A minimum startup delay equalization method for radar systems

Abstract

The present invention proposes a minimum startup delay equalization method for a radar system, which relates to the technical field of radar delay optimization. The method uses a pulse compression and signal cross-correlation method to estimate the radar system delay parameters. Based on the obtained radar system delay parameters, a delay switching control module is introduced into the radar system to perform delay compensation on signals of different paths, so that the arrival time of signals on each path tends to be consistent. Based on the priority and delay requirements of the startup subtask, computing power resources are dynamically distributed to each subtask to ensure that resources can be allocated to where they are most needed, thereby improving system performance.

Description

Technical Field

[0001] The present invention relates to the technical field of radar delay optimization, and in particular to a minimum startup delay equalization method for a radar system. Background Art

[0002] Continuous advances in electronics and signal processing technologies have significantly improved radar system performance, but startup delay remains a key factor impacting overall performance. To fully realize the performance potential of new radar hardware, such as high-resolution sensors and high-speed signal processing chips, it is necessary to research methods for minimizing startup delay equalization, enabling the system to quickly enter a stable operating state and efficiently utilize hardware resources.

[0003] Low-power design requirements: Battery-powered radar devices, such as portable radars or radar nodes in the Internet of Things, often employ a sleep and wakeup mechanism to extend battery life. This makes the radar's startup delay during wakeup a critical issue. Minimum startup delay balancing helps ensure fast response times while maintaining low power consumption.

[0004] The current technical problems are as follows:

[0005] 1. Hardware: Radar systems typically consist of multiple components, such as transmitters, receivers, and signal processors, making clock synchronization of these components challenging. Clock deviations can cause delays in data transmission and processing, impacting startup speed and signal processing accuracy, making it difficult to achieve a balanced minimum startup delay. In complex radar circuits, signal transmission between different modules introduces delays, such as propagation delays in circuit components like filters, amplifiers, and mixers. The accumulation of these delays increases the overall startup time of the system. Furthermore, due to process differences, different batches of hardware may exhibit varying delay performance, posing challenges to achieving a balanced minimum startup delay.

[0006] 2. Software: Radar system software requires initialization and configuration of numerous hardware registers and parameters, making the initialization process cumbersome and time-consuming. Furthermore, the order in which different tasks and functional modules are initialized affects startup time, making optimization challenging. Signal processing algorithms, such as target detection and tracking, are computationally intensive, requiring extensive parameter calculations and model loading during startup, leading to startup delays. Existing algorithms may struggle to balance real-time performance with resource utilization, making it difficult to quickly complete initialization and enter operational mode.

[0007] 3. System Integration: Modern radar systems often need to handle multiple tasks simultaneously, such as target detection, identification, and tracking. Improper prioritization and scheduling strategies between different tasks can lead to resource contention, increasing startup delays for certain critical tasks and impacting overall system performance. Radar systems often require data exchange and collaboration with external systems, such as command and control systems and other sensors. The response speed and data transmission latency of external systems can hinder the startup speed of the radar system, making it difficult to achieve a minimum startup delay independent of external factors. Summary of the Invention

[0008] In order to solve the above technical problems, the present invention proposes a minimum startup delay equalization method for a radar system, comprising the following steps:

[0009] S1. Estimate radar system delay parameters using a method based on pulse compression and signal cross-correlation;

[0010] S2. Based on the radar system delay parameters estimated in step S1, a delay switching control module is introduced into the radar system to perform delay compensation on signals of different paths so that the arrival times of the signals of each path tend to be consistent;

[0011] S3. Dynamically distribute computing resources to each subtask based on the priority and latency requirements of the subtask.

[0012] In a preferred embodiment, step S3 includes:

[0013] S31. Construct a matrix topology model based on the radar system's hardware and software configuration items, evaluate the signal communication of commands, data, and status between configuration items, and predict startup conflicts among configuration item nodes.

[0014] S32, with the goal of maximizing time resource savings, constructing a time optimization strategy based on the element characteristics in the matrix topology model, and dividing the total task cycle into multiple subtasks executed in parallel;

[0015] S33. In each parallel subtask, traverse and find all unrelated nodes to form an unrelated group, distribute communication and allocate computing resources to the nodes in the group, and calculate the time saved after the division;

[0016] S34. Results of subtask division and time saved after division , complete time resource planning and realize dynamic allocation and optimization of computing resources.

[0017] In a preferred embodiment, in step S31, an M×M matrix topology model A is constructed to describe the signal transmission and reception relationship between radar configuration item nodes. The element a in the matrix topology model is mnThe duration of the signal transmission or reception of the configuration item node in row m and column n is a. mn :

[0018] When m=n, it means that node m is transmitting a signal, and a mm Equal to the maximum transmission or reception duration of each sub-array of the radar configuration item node;

[0019] When m≠n, it represents the reception time of the signal transmitted by configuration item node n to configuration item node m;

[0020] In the matrix topology model A, the Y axis represents the time growth change. When the fixed time length a is divided into m configuration item nodes in sequence on the time axis, mm , to ensure that there is no conflict response between the configuration item nodes, if a mm =0, indicating that no communication time needs to be allocated.

