through training by adopting a cross-validation and resampling method, and optimizing the decision tree prediction model according to an evaluation result;
s30, determining the high-level and low-level classification of the quantum communication participants according to the decision tree prediction model during communication, distributing corresponding channel particles for the quantum communication participants of different grades according to preset information, and starting communication;
s40, the receiver respectively executes corresponding measurement operations according to the specified rules for quantum communication participants of different grades, and after the measurement operations are executed, the receiver is informed of the measurement results through classical channels;
and S50, after receiving all the measurement results, the receiver executes a unitary operation on the collapsed state according to the corresponding results, and recovers the information sent by the sender.
2. The quantum communication method based on decision tree according to claim 1, wherein the process of step S10 comprises:
s11, presetting characteristic information required by a decision tree prediction model according to the target requirement of quantum communication;
s12, extracting corresponding characteristic information from the acquired communicator information according to preset requirements;
s13, digitizing the extracted characteristic information;
s14, taking the digitized characteristic information as a data set, and dividing the data set into a training set and a testing set according to a specified proportion;
s15, inputting data of the training set into a set decision tree prediction model for model training, and outputting results according to the set high-level and low-level classification of the quantum communication participants, thereby determining each parameter of the decision tree prediction model;
s16, checking the obtained decision tree prediction model by adopting a test set, carrying out statistical analysis and judgment on the error rate according to the output result of the test set, and carrying out pruning optimization on the decision tree prediction model obtained by training by adopting a loss function.
3. The quantum communication method based on decision tree according to claim 2, wherein the characteristic information required by the preset decision tree prediction model is the percentage of the times of integrity in all the communications, the times of integrity in the last three communications, the dishonest possibility judged by a third party notarization platform, and whether to be marketed.
4. The quantum communication method based on decision tree according to claim 2, wherein the decision tree prediction model set in step S15 adopts a keni index model, and the formula is:
D 1 ＝{(x i )∈D|A j (x)＝a},0≤a≤x i ，D 2 ＝D-D 1
Figure FDA0003952845470000021
Figure FDA0003952845470000022
wherein D is a data set, A j Represents the j-th characteristic information, x i An ith value representing a certain characteristic information, a represents a branching condition value of the corresponding characteristic information, and the data set D is divided into D according to the value of a 1 And D 2 Two sub-sets of branches; k represents the number of classes of the characteristic information in D, C k Representing a subset of samples belonging to the kth class in D; gini (D) represents the base index of the data set D, gini (D, a) represents the base index of the data set D when the characteristic information a is taken;
and selecting the characteristic information corresponding to the minimum Basni index as an optimal segmentation point, distributing samples in a training set into two branch subsets according to the characteristic information during model training, and performing repeated calculation and selection in a circulating way until the number of samples in the branch subsets is smaller than a preset threshold value or no characteristic is selected, wherein each set of the two similar branch subsets corresponds to a high-level quantum communication participant and a low-level quantum communication participant in an output result respectively.
5. The quantum communication method based on decision tree according to claim 2, wherein the loss function in step S16 is expressed as:
C α (T)＝C(T)+α|T|
wherein T represents any subtree, C (T) represents the prediction error of training data, namely the corresponding radix index, |T| represents the number of leaf nodes of the subtree, and C α And (T) represents the overall loss of the subtree T when the parameter alpha is the same, and the set parameter alpha is used for balancing the fitting degree of training data and the complexity of the model.
6. The quantum communication method based on decision tree according to claim 4, wherein the process of step S20 comprises:
s21, according to a cross-validation error kfodLoss function and a resampling error resubLoss function, reassigning a data set D, training and testing a set decision tree prediction model, and calculating a cross-validation error;
s22, repeating the process of the step S21 for a plurality of times, and calculating the average value of the cross verification errors after a plurality of times of distribution, wherein the average value is used for evaluating a decision tree prediction model for training optimization;
s23, pruning the training optimized decision tree prediction model under the minimum cross validation error is selected according to the evaluation result of the step S22, and the decision tree prediction model is adjusted to achieve re-optimization of the decision tree prediction model.
7. The decision tree based quantum communication method of claim 6, wherein the classification of the quantum communication participants into the advanced quantum communication participants and the low quantum communication participants is determined in step S30.
8. The quantum communication method according to claim 7, wherein the specifying in the step S40 performs the corresponding measurement operation as:
when the receiver is an advanced quantum communication participant, executing single particle measurement by any other advanced quantum communication participant, and informing the receiver of the measurement result through a classical channel;
when the receiver is a low-level quantum communication participant, single-event measurement is performed by all other quantum communication participants, and the measurement result is told to the receiver through a classical channel.
CN202211455561.1A 2022-11-21 2022-11-21 Quantum communication method based on decision tree Pending CN116132024A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211455561.1A CN116132024A (en) 2022-11-21 2022-11-21 Quantum communication method based on decision tree

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211455561.1A CN116132024A (en) 2022-11-21 2022-11-21 Quantum communication method based on decision tree

Publications (1)

Publication Number Publication Date
CN116132024A true CN116132024A (en) 2023-05-16

Family

ID=86303391

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211455561.1A Pending CN116132024A (en) 2022-11-21 2022-11-21 Quantum communication method based on decision tree

Country Status (1)

Country Link
CN (1) CN116132024A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117786544A (en) * 2024-02-28 2024-03-29 浪潮通信信息系统有限公司 User satisfaction obtaining method and device, electronic equipment and storage medium

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117786544A (en) * 2024-02-28 2024-03-29 浪潮通信信息系统有限公司 User satisfaction obtaining method and device, electronic equipment and storage medium

Similar Documents

Publication Publication Date Title
CN105163326B (en) A kind of cell clustering method and system based on wireless network traffic feature
CN108874959A (en) A kind of user's dynamic interest model method for building up based on big data technology
CN111163487B (en) Communication waveform comprehensive transmission performance evaluation method and system
CN105472631A (en) Service data quantity and/or resource data quantity prediction method and prediction system
CN104837184A (en) Heterogeneous wireless network selection method based on interval triangular fuzzy number
CN116132024A (en) Quantum communication method based on decision tree
CN102722577A (en) Method and device for determining dynamic weights of indexes
CN110365603A (en) A kind of self adaptive network traffic classification method open based on 5G network capabilities
CN113569151B (en) Data recommendation method, device, equipment and medium based on artificial intelligence
CN115330201A (en) Power grid digital project pareto optimization method and system
CN107093005A (en) The method that tax handling service hall's automatic classification is realized based on big data mining algorithm
CN113726558A (en) Network equipment flow prediction system based on random forest algorithm
CN113420165A (en) Training of two-classification model and classification method and device of multimedia data
CN117312637A (en) Natural resource data system construction method based on data association identification
CN116578924A (en) Network task optimization method and system for machine learning classification
CN111343271A (en) DTU (data transfer unit) equipment-based artificial intelligence cloud control method and system
CN106550387B (en) A kind of wireless sensor network routing layer QoS evaluating method
CN112115991B (en) Mobile terminal change prediction method, device, equipment and readable storage medium
CN115169819A (en) Power supply business hall efficiency data processing method
CN113191569A (en) Enterprise management method and system based on big data
CN111858530A (en) Real-time correlation analysis method and system based on mass logs
CN112559854A (en) Classification method and device
CN111931966A (en) Power supply reliability prediction method based on decision tree regression
CN117171446B (en) Technical transaction recommendation method and recommendation system based on big data analysis
CN110109960A (en) A kind of data acquisition expansion control system and its collecting method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination