A Joint Optimization Method for Covert MISO Communication with Rotatable Antenna under Finite Character Input

By utilizing the lower bound of total mutual information and the upper bound of Kullback-Leibler divergence under finite character input, the joint optimization problem of rotatable antennas is transformed into a problem of maximizing legitimate received power, solving the problems of insufficient spatial degrees of freedom and high computational complexity, and realizing low-complexity covert MISO communication optimization.

CN123122610APending Publication Date: 2026-11-17ANHUI AGRICULTURAL UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611628038.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-21
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing technologies struggle to effectively utilize the spatial degrees of freedom of rotatable antennas under limited character input, making it difficult to directly process mutual information and concealment constraints, and resulting in high computational complexity for joint optimization.

Method used

By employing a lower bound on the total mutual information and an upper bound on the Kullback-Leibler divergence, the joint optimization problem of maximizing the total mutual information is transformed into a problem of maximizing the received power at the legitimate receiver. Furthermore, by alternately updating the transmit beamforming vector and the orientation of the rotatable antenna, the unit norm constraint is handled using a penalty function convex-concave procedure, thus avoiding the need to repeatedly solve the general convex optimization problem.

Benefits of technology

Without moving the array elements, spatial degrees of freedom are increased, a low-complexity joint optimization method is provided, the signal at the legitimate receiving end is enhanced and the signal leakage at the monitoring end is controlled, thus meeting the concealment requirements.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123122610A_ABST
    Figure CN123122610A_ABST
Patent Text Reader

Abstract

This invention discloses a joint optimization method for concealed MISO communication using rotatable antennas under finite character input. The transmitter is equipped with multiple rotatable antennas with independently adjustable zenith and azimuth angles. A legitimate link is established and the link channel is monitored based on directional gain. Using a lower bound on mutual information and an upper bound on Kullback-Leibler divergence, the joint design is transformed into receive power optimization. When the antenna orientation is fixed, beamforming is updated using second-order cone programming; when the beam is fixed, the antenna orientation is updated using differential convex approximation and a penalty function convex-concave procedure. Furthermore, a low-complexity pattern is formed using KKT closed beam updates and a finite set of candidate angles consisting of stationary points, concealment boundaries, and angular boundaries. This invention improves the mutual information of the legitimate receiver and reduces computational overhead while satisfying transmit power and concealment constraints.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the fields of wireless communication and physical layer security technology, and in particular to a joint optimization method for concealed MISO communication with rotatable antennas under limited character input. Background Technology

[0002] Covert wireless communication aims to reduce the ability of monitoring devices to detect communication activities while transmitting information from the transmitter to the legitimate receiver. Compared with traditional physical layer security technologies that only protect the transmitted content, covert communication further focuses on whether the transmission activity itself can be detected. Multi-antenna transmitters can use spatial degrees of freedom to enhance the signal of the legitimate receiver and suppress the received power of the monitoring device; therefore, beamforming has become an important means to improve the performance of covert communication.

[0003] Existing covert communication schemes with finite character inputs are typically designed for fixed-direction antennas, smart reflectors, or power and beamforming. Finite character inputs utilize discrete constellations such as PSK and QAM, which better reflect real-world digital communication systems than ideal Gaussian inputs. However, their mutual information includes expectation, exponential summation, and logarithmic nesting, and the monitoring end exhibits a Gaussian mixture distribution during transmission, making it difficult to directly use mutual information and detection performance for joint optimization. Existing related schemes can address these issues using lower bounds on mutual information, upper bounds on Kullback-Leibler divergence, positive semidefinite relaxation, or deep reinforcement learning. However, the physical pointing of the transmitting antenna is usually fixed, making it difficult to further utilize the spatial degrees of freedom provided by the individual antenna pattern.

