Physical layer authentication method and system based on random beam scanning and power feedback

Abstract

The invention discloses a physical layer authentication method and system based on random beam scanning and power feedback. The method comprises the steps that a transmitting end randomly arranges a predefined angle set in each round of scanning and transmits test signals one by one, and a receiving end measures power and feeds back an optimal beam label. Although the eavesdropping end can monitor the same signal, the eavesdropping end can only obtain partial power sorting information and cannot accurately deduce the optimal beam. According to the invention, the security of physical layer authentication can be effectively improved, the disguise risk is reduced, and the method is suitable for large-scale MIMO and millimeter wave communication systems.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication technology, specifically, it relates to a physical layer authentication method and system based on random beam scanning and power feedback. Background Technology

[0002] In fifth-generation (5G) and future sixth-generation (6G) mobile communication systems, massive millimeter-wave multiple-input multiple-output (MIMO) technology has become a core means to achieve high-speed, low-latency communication. Beamforming technology improves the signal quality and coverage of communication links by adjusting the phase and amplitude of the antenna array at the transmitting or receiving end to focus signal energy in a spatial direction.

[0003] However, in open wireless channels, beam scanning and feedback mechanisms are easily exploited by malicious eavesdroppers. For example, a legitimate receiver (Bob) will report the index of the beam with the strongest signal during scanning, while an eavesdropper (Eve) can listen to the feedback information and analyze the signal strength they receive to attempt to impersonate a legitimate user, thus posing a threat to system security.

[0004] Most existing physical layer authentication schemes are based on Channel State Information (CSI) or statistical characteristics for decision-making. However, their security and robustness remain limited under conditions of random channel fluctuations and the eavesdropping of feedback signals. Especially in random beam scanning scenarios, the eavesdropper (Eve) can infer the optimal beam direction of the legitimate receiver (Bob) through side channel information, increasing the probability of successful spoofing.

[0005] Therefore, how to enhance the security of physical layer authentication through randomization and signal power characteristics while maintaining high communication performance is a key technical problem that urgently needs to be solved. Summary of the Invention

[0006] To address the shortcomings of the existing technologies, the present invention aims to provide a physical layer authentication method and system based on random beam scanning and power feedback. By introducing a random scanning sequence and power feedback mechanism during the beam scanning stage, the present invention enables legitimate users to reliably identify the optimal beam, while eavesdroppers can only obtain partial sorting information, thereby improving the security and anti-spoofing capabilities of authentication. This invention is applicable to large-scale MIMO and millimeter-wave communication systems.

[0007] The technical solution of the present invention is described in detail below.

[0008] This invention provides a physical layer authentication method based on random beam scanning and power feedback, comprising the following steps: Step 1: Initialization The transmitter predefines and generates a set of angle indices. A unique label is assigned to each direction, and antenna array parameters are set; where the angle index set... Antenna array parameters include the number of antennas. and antenna spacing ; Step 2: Beam Scanning and Measurement The transmitting end randomly arranges the predefined angle index set and transmits test signals in a random order; the receiving end receives signals from each direction of the transmitting end one by one and measures the received power in each beam direction. Step 3: Maximum Power Detection The receiver independently acquires multiple power samples in each beam direction and averages them. The optimal beam label is determined based on the angle corresponding to the maximum average power. ; Step 4: Feedback and Certification The receiver feeds back the optimal beam tag. , Each beam direction is assigned a unique label, and the transmitter verifies whether the label matches the statistics of the legitimate channels.

[0009] In this invention, in step two, the test signal is a complex Gaussian signal.

[0010] In this invention, in step three, the receiving channel of the receiving end is modeled as a Rician fading channel.

[0011] This invention also provides a physical layer authentication system based on random beam scanning and power feedback, which includes a transmitter and a receiver; wherein: At the transmitting end, generate an angle index set. A unique label is assigned to each direction, and antenna array parameters are set. Test signals are transmitted in a random order by randomly arranging a predefined set of angle indices to verify whether the optimal beam label fed back by the receiver matches the legal channel statistics. At the receiving end, signals from each direction at the transmitting end are received sequentially, and the received power in each beam direction is measured. Multiple power samples are independently collected in each beam direction and averaged. The optimal beam tag is determined and fed back based on the angle corresponding to the maximum average power. .

