Self-adaptive searching and tracking resource scheduling method for multifunctional phased array radar

Abstract

The invention discloses an adaptive searching and tracking resource scheduling method for a multifunctional phased array radar. The method comprises the following steps: generating a three-dimensional searching visual direction request; obtaining all view direction requests in the current scheduling period; performing view direction request scheduling, and putting view direction requests which are not executed into a delay task queue; traversing and sequentially responding to the view direction requests in the execution queue, if the current view direction request is a three-dimensional search view direction request, performing trace point condensation, and creating a prompt search view direction request; if the current visual direction request is a search prompting visual direction request, trace point condensation is carried out, a track head is created, and a target tracking visual direction request is created according to track information; if the current visual direction request is a target tracking visual direction request and a target exists, trace point condensation, trace point association, track filtering, track management and track prediction are carried out; if the target does not exist, track management is carried out, and a target tracking visual direction request is created. The method can improve the task execution efficiency, and gives consideration to the high-precision tracking of the maneuvering target.

Description

Technical Field

[0001] This invention belongs to the field of multi-functional phased array radar technology, specifically relating to an adaptive search and tracking resource scheduling method for multi-functional phased array radar. Background Technology

[0002] Multifunctional phased array radars can perform various tasks, such as detecting new targets and tracking known targets within the surveillance airspace. The execution of multiple tasks requires sharing the system's limited resources, including time, aperture, bandwidth, and energy. Therefore, how to allocate these limited resources is crucial to maximizing the system's detection performance. Adaptive resource scheduling technology, which rationally allocates system resources based on the radar's operating mode, detection mission, and sensing environment, has been widely studied both domestically and internationally.

[0003] Based on the rules governing the partitioning of time and aperture resources, existing scheduling modes can be broadly categorized into sequential multibeam, interleaved multibeam, and simultaneous multibeam resource scheduling. Sequential multibeam scheduling rapidly revisits multiple tasks sequentially through time partitioning and adjusts the dwell time for each revisit, exhibiting high gain and suitability for long-range detection scenarios. Under this mode, existing technologies have proposed numerous heuristic or optimization scheduling methods based on criteria such as task priority, time windows, and detection performance, improving resource utilization. Interleaved multibeam scheduling uses pulse interleaving technology to schedule short-range tasks during the pulse transmission and reception intervals of long-range detection events, improving scheduling flexibility and system target capacity. Simultaneous multibeam scheduling uses aperture partitioning to form multiple independently transmitting and receiving beams, simultaneously revisiting multiple tasks with different power-aperture products, exhibiting long dwell time and suitability for medium- and short-range detection scenarios. Existing research has considered resource allocation issues under a single scheduling mode. For the joint time and aperture scheduling problem, existing technologies propose a time-aperture two-dimensional resource management method, enabling radar to image targets while performing search and tracking tasks. However, the existing time-aperture two-dimensional scheduling method has the following shortcomings: (1) the virtualized task requests do not make full use of prior information; (2) there is a lack of evaluation of the performance of multi-task execution and the establishment of feedback from evaluation to scheduling. Summary of the Invention

[0004] The purpose of this invention is to provide a multifunctional phased array radar adaptive search and tracking resource scheduling method, computer equipment, computer-readable storage medium, and computer program product. It establishes a closed-loop framework based on detection-processing-evaluation-scheduling, improves the parallel execution efficiency of stereo search and cue search tasks through aperture segmentation, and takes into account the high-precision target tracking requirements. Through time segmentation, it performs adaptive time allocation to take into account the tracking of both maneuvering and non-maneuvering targets.

[0005] To achieve the above objectives, one aspect of the present invention provides a method for adaptive search and tracking resource scheduling of a multifunctional phased array radar, comprising: Step S1: Divide the airspace into multiple regions, set different search parameters for the wave position of each region, and generate a stereo search line of sight request; Step S2: Set the scheduling interval, and put the delayed line-of-sight requests, the tracking line-of-sight requests in the current scheduling interval, the prompting search line-of-sight requests, and the stereo search line-of-sight requests numbered by wave position into the scheduling interval to obtain all line-of-sight requests in the current scheduling cycle. Step S3: Perform view request scheduling and place the view requests that failed to be executed into the delayed task queue; Step S4: Traverse and respond to the view requests in the execution queue in sequence. If the current view request is a stereo search view request, then execute step S5; if the current view request is a cue search view request, then execute step S6; if the current view request is a target tracking view request, then execute step S7. Step S5: Perform signal processing on the echo signal. If a target exists, perform point cache and aggregate intra-frame and inter-frame points. Create a prompt search view request based on the aggregated points. Step S6: Perform signal processing on the echo signal. If a target exists, perform point cache and converge multiple points within the prompt search beam. Create a track head based on the converged points and create a target tracking line-of-sight request based on the track information. Step S7: Perform signal processing on the echo signal. If a target exists, perform point convergence, point-to-path correlation, path filtering, path management, and path prediction. If no target exists, perform path management, terminate paths that have not been updated for a long time, and perform path prediction on confirmed paths that have not been updated for longer than the longest path update interval and have not been revoked. Create a target tracking line-of-sight request based on the path prediction results.

[0006] Another aspect of the present invention provides a computer device including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method described above.

[0007] Another aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0008] Another aspect of the present invention provides a computer program product including a computer program that, when executed by a processor, implements the steps of the method described above.

[0009] According to the multi-functional phased array radar adaptive search and tracking resource scheduling method, computer equipment, computer-readable storage medium and computer program product of the present invention, the parallel execution efficiency of stereo search and cue search tasks is improved by aperture segmentation, and the high-precision target tracking requirements are also taken into account. By time segmentation, adaptive time allocation is performed, which takes into account the tracking of both maneuvering and non-maneuvering targets. Attached Figure Description

[0010] To more clearly illustrate the technical solutions of the present invention, the accompanying drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort: Figure 1 This is a flowchart of a multi-functional phased array radar adaptive search and tracking resource scheduling method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a multi-functional phased array radar adaptive search and tracking resource scheduling method according to an embodiment of the present invention; Figure 3 This is a flowchart of a view request scheduling according to an embodiment of the present invention; Figure 4 This is a schematic diagram illustrating the principle of look-to-look request scheduling according to an embodiment of the present invention; Figure 5 This is a 2D schematic diagram of the target motion trajectory according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the filtering results of the confirmed flight track in the XY plane according to an embodiment of the present invention; Figure 7 This is a schematic diagram of the data rate optimization result according to an embodiment of the present invention; Figure 8 This is a schematic diagram illustrating the optimized dwell time result of one embodiment of the present invention; Figure 9 This is a schematic diagram of the power aperture product allocation result according to an embodiment of the present invention; Figure 10 This is a schematic diagram illustrating the optimized dwell time result of one embodiment of the present invention; Figure 11 This is a structural diagram of a computer device according to an embodiment of the present invention. Detailed Implementation

[0011] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0012] One embodiment of the present invention provides an adaptive search and tracking resource scheduling method for a multifunctional phased array radar, such as... Figure 1 As shown, the adaptive search and tracking resource scheduling method for multi-functional phased array radar according to an embodiment of the present invention includes steps S1 to S7. The following is in conjunction with... Figure 2 The schematic diagram illustrates in detail each step of the adaptive search and tracking resource scheduling method for multifunctional phased array radar according to an embodiment of the present invention.

[0013] Step S1: Generate stereo search view request Based on the site information, the airspace is divided into multiple regions, and different search parameters are set for the wavefront of each region to generate a stereo search line-of-sight request. Let the first... The distance range of each search space is The azimuth range is The pitch angle range is To improve the detection probability of targets deviating from the beam center, low-loss point beams are used for beam positioning, with the azimuth and elevation angle spacing between adjacent beam centers being as follows: and Based on the detection range, peak radar transmit power, beamwidth, detection probability, false alarm probability, and set RCS (Radar Cross Section) for each wave position, the pulse width, pulse repetition period, number of pulses, dwell time, and power aperture product are calculated.

[0014] Power aperture product (PAP) is the average transmit power. Subarray aperture size The product of these can also be expressed as peak transmit power. Pulse width and pulse repetition frequency It is expressed as the reciprocal of the pulse repetition period.