[0004] Movable antennas adapt to spatial channels by changing the positions of array elements, and reconfigurable surfaces reconstruct the propagation environment by adjusting reflection or transmission coefficients. However, their deployment, control variables, and structural conditions differ from those of directly rotating transmitting antennas. Rotatable antennas can adjust the main lobe direction of the beam without changing the element mounting positions. Their directional gain, along with the transmit beamforming, jointly determines the effective channel for legitimate receivers and monitoring ends. This results in a non-convex optimization problem with strong coupling between the beamforming vector and the orientation of multiple antennas; solving a general convex optimization problem for each update would incur significant computational overhead.

[0005] Therefore, a joint optimization method for rotatable antenna concealed MISO communication with limited character input is needed. This method should optimize the transmit beamforming and antenna orientation while meeting the requirements for transmit power and concealment, and provide a low-complexity implementation suitable for limited computing resources. Summary of the Invention

[0006] The purpose of this invention is to provide a joint optimization method for concealed MISO communication with rotatable antennas under finite character input, so as to solve the problems of insufficient spatial degrees of freedom of fixed-direction antennas, difficulty in directly handling mutual information and concealment constraints of finite character inputs, and high computational complexity of joint optimization.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] A joint optimization method for concealed MISO communication with a rotatable antenna under finite character input includes the following steps:

[0009] S1. Construct a covert MISO communication system including a transmitter, a legitimate receiver, and a monitoring end. Obtain the positions of each antenna, the position of the legitimate receiver, the position of the monitoring end, a finite character constellation set, noise power, maximum transmission power, and maximum rotation angle. The transmitter is equipped with K rotatable antennas, each with an adjustable zenith angle and an adjustable azimuth angle. The legitimate receiver and the monitoring end are each equipped with a single antenna.

[0010] S2. Convert the zenith angle and azimuth angle of each rotatable antenna into a three-dimensional unit pointing vector. Determine the cosine power directional gain based on the inner product of the pointing vector and the node direction vector. Establish equivalent channels from the transmitter to the legal receiver (i.e., the legal link) and from the transmitter to the monitoring end (i.e., the monitoring link).

[0011] S3. Select the transmitted symbols from the normalized finite character constellation set with equal probability, determine the total mutual information based on the received signals of the legal receivers under finite character input, determine the minimum total detection error probability and concealment constraint based on the observation distribution of the monitoring end in the non-transmitting state and the transmitting state, and establish the total mutual information maximization problem.

[0012] S4. By using the lower bound of total mutual information and the upper bound of Kullback-Leibler divergence, the joint optimization problem of maximizing total mutual information is transformed into the problem of maximizing the received power of the legitimate receiver. That is, under the constraints of transmit power, monitoring end received power and rotatable antenna orientation, the legitimate receiver's received power is maximized.

[0013] S5. Alternately update the transmit beamforming vector and the orientation of each rotatable antenna: when the antenna orientation is fixed, update the beamforming vector with second-order cone programming; when the beamforming vector is fixed, update the antenna orientation with differential convex approximation and penalty function convex-concave program.

[0014] When the transmit beamforming vector is fixed, the received power of the legitimate receiver and the monitoring end is expressed as a differential convex function, and the lower bound of the received power of the legitimate receiver and the upper bound of the received power of the monitoring end are constructed at the linearization point. The unit norm constraint is decomposed into the upper bound of the norm constraint and the lower bound of the norm constraint after first-order linearization. Non-negative relaxation variables and incremental penalty parameters are introduced, and the three-dimensional pointing vector of each rotatable antenna is updated through the penalty function convex-concave procedure.

[0015] S6. When the change in received power of the legitimate receiver in two consecutive iterations is less than the preset convergence threshold or the number of iterations reaches the preset upper limit, output the transmit beamforming vector and the orientation of each rotatable antenna.

[0016] Compared with the prior art, the present invention has the following beneficial effects:

[0017] 1. The physical orientation of each rotatable antenna is used as an optimizable variable and designed in conjunction with transmit beamforming to increase spatial freedom without moving the array element positions.