[0021] In a preferred embodiment, in step S32, the total task cycle is divided into a plurality of sub-intervals corresponding to the subtasks, and unrelated nodes are executed in parallel in each sub-interval. All nodes in the matrix topology model are sent according to the signal transmission time t c Sort from large to small and get the node sequence S={s1,s2,...,s i ,...,s N},

[0022] Dynamically divide the subtask interval into decreasing steps, gradually exploring larger partition windows; select the first N nodes from the current unassigned nodes h nodes, check whether they meet the irrelevance condition, if they meet the irrelevance condition, calculate the time saved after partitioning ΔT h

[0023]

[0024] Execute the N h When there are multiple nodes, the total time is determined by the longest communication time;

[0025] like , record the partition position h, and use this N h nodes are divided into the current subtask interval;

[0026] like , reduce the step size N h and recheck;

[0027] Let the threshold be K, when the maximum communication time of the remaining unassigned nodes is When it is less than the threshold K, the division is stopped.

[0028] In a preferred embodiment, in step S34, the total communication time resource T t Expressed as:

[0029]

[0030] Among them, c h is the actual communication time of the hth subtask, T p For protection time, L is the number of subintervals into which the total task cycle is divided;

[0031] The total time resource T0 after optimization is:

[0032] ;

[0033] in, Time resources saved for optimization.

[0034] In a preferred embodiment, step S2 includes the following steps:

[0035] S21, separating the reference signal and the target signal, and balancing the path difference between the reference signal and the target signal;

[0036] S22. Design a software switching module based on a matrix topology model;

[0037] S23. Design a dynamic balance selection mechanism to select the signal transmission path by controlling the switch;

[0038] S24. Integrate the software exchange module based on the matrix topology model into the radar system and perform system integration and debugging.

[0039] In a preferred embodiment, step S1 includes the following steps:

[0040] S11, obtaining a received signal from the radar and performing signal preprocessing;

[0041] S12, performing pulse compression and signal matching processing on the preprocessed signal;

[0042] S13. Calculate the correlation function using the processed signal and estimate the delay parameter.

[0043] In a preferred embodiment, in each subtask, all unrelated nodes are traversed to form an unrelated group, and computing time slices are allocated for concurrent communication of nodes in the group to optimize the communication time of the subtask. The specific steps are as follows:

[0044] (1) In the first subtask, starting from the first configuration item node, according to the irrelevant judgment condition a 1n =a n1 =0 (n>1) traverse to find all irrelevant nodes in sequence and form the initial node group in the order of nodes;

[0045] (2) Starting with the second configuration item node in the initial node group, the node is judged to be irrelevant with the subsequent nodes in the group in turn. If they are all relevant, the second configuration item node communicates concurrently with the first configuration item node, marks the node as merged, and deletes the other nodes in the group, and considers them not merged. If there are irrelevant nodes, the new irrelevant node is used to continue to judge irrelevant with the remaining nodes in the group until the remaining nodes in the group are irrelevant to each other. Then all nodes in the group can communicate concurrently, and other nodes that do not meet the rules are deleted and considered not merged.

[0046] (3) Select the next unmerged node m, according to the irrelevant judgment condition a mn =a nm =0, n>m, traverse and obtain all irrelevant nodes in turn to form an initial node group m, repeat step (2) to complete irrelevant judgment and form a new irrelevant group;

[0047] (4) Repeat step (3) to complete the node traversal, form all unrelated groups, and obtain the optimal time saving for the subtask and the optimized matrix topology model.

[0048] Compared with the prior art, the present invention has the following beneficial technical effects:

[0049] 1. Introducing a delay line to compensate for the delay of signals along different paths, aligning the arrival times of signals along each path, effectively avoids phase errors and waveform distortion caused by time differences between signals, thereby improving the radar system's processing accuracy of target echo signals and enhancing the accuracy of target detection, positioning, and tracking. A delay-related parameter estimation algorithm based on median filtering improves the estimation accuracy of delay parameters, facilitates more precise signal processing and analysis, further enhances the accuracy and reliability of signal processing, and provides a more accurate data foundation for subsequent tasks such as target recognition.

[0050] 2. Dynamically allocating communication time slices based on the priority and latency requirements of communication tasks ensures that critical tasks receive priority access to communication resources, reducing communication latency and thus accelerating the overall response speed of the radar system. This allows the radar to quickly start up and enter operational mode, enabling target detection and tracking. The introduction of delay lines for signal delay compensation and precise delay parameter estimation help reduce delays in signal processing, enabling the radar system to react more quickly to received signals and improving the system's real-time performance.

