Intermittent interference-based dual-unmanned aerial vehicle cooperative security data acquisition scheme

Abstract

The invention provides a double-unmanned aerial vehicle cooperative security data acquisition method based on intermittent interference. Aiming at the eavesdropping risk caused by wireless channel openness and the problem of coexistence of unmanned aerial vehicle energy limitation and information age constraint, a traditional continuous interference scheme has the defects that energy consumption is too high, and safety and timeliness are difficult to consider at the same time. Therefore, a cooperative system composed of an acquisition unmanned aerial vehicle and an interference unmanned aerial vehicle is constructed, the acquisition unmanned aerial vehicle is responsible for data acquisition, and the interference unmanned aerial vehicle sends interference signals to potential eavesdroppers in an intermittent mode. On the basis, a multi-objective optimization framework is established, the flight path, time scheduling and interference parameters of the unmanned aerial vehicle are jointly optimized, and variable coupling and non-convex constraint are processed by adopting an iterative algorithm combining block coordinate descent and continuous convex approximation. The method can balance the bit error rate of the eavesdropper, the information age and the energy consumption, meanwhile, compared with continuous interference, the total energy consumption of the system is remarkably reduced, and the network operation efficiency and safety are effectively improved.

Description

Technical Field

[0001] This invention relates to the fields of unmanned aerial vehicle (UAV) assisted communication, Internet of Things (IoT) data acquisition, and physical layer security technology, and particularly to a dual-UAV collaborative secure data acquisition scheme based on intermittent interference. Background Technology

[0002] With the rapid development of IoT technology, drones, with their advantages of flexible deployment and high-probability line-of-sight transmission, have become efficient mobile data acquisition platforms. In application scenarios with high real-time requirements, such as environmental monitoring and industrial control, the timeliness of data is crucial. Therefore, Age of Information (AoI), as a key indicator for measuring the freshness of information, has become a core objective for optimizing drone communication networks. However, drone-assisted data acquisition systems face a severe dual challenge in practical applications: on the one hand, the openness of wireless channels makes communication between drones and ground nodes extremely vulnerable to malicious eavesdropping; on the other hand, drones are limited by onboard battery capacity, and their energy budget for flight propulsion and communication interference is very tight. How to complete tasks with limited energy and extend network lifespan is a challenge in system design.

[0003] Existing technologies typically utilize cooperative jamming techniques within physical layer security to ensure communication security. However, to counter eavesdroppers whose locations may be uncertain, current solutions mostly employ continuous jamming strategies, where the jammer continuously emits high-power artificial noise throughout the entire mission cycle. While this "all-time" jamming approach can improve the rate of confidentiality, it results in significant energy waste, leading to low system energy efficiency and severely shortening the drone's endurance. Furthermore, existing research often focuses on optimizing a single or dual objective, such as maximizing energy efficiency under security constraints or studying AoI optimization under no security threats. It lacks a method for jointly optimizing the three mutually constraining indicators—physical layer security (anti-eavesdropping capability), AoI, and total system energy consumption (network lifetime)—within a unified framework. Therefore, there is an urgent need for a technical solution that can achieve low-energy consumption and high-timeliness data acquisition while ensuring security through flexible jamming strategies. Summary of the Invention

[0004] The technical problem this invention aims to solve is to address the shortcomings of the prior art by providing a dual-UAV collaborative safe data acquisition method based on intermittent interference. Compared with previous research, this invention proposes an intermittent interference strategy that, by jointly optimizing UAV trajectories, time scheduling, and interference parameters, significantly reduces energy consumption and improves data freshness while ensuring safety.

[0005] To achieve the above objectives, the technical solution adopted by this invention is: a method for safe data acquisition through dual-UAV cooperative operation based on intermittent interference, comprising the following steps: S1. Establish an IoT data collection scenario assisted by a data collection drone (Bob) and a jamming drone (Jack) to obtain the current location of the drone in the current time slot, the location of the IoT node, and the estimated location information of potential eavesdroppers. S2. Based on the location and status information obtained in step S1, establish communication models and time scheduling constraint models to describe legitimate links and eavesdropping links, including Time Division Multiple Access (TDMA) scheduling constraint models and data acquisition volume constraint models; at the same time, establish system energy consumption models including flight propulsion energy consumption and jamming transmission energy consumption. S3. Construct an optimization function aimed at improving the overall performance of the system. This function integrates three core performance indicators: AoI, total energy consumption, and Bit Error Rate (BER). S4. Taking into account the kinematic constraints, energy budget, collision avoidance constraints, and TDMA scheduling constraints of the UAV, the final joint optimization problem is constructed. The optimization objective of this problem is to minimize the weighted sum of the system AoI and the total energy consumption, while maximizing the bit error rate of the eavesdropper. S5. Based on the final optimization problem, the non-convex problem constructed in step S4 is solved by using an iterative algorithm based on Block Coordinate Descent (BCD) and Successive Convex Approximation (SCA). The original problem is decomposed into sub-problems of interference parameter optimization, time scheduling optimization and trajectory optimization, which are solved alternately to finally obtain the optimal UAV trajectory and intermittent interference strategy.

[0006] The beneficial effects of this invention are as follows: By constructing a multi-objective optimization problem involving bit error rate, AoI, and energy consumption, this invention can adaptively adjust the interference ratio and power based on the relative positions of the UAV and the eavesdropper: activating interference during critical time slots when close to the eavesdropper, and remaining silent during other time slots to conserve energy. The iterative algorithm based on BCD and SCA proposed in this invention can transform the complex non-convex coupled problem into a solvable convex subproblem, effectively solving the joint optimization problem of trajectory and interference parameters. Compared with traditional continuous interference schemes, this method significantly reduces the total system energy consumption (experiments show energy savings exceeding 60%) while ensuring the same level of security, and effectively reduces the information age.

[0007] Furthermore, the expression for the communication model is as follows:

[0008]

[0009]

[0010]

[0011]

[0012]

[0013] in, Indicates time slot IoT Nodes To collect the channel gain of the drone Bob; Indicates time slot The channel gain from interfering with unmanned aerial vehicle Jack to collecting data from unmanned aerial vehicle Bob; Indicates time slot IoT Nodes Channel gain to Eve the eavesdropper; This represents the channel gain used to interfere with the drone Jack's communication with the eavesdropper Eve; For reference channel gain, and These are the positions of Bob and Jack, respectively. For node position, The location of the eavesdropper's center. Let the radius be the radius of positional uncertainty. This refers to the flight altitude of the drone. Indicates the stage state, specifically, This represents the interference phase, at which point... , This represents the non-interference (silent) phase, at which time... , Indicates the transmit power of INs. The term "interference duration ratio" represents the proportion of the total time slot duration in the nth time slot during which interference is transmitted. Therefore, This indicates Jack's actual interference power. This represents the average achievable data transmission rate of the m-th INs in the n-th time slot.

[0014] Furthermore, the expression for the time scheduling constraint model is as follows:

[0015]

[0016] in, This indicates the duration of the time slot, where m represents the IN number, M represents the number of INs, and n represents the time slot number. For Bob's data acquisition time from the Mth IN in the nth time slot, the UAV acquires a set of INs via TDMA, and Bob acquires data from each IN. At most one IN data entry can be collected. The amount of data Bob needs to collect for each IN should reach the total task volume requirement by the end of the entire task cycle.

[0017] Furthermore, the system energy consumption formula for flight propulsion energy consumption and jamming launch energy consumption is as follows: in, , These represent the propulsion energy consumption of the data collection drone and the jamming drone, respectively. and These are constants for airfoil power and induced power, respectively, during hovering. This indicates the tip velocity of the rotor blade. Indicates the fuselage drag ratio. Indicates the robustness of the rotor. A represents air density, and A represents the rotor disk area. This indicates the energy consumption required to interfere with the drone.

[0018] Furthermore, the expression for the overall system performance index is as follows:

[0019]

[0020]

[0021] in, This indicates the maximum age of the system's data. This represents the total energy consumption of a dual-drone collaborative system, a value that must meet the maximum energy budget. Constraints; This represents the average bit error rate (BER) of the eavesdropper in the nth time slot. Since this invention employs an intermittent interference strategy, the total BER in this time slot consists of two parts: the first part is the BER during the interference activation phase. Its weight is the proportion of interference duration. The second part is the bit error rate during the interference silence phase. Its weight is the proportion of non-interference time. , Standard Gaussian The function is used to calculate the bit error probability under BPSK modulation based on the signal-to-interference-plus-noise ratio.

[0022] The beneficial effects of the above-mentioned further solution are as follows: This invention constructs a comprehensive index function that integrates AoI, total system energy consumption, and eavesdropper BER. Specifically, this function quantifies three core performance indicators (i.e., data freshness, network operating energy consumption, and physical layer security) that are mutually restrictive and often conflicting in UAV-assisted IoT scenarios, and unifies them under the same mathematical framework. Through this comprehensive quantitative representation, the limitations of traditional single-objective or dual-objective optimization methods (such as pursuing only the minimization of energy consumption while ignoring data timeliness, or focusing only on maximizing interference security while leading to excessive energy waste) can be overcome, and the overall operational efficiency of the dual-UAV collaborative system can be more realistically reflected. This not only provides a clear and explicit joint optimization objective for subsequent iterative optimization algorithms based on BCD and SCA, but also transforms the complex, multi-dimensional security and energy efficiency trade-off problem into a mathematical problem that can be efficiently solved by advanced algorithms, thus providing a solid foundation for achieving system performance trade-offs.

[0023] Furthermore, the final optimization problem is expressed as follows:

[0024]

[0025]

[0026]

[0027]

[0028]

[0029]

[0030]

[0031]

[0032]

[0033]

[0034]

[0035]

[0036] in, It is a key optimization variable. Represents the normalized weighting factor. Indicates the maximum duration of the time slot. This represents the upper limit of the bit error rate for legitimate drones. Indicates the location of the drone warehouse. This indicates the location that interfered with the take-off and landing of the drone. Indicates the maximum speed of the drone. This represents the minimum distance between two drones.

[0037] The beneficial effects of the aforementioned further scheme are as follows: It breaks through the energy consumption limitation of "all-time high-power transmission" in traditional continuous jamming schemes. It can intelligently adjust the intermittent jamming strategy of the jamming drone based on the drone's location, channel quality, and the real-time spatial distribution of eavesdropping threats: jamming is activated during critical time slots when the eavesdropping channel quality is high or the security threat is approaching to suppress the eavesdropper, while maintaining low-interference or silent operation during safe time slots to conserve airborne energy. Secondly, based on this dynamically adjusted jamming mechanism, this scheme can formulate more efficient flight trajectory and time scheduling strategies, ensuring that the data acquisition task can be completed with the lowest total system energy consumption and the lowest information age while meeting strict bit error rate constraints. Most importantly, the global optimization perspective based on multivariate coupling ensures that the processes of "physical layer security defense" and "efficient data acquisition" are not isolated but coordinated and forward-looking, thus avoiding system energy efficiency collapse or data timeliness failure caused by excessive pursuit of a single security indicator, and maximizing the overall system performance in complex and constrained environments.

[0038] Furthermore, step S5 includes the following steps: Set initial values ​​for all optimization variables. At the same time, initialize the set of relaxation variables used to handle non-convex constraints (including { and convergence tolerance threshold Set the initial iteration count ; Then, repeat the following sub-steps until the objective function value changes. Less than the threshold , Update interference resources: With the current flight trajectory and time scheduling parameters fixed, linearize the coupling terms of interference power and interference ratio using a first-order Taylor expansion, construct and solve the interference resource optimization subproblem, and obtain the optimal interference power in the current iteration round. and the ratio of interference duration ; Update acquisition time: With the current flight trajectory, interference parameters, and updated interference resources fixed, construct and solve a linear programming subproblem to obtain the optimal node acquisition time that satisfies the data acquisition volume constraint in the current iteration round. ; Update slot length: With the current flight trajectory, interference parameters, and updated acquisition time fixed, construct and solve the slot length optimization subproblem to obtain the optimal slot length that satisfies the maximum information age constraint in the current iteration round. ; Update Bob's trajectory: Fix the updated time scheduling parameters and disturbance parameters, introduce auxiliary variables to handle non-convex rate constraints, and use continuous convex approximation techniques to transform the non-convex trajectory optimization problem into a convex problem and solve it, thereby updating and collecting the UAV's flight trajectory. ; Update Jack's trajectory: With other variables fixed, and considering the distance constraint between the jamming drone and the eavesdropper, use a first-order Taylor expansion to find its convex upper bound, construct and solve the subproblem of optimizing the jamming drone's trajectory, and update the flight trajectory of the jamming drone. ; Convergence check: Update iteration count Calculate the objective function value based on all updated variables, and determine whether the difference between the objective function values ​​of the two iterations is less than 1. If less than If the algorithm converges, it outputs the final global approximate optimal solution; otherwise, increment the iteration count by one and return to the step of updating the interfering resources until the algorithm converges. Finally, when the algorithm converges, it outputs the final global optimal solution. , as control parameters for the drone to perform its mission.

[0039] The beneficial effects of the above-mentioned further scheme are as follows: This invention proposes an efficient iterative solution algorithm based on the combination of block coordinate descent (BCD) and continuous convex approximation (SCA), effectively solving the problem of highly coupled variables and non-convex constraints in the original joint optimization problem. Specifically, this method decomposes the complex mixed-integer non-convex programming problem into several easily handled convex optimization subproblems such as disturbance resource optimization, time scheduling optimization, and flight trajectory optimization, achieving effective decoupling of multidimensional decision variables and significantly reducing the computational complexity of the algorithm. In particular, for the non-convex rate constraints and safety distance constraints in flight trajectory optimization that are difficult to solve directly, the SCA technique is used to transform them into a series of convex approximation constraints. This not only ensures that the solution generated in each iteration satisfies the physical constraints, but also guarantees the monotonically decreasing objective function value and the rapid convergence of the algorithm. Ultimately, this scheme can find a joint control strategy that is approximately globally optimal under limited computational overhead, achieving the best trade-off between system safety, timeliness, and energy efficiency. Attached Figure Description

[0040] Figure 1 This is a flowchart of the method of the present invention. Figure 2 A safety data acquisition model for drones.

[0041] Figure 3 The convergence performance of the proposed algorithm is given.

[0042] Figure 4 The optimized trajectory of the two drones.

[0043] Figure 5 The energy consumption of the proposed interference scheme is compared with that of the benchmark interference scheme. Detailed Implementation

[0044] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0045] Example like Figure 1 As shown, this invention provides a dual-UAV cooperative safe data acquisition scheme based on intermittent interference, the implementation method of which is as follows: S1. Under the current time slot, collect the real-time location coordinates of the drone (Bob) and the jamming drone (Jack), the distribution location of the IoT nodes, and the location estimation information of potential eavesdroppers. This location information is used to calculate the channel gain of the legitimate communication link and the eavesdropping link, and then derive the safe transmission rate under intermittent interference to ensure that the amount of data collected in the specified flight time meets the mission requirements. S2. Based on the location and status information obtained in step S1, establish communication models and time scheduling constraint models to describe legitimate links and eavesdropping links, including TDMA scheduling constraint models and data acquisition quantity constraint models; at the same time, establish system energy consumption models including flight propulsion energy consumption and jamming transmission energy consumption. S3. Construct an optimization function aimed at improving the overall performance of the system. This function integrates three core performance indicators: the system's information age, total energy consumption, and the eavesdropper error rate. S4. Taking into account the kinematic constraints, energy budget, collision avoidance constraints, and TDMA scheduling constraints of the UAV, the final joint optimization problem is constructed. The optimization objective of this problem is to minimize the weighted sum of the system AoI and the total energy consumption, while maximizing the bit error rate of the eavesdropper. S5. Based on the final optimization problem, an iterative algorithm based on block coordinate descent and continuous convex approximation is used to solve the non-convex problem constructed in step S4. The original problem is decomposed into sub-problems of interference parameter optimization, time scheduling optimization, and trajectory optimization, which are solved alternately to finally obtain the optimal UAV trajectory and intermittent interference strategy. The implementation method is as follows: Set initial values ​​for all optimization variables. At the same time, initialize the set of relaxation variables used to handle non-convex constraints (including { and convergence tolerance threshold Set the initial iteration count ; Then, repeat the following sub-steps until the objective function value changes. Less than the threshold , Update interference resources: With the current flight trajectory and time scheduling parameters fixed, linearize the coupling terms of interference power and interference ratio using a first-order Taylor expansion, construct and solve the interference resource optimization subproblem, and obtain the optimal interference power in the current iteration round. and the ratio of interference duration ; Update acquisition time: With the current flight trajectory, interference parameters, and updated interference resources fixed, construct and solve a linear programming subproblem to obtain the optimal node acquisition time that satisfies the data acquisition volume constraint in the current iteration round. ; Update slot length: With the current flight trajectory, interference parameters, and updated acquisition time fixed, construct and solve the slot length optimization subproblem to obtain the optimal slot length that satisfies the maximum information age constraint in the current iteration round. ; Update Bob's trajectory: Fix the updated time scheduling parameters and disturbance parameters, introduce auxiliary variables to handle non-convex rate constraints, and use continuous convex approximation techniques to transform the non-convex trajectory optimization problem into a convex problem and solve it, thereby updating and collecting the UAV's flight trajectory. ; Update Jack's trajectory: With other variables fixed, and considering the distance constraint between the jamming drone and the eavesdropper, use a first-order Taylor expansion to find its convex upper bound, construct and solve the subproblem of optimizing the jamming drone's trajectory, and update the flight trajectory of the jamming drone. ; Convergence check: Update iteration count Calculate the objective function value based on all updated variables, and determine whether the difference between the objective function values ​​of the two iterations is less than 1. If less than If the algorithm converges, it outputs the final global approximate optimal solution; otherwise, increment the iteration count by one and return to the step of updating the interfering resources until the algorithm converges. Finally, when the algorithm converges, it outputs the final global optimal solution. , as control parameters for the drone to perform its mission. In this embodiment, as Figure 2 As shown, this paper constructs a dual-drone-assisted Internet of Things (IoT) system, which consists of two drones, M IoT nodes (INs), and a ground eavesdropper (Eve) with uncertain location information. The legitimate data collection drone (Bob) starts from the initial hangar... The drone departs, traverses and collects all classified information from M nodes, and then returns to the hangar. To ensure communication security, another friendly jamming drone (Jack) simultaneously departs from the base. Take off, intermittently emit artificial noise to interfere with Eve, and return to base after completing the mission. Assume all devices in the system are equipped with a single antenna. The set of IoT nodes is represented as... .