[0018] 2. For finite character input, the mutual information and divergence constraints corresponding to the Gaussian mixture distribution at the monitoring end, which are difficult to calculate directly, are represented by tractable lower and upper bounds, respectively, and a receiving power optimization model is formed based on the monotonic relationship.

[0019] 3. In the orientation update, construct conservative approximations with correct orientation for both the legal received power and the received power at the monitoring end, and handle the unit norm constraint through a penalty function convex-concave procedure.

[0020] 4. Provides closed beam updates based on KKT conditions and finite candidate angle search based on stagnation points, concealed boundaries, and angular boundaries, avoiding repeated solutions to the general convex optimization problem.

[0021] 5. It can enhance the signal of the legitimate receiver under different antenna numbers, maximum rotation angles and concealment levels, while controlling signal leakage at the monitoring end. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the rotatable antenna concealed MISO communication system of the present invention.

[0023] Figure 2 This is a schematic diagram of the overall process of the joint optimization method of the present invention.

[0024] Figure 3 This is a schematic diagram of the alternating optimization and low-complexity update process of the present invention.

[0025] Figure 4 This is a schematic diagram showing the relationship between the total mutual information and the number of rotatable antennas under different modulation schemes of the present invention, where Figure a represents 16QAM modulation and Figure b represents 8QAM modulation.

[0026] Figure 5 This is a schematic diagram illustrating the relationship between the total mutual information and the maximum rotation angle under different concealment levels of the present invention.

[0027] Figure 6 This is a schematic diagram showing the relationship between array gain and observation angle under different numbers of rotatable antennas according to the present invention. Figure a shows the case when the number of antennas is 4, and Figure b shows the case when the number of antennas is 16. Detailed Implementation

[0028] This invention proposes a joint optimization method for concealed MISO communication with rotatable antennas under finite character input, including the following steps:

[0029] S1. Construct a covert MISO communication system including a transmitter, a legitimate receiver, and a monitoring end. Obtain the positions of each antenna, the position of the legitimate receiver, the position of the monitoring end, a finite character constellation set, noise power, maximum transmission power, and maximum rotation angle. The transmitter is equipped with K rotatable antennas, each with an adjustable zenith angle and an adjustable azimuth angle. The legitimate receiver and the monitoring end are each equipped with a single antenna.

[0030] S2. Convert the zenith angle and azimuth angle of each rotatable antenna into a three-dimensional unit pointing vector. Determine the cosine power directional gain based on the inner product of the pointing vector and the node direction vector. Establish equivalent channels from the transmitter to the legal receiver (i.e., the legal link) and from the transmitter to the monitoring end (i.e., the monitoring link).

[0031] S3. Select the transmitted symbols from the normalized finite character constellation set with equal probability, determine the total mutual information based on the received signals of the legal receivers under finite character input, determine the minimum total detection error probability and concealment constraint based on the observation distribution of the monitoring end in the non-transmitting state and the transmitting state, and establish the total mutual information maximization problem.

[0032] S4. By using the lower bound of total mutual information and the upper bound of Kullback-Leibler divergence, the joint optimization problem of maximizing total mutual information is transformed into the problem of maximizing the received power of the legitimate receiver. That is, under the constraints of transmit power, monitoring end received power and rotatable antenna orientation, the legitimate receiver's received power is maximized.

[0033] S5. Alternately update the transmit beamforming vector and the orientation of each rotatable antenna: when the antenna orientation is fixed, update the beamforming vector with second-order cone programming; when the beamforming vector is fixed, update the antenna orientation with differential convex approximation and penalty function convex-concave program.

[0034] S6. When the change in received power of the legitimate receiver in two consecutive iterations is less than the preset convergence threshold or the number of iterations reaches the preset upper limit, output the transmit beamforming vector and the orientation of each rotatable antenna.