[0051] 3. The consistency of signal arrival times on each path and accurate delay parameter estimation make the radar system more stable during signal processing, reducing system misjudgments and missed judgments caused by inconsistent signals or inaccurate parameter estimation, improving system reliability, optimizing the use of communication resources, ensuring smooth and efficient communication, reducing system failures caused by communication congestion or unreasonable resource allocation, enhancing the stability of radar network communications, and thus improving the reliability of the entire radar system.

[0052] 4. Dynamic allocation of communication time slices can reasonably allocate communication resources according to actual task requirements, avoid waste and excessive occupation of resources, and improve the efficiency of communication resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0054] Figure 1 The figure is a flow chart of the minimum startup delay equalization method for a radar system according to the present invention.

[0055] Figure 2 This is a flow chart of the median filtering method for estimating radar system delay parameters according to the present invention.

[0056] Figure 3 This is a flow chart of delay compensation for signals on different paths according to the present invention.

[0057] Figure 4 Schematic diagram of delay simulation after the delay line component is integrated into the radar system.

[0058] Figure 5 The present invention dynamically distributes communication time resources to each subtask.

[0059] Figure 6 Transmits a beam pattern for signals between nodes of different configuration items.

[0060] Figure 7 This is the functional module architecture diagram of the radar system. DETAILED DESCRIPTION

[0061] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0062] In the drawings of the specific embodiments of the present invention, in order to better and more clearly describe the working principles of the various components in the system, the connection relationship of the various parts in the device is shown, which only clearly distinguishes the relative position relationship between the various components, and does not constitute a limitation on the signal transmission direction, connection sequence and size, dimension and shape of the components or structures.

[0063] Example 1

[0064] like Figure 1 FIG. 1 is a flow chart of a minimum startup delay equalization method for a radar system according to the present invention. The minimum startup delay equalization method includes the following steps:

[0065] S1. The radar system delay parameters are estimated using a method based on pulse compression and signal cross-correlation.

[0066] In order to precisely control and adjust the system and achieve equalization of startup delay, it is necessary to estimate the delay-related parameters in the current radar system, such as delay time, delay step size, etc. The delay-related parameter estimation algorithm based on median filtering performs noise reduction processing on the radar signal through median filtering to improve the estimation accuracy of delay-related parameters, thereby facilitating the optimization of the startup delay of the radar system.

[0067] The selection of the median filter window width plays an important role in estimating radar performance and calculating system complexity. Through theoretical analysis and simulation examples, the appropriate filter window width can be determined to ensure the estimation accuracy while reducing the calculation complexity, such as Figure 2 As shown, specifically including:

[0068] S11. Obtain the radar receiving signal and perform signal preprocessing.

[0069] First, the received signal of the intermediate frequency radar is obtained and preprocessed, such as removing the DC component and normalizing it, to improve the signal quality and the accuracy of subsequent processing.

[0070] S12. Perform pulse compression and signal matching processing on the preprocessed signal.

[0071] In signal processing, before performing pulse compression and signal matching, the preprocessed signal must first be further de-noised using median filtering. Median filtering effectively removes noise by dividing the signal data into multiple data groups based on a specific window width and taking the median of each data group as the filtered output.

[0072] It's important to note that the choice of window width is crucial: for signals with high noise and large fluctuations, a wider window is needed to cover more data points and smooth the signal. For signals rich in detail, a narrower window is more appropriate to avoid oversmoothing and loss of detail. Ideally, based on experience or experimentation, observe how the signal changes with different window widths and choose a width that improves the signal-to-noise ratio without losing critical information.

[0073] To perform pulse compression, a matched filter must be designed based on the characteristics of the transmitted signal. If the transmitted signal is a linear frequency modulated signal, the impulse response of the matched filter should be conjugate to that of the transmitted signal and exhibit time-reversal characteristics. The goal is to compress the energy dispersed over time into a narrow pulse at the receiving end.

[0074] The median-filtered signal is convolved with a matched filter. Through convolution, the signal and the filter gradually become correlated in time. Each sample in the signal is multiplied and accumulated with the corresponding sample in the filter, so that the scattered energy is superimposed at a specific moment, forming a high-amplitude narrow pulse, thus completing pulse compression.

[0075] Signal matching requires a reference signal, which can be the original transmitted signal or a known desired signal signature. The choice of reference signal depends on the purpose and scenario of signal matching. A correlation operation is performed between the pulse-compressed signal and the reference signal. The correlation between the signal and the reference signal at different time delays is calculated to determine the degree of similarity between the two. As the time delay changes, the correlation value changes. The delay at which the correlation value is maximized indicates the matching position between the signal and the reference signal, thereby achieving signal matching.

[0076] Through the above processing, not only can the noise in the signal be removed and the signal-to-noise ratio be improved, but the arrival time and other relevant parameters of the signal can also be determined more accurately in the subsequent delay-related parameter estimation.