[0012] Furthermore, in the system of the present invention, the workflow of the transmitting end includes: 1. Generate an angle index set The transmitter establishes a set of angle indexes containing multiple preset directions and assigns a unique label to each direction; 2. Set antenna array and beamforming parameters The transmission weights are configured according to the array structure, enabling the transmitter to form beams in all preset directions; 3. Transmit signals in random order. At the start of each scan, the set of angle indices is randomly arranged (the correspondence between labels and directions remains unchanged) to obtain a new transmission order, and beams are formed in each direction in the random order to send complex Gaussian modulated signals. 4. Receive feedback and verify it. After receiving the tag from the receiver, the transmitter uses a pre-defined tag table to determine whether the feedback conforms to the characteristics of a legitimate channel for identity verification.

[0013] Furthermore, in the system of the present invention, the receiving end workflow includes: 1. Receive signals from all transmitted beams in sequence. The receiving end receives signals from each direction one by one according to the scanning order of the transmitting end; 2. Collect multiple power samples independently for each direction and calculate the average. The received power is collected multiple times in each direction, and the fluctuation caused by instantaneous fading is reduced by averaging. 3. Determine the optimal direction Compare the average power in all directions and find the direction with the highest power. 4. Feedback on corresponding tags The tag corresponding to the optimal direction is sent back to the transmitter for identity authentication.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention achieves secure authentication in the physical layer of large-scale MIMO by introducing a random scanning order and power statistical analysis mechanism during beam scanning. Compared with traditional fixed scanning and direct feedback methods, this invention has significant technical advantages. The transmitter randomly permutes the angle codebook before each scan, making the beam scanning order completely independent in different scans. This breaks the fixed correspondence between angle indices and beam directions, preventing eavesdroppers from deducing the optimal beam direction by listening to feedback information.

[0015] On the other hand, although the eavesdropping terminal can receive the same scanning sequence, because its channel is independent of the legitimate receiver and has different statistical characteristics, it can only obtain ranking information of the received power, which differs from that of the legitimate receiver. Therefore, it cannot accurately determine the optimal beam selected by the legitimate receiver. Consequently, the probability of the eavesdropper's successful spoofing is limited by statistical mismatch, and its authentication information remains at a fuzzy ranking level, unable to form a deterministic judgment. Through this mechanism, this invention significantly improves the security and robustness of physical layer authentication while maintaining low system complexity, making it suitable for scenarios requiring high-security channel feedback, such as large-scale MIMO and millimeter-wave communication. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the overall structure of the physical layer authentication system of the present invention.

[0017] Figure 2 This is a schematic diagram of the physical layer authentication system based on random beam scanning and power feedback according to the present invention. Detailed Implementation

[0018] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0019] This invention is based on a physical layer authentication system comprising a transmitter (Alice), a legitimate receiver (Bob), and a potential eavesdropping device (Eve). The transmitter has... A uniform linear array antenna with an antenna spacing of . ,in The carrier wavelength is [wavelength]. The system structure is as follows: Figure 1 As shown.

[0020] Alice sends a complex Gaussian signal to Bob. Assume its signal power During beam scanning, Alice sets the angles... , Perform a scan one by one. The beamforming vector for each beam direction is defined as: , Both Bob's and Eve's receive channels are modeled as Rician fading channels, with the following expression: , in, This is the path loss coefficient; Rice factor; For the line-of-sight component, ; This is the non-line-of-sight component.

[0021] The expressions for the received signals by Bob and Eve are:

[0022] in Pointing angle The conjugate transpose of the beamforming weight vector For Bob and Eve in the direction The channel vector is given by s, which is the complex Gaussian modulated signal transmitted by the transmitter, and the noise term is given by s. , Its variance is determined by the signal-to-noise ratio. Decide.

[0023] To enhance authentication security, Alice updates the angle index set before each round of scanning. A random scan is performed, and this random sequence ensures that the beam direction between different scan rounds is unpredictable.

[0024] At the same time, each beam direction corresponds to a unique label. To form a tag set .