[0035] The present invention will be further described in detail below with reference to the embodiments and accompanying drawings.

[0036] Example 1: Alternating Optimization Mode

[0037] like Figure 1 and Figure 2As shown, the transmitter Alice is configured with K rotatable antennas, while the legitimate receiver Bob and the monitoring receiver Willie are each configured with a fixed antenna. Each link experiences quasi-static flat fading within a single transmission block, and the transmitter acquires the channel state information of both the legitimate and monitoring links. Transmitted symbols From a set containing M constellation points The constellation is selected with moderate probability and satisfies the normalization of average constellation power.

[0038] S1. Establish the antenna orientation and directional gain model. The orientation variable of the k-th rotatable antenna is denoted as... ,in The zenith angle is relative to the z-axis. Let be the azimuth angle in the xy plane, and its three-dimensional unit pointing vector be:

[0039] (1)

[0040] make and These represent the positions of node r and the k-th antenna, respectively. Let represent the unit direction vector pointing from the k-th antenna to node r. The antenna pattern is in cosine power form. This assumes the node is always located in the forward-radiating hemisphere of the feasible antenna orientation. The direction-dependent channel coefficients are expressed as: where, Let be the distance between node r and the k-th antenna, and denote it as... .

[0041] (2)

[0042] in, b represents Bob, and w represents Willie; ; It follows a complex Gaussian distribution with zero mean and unit variance; Control the width of the main lobe; .

[0043] Stacking the K channel coefficients yields When the application scenario allows nodes to enter the backward hemisphere, the inner product can be replaced with its positive part. To maintain consistency in the definition of the radiation pattern.

[0044] in, This is the orientation matrix for all rotatable antennas. and These represent the equivalent channel vectors for the legitimate link and the monitored link, respectively.

[0045] S2. Establish a valid receiver mutual information model. The signal received by Bob during the i-th channel usage is:

[0046] (3)

[0047] in, For beamforming vectors, variance Complex Gaussian white noise. Let Then the precise mutual information used by the channel each time is:

[0048] (4)

[0049] In equation (4), Let n be the index of the constellation points, and n be the number of points that follow a pattern with a mean of 0 and a variance of 0. Complex Gaussian noise samples, Indicates to Take the expected value.

[0050] When the transport block contains When using a secondary independent channel, the total mutual information is This precise mutual information is used for evaluating implementation effectiveness. To form a tractable optimization model, the following lower bound on the mutual information is adopted:

[0051] (5)

[0052] For a given constellation and noise variance, the lower bound of mutual information follows... Since it increases monotonically, Bob's received power can be used as the optimization target.

[0053] S3. Establish a binary hypothesis test and concealment constraints for the monitoring end. Willie's observation model is as follows:

[0054] ,

[0055] (6)

[0056] in, This indicates that the transmitter did not send a message. Indicates that the transmitter is sending; variance Complex Gaussian white noise. Below, due to The observed signal, derived from a finite set of character constellations, follows a Gaussian mixture distribution. Let... and They represent L observations at... and The joint distribution under the following conditions minimizes the total detection error probability. satisfy:

[0057] (7)

[0058] To ensure , apply Using the upper bound of divergence under the condition of finite character input, we can obtain:

[0059] (8)

[0060] Thus, with sufficient conditions To ensure concealment, among which This process forms a conservative approximation of the original hidden constraints. Among them, Indicates the Kullback-Leibler divergence. This represents the upper bound of the divergence. For concealment tolerance parameters, and , This represents the upper limit of the power received by the monitoring terminal.

[0061] S4. Constructing a joint optimization problem for received power:

[0062] ,

[0063] (9)

[0064] In the above problem, the beamforming vector and the antenna orientation pass through and The coupling is addressed using alternating optimization techniques. This is the maximum transmission power.