[0077] S13. Calculate the correlation function using the processed signal and estimate the delay parameter.

[0078] The processed signal is used to calculate the correlation function. The correlation function is an important basis for estimating the delay-related parameters. By calculating the correlation of the signal at different time delays, a preliminary estimate of the delay-related parameters can be obtained.

[0079] Preferably, the correlation function is a function used to describe the correlation between signals at different time points, and is used to measure the change in the similarity between two signals with time delay. Preferably, an autocorrelation function or a cross-correlation function is used.

[0080] According to the calculated correlation function, a delay-related parameter estimation algorithm based on median filtering is used to accurately estimate delay-related parameters, including delay time and delay step size. This algorithm improves the estimation accuracy of delay parameters by optimizing the shape of the correlation function.

[0081] Specifically, there are two discrete signals x[v] and y[v], and the cross-correlation function R xy [M] is defined as:

[0082]

[0083] M is the time delay, v is a discrete time variable, and this formula is used to measure how the similarity between two different signals x[v] and y[v] changes with the time delay M.

[0084] After calculating the correlation function, find the peak position of the function. Assuming that the correlation function reaches its peak at M=M0, the time corresponding to M0 is the initial estimated delay time T C , that is, T C = M0T Q , where T Q is the sampling period.

[0085] In order to improve the estimation accuracy, median filtering is used to process the correlation function.

[0086] Let R[M] be the original correlation function, is the correlation function after median filtering. Median filtering is to take the median of the data in a window centered at M in R[M] as Then in Find the peak position , then the optimized delay time estimate is .

[0087] Delay step It can be estimated by analyzing the periodicity or variation of the correlation function. If the correlation function has obvious periodicity, let its period be W, then the delay step can be preliminarily estimated as .

[0088] After median filtering, Perform the same cycle analysis to find a more accurate cycle , thus obtaining the optimized delay step estimate .

[0089] In this way, the shape of the correlation function is optimized by using median filtering, thereby improving the estimation accuracy of the delay-related parameters.

[0090] In the preferred embodiment, performance evaluation and optimization are conducted. System performance, such as the weighted mean square error of wind speed estimation, is evaluated through theoretical analysis, simulation examples, and measured data. Based on the evaluation results, the filter window width and algorithm parameters are further optimized to improve overall system performance and balance startup delays.

[0091] S2. Based on the radar system delay parameters estimated in step S1, a delay switching control module is introduced into the radar system to perform delay compensation on signals of different paths so that the arrival times of the signals of each path tend to be consistent.

[0092] In the delay component, delays of varying magnitudes are achieved by designing a combination of delay line branches and equal-loss through-links. The delay line branches introduce a specific delay time, while the equal-loss through-links ensure that the signal's transmission loss before and after the delay line is the same, thus ensuring balanced signal processing.

[0093] By controlling the switch through the driving circuit, it is possible to select whether the signal passes through the delay line branch or the equal loss direct branch. This selection mechanism enables the system to flexibly adjust the signal delay time as needed, thereby achieving delay compensation for signals in different paths, such as Figure 3 As shown, the specific steps include:

[0094] S21. Separate the reference signal and the target signal, and balance the path difference between the reference signal and the target signal.

[0095] In the signal transmission chain, the precise location of the signal source is determined by analyzing the system architecture and signal flow. Simultaneously, the measurement requirements and equipment layout are combined to identify a location that effectively separates the reference and target signals. Based on the signal type, power, and transmission characteristics, a coupler is selected to ensure even power distribution during signal separation. Preferably, a directional coupler can be used in the RF circuit. Appropriate selection ensures effective separation of the reference and target signals.

[0096] By using adjustable delay components, the delay of the reference signal is precisely adjusted by changing the control parameters of the electrical signal, and the transmission time of the reference signal is adjusted to minimize the delay difference between the reference signal and the target signal, ensuring that the signals of the two paths are synchronized in time, effectively balancing the path differences, and providing reliable protection for subsequent signal analysis and processing.

[0097] S22. Design a delayed switching control module.

[0098] Design a delay switching control module to ensure that the delay line branch and the equal-loss through-branch have consistent signal transmission losses. The delay time of the delay line branch should be precisely calculated and designed based on system requirements to achieve the desired delay step.

[0099] Specifically, to ensure consistent signal transmission losses between the delay line branch and the equal-loss through-branch, a detailed analysis of the signal transmission losses is required. When constructing the equal-loss through-branch, components equivalent to the key components in the delay line branch are selected and their parameters are appropriately configured.

[0100] It should be noted that while the losses of the two branches are designed to be consistent, slight differences may still exist in practice. To address this, adjustable attenuators are added to both branches. During the system debugging phase, a vector network analyzer is used to accurately measure the losses of the two branches. Based on the measurement results, the attenuation of the attenuators is adjusted to achieve highly consistent losses in both branches.

