Marine robust safety sensing joint design method based on unmanned aerial vehicle-carried STAR-RIS

Abstract

The invention discloses a maritime robust safety sensing joint design method based on unmanned aerial vehicle-carried STAR-RIS, and belongs to the field of safety sensing integration. The method comprises the following steps: constructing a communication and sensing network model among a satellite, an unmanned aerial vehicle carrying STAR-RIS, a communication ship and a target ship; constructing a joint optimization problem model which takes maximization of the average secrecy rate of the system as an objective function and takes perception performance requirements as constraint conditions; and solving the joint optimization problem model through an optimization algorithm to obtain the optimal three-dimensional trajectory of the unmanned aerial vehicle and the optimal phase shift matrix configuration of the STAR-RIS so as to control the flight of the unmanned aerial vehicle and configure the STAR-RIS. According to the method, cooperative improvement of communication security and perception robustness can be realized under the condition of unknown suspicious target position and strong dynamic channel, and a solution is provided for security communication, target detection and intelligent monitoring in an offshore low-altitude network.

Description

Technical Field

[0001] This invention relates to the field of integrated safety sensing, and particularly to a robust maritime safety sensing co-design method based on STAR-RIS mounted on an unmanned aerial vehicle. Background Technology

[0002] With the continuous growth of marine communication, maritime law enforcement, and intelligent monitoring missions, integrated marine sensing technology has become a core support for these application scenarios. Due to the limited coverage of shore-based base stations and the high path loss and latency of satellite signals, unmanned aerial vehicles (UAVs), with their flexible deployment and wide-area coverage, can provide numerous benefits for the design of integrated marine sensing systems. However, compared to land-based scenarios, the marine environment is characterized by a lack of infrastructure, strong channel time-varying characteristics, significant line-of-sight propagation, and high scene dynamics, making the system vulnerable to multiple threats during communication and sensing processes, including illegal interception, channel disturbances, and environmental noise.

[0003] Most current security sensing methods are designed for static land scenarios. Common optimization strategies include: (1) physical layer security enhancement, which suppresses illegal signal reception through beamforming or artificial noise; (2) sensing co-optimization, which improves sensing accuracy through fixed beams or path planning; and (3) robust security design, which improves algorithm stability by modeling uncertain positions within a known range. However, these methods are difficult to apply directly to dynamic maritime scenarios. This is because in the actual marine environment, suspicious targets are usually in a drifting state due to factors such as wave motion, ocean currents, and wind disturbances. Their positions exhibit irregular and non-static dynamic changes, making it difficult for traditional robust optimization models based on regular regions to accurately describe the uncertainty of the target's position, resulting in a significant decrease in both communication and sensing performance. In addition, maritime links are often affected by the Doppler effect and attitude disturbances, making it difficult to obtain channel state information in real time. This leads to significant performance degradation of traditional static robust algorithms in terms of both security and sensing accuracy.

[0004] To overcome the above problems, there is an urgent need for a joint design method for secure communication and sensing that can adapt to the dynamic, irregular and uncertain marine environment, so that UAVs can achieve simultaneous optimization of communication security and perception robustness under complex disturbance conditions when performing maritime law enforcement and suspicious target monitoring missions. Summary of the Invention

[0005] Purpose of the invention: To address the above problems, the purpose of this invention is to provide a robust maritime safety sensing co-design method based on STAR-RIS mounted on an unmanned aerial vehicle (UAV).

[0006] Technical solution: The present invention provides a robust maritime safety sensing co-design method based on STAR-RIS mounted on an unmanned aerial vehicle (UAV), comprising the following steps:

[0007] Step 1: Construct a communication and sensing network model between a satellite, a drone equipped with STAR-RIS, a communication ship, and a target ship; wherein, the satellite transmits signals, the drone receives the signals and simultaneously reflects the signals to the communication ship and transmits them to the target ship through the STAR-RIS, so as to realize communication services to the communication ship and synchronous sensing of the target ship;

[0008] Step 2: Based on the communication and sensing network, construct a joint optimization problem model with the objective function of maximizing the system's average security rate and the constraint of sensing performance requirements; wherein, the joint optimization problem model considers the position uncertainty of the target ship and introduces robust constraints for processing;

[0009] Step 3: Solve the joint optimization problem model using an optimization algorithm to obtain the optimal three-dimensional trajectory of the UAV and the optimal phase shift matrix configuration of STAR-RIS;

[0010] Step 4: Based on the optimal three-dimensional trajectory and the optimal phase shift matrix configuration, control the UAV to fly and configure the STAR-RIS to perform maritime safety communication and target perception tasks.