[0046] System communication model: Assuming the channel power gain between any two nodes follows a free-space path loss model, the channel gain from the IoT node to the collecting drone in the nth time slot is expressed as:

[0047] in For reference channel gain, For Euclidean distance. Similarly, the channel gain from the friendly jammer to the collecting drone is expressed as... To ensure system robustness against the uncertainty of Eve's location, this invention considers the worst-case scenario. Specifically, to ensure a lower bound on security performance, it is assumed that Eve is located closest to INs and farthest from the interfering drone. Therefore, the channel gain from the eavesdropper to the IoT node is expressed as... The channel gain from the friendly jammer to the EVE is expressed as:

[0048] make Let Bob and Eve represent these two individuals, respectively. Consider the proposed intermittent interference scheme, and the receiver... The signal-to-noise ratio (SINR) at that point is expressed as:

[0049] in This indicates a stage or state. Specifically, This represents the interference phase, at which point... , This represents the non-interference (silent) phase, at which time... . Indicates the transmit power of INs. The term "interference duration ratio" represents the proportion of the total time slot duration in the nth time slot during which interference is transmitted. Therefore, This indicates Jack's actual interference power.

[0050] The average achievable data transmission rate of the m-th INs in the n-th time slot The channel capacity is determined by a weighted average of the interference phase and the non-interference phase, expressed as:

[0051] Time scheduling model: set up This refers to the time Bob acquires data from the Mth IN in the nth time slot. The UAV acquires a set of INs via TDMA, and Bob acquires data from each IN in each time slot. At most one IN data entry can be collected per session. The allocation of time resources should meet the following constraints:

[0052] Define the amount of data to be collected for each IN operation as follows: At the end of the entire task cycle, the cumulative data volume of each IN collected by Bob should reach the total task volume requirement:

[0053] Bit error rate model for eavesdroppers: Define receiver The bit error rate (BER) under BPSK modulation is the time-weighted sum of the BERs of the two stages, specifically expressed as:

[0054] in, It is the Gaussian Q function.

[0055] AOI model: To quantify the timeliness of perceived data, this invention uses AoI (Aspect-Oriented Intelligence) as a performance metric. Considering the characteristics of data collection tasks, it is assumed that all IoT nodes are... Status update data is generated continuously, and the drone only unloads the data once it has completed all data collection tasks and returned to the warehouse. Therefore, for the receiving end, the maximum AoI of the data is determined by the total mission duration of the drone, defined as follows: :

[0056] Energy consumption model: This invention specifies that propulsion energy consumption is much greater than communication energy consumption. The total energy consumption of the system mainly consists of the propulsion energy consumption of the two UAVs and the transmission energy consumption of the jammer. Therefore, the total energy consumption includes propulsion energy consumption. With interference energy consumption Given the limitations of onboard battery capacity, the system's total energy consumption during mission execution must meet the maximum energy budget. The total energy consumption model is defined as follows:

[0057] Ultimately, the overall performance can be expressed as: ,in These represent the weighting coefficients for AOI, total energy consumption, and the bit error rate of the eavesdropper, respectively.

[0058] In summary, the optimization objective of this invention aims to maximize overall performance by optimizing interference power, interference duration ratio, time slot duration, data acquisition time, and the flight trajectories of the two UAVs. Therefore, the specific optimization problem is stated as follows:

[0059]

[0060]

[0061]

[0062]

[0063]

[0064]

[0065]

[0066]

[0067]

[0068]

[0069]

[0070]

[0071] in, It is a key optimization variable. Represents the normalized weighting factor. Indicates the maximum duration of the time slot. This represents the upper limit of the bit error rate for legitimate drones. Indicates the location of the drone warehouse. This indicates the location that interfered with the take-off and landing of the drone. Indicates the maximum speed of the drone. This represents the minimum distance between two drones.

[0072] Since the problem is non-convex and the variables are coupled with each other, this invention uses efficient iterative algorithms of block coordinate descent (BCD) and successive convex approximation (SCA) to divide the original problem into 5 word problems for solving.