[0015] (1) In the formula, The azimuth and elevation beamwidths are respectively and When the aperture is used, with antenna weighting or windowing, the beamwidth factor for the radar's azimuth and elevation angles is 1–5; otherwise, the aperture... The pulse repetition frequency simultaneously satisfies as well as , At the speed of light, The frequency used. Blind spots Decide Maximum , λ represents the duty cycle. λ represents the operating wavelength.

[0016] For the line of sight located at distance (3D search is) ), Azimuth beamwidth and pitch angle beamwidth For radar observation of a target, the number of transmitted pulses can be calculated using radar equations. (2) In equation (2), For target RCS; The azimuth and elevation angle widths are respectively and Subarray gain at time; This is the radar detectability factor, which determines the probability of a given false alarm. Under these conditions, in order to achieve the desired detection probability Required signal energy to receive; This indicates that the scan angle deviates from the normal. and Antenna scanning loss resulting from this; The system noise temperature; is Boltzmann's constant.

[0017] Step S2: Obtain all look-ahead requests within the current scheduling period. Set the scheduling interval. Place delayed look-ahead requests, tracking look-ahead requests within the current scheduling interval, cue search look-ahead requests, and stereo search look-ahead requests numbered by wave position into the scheduling interval to obtain all look-ahead requests within the current scheduling period. Initially, the task only contains stereo search look-ahead requests. After a target is detected at a stereo search wave position, a cue search look-ahead request is generated (see subsequent step S5). After the cue search look-ahead request detects a target, a target confirmation look-ahead request is generated. After confirmation, a target tracking look-ahead request is generated (see step S6). After the look-ahead requests are scheduled (see step S3), any unexecuted look-ahead requests are placed in the delayed task queue.

[0018] View requests have multiple attributes, the first being... A view request can be represented as (3) Here, TP represents the mission type. ST represents the arrival time, which is also the optimal time to execute the observation request from the mission's perspective. For stereo search missions, ST is determined by the preset positional arrangement and beam scanning order; for cue search missions, ST is determined by the detection value of the previous scheduling cycle and the initial time of the current scheduling; for target tracking missions, ST is the sum of the track update time and the revisit interval (data rate). The mission execution time is allowed to slide around the request time (time window), but not exceeding the current scheduling interval. PRTY represents the mission priority; observations with the same ST but higher priority will be scheduled first. The priorities of target confirmation, target update, cue search, and stereo search are from highest to lowest. The line-of-sight angle is composed of azimuth and elevation; PRF represents the pulse repetition frequency. For stereo search, the PRF is determined by the preset detection range; while for cue search and target tracking, the target range estimate is known. The selection of PRF must simultaneously meet the following requirements. as well as ,in, Represents the speed of light. This indicates the sampling frequency, and the pulse width is determined by the PRF and duty cycle. This represents the estimated value of the target state. The dwell time is the product of the number of pulses and the pulse repetition time. PAP represents the power aperture product, and BAW and BEW represent the bandwidth and azimuth / elevation beamwidth, respectively.

[0019] Step S3: Lookout Request Scheduling Due to resource constraints, not all look-ahead requests can be executed within a single scheduling cycle. The resource scheduler selects a subset of observations that can be executed within the schedule without violating any resource constraints and arranges them in time order according to their priority. Stereo search and cue search can be executed concurrently, and the scheduler schedules these two tasks in parallel. For look-ahead requests that cannot be executed in the current scheduling cycle, the scheduler places them in a deferred task queue, appropriately increases the priority of the request within the same task type, and then schedules them in the next cycle. Figure 3 and Figure 4 As shown, the scheduling process is as follows.

[0020] S11: Retrieve delay time within the specified time All view requests within.

[0021] S12: Extract non-cue search type view requests, sort non-repeating priority view requests from high to low, and iterate through the non-repeating priority.

[0022] S13: Sort the STs of different look-ahead requests with the same priority from smallest to largest, traverse them, and fill them sequentially into the current scheduling interval (or time interval). The execution time of the first look-ahead request is the initial time of the current scheduling interval. The following is the first Execution time of each request It equals the end time of the previous request, that is... For view requests that cannot be executed within the current scheduling period (whose end time exceeds the end time of the current scheduling interval). ( ), needs to be placed in the delay queue.