[0011] Furthermore, the mathematical expression of the joint optimization problem model described in step 2 is:

[0012] ,

[0013] ,

[0014] ,

[0015] ,

[0016] ,

[0017] ,

[0018] ,

[0019] The first three constraints, C1, C2, and C3, are kinematic constraints for the UAV, while the fourth constraint, C4, indicates that the sensing power transmitted to the target vessel K must meet a threshold. The fifth constraint C5 is the constraint on the STAR-RIS reflection phase, and the sixth constraint C6 is the constraint on the STAR-RIS transmission phase. Let n be the horizontal coordinates of the UAV in the nth time slot. The altitude of the UAV in the nth time slot. The transmission matrix of STAR-RIS in the nth time slot. The reflection matrix of STAR-RIS in the nth time slot. The number of time slots , It is the error set of the elliptic domain. It is the rate of a legitimate link. It is the rate of the suspicious link; and These are the maximum horizontal and vertical speeds of the drone, respectively. and These are the upper and lower limits of flight altitude. Where T is the length of a single time slot and T is the total mission duration; The equivalent complex channel for satellite-UAV-target ship; This indicates the power at which the satellite signal is transmitted to the drone and then to the ship. Indicates the first The transmission phase shift angle of each unit in time slot n; Indicates the first The reflection phase shift angle of each unit in time slot n.

[0020] Furthermore, the joint optimization problem model is solved using an alternating optimization algorithm, which includes:

[0021] The three-dimensional trajectory of the UAV is fixed, and the phase shift matrix of STAR-RIS is optimized;

[0022] Furthermore, by fixing the phase shift matrix of the STAR-RIS, the three-dimensional trajectory of the UAV is optimized;

[0023] Repeat the above steps iteratively until convergence.

[0024] Furthermore, the step of fixing the three-dimensional trajectory of the UAV and optimizing the phase shift matrix of STAR-RIS includes:

[0025] For any given drone location First, transform the phase diagonal matrix into a unit modulus form, as follows:

[0026] ,

[0027] and utilize and Introduced into the channel coefficient matrix, its squared modulus is written as a quadratic form, expressed as:

[0028] ,

[0029] ,

[0030] in, , ;

[0031] When fixing the UAV trajectory, the joint optimization problem model is transformed into the following form:

[0032] ,

[0033] ,

[0034] ,

[0035] ,

[0036] in, , ;

[0037] In reality, the location of the target vessel K is inaccurate; only its center point is known. The position, introduced and ,make This indicates its possible location within the elliptic domain. Therefore, the gain at the sensing end depends on the perturbed receiving column vector, which is expressed as:

[0038] ,

[0039] in, ;

[0040] The corresponding matrix can be written as follows:

[0041] ,

[0042] Since the trajectory of the communication ship is known, the joint optimization problem model can be reconstructed as follows:

[0043] ,

[0044] ,

[0045] ,

[0046] ;

[0047] Here, the constraint of the model means that, under any possible location of the eavesdropping ship, the leakage gain at the reflective end must exceed the perception threshold to ensure that the eavesdropping target can always be illuminated by the drone;

[0048] Introducing positive semidefinite matrix variables ,use The joint optimization problem model is relaxed to the following form:

[0049] ,

[0050] ,

[0051] ,

[0052] ,

[0053] ,

[0054] ;

[0055] Gaussian randomization is used from The feasible solution is obtained by randomly generating and projecting the STAR-RIS phase control vector back to the physically feasible set from the optimal positive semidefinite relaxation matrix. The feasible solution is denoted as:

[0056] .