[0073] (1) The flight trajectories, time slot lengths, and acquisition times of the fixed interference UAV and the acquisition UAV are determined. A first-order Taylor expansion is used to process the non-convex terms, resulting in sub-problems for optimizing interference power and interference ratio. These sub-problems are then transformed into equivalent convex optimization problems. First, for the non-convex terms... The present invention at the k-th iteration point Using a first-order Taylor expansion, we obtain its global upper bound at point (). for:

[0074] , Then, regarding the bit error rate function in the objective function, this invention addresses its first... At the next iteration point, a first-order Taylor expansion is performed, and its linearization linearizes the bit error rate to... Thus, a linear approximation function for the bit error rate is constructed as follows:

[0075] , Among them, the partial derivative terms , This is calculated based on the derivative properties of the Q function. Similarly, the bit error rate of a legitimate receiver can be approximated linearly as follows: Finally, we arrive at the convex subproblem ( ):

[0076]

[0077]

[0078]

[0079] (2) The flight trajectories, interference power, interference ratio and time slot length of the fixed interference UAV and the data acquisition UAV are used to obtain the data acquisition time optimization subproblem. The CVX toolkit is used directly to solve this convex subproblem (P3).

[0080]

[0081]

[0082]

[0083]

[0084] In this problem, given Bob's flight time and trajectory, his flight speed can obviously be obtained. Therefore, the time allocation optimization in this problem becomes a linear optimization problem, which can be directly solved using CVX.

[0085] (3) The flight trajectories, interference power, interference ratio and acquisition time of the fixed interference UAV and the acquisition UAV are used to obtain the time slot length optimization subproblem. The CVX toolkit is used directly to solve this convex subproblem (P4).

[0086]

[0087]

[0088]

[0089]

[0090] This subproblem is a standard convex optimization problem, and the optimal time slot length can be obtained directly using the CVX toolkit.

[0091] (4) By fixing the flight trajectories, interference power, interference ratio, and acquisition time of the jamming UAV and the data acquisition UAV, a time slot length optimization subproblem is obtained, which is then transformed into an equivalent convex optimization problem. First, for the rate constraint, this invention introduces slack variables. The rewrite rate constraint is:

[0092] Introducing slack variables The variable must satisfy the following inequality constraints. In order to transform the original problem into a solvable form, auxiliary variables are defined. and The relaxation constraints are respectively and In addition, an auxiliary variable is introduced to determine the relative distance between the two drones. Its relaxation constraint is expressed as .

[0093] For key nonconvex terms The present invention is at point The convex lower bound is obtained by applying a first-order Taylor expansion. Transform the constraints into:

[0094]

[0095] Similarly, convexity can be achieved using a first-order Taylor expansion. , thus obtaining the convex lower bound .

[0096] For collision constraints, this invention addresses local points. Using a first-order Taylor expansion, we obtain the lower bound of the norm square function in (1d) and (5) as follows: The final problem obtained is the convex subproblem (P5).

[0097]

[0098]

[0099]

[0100]

[0101]

[0102]

[0103]

[0104]

[0105]

[0106]

[0107] (5) By fixing the time slot length, acquisition time, interference parameters, and the trajectory of the acquired UAV, the subproblem of interfering UAV is obtained, and it is transformed into an equivalent convex optimization problem. First, slack variables are introduced. ,make And Defined as signal-to-interference-plus-noise ratio (SIR / NNR):

[0108] Where a and b represent non-negative coefficients. Since It is about The concave increasing function, and It is a convex decreasing function, according to the properties of composite functions. = It is about Decreasing concave function, the present invention in Find its convex upper bound by performing a first-order Taylor expansion:

[0109] Similar to subproblem 4, this invention linearizes the collision constraint as follows: This ultimately leads to the convex subproblem ( )

[0110]

[0111]

[0112]

[0113] (6) Repeat the above steps until the change in the objective function value is less than the preset threshold. .

[0114] The effectiveness of the algorithm proposed in this invention was verified by experimental results. Figure 3 The convergence performance curves of the algorithm are plotted in the figure. It can be seen that as the number of iterations increases, the total objective function value of the system decreases rapidly and tends to stabilize (circle-shaped curve), converging quickly on the 13th iteration. Simultaneously, the AOI (triangle-shaped curve) and total energy consumption (square-shaped curve) show a decreasing trend, while the eavesdropper error rate (diamond-shaped curve) increases and stabilizes at approximately 0.35, demonstrating the algorithm's efficiency.

[0115] exist Figure 4 The image shows the optimized horizontal flight trajectories of the drones. The data acquisition drone (circular trajectory) plans a circuitous path to approach the IoT node and obtain better line-of-sight channel gain; while the cooperative jamming drone (triangular trajectory) flies directly to the potential area of ​​the eavesdropper and hovers above it, maximizing the suppression effect by shortening the jamming distance.