[0023] S14: If there is no stereo search task in the task queue, then prompt for a search view request according to S13 after the last scheduled tracking view request; otherwise, proceed to S15.

[0024] S15: If a prompt for a search view request exists, proceed to S16; otherwise, proceed to S18 and generate a new task request based on the task execution result, and then proceed to the next scheduling interval.

[0025] S16: Find consecutive 3D search time intervals in the task execution queue. Iterate through each time interval and schedule the lookout request according to S13. For delayed tasks that cannot be scheduled in the current time interval, continue scheduling in the next time interval. If there are still delayed lookout tasks, proceed to S17.

[0026] S17: Use the end time of the last scheduled hint search as the initial scheduling time. Given the end time of the scheduling, delayed prompt search tasks are scheduled within this time interval according to S13.

[0027] S18: Compare the end time of the last scheduled hint search with the initial scheduling time. Given the end time of the scheduling, delayed prompt search tasks are scheduled within this time interval according to S13.

[0028] S19: Increase the priority of lookout requests in the delayed task queue by a factor equal to the difference in task priorities.

[0029] As can be seen, cue search can be arranged in parallel with 3D search, or it can be arranged sequentially with tracking tasks.

[0030] Step S4: Traverse and respond to the lookout requests in the execution queue in sequence. Iterate through and respond to the look-ahead requests in the execution queue in sequence. For the beam pointing to the preset width in the look-ahead angle direction of the look-ahead request, emit a specified number of pulses with a specified pulse width and bandwidth. If the current look-ahead request is a stereo search, proceed to step S5; if the current look-ahead request is a cue search, proceed to step S6; if the current look-ahead request is a target tracking, proceed to step S7.

[0031] Step S5: Perform signal processing on the echo signal. If a target exists, perform point cache and aggregate intra-frame and inter-frame points. Create a prompt search look-ahead request based on the aggregated points.

[0032] Target detection can be performed using traditional methods, including pulse compression, coherent accumulation, sum-difference angle measurement, and constant false alarm rate (CFAR) detection. The same target may be detected multiple times within adjacent beams of a stereo search, thus requiring clustering. Intra-frame clustering is performed on the measurements obtained from the stereo search within the current scheduling interval. Then, a buffer window of two scheduling intervals is set for the measurements obtained from the stereo search, and inter-frame clustering is performed by windowing. The clustering algorithm used is DBSCAN, with the distance, azimuth, and elevation resolutions of different tasks used as clustering thresholds. When the stereo search task produces a detection in the previous scheduling cycle, it prompts the search task to create a look-ahead request in the current scheduling cycle. Otherwise, it indicates that the search task is idle and will not create any search requests. After a target is detected in the stereo search, n prompt search requests are generated. The prompt search request corresponds to n (5 / 7 / 9) quincunx wave positions around the detection point. Based on the distance to the detection point, the prompt search wave position width, detection probability, false alarm probability, and set RCS, the pulse width, pulse repetition time, number of pulses, dwell time, and power aperture product parameters of each prompt search line-of-sight request are calculated.

[0033] Step S6: Perform signal processing on the echo signal. If a target is present, perform point cache and converge the points within multiple cue search beams. Create a track header based on the converged points and generate a target tracking line-of-sight request based on the track information.

[0034] The prompt search requires multi-beam scanning around the detection points after stereo search convergence, and point convergence is performed on all scan results using the DBSCAN algorithm. A track header is created based on the converged point tracks. Based on the track information, a tracking confirmation look-ahead request is created. The tracking confirmation data rate is equal to the reciprocal of the scheduling interval, which is the maximum update data rate. After three consecutive frames of confirmation, a track update look-ahead request is generated. Whether it is target confirmation or target update, the target state needs to be predicted to the revisit time, as shown in subsequent step S7. Based on the predicted distance, tracking beamwidth, detection probability, false alarm probability, and the set RCS, the pulse width, pulse repetition time, number of pulses, dwell time, and power aperture product parameters of each prompt search look-ahead request are calculated. The line-of-sight angle of beam illumination is calculated based on the predicted azimuth and predicted elevation angles, thereby generating the look-ahead request.

