A wireless channel key generation method and system based on a flow antenna

Abstract

This invention discloses a wireless channel key generation method and system based on a fluidic antenna. First, a physical layer key generation (PLKG) model based on a fluidic antenna system (FAS) is constructed, completing channel modeling adapted to FAS characteristics. Second, for independent and identically distributed (AID) and spatially correlated channel scenarios, based on the PLKG secret key rate, a key rate optimization problem with transmit power constraints and sparse port activation constraints is further established, and a solution method is proposed to achieve joint optimization of sparse port selection and transmit beamforming vector. Finally, through the entire process of channel feature sequence detection, quantization, information negotiation, and privacy amplification, a symmetric encryption key shared by legitimate parties is generated. This invention achieves low-overhead, high-security, and highly adaptable wireless channel key generation through the dynamic port reconfiguration capability of the fluidic antenna.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication technology, and mainly relates to a method and system for generating wireless channel keys based on a fluid antenna. Background Technology

[0002] With the rapid development of wireless communication technology, the demand for data transmission security in fields such as the Internet of Things and industrial control is becoming increasingly urgent. Traditional cryptography relies on pre-allocated keys and centralized key management, which faces problems such as high computational complexity and difficulty in key updates in resource-constrained wireless devices. Physical Layer Key Generation (PLKG) technology utilizes the reciprocity, time-varying nature, and spatial uniqueness of wireless channels to generate confidential keys by extracting random entropy from channel state information. It eliminates the need for pre-allocated key infrastructure, providing a new solution for wireless communication security.

[0003] However, existing PLKG technology still faces many challenges: on the one hand, fixed antenna systems have limited spatial degrees of freedom, making it difficult to guarantee the key generation rate in resource-constrained scenarios; on the other hand, in spatially correlated channel environments, eavesdroppers may obtain key information through channel correlation, reducing system security; in addition, port selection and beamforming optimization in traditional PLKG schemes are mostly non-convex problems with high solution complexity, making it difficult to meet practical deployment requirements.

[0004] Fluidic antenna systems (FAS), as a novel reconfigurable antenna technology, dynamically select antenna ports through software control, providing abundant spatial degrees of freedom while reducing the number of RF chains, thus offering a new approach to solving the aforementioned problems. In existing technologies, FAS is mainly used to improve communication capacity and anti-interference performance, but it has not fully utilized the characteristics of PLKG technology and lacks models and optimization methods for physical layer key generation, making it difficult to fully leverage the advantages of fluidic antennas.

[0005] Therefore, there is an urgent need for a physical layer key generation method that integrates fluid antenna technology. By using dynamic port selection and beamforming optimization, this method can improve key generation rate and security while reducing resource overhead and computational complexity, thus meeting the secure communication needs of resource-constrained scenarios. Summary of the Invention

[0006] This invention addresses the problems of high hardware resource overhead, insufficient key generation efficiency and security in complex scenarios, and high solution complexity in existing physical layer key generation technologies. It provides a wireless channel key generation method and system based on a fluidic antenna. First, a physical layer key generation (PLKG) model based on a fluidic antenna system (FAS) is constructed, completing channel modeling adapted to FAS characteristics. Second, for independent and identically distributed and spatially correlated channel scenarios, based on the PLKG secret key rate, a key rate optimization problem with transmit power constraints and sparse port activation constraints is further established and a solution method is proposed, achieving joint optimization of sparse port selection and transmit beamforming vector. Finally, through the entire process of channel feature sequence detection, quantization, information negotiation, and privacy amplification, a symmetric encryption key shared by legitimate parties is generated. This invention achieves low-overhead, high-security, and highly adaptable wireless channel key generation through the dynamic port reconfiguration capability of the fluidic antenna.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is: a wireless channel key generation method based on a flowing antenna, comprising the following steps:

[0008] S1. Channel Modeling and Bidirectional Channel Probing: A multi-input single-output (MIMO) wireless communication system model is constructed. The core nodes of the system include a base station equipped with a streaming antenna, a legitimate user, and an eavesdropper. The streaming antenna on the base station side is configured with multiple preset antenna ports, and is equipped with a radio frequency link with a number less than the total number of ports. The dynamic selection and activation of the antenna ports are achieved through software control. Only the activated ports are connected to the radio frequency link to complete signal transmission and reception. The system adopts a time-division duplex communication protocol and utilizes the time reciprocity of the wireless channel to complete uplink and downlink bidirectional channel probing within the channel coherence time, ensuring that the channel state information obtained by the base station and the legitimate user has strong reciprocity.