[0116] exist Figure 5 The energy consumption of the intermittent interference scheme proposed in this invention was compared with that of a benchmark continuous interference scheme. The results show that, while meeting the same data acquisition workload, the final energy consumption of the proposed scheme is significantly lower than that of the benchmark scheme, achieving an energy saving effect of over 60%. This verifies that optimizing the interference ratio... By activating interference only during critical time slots, unnecessary energy waste can be effectively avoided.

[0117] In summary, compared with the prior art, the present invention has the following advantages: This invention breaks through the energy consumption bottleneck of traditional continuous jamming, achieving precise and on-demand allocation of jamming energy. It innovatively introduces the jamming duration ratio as an optimization degree of freedom, breaking the rigid pattern of continuous noise transmission throughout the entire period in existing technologies. By implementing jamming only during critical time slots with poor channel quality and high eavesdropping risk, this invention significantly reduces the total system energy consumption compared to traditional solutions while ensuring basic communication security, effectively solving the problem of limited onboard energy for UAVs.

[0118] A unified framework for joint optimization of multi-dimensional performance indicators was constructed, achieving a trade-off in system efficiency. For three mutually constraining core indicators—physical layer security, Information Age (AoI), and system energy consumption—this invention establishes a unified mathematical optimization model. This framework overcomes the limitations of single-objective optimization and, through a weight adjustment mechanism, can achieve a trade-off between security, timeliness, and energy efficiency in complex and dynamic IoT environments.

[0119] An efficient algorithm for solving non-convex coupled problems is proposed, ensuring the robustness of trajectory planning and resource scheduling. Addressing the strong non-convexity and strong variable coupling characteristics inherent in optimization problems, this invention designs an alternating iterative algorithm based on block coordinate descent (BCD) and continuous convex approximation (SCA). This algorithm decomposes the complex original problem into several convex subproblems for sequential solution, exhibiting not only low computational complexity and good convergence performance, but also generating robust flight trajectories and interference strategies adaptable to the uncertainty of the eavesdropper's location.

Claims

1. A method for safe data acquisition through dual-UAV cooperative operation based on intermittent interference, characterized in that, Includes the following steps: S1. Establish an IoT data collection scenario assisted by a data collection drone (Bob) and a jamming drone (Jack) to obtain the current location of the drone in the current time slot, the location of the IoT node, and the estimated location information of potential eavesdroppers. S2. Based on the location and status information obtained in step S1, establish communication models and time scheduling constraint models to describe legitimate links and eavesdropping links, including TDMA scheduling constraint models and data acquisition quantity constraint models; at the same time, establish system energy consumption models including flight propulsion energy consumption and jamming transmission energy consumption. S3. Construct an optimization function aimed at improving the overall performance of the system. This function integrates three core performance indicators: Information Age (AoI), total energy consumption, and Bit Error Rate (BER). S4. Taking into account the kinematic constraints, energy budget, collision avoidance constraints, and TDMA scheduling constraints of the UAV, the final joint optimization problem is constructed. The optimization objective of this problem is to minimize the weighted sum of the system AoI and the total energy consumption, while maximizing the bit error rate of the eavesdropper. S5. Based on the final optimization problem, the non-convex problem constructed in step S4 is solved using an iterative algorithm based on block coordinate descent (BCD) and continuous convex approximation (SCA). The original problem is decomposed into sub-problems of interference parameter optimization, time scheduling optimization, and trajectory optimization, which are solved alternately to finally obtain the optimal UAV trajectory and intermittent interference strategy.

2. The method for cooperative and secure data acquisition of two unmanned aerial vehicles based on intermittent interference according to claim 1, characterized in that, The expression for the communication model is as follows: in, Indicates time slot IoT Nodes To collect the channel gain of the drone Bob; Indicates time slot The channel gain from interfering with unmanned aerial vehicle Jack to collecting data from unmanned aerial vehicle Bob; Indicates time slot IoT Nodes Channel gain to Eve the eavesdropper; This represents the channel gain used to interfere with the drone Jack's communication with the eavesdropper Eve; For reference channel gain, and These are the positions of Bob and Jack, respectively. For node position, The location of the eavesdropper's center. Let the radius be the radius of positional uncertainty. The flight altitude of the drone. in Indicates the stage state, specifically, This represents the interference phase, at which point... , This represents the non-interference (silent) phase, at which time... , Indicates the transmit power of INs. The term "interference duration ratio" represents the proportion of the total time slot duration during which interference is transmitted in the nth time slot. Therefore, This indicates Jack's actual interference power. This represents the average achievable data transmission rate of the m-th INs in the n-th time slot.