[0065] S5, fixed in the t-th outer iteration Taking advantage of the property that the common phase of u does not affect the power constraint and concealment constraint, let Let be a non-negative real number. The beamforming subproblem can be written as:

[0066] ,

[0067] (10)

[0068] This problem is a second-order cone programming problem, which can be solved by a convex optimization solver to obtain the current block solution that is globally optimal.

[0069] S6, Fixed Beamforming Vector Update antenna orientation matrix Define nodes The received power is It is expressed as the difference between two convex functions. in, ,matrix With orientation matrix One-to-one correspondence.

[0070] In the Second inner iteration point At this point, utilizing the first-order supporting hyperplane property of convex functions, for and By performing first-order linearization on each, a conservative lower bound on Bob's received power is obtained:

[0071] (11)

[0072] And a conservative upper bound on Willie's receive power:

[0073] (12)

[0074] Therefore, there is .

[0075] For the unit norm constraint This is decomposed into an upper and lower norm bound, and the lower bound is linearized to first order by introducing a nonnegative slack variable. ,Right now .

[0076] Therefore, the first The antenna orientation update can be written as:

[0077]

[0078] in, The penalty parameter is gradually increased. This problem is solved numerically using the CVX toolbox in MATLAB. The above convex subproblem is solved repeatedly until the antenna orientation and relaxation variables converge. Then, the beamforming update is returned until the outer layer alternating optimization converges.

[0079] In equations (11) and (12) Represents a vector Find the gradient; in equation (13) It is the unit vector along the z-axis.

[0080] Example 2: Low-complexity update mode

[0081] At the transmitting end where computing power is limited, the following can be used: Figure 3 The low-complexity pattern is shown. With the antenna facing backward, let... and These are the KKT multipliers for the transmit power constraint and the concealment constraint, respectively. If the maximum ratio transmission beam satisfies the monitoring end constraint, then it is directly taken. Otherwise, the concealment constraint is activated, and the beamforming vector is:

[0082] (14)

[0083] in , which can be determined by The answer can be obtained by combining the Sherman-Morrison identity in a closed form. and When collinear, along Launch in the direction of the direction, and take no more than And the power meets the power limit of the monitoring terminal. In equation (14) It is a K-order identity matrix.

[0084] After fixing u, the zenith angle and azimuth angle of the k-th rotatable antenna are updated sequentially. The composite components of the other antennas are denoted as... Let the coefficient of the k-th antenna be denoted as ,make Then the received power of node r is:

[0085] (15)

[0086] in , , For the current corner to be updated Bob's received power stagnation point is determined by , or The monitoring terminal concealment boundary is given by... Given the above feasible roots, angle boundaries, and current feasible angles, we merge them into a candidate set and select those that satisfy the following conditions: And make The largest angle.

[0087] After completing one round of updates for all K antennas, the closed-beam update is executed again; the process ends when the relative change in received power is less than a threshold or the maximum number of iterations is reached. This mode avoids repeatedly solving the second-order cone programming and penalty function subproblems, thus reducing the solution overhead for large-scale antenna arrays.

[0088] Example 3: Simulation Verification of Parameters and Results

[0089] In a set of exemplary simulations, the transmitter uses a uniform linear array centered at the origin with an element spacing of 0.0625m and an operating wavelength of 0.125m. Bob and Willie are located at distances of 4000m and 6000m from the transmitter, respectively, with z-axis angles of 20° and -10°. The transmit block length L is 100, the maximum transmit power is 30dBm, the noise power of Bob and Willie is -90dBm, the path loss index is 3.6, the directivity factor p is 2, the maximum rotation angle is 30°, and the concealment level is... The value is set to 0.05, and 16QAM is used. The above values ​​are for illustrative purposes only and do not constitute a limitation on the scope of protection.