[0009] A wireless channel model adapted to the characteristics of the fluid antenna is constructed to characterize the transmission characteristics of the legitimate channel from the base station to the legitimate user and the eavesdropping channel from the base station to the eavesdropper. At the same time, a spatial correlation matrix is ​​introduced to describe the channel spatial correlation between each preset port of the fluid antenna, thus fully characterizing the channel propagation characteristics of the fluid antenna array.

[0010] Channel detection is completed through uplink and downlink bidirectional pilot transmission process. Channel state information of base station and legitimate users is obtained based on channel estimation algorithm. The equivalent channel estimation result on the base station side is constructed using beamforming vector to ensure that the channel estimation of legitimate transceivers has strong reciprocity.

[0011] S2. Key Rate Modeling and Optimization Problem Construction: Based on the mutual information and entropy analysis theories in information theory, the calculation expression for the system's secure key rate is derived for two typical wireless scenarios: independent and identically distributed channels and spatially correlated eavesdropping channels. With optimizing the system's secure key rate as the core optimization objective, a joint optimization problem for key generation performance is constructed by combining the maximum transmit power constraint of the base station and the constraint on the number of active antenna ports. The constraint on the number of active ports is used to ensure that the number of active antenna ports does not exceed the number of radio frequency links configured in the system, adapting to the hardware configuration characteristics of the flowing antenna.

[0012] S3. Port Selection and Beamforming Optimization: For the constructed non-convex optimization problem, an optimization algorithm adapted to the characteristics of the fluid antenna is used to solve it. The non-convex optimization problem is transformed into an efficient convex optimization subproblem through convex approximation. At the same time, the sparse constraint processing algorithm is used to realize the sparse activation of the antenna ports. A low-complexity port selection strategy is designed in combination with the channel spatial correlation characteristics of the fluid antenna. The optimal, suboptimal, random optimization or antenna port combination that meets the performance threshold is selected. The adapted port activation scheme and transmit beamforming vector are obtained by iterative solution, and the channel detection and signal transmission and reception scheme in the key generation process is determined.

[0013] S4. Secret Key Generation: Based on the port activation scheme and beamforming vector determined in S3, both communicating parties synchronously collect full-cycle channel feature sequences and construct channel feature sequences. The channel feature sequences are then sequentially quantized, negotiated, and privacy-enhancing processed to finally generate a symmetric encryption key shared by both legitimate senders and receivers. The specific steps are as follows:

[0014] S41. Full-cycle channel feature sequence acquisition: The base station and the legitimate user continuously and synchronously conduct bidirectional channel detection during the communication cycle, acquire multiple sets of continuous channel state information, and construct a channel feature sequence for key generation.

[0015] S42. Channel Feature Quantization and Initial Key Generation: The base station and the legitimate user respectively quantize the channel feature sequences they collect, set an adaptive quantization threshold based on the statistical characteristics of the channel feature sequences, map continuous channel feature values ​​into discrete binary bit sequences, and generate initial key sequences respectively.

[0016] S43. Key information negotiation: The base station and the legitimate user exchange verification information through a public channel, and use a preset error correction mechanism to locate and correct the erroneous bits in the initial key sequence to obtain a completely consistent negotiated key sequence.

[0017] S44. Privacy Amplification and Shared Key Generation: The base station and the legitimate user use a pre-agreed randomness enhancement algorithm to compress and purify the key sequence negotiated in step S43, removing redundant and leakable information from the sequence, and finally generating a shared, highly random, and highly secure symmetric encryption key.

[0018] Compared with the prior art, the present invention has the following beneficial effects:

[0019] (1) This invention makes full use of the dynamic port reconfiguration capability of the fluid antenna, which significantly reduces the hardware overhead of the radio frequency link while improving the key generation rate, achieving a balance between hardware cost and key performance, and can be adapted to the deployment requirements of resource-constrained scenarios.

[0020] (2) This invention completes key rate modeling and optimization design for multiple typical channel scenarios, fully considers the impact of channel spatial correlation on key security, can effectively suppress eavesdropping behavior, and improve the anti-eavesdropping capability and key security of the system in complex wireless environments.

[0021] (3) The present invention designs a low-complexity port selection and beamforming optimization method for adapting to fluid antennas, which effectively reduces the solution complexity of the joint optimization problem, improves the algorithm convergence efficiency and engineering deployment feasibility, and can meet the real-time requirements of actual wireless communication systems.