[0035] Step S7: Process the echo signal. If a target is present, perform point convergence, point-to-path correlation, path filtering, path management, and path prediction. If no target is present, perform path management, terminate paths that have not been updated for a long time, and perform path prediction for confirmed paths that have not been updated for a period greater than the longest path update interval (longest tracking data rate) and have not been revoked. Based on the path prediction results, create a target tracking line-of-sight request.

[0036] The echo signal undergoes signal processing. If a target is present, point aggregation is performed. The aggregated points are then used for track initiation and point association, employing the global nearest neighbor algorithm. After track initiation, the first two points of the track are used for filter initialization to obtain an initial target state estimate. and estimate covariance Assuming a model The target state is conditional The conditional probability density function is a Gaussian distribution, i.e. ,in, and The first The state estimates and covariance corresponding to each model Indicates measurement, This represents a normal distribution. The initial model probability is... , and The initial value is obtained through and The results show that three models are considered: uniform speed, uniform acceleration, and uniform speed turning. For the uniform speed and uniform acceleration models, an extended Kalman filter is used to reduce the computational load while meeting the filtering accuracy requirements. For the uniform speed turning model, a square root cumulative Kalman filter algorithm is used to meet the requirement of high-precision estimation of the turning rate.

[0037] Adaptive tracking updates non-maneuvering targets less frequently than maneuvering targets, thus improving system resource utilization. Tracking data rates are categorized from low to high as follows: file, represented as Data rate optimization depends on the relationship between predicted tracking accuracy and a specified accuracy threshold, and predicted accuracy can be evaluated by predicted covariance. In the IMM algorithm framework, each model outputs a predicted covariance, but a final predicted covariance is lacking. Therefore, based on the best-fit Gaussian model, the first and second moments (states and covariance) of the multi-model model are equivalent to the first and second moments of a single model, resulting in the combined transition matrix and predicted covariance. (4) (5) In equations (4) and (5), and All are data rates The function. For the number of models, For the first The probability of the predicted model for each model. Indicates data rate Time model The state transition matrix, Indicates data rate Time model The process noise covariance matrix.

[0038] Since most maneuvering gaps are caused by targets falling outside the beam illumination range, the fraction of the azimuth beamwidth is defined as the accuracy threshold, known as track sharpness. When a tracking mission requests track update observations, the tracker selects the lowest track revisit rate whose predicted track azimuth accuracy does not exceed the specified track sharpness. The optimization problem can be formulated as follows: (6) In equation (6), This is a track sharpness adjustment factor. Indicates the azimuth beamwidth used for the tracking mission. This represents a function that maps positional errors to azimuth accuracy, primarily involving coordinate transformations and square root operations on the corresponding elements. Due to the small data set, and... about The data rate decreases monotonically, and the search can be performed sequentially from the lowest data rate until the track sharpness requirement is met. As shown in (6), maneuvering targets will exceed the track sharpness faster than non-maneuvering targets, so the track needs to be updated more frequently.

[0039] according to and Calculate the target predicted state as follows (7) The predicted distance, predicted azimuth, and predicted elevation angle are obtained through coordinate transformation.

[0040] If no target exists, perform track management, terminate tracks that have not been updated for a long time, and for confirmed tracks that have not been updated and have not been revoked within a time interval greater than the longest track update interval (longest tracking data rate), perform track prediction, with the prediction time being the start time of the next scheduling. Based on the track prediction results, create a target tracking look-ahead request.

[0041] The following example illustrates the outstanding effect of the multi-functional phased array radar adaptive search and tracking resource scheduling method of the present invention.