[0025] During the scan, Bob calculates the average received power for each beam direction: , Where E represents the expected value (mean), which is the statistical average of the squares of the received signal amplitude. Bob calculates based on the maximum average received power. Determine the optimal beam label , in, Bob then sends the tag corresponding to the maximum power to Alice, who then passes the safety verification.

[0026] Eve simultaneously monitors the scan sequence and forms a power vector: The sorting result is .

[0027] The position of the beam returned by Bob in Eve's sort is: .

[0028] Therefore, Bob can directly identify its strongest beam, while Eve can only know the beam's ranking in her own observations. This information asymmetry means that Eve can only rely on incomplete and uncertain channel information when carrying out an impersonation attack.

[0029] To analyze the distribution of the received signals by Bob and Eve and suppress the effects of transient small-scale fading and noise, Bob and Eve independently acquired data in each beam direction. A power sample and average: .

[0030] According to the central limit theorem, when N is sufficiently large, the average power... It approximately follows a Gaussian distribution. Its mean With variance They are determined by the following formulas respectively: , , in, and These are the second and fourth moments of the channel gain, respectively, reflecting the statistical characteristics of the channel energy. , .

[0031] Therefore, it can be seen that the mean and variance of the average power are simultaneously affected by the channel statistical parameters (Rician factor). Path loss ) and beam direction characteristics ( The joint decision of ).

[0032] when When the sample size is large, averaging the samples can effectively mitigate the effects of small-scale fading and noise. This can be achieved through analysis. Based on the Gaussian statistical characteristics, Bob is able to stably determine the optimal beam label. Eve can only obtain limited statistical information from the power sort and the position of Bob's chosen index in that sort. .

[0033] During the authentication phase, Eve attempts to deduce the ranking based on its own observation strategy. Corresponding tags The probability of successfully impersonating Bob can be expressed as: .

[0034] Since Eve's inference depends on Bob's optimal index. Its position in the sort, the probability can be expressed as , The first item indicates the position of Bob's selected index in Eve's sort. The first term represents the probability of the digit being inferred, and the second term represents the probability that Eve will be able to make a successful deduction given that this information is known.

[0035] This formula illustrates that Eve's success probability is essentially limited by the statistical difference between Bob's and Eve's channel observations. Even if Eve utilizes complete ordering information, it cannot guarantee that its inferred index will match Bob's strongest beam index. Furthermore, when the channel differences are significant, the information Eve can obtain is limited to probabilistic inference, making it difficult to bypass authentication.

[0036] like Figure 2 As shown, the physical layer authentication system provided by this invention includes: The transmitting unit Alice includes a random scan control module, a beamforming module, and a tag allocation and matching module; The workflow of each module is as follows: 1. Random Scan Control Module This module is used to randomly arrange a predefined set of angle indices before each scan begins, generating an unpredictable scan order, thereby ensuring that the transmission order of each beam direction is random during beam scanning.

[0037] 2. Beamforming module This module selects the corresponding beam direction and generates the corresponding beamforming vector based on the randomly arranged angle index. Test signals are transmitted sequentially in each direction, enabling the receiver to measure the received power in that direction.

[0038] 3. Tag assignment and matching module This module pre-assigns a unique tag to each beam direction and, during the authentication phase, matches and verifies the tag returned by Bob with the legal tags recorded locally to determine whether the receiver is a legitimate user.

[0039] The legitimate receiver Bob includes a power measurement module, an optimal beam detection module, and a feedback module. The workflow of each module is as follows: 1. Power Measurement Module This module is used to receive test signals transmitted sequentially by Alice, collect multiple received power samples in each beam direction, and average the samples to suppress the effects of transient noise and small-scale fading.

[0040] 2. Optimal Beam Detection Module Based on the average received power of each beam direction, this module determines the direction with the highest power and obtains the unique tag corresponding to that direction as the optimal beam identifier for this round of scanning.

[0041] 3. Feedback Module This module feeds back the label of the optimal beam to Alice for subsequent legitimacy authentication.

[0042] Eve, the eavesdropping module, includes a monitoring module, a power sorting module, and a feedback module.

[0043] The workflow of each module is as follows: 1. Listening module This module passively receives Alice's test signals during the scanning process and listens for the optimal beam tag returned by Bob.