[0090] like Figure 4 As shown, normalized 16QAM and 8QAM finite character constellations are used respectively. 16QAM contains 16 equally probable constellation points, and 8QAM contains 8 equally probable constellation points. Each constellation undergoes unit average power normalization. With the increase in the number of rotatable antennas, the total mutual information of both the alternating optimization mode and the low-complexity mode increases. The alternating optimization mode outperforms the low-complexity mode and is superior to both the fixed-orientation antenna scheme and the isotropic antenna scheme. The finite character input causes the total mutual information to tend to saturate with a larger number of antennas. Figure 4 The performance evaluation uses an exact mutual information expression, and the optimization process uses the received power target corresponding to the lower bound of mutual information.

[0091] like Figure 5 As shown, increasing the maximum rotation angle increases the available orientation degrees of freedom and improves the total mutual information at legitimate receivers; increasing the concealment level parameter relaxes the constraints at the monitoring end, correspondingly improving the mutual information. Different algorithms degenerate into a fixed-orientation beamforming problem under the condition of zero rotation angle and the same concealment level.

[0092] like Figure 6 As shown, the combined adjustment of the rotatable antenna and beamforming enhances the array gain in the Bob direction and results in a lower response at Willie. With the increase in the number of antennas (from 4 to 16), the spatial degrees of freedom available for legitimate link enhancement and monitoring link destructive events increase. In practice, the actual monitored link received power should not exceed... As a standard for covert verification.

[0093] In summary, this invention introduces three-dimensional orientation control of a rotatable antenna into finite character input covert MISO communication, and achieves joint updating of beamforming and antenna orientation through alternating optimization mode or low complexity mode, thereby improving the communication performance of the legitimate receiver while satisfying the covertness constraint.

Claims

1. A joint optimization method for concealed MISO communication with a rotatable antenna under finite character input, characterized in that, Includes the following steps: S1. Construct a covert MISO communication system including a transmitter, a legitimate receiver, and a monitoring end. Obtain the positions of each antenna, the position of the legitimate receiver, the position of the monitoring end, a finite character constellation set, noise power, maximum transmission power, and maximum rotation angle. The transmitter is equipped with K rotatable antennas, each with an adjustable zenith angle and an adjustable azimuth angle. The legitimate receiver and the monitoring end are each equipped with a single antenna. S2. Convert the zenith angle and azimuth angle of each rotatable antenna into a three-dimensional unit pointing vector. Determine the cosine power directional gain based on the inner product of the pointing vector and the node direction vector. Establish equivalent channels from the transmitter to the legal receiver (i.e., the legal link) and from the transmitter to the monitoring end (i.e., the monitoring link). S3. Select the transmitted symbols from the normalized finite character constellation set with equal probability, determine the total mutual information based on the received signals of the legal receivers under finite character input, determine the minimum total detection error probability and concealment constraint based on the observation distribution of the monitoring end in the non-transmitting state and the transmitting state, and establish the total mutual information maximization problem. S4. By using the lower bound of total mutual information and the upper bound of Kullback-Leibler divergence, the joint optimization problem of maximizing total mutual information is transformed into the problem of maximizing the received power of the legitimate receiver. That is, under the constraints of transmit power, monitoring end received power and rotatable antenna orientation, the legitimate receiver's received power is maximized. S5. Alternately update the transmit beamforming vector and the orientation of each rotatable antenna: when the antenna orientation is fixed, update the beamforming vector with second-order cone programming; when the beamforming vector is fixed, update the antenna orientation with differential convex approximation and penalty function convex-concave program. When the transmit beamforming vector is fixed, the received power of the legitimate receiver and the monitoring end is expressed as a differential convex function, and the lower bound of the received power of the legitimate receiver and the upper bound of the received power of the monitoring end are constructed at the linearization point. The unit norm constraint is decomposed into the upper bound of the norm constraint and the lower bound of the norm constraint after first-order linearization. Non-negative relaxation variables and incremental penalty parameters are introduced, and the three-dimensional pointing vector of each rotatable antenna is updated through the penalty function convex-concave procedure. S6. When the change in received power of the legitimate receiver in two consecutive iterations is less than the preset convergence threshold or the number of iterations reaches the preset upper limit, output the transmit beamforming vector and the orientation of each rotatable antenna.