[0042] Set the peak transmit power of the phased array radar to kw, wavelength is m, noise figure is dB, scheduling period of 50 ms, stereo search PRF of 750 Hz, and detection probability of The false probability is The minimum detectable target SNR is 12.8 dB, and the reference target RCS is [missing value]. Under the given conditions, the maximum detection range is 200 km. The task priorities for target confirmation, target update, cue search, and stereo search are 4, 3, 2, and 1, respectively, with delayed tasks having a priority increased to 0.2. To avoid missed target detections, low-loss point beamforming is used in the beamforming, arranged in a sinusoidal coordinate system. The azimuth and elevation beamwidth for stereo search is... The prompt indicates that the azimuth and elevation beamwidth for the search is... The tracking beamwidth is The azimuth beamwidth difference between adjacent beam positions is 0.866 azimuth beamwidths, and the elevation beamwidth difference is 0.75 elevation beamwidths. The azimuth scanning range of the phased array radar is set as follows: The elevation scan range is The bandwidths for stereo search, boost search, and target tracking are 0.5MHz, 1MHz, and 2MHz, respectively. The track revisit interval is... Therefore, the highest tracking data rate is 5 Hz, the lowest data rate is 0.5 Hz, and the track confirmation data rate is 20 Hz. The track sharpness factor is 0.03.

[0043] The site of the phased array radar is m, assuming there are 6 targets distributed at different altitudes in its detection airspace, the radar position and target trajectories are as follows: Figure 5 As shown.

[0044] By confirming the points on the track, the correlation, and the filtering results, it can be seen that the adaptive tracking method of this invention achieves correct correlation and continuous stable tracking of all targets in the airspace. Furthermore, after forming a confirmed track, the points prompted for searching are only correlated, not filtered, thereby avoiding the same target from repeatedly starting a new track. Only when a target is lost is the target track updated using the points prompted for searching.

[0045] To more clearly demonstrate the effect of adaptive data rate, Figure 6 The filtered results of the confirmed track in the XY plane are presented. It is evident that the data rate for the target during maneuvering turns is significantly higher than that during non-maneuvering segments (straight-line motion). The method of this embodiment improves system resource utilization by reducing the update data rate of the track during non-maneuvering segments, allowing resources saved from the tracking task to be used for new target discovery.

[0046] Figure 7 The data rate optimization results for tracks 1 and 9 are presented. It can be seen that track 1 maneuvers around 40 s and 110 s, thus receiving a higher revisit data rate. After 120 s, the target moves primarily in a straight line, resulting in a generally lower data rate. Track 9 maneuvers around 40 s and 80 s, also receiving a higher revisit data rate, while the data rate is generally lower in the non-maneuvering segments. This further verifies the effectiveness of the adaptive tracking data rate. Due to the difficulty in predicting target maneuvers, the data rate adjustment has a lag, manifested in the target's data rate jumping from minimum to maximum to quickly reduce lag error.

[0047] The results of the time of stay allocation are as follows Figure 8 As shown, the target's dwell time is mainly determined by distance; the greater the distance, the lower the SNR, and the longer the dwell time under the same detection requirements. Track 10 moves closer to the phased array radar for the first 80 seconds and then moves away from the target. Therefore, the target's dwell time exhibits a certain symmetry, verifying the effectiveness of the dwell time allocation. Due to the slow change in target distance, the dwell time allocation results show a stepped pattern, which is also related to the assumption that the target's RCS does not fluctuate. In addition, there are cases where the dwell times of two targets "interleave," which is caused by one beam detecting two targets simultaneously.

[0048] Figure 9 A graph showing the PAP allocation results is presented. It can be seen that although the instantaneous PAP of the phased array radar varies significantly over time, the average PAP remains close to 12460 W·m² within each scheduling cycle. This indicates that although the phased array radar allocates some resources for cue search and target tracking, its primary task remains stereo search.

[0049] Figure 10The time occupancy rates of various tasks in each scheduling cycle are presented, defined as the ratio of the total dwell time consumed by each task in that cycle to the scheduling interval. The results show that the phased array radar almost always dedicates 90% or more of its time window to stereo search. Fluctuations in the stereo search time proportion are due to 1) triggering a cue search task whenever a stereo search produces a detection, even if the detected target is already being tracked; and 2) variations in the number of tracked targets and target distances, leading to fluctuations in the resources allocated to all tracking tasks. Overall, this result indicates that tracking six targets simultaneously performing stereo search and cue search does not significantly strain the phased array radar's resources. However, if the number of targets in the search airspace increases, more and more resources will be shifted from search tasks to tracking tasks. Since the average resource consumption for tracking six targets is approximately 1.38%, the maximum target capacity of the phased array radar under the same conditions can be estimated to be approximately 434. In reality, target capacity is closely related to the number of false targets and target dynamics. When the false rate increases, the radar will consume a lot of resources even without targets, and high-frequency target maneuvers will also cause the phased array radar to revisit frequently, increasing resource consumption.