[0022] (4) The steps of the present invention construct a full-process key generation system, which can be widely used in the security protection of various wireless communication scenarios. While significantly reducing the hardware overhead of the radio frequency link, it improves the key generation rate and the anti-eavesdropping ability of the system in complex channel scenarios, reduces the optimization complexity, and is suitable for wireless security communication in resource-constrained scenarios such as the Internet of Things and industrial control. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating the application scenarios and process steps of the method of the present invention;

[0024] Figure 2 This is a flowchart of the steps of the method of the present invention;

[0025] Figure 3 This is a schematic diagram of the structure of the wireless communication system constructed in step S1 of the method of the present invention;

[0026] Figure 4 This is a comparison chart of the confidential key rate under different transmit power for different port selection schemes in an independent and co-distributed channel scenario according to an embodiment of the present invention.

[0027] Figure 5 This is a comparison chart of the confidentiality key rate under different transmission powers for different port selection schemes in space-related eavesdropping scenarios according to embodiments of the present invention. Detailed Implementation

[0028] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are for illustrative purposes only and are not intended to limit the scope of the invention.

[0029] Example 1

[0030] This embodiment provides a wireless channel key generation method based on a fluid antenna, which can be applied to secure communication scenarios for IoT terminal access, as shown in Figure 1. By leveraging the dynamic reconfiguration capability of the fluid antenna, low-overhead, high-security key generation is achieved, as shown in Figure 2. The specific steps include:

[0031] Step S1, Channel Modeling and Two-Way Channel Probe:

[0032] Constructing a multi-input single-output wireless communication system, such as Figure 3 As shown, the core nodes of the wireless communication system include Alice, a base station equipped with a fluid antenna, Bob, a legitimate user, and Eve, an eavesdropper. The fluid antenna configuration on the base station side... A set of pre-defined antenna ports evenly distributed along a linear space, and equipped with a number of fewer than the total number of ports. The system uses a single radio frequency (RF) link, with software control enabling dynamic selection and activation of antenna ports. Only activated ports are connected to the RF link for signal transmission and reception. The wireless communication system employs a time-division duplex (TDM) communication protocol, leveraging the time reciprocity of the wireless channel to complete uplink and downlink bidirectional channel detection within the channel coherence time, ensuring strong reciprocity of channel state information acquired by base station Alice and legitimate user Bob.

[0033] The wireless channel model constructed in this invention is a mathematical model used to represent signal transmission from the transmitting end of a streaming antenna to the legitimate receiving end and the eavesdropping receiving end. This model includes two types of transmission models: the legitimate channel model and the eavesdropping channel model. Its specific structure is as follows:

[0034]

[0035] In the formula This is the channel vector from the base station to the legitimate user. This is the channel vector from the base station to the eavesdropper. , This represents the path loss factor for the corresponding channel. , Let them be independent and identically distributed complex Gaussian random vectors. The model represents the spatial correlation matrix. It integrates three transmission characteristics: path loss, antenna port spatial correlation, and channel random fading. Furthermore, it leverages the time reciprocity of the time-division duplex protocol to ensure consistent uplink and downlink channel statistical characteristics. This provides a basis for subsequent channel estimation, secret key rate modeling, and joint port and beam optimization, forming the foundation for key generation in this invention.

[0036] The spatial correlation matrix used in this invention A symmetric positive semi-definite complex matrix, which is used to characterize the spatial correlation between all preset ports of the fluid antenna. The matrix contains the ... Line number The elements of the column are calculated using the zeroth-order Bessel function, specifically expressed as:

[0037]

[0038] In the formula It is a zero-order Bessel function of the first kind. , For port indexing, This represents the normalized size of the fluid antenna. The spatial correlation between ports is reflected by the port spacing. The smaller the port spacing, the closer the corresponding matrix elements are to 1, and the stronger the port correlation. The larger the port spacing, the more the corresponding matrix elements show an oscillating decay trend, and the port correlation decreases rapidly. The diagonal elements of this matrix are always 1, representing a constant value for the correlation of the port itself. The off-diagonal elements show a regular decay with the change of port spacing, which can accurately reflect the spatial coupling characteristics of the fluid antenna ports.

[0039] After the channel model is constructed, this step completes channel estimation through uplink and downlink bidirectional pilot signal interaction and the LS algorithm. First, it enters the downlink detection phase, where the base station utilizes beamforming. Send downlink pilot sequence The legitimate user and the eavesdropper receive the transmitted signal simultaneously. The expression for the received signal is:

[0040]

[0041] in, It is Gaussian white noise, and obeys .