3. The method for cooperative safe data acquisition of two unmanned aerial vehicles based on intermittent interference according to claim 1, characterized in that, The expression for the time scheduling constraint model is as follows: in, This indicates the duration of the time slot, where m represents the IN number, M represents the number of INs, and n represents the time slot number. This refers to the time Bob acquires data from the Mth IN in the nth time slot; the UAV acquires a set of INs via TDMA, and Bob acquires data from each IN. If at most one IN data is collected, then there are: in, The amount of data Bob needs to collect for each IN should reach the total task volume requirement by the end of the entire task cycle.

4. The method for cooperative safe data acquisition of two unmanned aerial vehicles based on intermittent interference according to claim 1, characterized in that, The expression for the overall system performance index is as follows: in, This indicates the maximum information age of the system's data, used to quantify the timeliness of data acquisition at the receiving end. This indicates the total number of time slots divided into the drone flight mission cycle. This represents the duration of the nth time slot; This represents the total energy consumption of a dual-drone collaborative system, a value that must meet the maximum energy budget. Constraints The total propulsion energy consumption of the two drones (Bob, the data acquisition drone, and Jack, the jamming drone) was characterized throughout the entire mission cycle. This characterizes the communication transmission energy consumed by the intermittent jamming of unmanned aerial vehicles (UAVs), the magnitude of which is determined by the jamming power. and the proportion of interference duration Joint decision; This represents the average bit error rate (BER) of the eavesdropper in the nth time slot. Since this invention employs an intermittent interference strategy, the total BER in this time slot consists of two parts: the first part is the BER during the interference activation phase. Its weight is the proportion of interference duration. The second part is the bit error rate during the interference silence phase. Its weight is the proportion of non-interference time. , It is a standard Gaussian Q function used to calculate the bit error probability under BPSK modulation based on the signal-to-interference-plus-noise ratio.

5. The method for cooperative safe data acquisition of two unmanned aerial vehicles based on intermittent interference according to claim 1, characterized in that, The final optimization problem is expressed as follows: in, It is a key optimization variable. Represents the normalized weighting factor. Indicates the maximum duration of the time slot. This represents the upper limit of the bit error rate for legitimate drones. Indicates the location of the drone warehouse. This indicates the location that interfered with the take-off and landing of the drone. Indicates the maximum speed of the drone. This represents the minimum distance between two drones.

6. The method for cooperative and secure data acquisition of two unmanned aerial vehicles based on intermittent interference according to claim 1, characterized in that, S5 includes the following steps: Set initial values ​​for all optimization variables. At the same time, initialize the set of relaxation variables used to handle non-convex constraints (including { and convergence tolerance threshold Set the initial iteration count ; Then, repeat the following sub-steps until the objective function value changes. Less than the threshold ; Update interference resources: With the current flight trajectory and time scheduling parameters fixed, linearize the coupling terms of interference power and interference ratio using a first-order Taylor expansion, construct and solve the interference resource optimization subproblem, and obtain the optimal interference power in the current iteration round. and the ratio of interference duration ; Update acquisition time: With the current flight trajectory, interference parameters, and updated interference resources fixed, construct and solve a linear programming subproblem to obtain the optimal node acquisition time that satisfies the data acquisition volume constraint in the current iteration round. ; Update slot length: With the current flight trajectory, interference parameters, and updated acquisition time fixed, construct and solve the slot length optimization subproblem to obtain the optimal slot length that satisfies the maximum information age constraint in the current iteration round. ; Update Bob's trajectory: Fix the updated time scheduling parameters and disturbance parameters, introduce auxiliary variables to handle non-convex rate constraints, and use continuous convex approximation techniques to transform the non-convex trajectory optimization problem into a convex problem and solve it, thereby updating and collecting the UAV's flight trajectory. ; Update Jack's trajectory: With other variables fixed, and considering the distance constraint between the jamming drone and the eavesdropper, use a first-order Taylor expansion to find its convex upper bound, construct and solve the subproblem of optimizing the jamming drone's trajectory, and update the flight trajectory of the jamming drone. ; Convergence check: Update iteration count Calculate the objective function value based on all updated variables, and determine whether the difference between the objective function values ​​of the two iterations is less than 1. If less than If the algorithm converges, it outputs the final global approximate optimal solution; otherwise, increment the iteration count by one and return to the step of updating the interfering resources until the algorithm converges. Finally, when the algorithm converges, it outputs the final global optimal solution. , as control parameters for the drone to perform its mission.

Images