[0050] Following the above process, Table 1 compares the performance of the method of this embodiment and the traditional non-adaptive scheduling method. In the table, the average distance estimation error is defined as in equation (8), and the average azimuth and elevation angle estimation errors are defined similarly.

[0051] (8) In the formula, This indicates the number of confirmed flight tracks. Indicates the first Confirm the length of the flight. and They represent the first The estimated and actual values ​​of the dwell distance.

[0052] Table 1 Comparison of Target Tracking Performance

[0053] Combining the results in Table 3, it can be seen that the performance of the method of the present invention is a compromise between the results of the traditional method at data rates of 5Hz and 2.5Hz. Considering the actual situation, the system tends to use a higher data rate ( TAS tracking is performed at Hz. In this case, the method of the present invention can improve the system target capacity by more than 1.5 times while meeting the intelligence quality requirements.

[0054] In summary, the method of this invention, targeting phased array radars with shared apertures for stereo search, cue search, and tracking, establishes a closed-loop framework based on detection-processing-evaluation-scheduling. Through aperture and bandwidth segmentation, it improves the parallel execution efficiency of stereo search and cue search tasks while also ensuring high-precision target tracking. Through time segmentation and target dynamic perception, it adaptively adjusts dwell and revisit times, accommodating both maneuvering and non-maneuvering target tracking. Specifically, dwell time allocation is based on prior target distance information and detection probability, allocated on demand. Revisit time is based on optimal Gaussian fitting and interactive multi-model prediction of target dynamics, increasing revisit frequency during maneuvers and decreasing it during non-maneuvering. Compared to traditional TAS fixed high data rate (≥5Hz) scheduling methods, this invention's method can increase system target capacity by approximately 1.5 times while maintaining clear situational awareness and high-precision detection.

[0055] Embodiments of the present invention also provide a computer device, which may be a server, and its internal structure diagram may be as follows: Figure 11 As shown. The computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores operating parameter data for various components. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer program implements the steps of the method according to embodiments of the present invention.

[0056] Those skilled in the art will understand that Figure 11 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0057] Embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method of the embodiments of the present invention.

[0058] Embodiments of the present invention also provide a computer program product, including a computer program that, when executed by a processor, implements the steps of the method of the embodiments of the present invention.

[0059] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for adaptive search and tracking resource scheduling in a multi-functional phased array radar, characterized in that, include: Step S1: Divide the airspace into multiple regions, set different search parameters for the wave position of each region, and generate a stereo search line of sight request; Step S2: Set the scheduling interval, and put the delayed line-of-sight requests, the tracking line-of-sight requests in the current scheduling interval, the prompting search line-of-sight requests, and the stereo search line-of-sight requests numbered by wave position into the scheduling interval to obtain all line-of-sight requests in the current scheduling cycle. Step S3: Perform view request scheduling and place the view requests that failed to be executed into the delayed task queue; Step S4: Traverse and respond to the view requests in the execution queue in sequence. If the current view request is a stereo search view request, then execute step S5; if the current view request is a cue search view request, then execute step S6; if the current view request is a target tracking view request, then execute step S7. Step S5: Perform signal processing on the echo signal. If a target exists, perform point cache and aggregate intra-frame and inter-frame points. Create a prompt search view request based on the aggregated points. Step S6: Perform signal processing on the echo signal. If a target exists, perform point cache and converge multiple points within the prompt search beam. Create a track head based on the converged points and create a target tracking line-of-sight request based on the track information. Step S7: Perform signal processing on the echo signal. If a target is present, perform point convergence, point-to-path correlation, path filtering, path management, and path prediction. If no target exists, perform track management, terminate tracks that have not been updated for a long time, and for confirmed tracks that have not been updated for longer than the longest track update interval and have not been revoked, perform track prediction and create a target tracking sight request based on the track prediction results.