[0101] The delay time required for the delay line branch is accurately calculated based on the target detection range and range resolution. In pulse radar, to detect a target at a specific distance, the delay time is calculated based on the radar equation and the propagation speed of electromagnetic waves. Assume that the radar needs to detect a target at a distance of R, and the delay time t = 2R / c, where c is the speed of light.

[0102] Select a digital delay line to achieve the desired delay step. Digital delay lines achieve discrete, high-precision delay steps through precise digital control. Preferably, a multi-stage delay line cascade can be used in the design, combining delay lines with different delay amounts to achieve flexible and precise delay steps. Preferably, a high-precision clock source is used to synchronize the delay line control circuitry to ensure delay time stability and accuracy.

[0103] S23. Design a dynamic balance selection mechanism to select the signal transmission path by controlling the switch.

[0104] A dynamic balance selection mechanism is designed to select the signal transmission path by controlling the switch. The switch selection should be based on the principles of fast response and low insertion loss to ensure efficient signal transmission and accurate delay compensation.

[0105] Preferably, multiple signal sources, transmission paths and a switch are created to simulate the signal transmission process. In each cycle, a signal source is selected to generate a signal, a transmission path is dynamically selected, and the signal is transmitted to the path.

[0106] Specifically, the switch selection should follow the principles of fast response and low insertion loss. insertion Refers to the power loss when the signal passes through the switch, which can be expressed as:

[0107]

[0108] Among them, P input is the input signal power, P output is the output signal power. The response time is the time it takes for the switch to complete the switching action from receiving the control signal.

[0109] In a preferred embodiment, PIN diode switches and field-effect transistor switches can be selected. PIN diode switches have low insertion loss and high isolation at high frequencies, while field-effect transistor switches have fast switching response speeds and are suitable for scenarios requiring fast switching.

[0110] Assume there are n' signal sources and m' transmission paths. In each cycle, a round-robin algorithm is used to ensure load balancing.

[0111] In the polling mode, a counter is set. After each cycle is completed, k1 = (k1 + 1) mod n', and the k1th signal source is selected to generate the signal.

[0112] For the transmission path, a counter is also set. After each cycle, h = (h + 1) mod m', and the hth transmission path is selected.

[0113] In a preferred embodiment, a simulation system is constructed, comprising multiple signal sources, transmission paths, and a switch. During each simulation cycle, the signal source and transmission path are selected according to the aforementioned method, and various parameters during signal transmission, such as signal strength and delay, are monitored. By analyzing the simulation data, it is evaluated whether the selected switch and dynamic selection mechanism meet the requirements of fast response and low insertion loss, and whether the delay compensation is accurate. Based on the evaluation results, the switch selection or dynamic selection algorithm is optimized and adjusted.

[0114] S24. Integrate the delay switching control module into the radar system and perform system integration and debugging.

[0115] During signal transmission, the driver circuit controls the switch, selecting the appropriate delay line branch or equal-loss through branch to compensate for signal delays along different paths. This approach aligns the arrival times of signals along each path, achieving equalized startup delays.

[0116] Integrate the delay line component into the radar system and conduct overall debugging. Through testing and optimization, ensure that the delay line component works in coordination with other system components to achieve the expected delay equalization effect, such as Figure 4 Figure 2 shows a schematic diagram of delay simulation after the delay line component is integrated into the radar system.

[0117] Through this step S2, the startup delay can be effectively balanced in the radar system, thereby improving the overall performance of the system and the signal processing efficiency.

[0118] S3. Dynamically distribute computing resources to each subtask based on the priority and latency requirements of the subtask.

[0119] In order to rationally allocate communication resources, computing resources, and storage resources in the radar system, optimize resource utilization efficiency, reduce resource competition and waiting time, and achieve balanced startup delays, this step dynamically allocates computing time slices to each subtask based on the priority and latency requirements of the communication task, optimizes the use of communication resources, and reduces startup delays.

[0120] Preferably, by establishing a matrix topology model based on radar time-division communication, the prediction and avoidance of configuration item node response conflicts are achieved, and a time optimization strategy is constructed with the goal of maximizing time resource savings to obtain an optimized time resource scheduling solution.

[0121] In the matrix topology model, the communication transmission time and dependency between configuration item nodes constitute the core constraints of time resource allocation. The goal is to maximize the time resource savings (i.e., minimize the total time resource T) by dynamically dividing the subtask intervals while ensuring the task execution logic. t ),like Figure 5 The specific steps are as follows:

[0122] S31. Construct a matrix topology model based on the hardware and software configuration items of the radar system, evaluate the signal communication of commands, data, and status between configuration items, and predict startup conflicts of configuration item nodes.

[0123] The configuration item nodes of the matrix topology model are used to represent subtasks, and the edges of the matrix topology model represent the communication dependencies of subtasks. Unrelated configuration item nodes, that is, configuration item nodes without direct communication dependencies, can be executed in parallel. The communication transmission time represents the time resource T for triggering communication after the configuration item node completes the calculation. t .