[0042] Subsequently, the legitimate user and the eavesdropper completed downlink channel estimation based on the LS algorithm, utilizing... The estimation result of the corresponding channel is directly calculated, specifically as follows:

[0043]

[0044] During the uplink detection phase, legitimate users send unity-power pilot sequences. The signals received by the base station and the eavesdropper are:

[0045]

[0046] in, The transmit power for legitimate users.

[0047] The base station obtains the channel estimation results using the LS algorithm. Then, an equivalent channel estimate is constructed using beamforming vectors. Specifically:

[0048]

[0049]

[0050] in, This is the equivalent noise term.

[0051] Step S2, Key Rate Modeling and Optimization Problem Construction:

[0052] Based on the channel estimation results obtained in step S1, and combined with information theory, rate modeling is performed for two deployment scenarios. The wireless channel scenarios include independent and co-distributed channel scenarios and spatially correlated eavesdropping channel scenarios. In the independent and co-distributed channel scenario, the spatial distance between the eavesdropper Eve and the legitimate user Bob satisfies the channel decorrelation condition, and the eavesdropping channel is statistically independent of the legitimate channel. In the spatially correlated eavesdropping channel scenario, the spatial distance between the eavesdropper Eve and the legitimate user Bob is relatively short, and the eavesdropping channel is spatially correlated with the legitimate channel. In the independent and co-distributed channel scenario, the secret key rate is calculated based on the mutual information of the channel estimates from both legitimate sender and receiver. The specific calculation formula is as follows:

[0053] In space-related eavesdropping channel scenarios, the confidential key rate is calculated based on the conditional mutual information estimated by the legitimate sender and receiver, minus the information entropy obtainable by the eavesdropper. The specific calculation formula is as follows:

[0054] in, For noise variance,

[0055] Furthermore, a joint optimization problem of sparse port selection and transmit beamforming for the fluid antenna is constructed. This optimization problem aims to maximize the secret key rate, while imposing dual constraints on the beamforming vector. The first constraint is the transmit power constraint. The squared L2 norm of the beamforming vector must not exceed the maximum transmit power of the base station. To ensure that the base station's transmission power does not exceed the hardware's capacity limit, the second constraint is a sparse port activation constraint. The beamforming vector's zero norm must not exceed the number of radio frequency links. To ensure that the number of active ports matches the hardware configuration of the fluid antenna, the above two types of constraints together limit the feasible domain of the optimization problem, ensuring that the optimization results meet the engineering implementation requirements.

[0056] The specific optimization problem expression is as follows:

[0057]

[0058]

[0059]

[0060] Step S3, Port Selection and Beamforming Optimization Design:

[0061] For the optimization problem constructed in step S2, an optimization algorithm adapted to the characteristics of the fluid antenna is used for solution. The non-convex optimization problem is transformed into an efficiently solvable convex subproblem through convex approximation. Simultaneously, a reweighting algorithm is used to promote the beamforming vector... The sparsity of the antenna port is used to achieve sparse activation. This is combined with the channel spatial correlation matrix of the fluid antenna. Based on the characteristics of the channel, a low-complexity port selection strategy is designed to screen port combinations with the best, second-best, or performance thresholds for channel quality. The appropriate antenna port activation scheme and transmit beamforming vector are obtained through iterative solution, and the channel detection and signal transmission scheme in the key generation process is determined.

[0062] S31. Convex Approximation of Non-Convex Objective Functions: For non-convex logarithmic determinant terms and higher-order coupling terms in the optimization objective, in the... In the next iteration, based on the beamforming vector of the current iteration We construct a convex approximate lower bound for the objective function by performing a first-order Taylor expansion on the non-convex objective function; simultaneously, we relax the coupling terms using the Cauchy-Schwarz inequality, transforming the original non-convex optimization problem into a convex optimization subproblem, specifically:

[0063]

[0064] in, Spatial correlation matrix The eigenvectors and eigenvalue matrices obtained from singular value decomposition The transformed subproblem uses slack variables. With beamforming vector By jointly constructing an optimization objective while retaining the transmit power constraint and transforming the objective function into a solvable linear constraint form, the original non-convex problem becomes capable of convex optimization solutions.