[0057] Furthermore, the step of optimizing the three-dimensional trajectory of the UAV by fixing the phase shift matrix of the STAR-RIS includes:

[0058] When the optimal solution is obtained When considering the influence of the actual marine environment, the uncertainty domain of the target vessel's actual position is expressed as follows:

[0059] ,

[0060] ,

[0061] in, , , ;

[0062] Therefore, the joint optimization problem model is transformed into the following form:

[0063] ,

[0064] ,

[0065] ,

[0066] ,

[0067] ;

[0068] Introducing slack variables At this point, the joint optimization problem model is reconstructed into the following form:

[0069] ,

[0070] ,

[0071] ,

[0072] ,

[0073] ,

[0074] ,

[0075] C6: ,

[0076] C7: ;

[0077] Then, through a first-order Taylor expansion, the objective function in the reconstructed joint optimization problem model is approximately transformed into the following form:

[0078] ,

[0079] ,

[0080] in, , , , ;

[0081] At this point, the sensing power constraint in the reconstructed joint optimization problem model changes accordingly, and the problem becomes equivalent to:

[0082] ,

[0083] ,

[0084] ,

[0085] ,

[0086] ;

[0087] The joint optimization problem model described here is a standard linear programming problem. The optimal solution to this maximization problem can be obtained directly through convex programming, which is the optimal coordinate point of the UAV in each time slot, thereby achieving trajectory optimization.

[0088] Beneficial effects: Compared with the prior art, the significant advantages of this invention are:

[0089] 1. This invention addresses the dynamic drift characteristics of suspicious targets in the marine environment by establishing a simplified model of irregular position uncertainty sets, which can effectively describe the target position distribution characteristics affected by ocean waves and current disturbances. Compared with traditional robust modeling methods that are only applicable to regular geometric regions such as circular and rectangular domains, this invention can maintain system stability and physical interpretability under more complex dynamic conditions.

[0090] 2. This invention achieves adaptive signal control in a dual-half-space by equipping the UAV platform with STAR-RIS, which can simultaneously transmit and reflect signals; and jointly optimizes the UAV's three-dimensional trajectory and the STAR-RIS phase shift matrix, so that the system can maintain both high security rate and target perception accuracy under conditions of Doppler effect, wave disturbance and irregular target distribution.

[0091] 3. By constructing a joint optimization model based on worst-case scenario constraints, this invention can adaptively adjust communication and sensing strategies under conditions of incomplete channel state information and dynamic location disturbances.

[0092] 4. This invention does not rely on ground infrastructure. STAR-RIS can be flexibly deployed in different sea areas by drones to achieve rapid networking and wide-area coverage. It has low deployment costs and is suitable for various mission scenarios such as maritime law enforcement, maritime search and rescue, and suspicious target monitoring. It provides a feasible engineering solution for maritime safety sensing in future 6G low-altitude networks. Attached Figure Description

[0093] Figure 1 This is a flowchart of the present invention;

[0094] Figure 2 A schematic diagram illustrating the STAR-RIS sensor for robust maritime safety equipped on a drone.

[0095] Figure 3 Three-dimensional trajectory diagrams for different energy splitting coefficients and different sensing power thresholds;

[0096] Figure 4 Three-dimensional trajectory plots for different error ranges;

[0097] Figure 5 Average security rate versus iteration number under different parameter settings;

[0098] Figure 6 The graph shows the variation of M in the safety sensing schemes under four different optimization schemes. Detailed Implementation

[0099] The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and not intended to limit the scope of the invention. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the embodiments of the present invention, and not all structures.

[0100] In the following description, specific details such as target system architecture and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of this application with unnecessary detail.

[0101] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0102] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0103] Furthermore, in the description of this application and the appended claims, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0104] References to "one embodiment" or "some embodiments" in this specification mean that one or more embodiments of this application include the target features, structures, or characteristics described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized.

[0105] The flowchart of the maritime robust safety sensing co-design method based on STAR-RIS on a UAV described in this embodiment is as follows: Figure 1 As shown, the method includes the following steps:

[0106] Step 1: Construct a communication and sensing network model between a satellite, a drone equipped with STAR-RIS (Simultaneously transmitting and reflecting reconfigurable intelligent surface), a communication ship, and a target ship; wherein, the satellite transmits a signal, which is simulated using a simplified free-space channel model, the drone receives the signal and simultaneously reflects the signal to the communication ship and transmits it to the target ship through the STAR-RIS, so as to realize communication services to the communication ship and synchronous sensing of the target ship.

[0107] Step 2: Based on the communication and sensing network, construct a joint optimization problem model with the objective function of maximizing the system's average security rate and the constraint of sensing performance requirements; wherein, the joint optimization problem model considers the position uncertainty of the target ship and introduces robust constraints for processing.