2. The method as described in claim 1, characterized in that, In step S7, if a target exists, a trajectory is initiated on the aggregated point track. After the trajectory is initiated, the first two points of the trajectory are used for filter initialization to obtain the target's initial state estimate. and estimate covariance ,in k Indicates time; The target predicted state is: ， The integrated transition matrix of multiple models and predicted covariance They are respectively: ， ， in, and All are data rates The function, For the number of models, For the first The probability of the predicted model for each model. Indicates data rate Time model The state transition matrix, Indicates data rate Time model The process noise covariance matrix; Based on the target's predicted state, the predicted distance, predicted azimuth, and predicted elevation angle are obtained through coordinate transformation.

3. The method as described in claim 1 or 2, characterized in that, In step S6, a tracking confirmation line-of-sight request is created based on the track information. After three consecutive frames of confirmation, a track update line-of-sight request is generated. The target state is predicted to the revisit time. The search parameters for each prompt search line-of-sight request are calculated based on the predicted distance. The line-of-sight angle of beam illumination is calculated based on the predicted azimuth and predicted elevation angles, thereby generating a target tracking line-of-sight request.

4. The method as described in claim 1 or 2, characterized in that, Based on the predicted distance, tracking beamwidth, detection probability, false alarm probability, and set RCS, the pulse width, pulse repetition time, number of pulses, dwell time, and power aperture product parameters for each cue search look-ahead request are calculated.

5. The method as described in claim 1 or 2, characterized in that, In step S5, after the target is detected by the stereo search, a prompt search request is generated. Waveforms are arranged around the detection point. Based on the distance of the detection point, the prompt search waveform width, the detection probability, the false alarm probability, and the set RCS, the pulse width, pulse repetition time, pulse number, dwell time, and power aperture product parameters of each prompt search line-of-sight request are calculated.

6. The method as described in claim 1 or 2, characterized in that, Step S3 includes: S11: Retrieve all view requests with a delay time within the specified time; S12: Extract non-cue search type view requests, sort non-repeating priority view requests from high to low, and iterate through the non-repeating priority. S13: Sort the arrival times of different view requests with the same priority from smallest to largest, and iterate through them, filling them into the current scheduling interval in order. For view requests that cannot be executed within the current scheduling period, put them into the delay queue. S14: If there is no stereo search task in the task queue, then after the last scheduled tracking gaze request, a prompt search gaze request will be scheduled according to step S13; otherwise, proceed to step S15. S15: If a prompt search view request exists, proceed to step S16; otherwise, proceed to step S18, and generate a new task request based on the task execution result, and then proceed to the next scheduling interval. S16: Find consecutive 3D search time intervals in the task execution queue, traverse each time interval, arrange the view request according to step S13, and continue to arrange the delayed task in the next time interval if there are still delayed prompt search tasks. If there are still delayed prompt search tasks, proceed to step S17. S17: Take the end time of the last scheduled hint search as the initial scheduling time, and the end time of the current scheduling interval as the scheduling end time. Schedule delayed hint search tasks within this time interval according to step S13. S18: Compare the end time of the last scheduled hint search with the initial scheduling time, and the end time of the current scheduling interval with the scheduling end time. Then, schedule delayed hint search tasks within this time interval according to step S13. S19: Increase the priority of lookout requests in the delayed task queue by a factor equal to the difference in task priorities.

7. The method as described in claim 1 or 2, characterized in that, In step S1, the first A view request is represented as In this context, TP represents the mission type, ST represents the arrival time, and for stereo search missions, ST is determined by the preset beam position arrangement and beam scanning order; for cue search missions, ST is determined by the detection value of the previous scheduling cycle and the initial time of the current scheduling; for target tracking missions, ST is the sum of the track update time and the revisit time interval, and PRTY represents the mission priority, with target confirmation, target update, cue search, and stereo search having priorities from highest to lowest. The line-of-sight angle is composed of azimuth and elevation; PRF represents the pulse repetition frequency. The dwell time is indicated by PAP, the power aperture product is indicated by BAW, and the azimuth and elevation beamwidths are indicated by BAW and BEW, respectively.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1-7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-7.

Images