[0065] S32. Sparse Constraint Handling: Reweighting Norm substitutes for nonconvex Norm constraints The sparsity of the approximate beamforming vector. In the first... In this iteration, a diagonal weighted matrix is ​​introduced. The weighting function is defined as follows:

[0066]

[0067] in, For the first The nth iteration of the beamforming vector One element, This is a regularization parameter used to avoid zero denominators;

[0068] Further in step S2 Norm constraints Replace with weighted Norm constraints:

[0069]

[0070] S33. Iterative Solution Process: Construct an iterative algorithm to solve the convex optimization subproblem. The convex optimization subproblem in this step is solved iteratively. During the iterative execution process, the iteration index is initialized first. Convergence threshold Initial beamforming vector With weighted matrix Then, based on the beamforming vector of the current iteration, the convex optimization subproblem is solved to obtain the updated beamforming vector. Then calculate the confidential key rate corresponding to the current round. After completing the iterative index update, the solution process is repeated until the difference in the secret key rate between two adjacent iterations satisfies the convergence condition. At this point, the algorithm is determined to have converged and the iteration terminates, outputting the current beamforming vector as the optimal solution. The specific process is as follows:

[0071] S33-1: Initialize the iteration index Set convergence threshold Initialize beamforming vector With weighted matrix ;

[0072] S33-2: Beamforming vector based on the current iteration Construct a convex optimization subproblem and solve it to obtain the updated beamforming vector. ;

[0073] S33-3: Calculate the objective function value for the current iteration. , that is, the confidentiality key rate in the corresponding scenario;

[0074] S33-4: Update Iteration Index ;

[0075] S33-5: Repeat steps S33-2 to S33-4 until the convergence condition is met.

[0076] S34. Sliding window port selection strategy: Based on the distribution characteristics of channel spatial correlation of fluid antennas, it selects antenna port combinations with optimal, suboptimal, or performance threshold-compliant channel quality. While meeting the constraints on the number of RF links, it optimizes the key generation rate to obtain an appropriate active port configuration and beamforming vector. The specific process is as follows:

[0077] S34-1, Generate a candidate window set: In The sliding length on each preset port is The window generates A set of candidate windows, each window corresponding to a consecutive set of windows. One preset port;

[0078] S34-2. Selecting the optimal window: For the matrix Perform eigenvalue decomposition to extract the eigenvector corresponding to the largest eigenvalue. . The feature vector The elements corresponding to the ports are assigned to the initial beamforming vector. Set the remaining port elements to zero, and calculate the corresponding values ​​for each window. Then, the power is normalized, and the window whose value meets the performance threshold is selected as the set of adapted active ports;

[0079] S34-3, Iterative optimization of beamforming: Input the generated initial beamforming vector into the iterative optimization algorithm in step S33, and output the adapted activation port configuration and beamforming vector.

[0080] S4. Secret Key Generation: Based on the port activation scheme and beamforming vector determined in S3, both communicating parties synchronously collect full-cycle channel feature sequences and construct channel feature sequences. The channel feature sequences are then sequentially quantized, negotiated, and privacy-enhancing processed to finally generate a symmetric encryption key shared by both legitimate senders and receivers. The specific steps are as follows:

[0081] S41. Full-cycle channel feature sequence acquisition: During the complete key generation cycle, the streaming antenna completes port activation according to the port scheme optimized in step S3. The base station and legitimate users continuously and synchronously perform bidirectional channel probing during the communication cycle, acquiring multiple sets of continuous channel state information to construct a channel feature sequence for key generation, thus completing the process. After sampling, discrete channel feature sequences for base station side and legitimate user side used for key generation are obtained:

[0082]

[0083]

[0084] S42. Channel Feature Quantization and Initial Key Generation: The base station and the legitimate user quantize their respective collected channel feature sequences, set an adaptive quantization threshold based on the statistical characteristics of the channel feature sequences, map continuous channel feature values ​​into discrete binary bit sequences, and generate initial key sequences respectively;

[0085] In this embodiment, Alice and Bob respectively perform statistical analysis on the channel feature sequences they collected, based on the mean of the sequences. An adaptive quantization threshold is set based on variance, and a multi-bit quantization rule is used to map continuous channel feature values ​​into discrete binary bit sequences. Taking two-bit quantization as an example, three quantization thresholds are set. in To quantize the coefficients, the channel feature values ​​are further divided into four quantization intervals, and the quantization mapping rule is as follows:

[0086]

[0087] Using the quantization rules described above, Alice and Bob each extracted highly random initial binary key sequences. As the initial key;

[0088] S43. Due to factors such as environmental noise and non-ideal hardware characteristics, the initial key sequence generated by the two communicating parties may contain a small number of inconsistent bits. The base station and the legitimate user exchange verification information through a public channel, and use a preset error correction mechanism to locate and correct the erroneous bits in the initial key sequence, thereby obtaining a completely consistent negotiated key sequence.