[0124] Construct an M×M matrix topology model A to describe the signal transmission and reception relationship between radar configuration item nodes. Figure 6 The following figure shows the signal transmission beam patterns between different configuration item nodes (corresponding to different colors in the figure). Based on the transmission and reception beam patterns, the signal transmission or reception duration of the item node is calculated.

[0125] Element a in the matrix topology model mn The duration of the signal transmission or reception of the configuration item node in row m and column n is a. mn :

[0126] When m=n, it means that the configuration item node m is transmitting a signal, and a mm Equal to the maximum transmission or reception duration of each sub-array of the radar configuration item node;

[0127] When m≠n, it represents the reception time of the signal transmitted by configuration item node n to configuration item node m;

[0128] In the matrix topology model A, the Y axis represents the time growth change. When the fixed time length a is divided into m configuration item nodes in sequence on the time axis, mm , which ensures that there is no conflict response between configuration item nodes. mm =0, indicating that no communication time needs to be allocated.

[0129] The matrix topology model avoids communication conflicts and optimizes the efficiency and accuracy of radar time-division communication by matrix-describing and time-dividing the transmission and reception durations of configuration item nodes.

[0130] S32, with the goal of maximizing time resource savings, constructing a time optimization strategy based on the element characteristics in the matrix topology model, and dividing the total task cycle into multiple subtasks executed in parallel;

[0131] Divide the total task cycle into multiple subintervals L corresponding to subtasks. In each subinterval, execute unrelated nodes in parallel as much as possible to reduce serial waiting time. The specific steps are as follows:

[0132] For all configuration item nodes in the matrix topology model, the signal transmission time t c Sort from large to small and get the configuration item node sequence S={s1,s2,...,s i ,...,s N}, give priority to processing configuration item nodes with large communication resources to reduce their blocking effect on subsequent tasks.

[0133] Dynamically divide the subtask interval into decreasing steps: the initial step size is the maximum division granularity (such as a single node), and gradually try larger division windows.

[0134] Division rule: Select the first N configuration item nodes from the currently unassigned ones h Check whether the configuration item node meets the irrelevance condition (no communication dependency or communication dependency has been assigned).

[0135] If the irrelevance condition is met, calculate the time saved after partitioning ΔT h

[0136]

[0137] Execute the N h When there are multiple configuration item nodes, the total time is determined by the longest communication time;

[0138] like , record the partition position h, and use this N h Configuration item nodes are assigned to the current subtask interval.

[0139] like , reduce N h Step length and recheck.

[0140] Set the threshold to K, when the maximum communication time of the remaining unassigned configuration item nodes When it is less than the threshold K, the partitioning stops. If the communication resources of the remaining configuration item nodes are already small enough, the benefit of continuing the partitioning may be lower than the additional resources for managing parallel tasks.

[0141] In a preferred embodiment, the step size is adaptively adjusted according to historical partitioning results to avoid frequent trials.

[0142] In a preferred embodiment, independent subgraphs are identified in advance through topological sorting, reducing runtime checking resources.

[0143] In a preferred embodiment, a weight of time saving and resource consumption is introduced to avoid over-partitioning.

[0144] S33. In each parallel subtask, traverse and search for all unrelated configuration item nodes to form unrelated groups, allocate computing resources to the nodes in the group, communicate concurrently, and calculate the time saved after the division.

[0145] In each subtask, all unrelated configuration item nodes are traversed to form an unrelated group. The computing power time slices are allocated to the configuration item nodes in the group for concurrent communication to optimize the communication time of the subtask. The specific steps are as follows:

[0146] (1) In the first subtask, starting from the first configuration item node, according to the irrelevant judgment condition a 1n =a n1 =0 (n>1) traverse to find all irrelevant configuration item nodes in sequence, and form an initial node group according to the order of configuration item nodes.

[0147] (2) Starting with the second configuration item node in the initial node group, the irrelevant judgment is performed with the subsequent points in the group in turn. If they are all relevant, the second configuration item node communicates concurrently with the first configuration item node, and marks the configuration item node as merged, and deletes the other configuration item nodes in the group and considers them not merged; if there are irrelevant configuration item nodes, the new irrelevant configuration item node continues to perform irrelevant judgment with the remaining configuration item nodes in the group until the remaining configuration item nodes in the group are irrelevant to each other. Then all configuration item nodes in the group can communicate concurrently, and other configuration item nodes that do not meet the rules are deleted and considered not merged. The initial node group becomes an irrelevant group.

[0148] (3) Select the next unmerged configuration item node m, and determine the irrelevant condition a mn =a nm =0 (n>m) traverse and obtain all irrelevant configuration item nodes in sequence to form an initial node group m, repeat step (2) to complete irrelevant judgment and form a new irrelevant group.