2. The joint optimization method for rotatable antenna concealed MISO communication under finite character input as described in claim 1, characterized in that, In step S2, the orientation of the k-th rotatable antenna is changed by... It means that among them , The corresponding three-dimensional pointing vector is: (1) in, The maximum rotation angle; The unit direction vector of the k-th rotatable antenna pointing towards node r is: r is the identifier of the legitimate receiving end b or the identifier of the monitoring end w.

3. The joint optimization method for rotatable antenna concealed MISO communication under finite character input as described in claim 2, characterized in that, In step S2, assuming the node is located in the forward-radiating hemisphere of the rotatable antenna, the channel coefficient from the k-th rotatable antenna to node r is: (2) in, For distance-dependent large-scale channel power gain, This is the path loss index. This represents the small-scale fading coefficient. As a directional factor, This represents the maximum antenna gain. For nodes With the The distance between the rotatable antennas.

4. The joint optimization method for rotatable antenna concealed MISO communication under finite character input as described in claim 3, characterized in that, In steps S3 and S4, the finite character constellation set contains Each constellation point and its average power is normalized; Use Pinsker's inequality to minimize the total detection error probability. The Kullback-Leibler divergence constraint is conservatively transformed into a monitoring end received power constraint: (3) in, This is the orientation matrix for all rotatable antennas. For the transmit beamforming vector, To monitor the noise variance, The number of times a channel is used within a transport block; For concealment tolerance parameters, and ; This represents the upper limit of the power received by the monitoring terminal. and These are the equivalent channel vectors from the transmitter to the legitimate receiver and from the monitoring end, respectively.

5. The joint optimization method for rotatable antenna concealed MISO communication under finite character input as described in claim 4, characterized in that, In step S5, with the rotatable antenna orientation fixed, the equivalent received signal at the legitimate receiver is made non-negative real by selecting the common phase of the transmit beamforming vector, and the beamforming subproblem is transformed into a second-order cone programming problem: , (4) in, , ; This is the maximum transmission power; and These represent taking the real part and the imaginary part, respectively. This indicates the conjugate transpose. This represents the L2 norm.

6. The joint optimization method for rotatable antenna concealed MISO communication under finite character input as described in claim 5, characterized in that, Step S5 also includes a low-complexity update mode; when the orientation of the fixed rotatable antenna is fixed, the monitoring end received power constraint is determined according to the KKT condition: when it is not activated, the maximum ratio transmission beamforming is used; when it is activated, the closed beamforming vector containing the inverse operation of the outer product matrix of the monitoring end channel is used, and the corresponding multiplier ratio is calculated by the Sherman-Morrison identity.

7. The joint optimization method for rotatable antenna concealed MISO communication under finite character input as described in claim 6, characterized in that, With the beamforming vector fixed and the k-th rotatable antenna updated sequentially, and its azimuth or zenith angle fixed respectively, the received power of node r is expressed as: (5) in, ; respectively determine the stationary point of the received power at the legitimate receiver and the received power at the monitoring end. The boundary point, the boundary of the feasible region of angles, and the current feasible angle are used to form a finite set of candidate angles. The angle that maximizes the received power of the legitimate receiver is selected from the candidate angles that satisfy the concealment constraint. In equation (5), ; The composite components excluding the k-th antenna. The coefficient of the k-th antenna; .

8. The joint optimization method for rotatable antenna concealed MISO communication under finite character input according to any one of claims 1 to 7, characterized in that, The finite character constellation set is either a phase shift keying constellation set or a quadrature amplitude modulation constellation set; the transmitter knows the channel state information of the legitimate link and the monitoring link, and performs joint optimization before the start of each quasi-static flat fading transport block.