[0089] In this embodiment, Alice and Bob exchange a small amount of verification data through a public wireless channel, using BCH linear block error correction code to update their initial key sequences. The inconsistent bits in the key are located and corrected to eliminate bit differences caused by environmental noise and hardware non-ideal characteristics, thereby obtaining a completely consistent negotiated key sequence between the two parties.

[0090] S44. Privacy Amplification and Shared Key Generation: To eliminate the information entropy that may be leaked to eavesdroppers during the information negotiation process in step S43 via public channel interaction, the base station and the legitimate user use a pre-agreed randomness enhancement algorithm to compress and purify the negotiated key sequence, removing redundant and leakable information to ultimately generate a shared, highly random, and highly secure symmetric encryption key. In this embodiment, Alice and Bob use a pre-agreed one-way hash function. The negotiated key sequence is compressed and purified to eliminate information entropy that might be leaked to Eve during the negotiation process. Redundant information in the sequence is also removed to increase the randomness and unpredictability of the key, ultimately generating a shared symmetric encryption key. .

[0091] Test case

[0092] The communication scenario parameters for this test are consistent with those in Embodiment 1 of this invention. The positions of the base station (Alice) and the legitimate user (Bob) are set to the origin (0 m, 0 m) and (70 m, 0 m) respectively, with a straight-line distance of 70 m between them. The location of the eavesdropper (Eve) is set for two typical eavesdropping scenarios: In the independent and identically distributed scenario, Eve's position is set in a region statistically independent of Bob, such as (35 m, 35 m), ensuring that the eavesdropping channel and the legitimate channel satisfy the statistical independence characteristic; In the spatially correlated eavesdropping scenario, Eve's position is evenly distributed within a circular region centered on Bob with a radius of half a wavelength, ensuring that the eavesdropping channel and the legitimate channel have significant spatial correlation.

[0093] This test selects two mainstream fixed antenna physical layer key generation methods in existing technologies as comparison objects, and sets up two implementations of the present invention for performance verification. The hardware and algorithm configurations of each scheme are as follows: The first comparison scheme is the traditional fixed antenna optimized beamforming scheme "FA OPT". The base station is equipped with a traditional fixed antenna array, without dynamic port selection capability, and only uses an iterative optimization algorithm to solve for the optimal beamforming vector, corresponding to the joint optimization scheme based on fixed antennas in existing technologies. The second comparison scheme is the traditional fixed antenna maximum ratio combining scheme "FA MRC". The base station is equipped with a traditional fixed antenna array, and uses maximum ratio combining technology to design the beamforming vector, which is a typical low-complexity implementation scheme in existing technologies. The two test schemes of the present invention are the fluid antenna joint optimization scheme "FAS OPT" and the fluid antenna sliding window optimization scheme "FAS Sliding Window". The fluid antenna joint optimization scheme adopts the port selection and beamforming joint optimization method of the present invention, and the fluid antenna sliding window optimization scheme adopts the low-complexity sliding window port selection strategy of the present invention combined with the beamforming optimization method. The base stations of both schemes are equipped with fluid antennas, and the total number of ports is... The number of RF links is set to and .

[0094] This test uses base station transmit power as a variable, with a value range of 15 dBm to 35 dBm and a step size of 5 dBm, to test the encryption key rate of different schemes at various power points. Specifically, as follows... Figure 4 and Figure 5 As shown. Figure 4 This is a comparison chart of the confidential key rate under different transmit power for different port selection schemes in an independent and co-distributed channel scenario according to an embodiment of the present invention. Figure 5 This is a comparison chart of the confidentiality key rate under different transmit power for different port selection schemes in a space-related eavesdropping scenario according to an embodiment of the present invention. The chart shows that, at all base station transmit power points, the confidentiality key rate of the fluid antenna joint optimization scheme of the present invention is significantly higher than that of the two traditional fixed antenna schemes. For example, in... Figure 5 In this invention, at 35 dBm, the key rate under the N=5 configuration is 2.2 bit / channel use, which is approximately 47% higher than the 1.5 bit / channel use of the traditional fixed antenna maximum ratio combining scheme. This improvement is achieved with the increase in the number of RF links. With the improvement, the performance advantages of the solution of this invention are further expanded. The key rate is 2.5 bits / channel, compared to The speed improvement is approximately 14%, while traditional fixed antenna solutions are limited by the number of ports and cannot be further improved through dynamic port selection. Furthermore, as the base station transmit power increases, the key rate of all solutions shows an upward trend, but the growth slope of the proposed solution is greater, and its performance advantage is more obvious in high-power scenarios. This indicates that the joint optimization mechanism of port selection and beamforming in this invention can utilize transmit power more efficiently and improve key generation efficiency.