[0108] Furthermore, the mathematical expression of the joint optimization problem model described in step 2 is:

[0109] ,

[0110] ,

[0111] ,

[0112] ,

[0113] ,

[0114] ,

[0115] ,

[0116] The first three constraints, C1, C2, and C3, are kinematic constraints for the UAV, while the fourth constraint, C4, indicates that the sensing power transmitted to the target vessel K must meet a threshold. The fifth constraint C5 is the constraint on the STAR-RIS reflection phase, and the sixth constraint C6 is the constraint on the STAR-RIS transmission phase. Let n be the horizontal coordinates of the UAV in the nth time slot. The altitude of the UAV in the nth time slot. The transmission matrix of STAR-RIS in the nth time slot. The reflection matrix of STAR-RIS in the nth time slot. The number of time slots , It is the error set of the elliptic domain. It is the rate of a legitimate link. It is the rate of the suspicious link; and These are the maximum horizontal and vertical speeds of the drone, respectively. and These are the upper and lower limits of flight altitude. Where T is the length of a single time slot and T is the total mission duration; The equivalent complex channel for satellite-UAV-target ship; This indicates the power at which the satellite signal is transmitted to the drone and then to the ship. Indicates the first Each unit in the time slot The transmission phase shift angle; Indicates the first The reflection phase shift angle of each unit in time slot n.

[0117] Step 3: Solve the joint optimization problem model using an optimization algorithm to obtain the optimal three-dimensional trajectory of the UAV and the optimal phase shift matrix configuration of STAR-RIS.

[0118] Furthermore, the joint optimization problem model is solved using an alternating optimization algorithm, which includes:

[0119] The three-dimensional trajectory of the UAV is fixed, and the phase shift matrix of STAR-RIS is optimized;

[0120] Furthermore, by fixing the phase shift matrix of the STAR-RIS, the three-dimensional trajectory of the UAV is optimized;

[0121] Repeat the above steps iteratively until convergence.

[0122] Furthermore, the step of fixing the three-dimensional trajectory of the UAV and optimizing the phase shift matrix of STAR-RIS includes:

[0123] For any given drone location First, transform the phase diagonal matrix into a unit modulus form, as follows:

[0124] ,

[0125] and utilize and Introduced into the channel coefficient matrix, its squared modulus is written as a quadratic form, expressed as:

[0126] ,

[0127] ,

[0128] in, ,

[0129] ;

[0130] When fixing the UAV trajectory, the joint optimization problem model is transformed into the following form:

[0131] ,

[0132] ,

[0133] ,

[0134] ,

[0135] in, , ;

[0136] In reality, the location of the target vessel K is inaccurate; only its center point is known. The position, introduced and ,make This indicates its possible location within the elliptic domain. Therefore, the gain at the sensing end depends on the perturbed receiving column vector, which is expressed as:

[0137] ,

[0138] in, ;

[0139] The corresponding matrix can be written as follows:

[0140] ,

[0141] Since the trajectory of the communication ship is known, the joint optimization problem model can be reconstructed as follows:

[0142] ,

[0143] ,

[0144] ,

[0145] ;

[0146] Here, the constraint of the model means that, under any possible location of the eavesdropping ship, the leakage gain at the reflective end must exceed the perception threshold to ensure that the eavesdropping target can always be illuminated by the drone;

[0147] Introducing positive semidefinite matrix variables ,use The joint optimization problem model is relaxed to the following form:

[0148] ,

[0149] ,

[0150] ,

[0151] ,

[0152] ,

[0153] ;

[0154] Gaussian randomization is used from A feasible solution is obtained, and the feasible solution is denoted as:

[0155] .

[0156] Furthermore, the step of optimizing the three-dimensional trajectory of the UAV by fixing the phase shift matrix of the STAR-RIS includes:

[0157] When the optimal solution is obtained When considering the influence of the actual marine environment, the uncertainty domain of the target vessel's actual position is expressed as follows:

[0158] ,

[0159] ,

[0160] in, , , ;

[0161] Therefore, the joint optimization problem model is transformed into the following form:

[0162] ,

[0163] ,

[0164] ,

[0165] ,

[0166] ;

[0167] Introducing slack variables At this point, the joint optimization problem model is reconstructed into the following form:

[0168] ,

[0169] ,

[0170] ,

[0171] ,

[0172] ,

[0173] ,

[0174] C6: ,

[0175] C7: ;

[0176] Then, through a first-order Taylor expansion, the objective function in the reconstructed joint optimization problem model is approximately transformed into the following form:

[0177] ,

[0178] ,

[0179] in, , , , ;

[0180] At this point, the sensing power constraint in the reconstructed joint optimization problem model changes accordingly, and the problem becomes equivalent to:

[0181] ,

[0182] ,

[0183] ,

[0184] ,

[0185] ;

[0186] The joint optimization problem model described here is a standard linear programming problem. The optimal solution to this maximization problem can be obtained directly through convex programming, thereby achieving trajectory optimization.

[0187] Step 4: Based on the optimal three-dimensional trajectory and the optimal phase shift matrix configuration, control the UAV to fly and configure the STAR-RIS to perform maritime safety communication and target perception tasks.

[0188] The following examples further evaluate and analyze the system performance of the maritime robust safety sensing co-design scheme based on STAR-RIS mounted on an unmanned aerial vehicle (UAV) as described in this invention. For example... Figure 2 As shown, consider a low-Earth orbit satellite S, a drone U carrying STAR-RIS, and two surface vessels: a communication vessel E and a target vessel K, where the target vessel can be considered a potential eavesdropping device. Assuming the maritime scenario is time-varying, the total mission duration T is uniformly divided into N time slots, i.e., the length of a single time slot is... The simulation settings are as follows: the initial coordinates and final coordinates of the UAV U are located at... and The coordinates of the launched satellite are set to Furthermore, unless otherwise specified, this example sets up time slots separately. The number of components in STAR-RIS The horizontal flight speed of the drone Vertical flight speed of drones The minimum altitude for drone flight is The highest height is The speeds of the communication vessel E and the suspected target vessel K are: Path loss factor is , It is the spacing between adjacent STAR-RIS units. This is the estimated error range of the eavesdropping ship K.

[0189] like Figure 3 The example shown has different energy splitting coefficients. and sensing power threshold The obtained 3D trajectories of the UAV under the given conditions show significant differences. As can be seen from the figure, when the perception threshold is relatively strict (e.g., -55 dBm), the UAV's trajectory altitude increases during the initial and final phases to enhance the perception gain of the eavesdropping ship and reduce its ability to eavesdrop on legitimate signals. In the middle phase, due to the close proximity to the two ships, the trajectory shows a moderate downward trend to balance communication rate and perception accuracy. When the energy splitting coefficient increases, STAR-RIS allocates more energy to the communication link, resulting in a decrease in perception end gain. To meet the perception power constraint, the UAV will moderately approach the eavesdropping ship and increase its altitude to achieve leakage control. When the perception threshold is relaxed from -55 dBm to -60 dBm, the constraint condition weakens, and the UAV can get closer to the communication ship, thereby improving the communication rate. This result verifies that the trajectory optimization strategy proposed in this invention can achieve an adaptive trade-off between secure communication and perception performance.

[0190] like Figure 4 As shown, when the uncertainty range of the eavesdropping ship's location gradually expands from 0 to a certain radius, the spatial distribution of the UAV trajectory exhibits a regular change. When the eavesdropping ship's location is completely determined (error range is 0), the UAV can cancel the leaked signal through precise phase control, resulting in minimal changes in trajectory altitude. However, when the uncertainty range of the eavesdropping ship expands, the UAV needs to comprehensively consider all possible sets of the most difficult-to-detect points. To ensure the sensing power constraint in the worst-case scenario, its horizontal trajectory tends to move away from the boundary of the uncertain region, while simultaneously compensating for the sensing gain by reducing its flight altitude. This result demonstrates that the robust design mechanism of this invention can maintain the effectiveness of sensing constraints under unknown or dynamic target location conditions, enhancing the system's resistance to uncertainty.

[0191] like Figure 5 As shown, the Alternating Optimization (AO) algorithm of this invention operates under different parameters (including energy splitting coefficients). Sensing power threshold Error range Under the given settings, both exhibited good convergence and stability. Figure 5 The results show that the algorithm typically reaches a steady state after 3 to 5 iterations, verifying the efficiency of the optimization framework proposed in this invention. In an ideal scenario with no positional uncertainty, both a larger transmission coefficient and a more lenient sensing threshold can significantly improve the system's average security rate. As the positional uncertainty of the eavesdropping vessel increases, the system's security rate decreases accordingly, but it still maintains smooth convergence overall, demonstrating the method's ability to balance performance and robustness.