[0149] (4) Repeat step (3) to complete the traversal of the configuration item nodes and form all unrelated groups to obtain the optimal subtask and the optimized matrix topology model.

[0150] S34. Results of subtask division and time saved after division , complete time resource planning and realize dynamic allocation and optimization of computing resources.

[0151] Total communication time resources T t Expressed as:

[0152]

[0153] Among them, c h is the actual communication time of the hth subtask, T p To protect the time, L is the number of subintervals into which the total task cycle is divided.

[0154] The total time resource T0 after optimization is:

[0155]

[0156] in, Time resources saved for optimization.

[0157] Through the above steps and formulas, the construction of communication matrix topology model, the construction of time optimization strategy, subtask division, Optimization and time resource planning, thereby effectively saving time resources and realizing dynamic allocation and optimization of communication resources.

[0158] This step S3 optimizes the execution efficiency of tasks such as radar signal processing and data processing by rationally allocating computing resources, such as CPUs and GPUs, and reduces the waiting time and execution time of computing tasks. Computing resources can be dynamically allocated based on task priority and computing requirements, improving resource utilization.

[0159] Example 2

[0160] At the end of Example 1, after the dynamic allocation and optimization of computing resources are realized, the radar system can flexibly form a network, adjust the topology and optimize resource utilization without centralized management through self-organizing network and edge computing technology, thereby reducing startup delays.

[0161] The various subsystems or modules of the radar system are integrated and optimized to reduce the interface delay and coordination time between systems and achieve balanced overall startup delay.

[0162] Ad hoc networking technology: By designing an ad hoc networking protocol, the radar system can flexibly form networks, adjust topology, and optimize resource usage without centralized management. Based on a distributed algorithm, the ad hoc networking protocol enables local optimization and cluster management, allowing devices to dynamically adjust network topology, reducing latency and improving system robustness.

[0163] Edge computing technology shifts computing tasks from the cloud or central servers to edge devices on the network, reducing latency and improving real-time computing capabilities. By deploying computing modules on end devices, base stations, or other edge nodes, data processing and decision-making are decentralized to the edge of the network, reducing the burden on central servers.

[0164] Hardware and software collaborative design: Through the design of multifunctional antenna arrays and flexible T / R modules, efficient signal processing units, and dynamic resource scheduling and optimization, efficient utilization of hardware resources and flexible software control are achieved. Preferably, an antenna system that supports orthogonal waveforms and phased array beams, as well as a system with intelligent resource allocation capabilities, can intelligently allocate computing power between different subtasks.

[0165] like Figure 7 The following is a diagram of the functional module architecture of the radar system. The functions of each module are as follows:

[0166] Transmitter / Receiver Modules: These come in multiple groups (labeled 1, 2, ..., N, T) and are the fundamental components of the radar system. They amplify and transmit signals from the main control module, receive weak signals reflected from targets, and perform preprocessing such as low-noise amplification to provide signals of appropriate strength for subsequent signal processing.

[0167] Array control module: responsible for managing and controlling the working status of each transmitting / receiving module in the antenna array, such as controlling the transmission power, phase and other parameters to realize the scanning and pointing control of the antenna beam.

[0168] Main control module: It is the control core of the entire radar system, coordinating and managing the operation of each sub-module, sending control instructions, receiving processed data, and scheduling tasks according to system requirements.

[0169] Receiver processor: Performs digital beamforming processing on the received multi-channel signals. By performing weighted summation and other operations on the signals of each channel, it enhances the signal in the desired direction, suppresses interference in other directions, and forms a receiving beam with a specific direction.

[0170] Signal processor: further processes the processed signal, including pulse compression and Doppler processing, to extract information such as the target's distance and speed.

[0171] Data processor: Analyzes and solves the signal-processed data to complete target detection, tracking, identification and other functions, and converts the original signal data into meaningful target information.

[0172] Control terminal: On the one hand, it presents the target information processed by the data processor, such as target position and speed, to the operator in an intuitive manner (such as radar screen display); on the other hand, the operator can send control instructions to the main control module through this terminal to realize the operation and parameter setting of the radar system.

[0173] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted via the computer-readable storage medium. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrations. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0174] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present application, and such modifications or substitutions should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A minimum startup delay equalization method for a radar system, characterized in that: The steps include: S1. Estimate radar system delay parameters using a method based on pulse compression and signal cross-correlation; S2. Based on the radar system delay parameters estimated in step S1, a delay switching control module is introduced into the radar system to perform delay compensation on signals of different paths so that the arrival times of the signals of each path tend to be consistent; S3: Dynamically distribute computing resources to each subtask based on the priority and latency requirements of the subtask, including: S31. Construct a matrix topology model based on the hardware and software configuration items of the radar system, evaluate the signal communication of commands, data, and status between configuration items, and predict startup conflicts of configuration item nodes; S32, with the goal of maximizing time resource savings, constructing a time optimization strategy based on the element characteristics in the matrix topology model, and dividing the total task cycle into multiple subtasks executed in parallel; S33. In each parallel subtask, traverse and find all unrelated configuration item nodes to form an unrelated group, distribute computing power resources to the configuration item nodes in the group, and calculate the time saved after the division; S34. Results of subtask division and time saved after division , complete time resource planning and realize dynamic allocation and optimization of computing resources.