[0095] In summary, the fluid antenna physical layer key generation method of the present invention, compared with the traditional fixed antenna scheme, can significantly improve the confidentiality key rate under the same hardware resources and transmission power conditions, while adapting to complex eavesdropping scenarios and possessing higher security and engineering practicality.

[0096] In summary, the method of this invention is applicable to resource-constrained wireless communication and IoT access scenarios. Addressing the core problems of traditional physical layer key generation technologies—such as high RF link overhead in fixed antenna systems, insufficient key random entropy in spatially correlated channels, low key generation rate, and high computational complexity and difficulty in engineering deployment due to joint optimization of port selection and beamforming—this invention proposes a physical layer key generation method based on fluid antennas. Unlike existing traditional schemes that rely solely on fixed antenna arrays and only pursue key rate improvement while ignoring hardware resource overhead, this invention constructs a key generation optimization framework centered on dynamic channel customization using fluid antennas. It can output various feasible solutions, including optimal, suboptimal, and stochastic optimizations. Based on the spatial correlation characteristics of the channel, it jointly designs sparse port selection and beamforming, improving the key entropy of legitimate channels and significantly increasing the key generation rate in independent, identically distributed, and spatially correlated channel scenarios. This method is suitable for physical layer secure key generation and wireless access security protection in various scenarios such as IoT, industrial control, and 5G / 6G edge networks.

[0097] It should be noted that the above content merely illustrates the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. For those skilled in the art, various improvements and modifications can be made without departing from the principle of the present invention, and all such improvements and modifications fall within the scope of protection of the claims of the present invention.

Claims

1. A physical layer key generation method based on a fluidic antenna, characterized in that, Includes the following steps: S1. Channel Modeling and Two-Way Channel Probe: Construct a multi-input single-output communication system model in time-division duplex (TDD) mode. The communication system includes a base station equipped with a streaming antenna, a single-antenna legitimate user, and a single-antenna eavesdropper. The number of streaming antennas is greater than a preset number of RF links. Based on the communication system, a wireless channel model is constructed, and a spatial correlation matrix is ​​introduced to characterize the spatial correlation between the ports of the streaming antennas. Channel probe is completed through uplink and downlink bidirectional pilot transmission to obtain channel state information between the base station and the legitimate user. S2. Key Rate Modeling and Optimization Problem Construction: For independent and co-distributed channel scenarios and spatially correlated eavesdropping channel scenarios, the system's confidential key rate is calculated respectively. With the objective of maximizing the confidential key rate, a non-convex joint optimization problem of port selection and beamforming is constructed based on base station transmit power constraints and sparse port activation constraints of fluid antennas. S3. Port selection and beamforming optimization solution: The non-convex joint optimization problem constructed in step S2 is transformed into a convex approximation. The sparse activation of the fluid antenna port is achieved through sparse constraint processing. Based on the sliding window port selection method, the optimal port activation scheme and transmit beamforming vector are obtained by iterative solution. S4. Secret Key Generation: Based on the port activation scheme and beamforming vector determined by S3 optimization, both communicating parties synchronously collect full-cycle channel feature sequences and construct channel feature sequences. The channel feature sequences are then quantized, negotiated, and privacy amplified in sequence to finally generate a symmetric encryption key shared by the legitimate sending and receiving parties.

2. The physical layer key generation method based on a fluidic antenna according to claim 1, characterized in that: The wireless channel model in step S1 includes a legal channel model and an eavesdropping channel model, and its specific structure is as follows: ； In the formula This is the channel vector from the base station to the legitimate user. This is the channel vector from the base station to the eavesdropper. , This represents the path loss factor for the corresponding channel. , Let them be independent and identically distributed complex Gaussian random vectors. This is the spatial correlation matrix; The spatial correlation matrix Let be a symmetric positive semi-definite complex matrix, and let the nth matrix be a symmetric positive semi-definite complex matrix. Line number The elements of the column are calculated using the zeroth-order Bessel function, specifically: ； In the formula It is a zero-order Bessel function of the first kind. , For port indexing, This represents the normalized size of the fluid antenna.