[0192] Figure 6The average security rate of various strategies under different IRS cell numbers M is compared, including the alternating optimization of this invention, trajectory-only optimization (optimizing only the trajectory without changing the phase), phase-only optimization (optimizing only the phase without changing the trajectory), and fixed-altitude optimization (allowing the UAV trajectory to change only in two dimensions, without optimizing the altitude). As shown in the figure, all methods exhibit a monotonically increasing trend with increasing cell number M, verifying the passive beam gain and interference suppression capabilities brought by larger-scale arrays. Particularly noteworthy is that the method described in this invention achieves the highest security rate for all cell number values, and the gap with the baseline method widens with increasing cell number, indicating that the joint design can more fully utilize the degrees of freedom and gain of large-scale arrays. In comparison, trajectory-only optimization significantly outperforms phase-only optimization and fixed-altitude optimization, indicating that geometric degrees of freedom (closer to the legitimate end, farther from the eavesdropping end) contribute more directly to security performance; while the marginal benefit of phase optimization alone is limited and cannot replace the geometric isolation effect brought by three-dimensional trajectories. Overall, Figure 5 This demonstrates from the perspectives of scalability and combined benefits that, as array size increases, joint optimization is the most effective way to convert array gains into improved security rates, while single-dimensional optimization is unlikely to achieve the same level of improvement.

[0193] To address the technical problem that existing security sensing technologies are mainly applied to static land scenarios and are difficult to adapt to the dynamic marine environment, this invention aims to provide a robust maritime security sensing co-design method based on a UAV equipped with STAR-RIS, thus solving the problem of decreased communication security and perception reliability in complex marine environments. In this invention, a satellite transmits reflected and transmitted signals to a UAV equipped with STAR-RIS to communicate with legitimate ships while simultaneously sensing suspicious target ships (whose signals can be eavesdropped). A real-world scenario is considered where the location estimation of suspicious target ships is inaccurate due to wave motion, positioning issues, and harsh marine environments. To address these challenges, the robust maritime security sensing co-design method based on a UAV equipped with STAR-RIS achieves a three-layer collaborative architecture of satellite-UAV-ship through the UAV-equipped STAR-RIS. Based on this, a joint optimization model is constructed to maximize security rate and constrain perception performance. By adjusting and optimizing the energy splitting coefficient and perception power threshold, robust constraints and alternating optimization algorithms are used to simultaneously optimize the UAV's 3D trajectory and the STAR-RIS phase shift matrix. This solution can achieve a synergistic improvement in communication security and perception robustness under conditions of unknown suspicious target locations and highly dynamic channels, providing a feasible robust integrated sensing solution for secure communication, target detection and intelligent monitoring in maritime low-altitude networks.

Claims

1. A joint design method for robust maritime safety sensing based on STAR-RIS on an unmanned aerial vehicle (UAV), characterized in that, Includes the following steps: Step 1: Construct a communication and sensing network model between a satellite, a drone equipped with STAR-RIS, a communication ship, and a target ship; wherein, the satellite transmits signals, the drone receives the signals and simultaneously reflects the signals to the communication ship and transmits them to the target ship through the STAR-RIS, so as to realize communication services to the communication ship and synchronous sensing of the target ship; Step 2: Based on the communication and sensing network, construct a joint optimization problem model with the objective function of maximizing the system's average security rate and the constraint of sensing performance requirements; wherein, the joint optimization problem model considers the position uncertainty of the target ship and introduces robust constraints for processing; Step 3: Solve the joint optimization problem model using an optimization algorithm to obtain the optimal three-dimensional trajectory of the UAV and the optimal phase shift matrix configuration of STAR-RIS; Step 4: Based on the optimal three-dimensional trajectory and the optimal phase shift matrix configuration, control the UAV to fly and configure the STAR-RIS to perform maritime safety communication and target perception tasks.