[0044] 2. Power sorting module This module measures and averages the received power in each beam direction and sorts them according to the power values.

[0045] 3. Feedback Module When carrying out a spoofing attack, the module returns a speculative label to Alice based on power ranking and monitoring results. However, since its channel statistics are inconsistent with Bob's, the speculation contains errors and cannot be guaranteed to be consistent with Bob's actual choice.

[0046] Based on the above system, the implementation steps are as follows: Initialization: Alice random scan control module generates angle set A random scanning order mechanism is set, and the beamforming module uses predefined parameters. , Initialize the array.

[0047] Beam scanning and measurement: Alice transmits test signals one by one according to the order generated by the random scanning control module; the beamforming module forms a corresponding beam for each angle; Bob's power measurement module measures the power of the received signal.

[0048] Maximum Detection: Bob's optimal beam detection module finds the label corresponding to the maximum value from all average power values. .

[0049] Feedback and Authentication: Bob's Feedback Module Optimal Beam Label Feedback is sent to Alice; Alice's tag assignment and matching module verifies whether the tag matches a valid tag.

[0050] Eve's monitoring and inference: Eve's monitoring module captures Alice's scan sequence and the tags returned by Bob; the power sorting module sorts the power it receives and infers the sorting position of the legitimate end tags; the feedback module performs a spoofing attack and returns the tags in the inferred sorting position in the next round of verification.

[0051] Result verification: The inventors conducted simulation experiments on random beam scanning in a typical Rican channel environment. In the experiment, the transmitter was equipped with a 12-array antenna, with a scanning angle range of −90° to 90° and a step size of 10°. The legitimate receiver, Bob, was positioned at 30°, while the eavesdropping receiver, Eve, was positioned at 20°. The results showed that, under the combined effects of random scanning and power averaging, Bob could reliably identify its optimal beam direction. Eve, however, due to significant differences in channel statistical characteristics compared to Bob, could only make uncertain inferences based on the ranking of received power, resulting in a single-round success rate of only about 5.61%. Furthermore, repeated verification rounds exponentially reduced this probability.

[0052] In this invention, the transmitting end randomly arranges a predefined set of angles and transmits test signals one by one in each scan round, while the receiving end measures the power and feeds back the optimal beam tag. This invention introduces randomness and statistical isolation into the physical layer authentication mechanism. Through random beam indexing and power averaging, it ensures Bob's authentication remains robust, effectively reducing the probability of a successful eavesdropping attack.

Claims

1. A physical layer authentication method based on random beam scanning and power feedback, characterized in that, Includes the following steps: Step 1: Initialization The transmitter predefines and generates a set of angle indices. A unique label is assigned to each direction, and antenna array parameters are set; where the angle index set... Antenna array parameters include the number of antennas. and antenna spacing ; Step 2: Beam Scanning and Measurement The transmitting end randomly arranges the predefined angle index set and transmits test signals in a random order; the receiving end receives signals from each direction of the transmitting end one by one and measures the received power in each beam direction. Step 3: Maximum Power Detection The receiver independently acquires multiple power samples in each beam direction and averages them. The optimal beam label is determined based on the angle corresponding to the maximum average power. ; Step 4: Feedback and Certification The receiver feeds back the optimal beam tag. , Each beam direction is assigned a unique label, and the transmitter verifies whether the label matches the statistics of the legitimate channels.

2. The physical layer authentication method according to claim 1, characterized in that, In step two, the test signal is a complex Gaussian signal.

3. The physical layer authentication method according to claim 1, characterized in that, In step two, the receiving channel at the receiver is modeled as a Rician fading channel.

4. A physical layer authentication system implementing the method of claim 1, characterized in that, It includes a transmitter and a receiver; wherein: At the transmitting end, generate an angle index set. A unique label is assigned to each direction, and antenna array parameters are set. Test signals are transmitted in a random order by randomly arranging a predefined set of angle indices to verify whether the optimal beam label fed back by the receiver matches the legal channel statistics. At the receiving end, signals from each direction at the transmitting end are received sequentially, and the received power in each beam direction is measured. Multiple power samples are independently collected in each beam direction and averaged. The optimal beam tag is determined and fed back based on the angle corresponding to the maximum average power. .

Images