3. The physical layer key generation method based on a fluidic antenna according to claim 2, characterized in that: The bidirectional channel detection in step S1 includes downlink detection and uplink detection, wherein, During the downlink detection phase, the base station transmits a unity-power pilot sequence. Signals received by legitimate users and eavesdroppers for: ； in, For channel vectors, The base station's transmit beamforming vector. It is Gaussian white noise, and obeys ; Legitimate users and eavesdroppers exploit the LS algorithm. Channel estimation is obtained: ； in, and This is the equivalent noise term; During the uplink probing phase, legitimate users transmit unity-power pilot sequences. The signals received by the base station and the eavesdropper are: ； The base station utilizes the LS algorithm Channel estimation results Equivalent channel estimation results are constructed using beamforming vectors. The details are as follows: ； ； in, For channel vectors, This is the equivalent noise term.

4. The physical layer key generation method based on a fluidic antenna according to claim 3, characterized in that: In the independent and co-distributed channel scenario described in step S2, the secret key rate The mutual information calculation based on the channel estimation of legitimate sender and receiver is as follows: ； in, For legitimate users' transmission power, The channel gain between the base station and the user; In space-related eavesdropping channel scenarios, the secret key rate Based on the conditional mutual information calculation of the channel estimation between the legitimate sender and receiver, and after deducting the entropy of information obtainable by the eavesdropper, the specific calculation is as follows: ； in, For noise variance, These are equivalent intermediate variables; The base station transmit power constraint is specifically: the square of the second norm of the beamforming vector must not exceed the maximum transmit power of the base station; the sparse port activation constraint of the fluid antenna is specifically: the zero norm of the beamforming vector must not exceed the number of radio frequency links.

5. The physical layer key generation method based on a fluidic antenna according to claim 1, characterized in that, The sparsity processing in step S3 is to promote beamforming vectors through a reweighting algorithm. Due to the sparsity of the data, the reweighting algorithm specifically includes: S31. Introducing a diagonal weighted matrix The weighting function is defined as follows: ； in, For the first The nth iteration of the beamforming vector One element, For regularization parameters; S32, will Norm constraints replaced with weighted norms Norm constraints: ； in, This is the weight matrix. is the beamforming vector, and N is the number of RF chains; S33, Iteratively update the weight matrix and beamforming vector Until it converges.

6. The physical layer key generation method based on a fluidic antenna according to claim 5, characterized in that, The sliding window port selection method in step S3 specifically includes: S31. Generate a candidate window set: In The sliding length on each preset port is The window generates A set of candidate windows, each window corresponding to a consecutive set of windows. One preset port; S32. Selecting the optimal window: For the spatial correlation matrix Perform eigenvalue decomposition to extract the eigenvector corresponding to the largest eigenvalue. , to feature vector The elements corresponding to the ports are assigned to the initial beamforming vector. Set the remaining port elements to zero, and calculate the corresponding values ​​for each window. Then, the power is normalized, and the window whose value meets the performance threshold is selected as the set of adapted active ports; S33, Iterative Optimization Beamforming: Input the generated initial beamforming vector into the iterative optimization algorithm, and output the adapted activation port configuration and beamforming vector.

7. The wireless channel key generation method based on a flowing antenna as described in claim 1, characterized in that: Step S4 specifically includes the following steps: S41. Full-cycle channel feature sequence acquisition: The base station and the legitimate user continuously and synchronously conduct bidirectional channel detection during the communication cycle, acquire multiple sets of continuous channel state information, and construct a channel feature sequence for key generation. S42. Channel Feature Quantization and Initial Key Generation: The base station and the legitimate user respectively quantize the channel feature sequences they collect, set an adaptive quantization threshold based on the statistical characteristics of the channel feature sequences, map continuous channel feature values ​​into discrete binary bit sequences, and generate initial key sequences respectively. S43. Key information negotiation: The base station and the legitimate user exchange verification information through a public channel, and use a preset error correction mechanism to locate and correct the erroneous bits in the initial key sequence to obtain a completely consistent negotiated key sequence. S44. Privacy Amplification and Shared Key Generation: The base station and the legitimate user use a pre-agreed randomness enhancement algorithm to compress and purify the key sequence negotiated in step S43, removing redundant and leakable information from the sequence to generate a symmetric encryption key.

8. The wireless channel key generation method based on a flowing antenna as described in claim 7, characterized in that: In step S42, the quantization process involves setting an adaptive quantization threshold based on the mean and variance of the sequence, and using a multi-bit quantization rule to map continuous channel feature values ​​into discrete binary bit sequences to generate an initial key. The specific quantization mapping rule is as follows: ； Among them, 11, 10, 01, and 00 refer to double-bit quantization. There are three quantization thresholds. The mean of the sequence. This is the quantization coefficient.

9. A wireless channel key generation system based on a fluid antenna, comprising a computer program, characterized in that: When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1-8 above.

Images