2. The minimum startup delay equalization method for a radar system according to claim 1, characterized in that: In step S31, an M×M matrix topology model A is constructed to describe the signal transmission and reception relationship between radar configuration item nodes. The element a in the matrix topology model is mn Represents the duration a of the node signal transmission or reception for the configuration item in row m and column n mn ; When m=n, it means that the configuration item node m is transmitting a signal, and a mm Equal to the maximum transmission or reception duration of each subarray of the configuration item node; when m≠n, it represents the reception duration of the signal transmitted by configuration item node n to configuration item node m; In the matrix topology model A, when the time axis is divided into m configuration item nodes, the time length a mm , to ensure that there is no conflict response between the configuration item nodes, if a mm =0, indicating that no time allocation is required.

3. The minimum startup delay equalization method for a radar system according to claim 2, characterized in that: In step S32, the total task cycle is divided into a plurality of sub-intervals corresponding to the subtasks, and unrelated configuration item nodes are executed in parallel in each sub-interval. All configuration item nodes in the matrix topology model are sent according to the signal transmission time t c Sort from large to small and get the configuration item node sequence S={s1,s2,...,s i ,...,s N }, N is the total number of configuration item nodes, i is the sequence number of the configuration item node; dynamically divide the subtask interval into descending steps, gradually try a larger partition window; select the first N from the current unassigned nodes h nodes, check whether they meet the irrelevance condition, if they meet the irrelevance condition, calculate the time saved after partitioning ΔT h : in, Representation node Communication time; parallel execution of the N h When there are multiple nodes, the total time is determined by the longest communication time; if , record the partition position h, and use this N h nodes are included in the current subtask interval; if , reduce N h The step size is rechecked; let the threshold be K, when the maximum communication time of the remaining unassigned configuration item nodes When it is less than the threshold K, the division is stopped.

4. The minimum startup delay equalization method for a radar system according to claim 3, characterized in that: In step S34, the total communication time resource T t Expressed as: Among them, c h is the actual communication time of the hth subtask, T p To protect time, L is the number of subintervals into which the total task cycle is divided; the total time resource T0 after optimization is: .

5. The minimum startup delay equalization method for a radar system according to claim 1, characterized in that: Step S2 includes the following steps: S21, separating the reference signal and the target signal, and balancing the path difference between the reference signal and the target signal; S22. Design a delayed switching control module; S23. Design a dynamic balance selection mechanism to select the signal transmission path by controlling the switch; S24. Integrate the delay switching control module into the radar system and perform radar system integration and debugging.

6. The minimum startup delay equalization method for a radar system according to claim 1, characterized in that: Step S1 includes the following steps: S11, obtaining a received signal from the radar and performing signal preprocessing; S12, performing pulse compression and signal matching processing on the preprocessed signal; S13. Calculate the correlation function using the processed signal and estimate the delay parameter.

7. The minimum startup delay equalization method for a radar system according to claim 1, characterized in that: In step S33, in each subtask, all unrelated configuration item nodes are traversed to form an unrelated group. The computing power time slices are allocated to the configuration item nodes in the group for concurrent communication to optimize the communication time of the subtask. The specific steps are as follows: (1) In the first subtask, starting from the first configuration item node, according to the irrelevant judgment condition a 1n =a n1 = 0, n>1, traverse to find all irrelevant configuration item nodes in sequence, and form an initial node group according to the order of configuration item nodes; (2) Starting with the second configuration item node in the initial node group, the node performs irrelevant judgment with subsequent nodes in the group in turn. If they are all relevant, the second configuration item node communicates concurrently with the first configuration item node, marks the configuration item node as merged, and deletes other configuration item nodes in the group and considers them as not merged. If there are irrelevant configuration item nodes, the new irrelevant configuration item node continues to perform irrelevant judgment with the remaining configuration item nodes in the group until the remaining configuration item nodes in the group are irrelevant to each other. Then all configuration item nodes in the group communicate concurrently, and other configuration item nodes that do not meet the rules are deleted and considered as not merged. (3) Select the next unmerged configuration item node m, and determine the irrelevant condition a mn =a nm =0, n>m, traverse and obtain all irrelevant configuration item nodes in sequence to form an initial node group m, repeat step (2) to complete irrelevant judgment and form a new irrelevant group; (4) Repeat step (3) to complete the configuration item node traversal, form all unrelated groups, and obtain the optimal time saving for subtasks and the optimized matrix topology model.

Images