2. The maritime robust safety sensing co-design method based on STAR-RIS mounted on an unmanned aerial vehicle (UAV) as described in claim 1, characterized in that, The mathematical expression for the joint optimization problem model described in step 2 is: ， ， ， ， ， ， ， The first three constraints, C1, C2, and C3, are kinematic constraints for the UAV, while the fourth constraint, C4, indicates that the sensing power transmitted to the target vessel K must meet a threshold. The fifth constraint C5 is the constraint on the STAR-RIS reflection phase, and the sixth constraint C6 is the constraint on the STAR-RIS transmission phase. Let n be the horizontal coordinates of the UAV in the nth time slot. The altitude of the UAV in the nth time slot. The transmission matrix of STAR-RIS in the nth time slot. The reflection matrix of the nth time slot STAR-RIS is... The number of time slots , It is the error set of the elliptic domain. It is the rate of a legitimate link. It is the rate of the suspicious link; and These are the maximum horizontal and vertical speeds of the drone, respectively. and These are the upper and lower limits of flight altitude. Where T is the length of a single time slot and T is the total mission duration; The equivalent complex channel for satellite-UAV-target ship; This indicates the power at which the satellite signal is transmitted to the drone and then to the ship. Indicates the first The transmission phase shift angle of each unit in time slot n; Indicates the first The reflection phase shift angle of each unit in time slot n.

3. The maritime robust safety sensing co-design method based on STAR-RIS mounted on an unmanned aerial vehicle (UAV) as described in claim 2, characterized in that, The joint optimization problem model is solved using an alternating optimization algorithm, which includes: The three-dimensional trajectory of the UAV is fixed, and the phase shift matrix of STAR-RIS is optimized; Furthermore, by fixing the phase shift matrix of the STAR-RIS, the three-dimensional trajectory of the UAV is optimized; Repeat the above steps iteratively until convergence.

4. The maritime robust safety sensing co-design method based on STAR-RIS mounted on an unmanned aerial vehicle (UAV) as described in claim 3, characterized in that, The steps of fixing the three-dimensional trajectory of the UAV and optimizing the phase shift matrix of STAR-RIS include: For any given drone location First, transform the phase diagonal matrix into a unit modulus form, as follows: ， and utilize and Introduced into the channel coefficient matrix, its squared modulus is written as a quadratic form, expressed as: ， ， in, , ; When fixing the UAV trajectory, the joint optimization problem model is transformed into the following form: ， ， ， ， in, , ; In reality, the location of the target vessel K is inaccurate; only its center point is known. The position, introduced and ,make This indicates its possible location within the elliptic domain. Therefore, the gain at the sensing end depends on the perturbed receiving column vector, which is expressed as: ， in, ; The corresponding matrix can be written as follows: ， Since the trajectory of the communication ship is known, the joint optimization problem model can be reconstructed as follows: ， ， ， ； Here, the constraint of the model means that, under any possible location of the eavesdropping ship, the leakage gain at the reflective end must exceed the perception threshold to ensure that the eavesdropping target can always be illuminated by the drone; Introducing positive semidefinite matrix variables ,use The joint optimization problem model is relaxed to the following form: ， ， ， ， ， ； Gaussian randomization is used from The feasible solution is obtained by randomly generating and projecting the STAR-RIS phase control vector back to the physically feasible set from the optimal positive semidefinite relaxation matrix. The feasible solution is denoted as: 。 5. The maritime robust safety sensing co-design method based on STAR-RIS mounted on an unmanned aerial vehicle (UAV) as described in claim 4, characterized in that, The steps for optimizing the three-dimensional trajectory of the UAV by fixing the phase shift matrix of the STAR-RIS include: When the optimal solution is obtained When considering the influence of the actual marine environment, the uncertainty domain of the target vessel's actual position is expressed as follows: ， ， in, , , ; Therefore, the joint optimization problem model is transformed into the following form: ， ， ， ， ； Introducing slack variables At this point, the joint optimization problem model is reconstructed into the following form: ， ， ， ， ， ， C6： ， C7: ； Then, through a first-order Taylor expansion, the objective function in the reconstructed joint optimization problem model is approximately transformed into the following form: ， ， in, , , , ; At this point, the sensing power constraint in the reconstructed joint optimization problem model changes accordingly, and the problem becomes equivalent to: ， ， ， ， ； The joint optimization problem model described here is a standard linear programming problem. The optimal solution to this maximization problem can be obtained directly through convex programming, which is the optimal coordinate point of the UAV in each time slot, thereby achieving trajectory optimization.

Images