Radar detection sign parameter method and device based on local calculation

Abstract

The invention relates to the technical field of radar detection, and discloses a radar sign parameter detection method and device based on local calculation, and the method comprises the steps: carrying out the compression and distillation operation of a basic calculation model through a cloud, obtaining a to-be-deployed model corresponding to the basic calculation model, and deploying the to-be-deployed model to a device end corresponding to a radar, obtaining a post-deployment model corresponding to the equipment end; and receiving a target parameter of the target object collected by the radar, and judging whether the target parameter meets a sign parameter calculation condition, and if so, calculating a target sign parameter of the target object through the deployed model and the target parameter. Visibly, the target physical sign parameters of the target object can be calculated through the post-deployment model deployed at the equipment end and the target parameters acquired by the radar, so that when the radar detects the physical sign parameters, the off-line use of the equipment end is realized, the real-time performance of data transmission and the convenience of data detection are further improved, and the user experience is improved. Therefore, the detection efficiency of the radar on the physical sign parameters is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of radar detection, and particularly to a method and device for detecting physical sign parameters by radar based on local computing. Background Art

[0002] With the rapid development of smart home and health monitoring technologies, the demand for dynamic monitoring of human physical signs in the home scenario is increasing day by day. Millimeter-wave radar technology, as a non-contact sensing technology, has advantages such as high precision, strong penetration, and privacy protection, and has gradually become a research hotspot in the field of home health monitoring.

[0003] However, existing millimeter-wave radar monitoring systems usually adopt a cloud computing architecture, that is, the device end transmits the collected radar signals to the cloud, and the cloud algorithm performs inference and analysis. Although this architecture can utilize the powerful computing power of the cloud, it also has the following drawbacks: 1. Dependence on cloud computing results in the device being unable to work properly without a network connection, restricting its application scenarios; 2. Transmitting data to the cloud requires additional communication costs, and network latency may affect the real-time nature of data transmission. These problems seriously restrict the further development and popularization of millimeter-wave radar monitoring systems. It can be seen that it is particularly important to provide a method that can improve the convenience and efficiency of detecting physical sign parameters of millimeter-wave radar monitoring systems. Summary of the Invention

[0004] The present invention provides a method and device for detecting physical sign parameters by radar based on local computing. When detecting physical sign parameters by radar, offline use of the device end is realized, thereby improving the real-time nature of data transmission and the convenience of data detection, and thus improving the detection efficiency of the radar for physical sign parameters.

[0005] To solve the above technical problems, in the first aspect of the present invention, a method for detecting physical sign parameters by radar based on local computing is disclosed, and the method includes:

[0006] Performing compression and distillation operations on a pre-trained basic calculation model through the cloud to obtain a to-be-deployed model corresponding to the basic calculation model;

[0007] Deploying the to-be-deployed model to the device end corresponding to the radar to obtain a deployed model corresponding to the device end;

[0008] Receiving target parameters of a target object collected by the radar, and determining whether the target parameters meet a preset physical sign parameter calculation condition. If so, calculating target physical sign parameters of the target object through the deployed model and the target parameters; the target parameters of the target object include type parameters, distance parameters, body movement parameters, and basic physical sign parameters of the target object.

[0009] As an alternative implementation, in the first aspect of the present invention, before performing compression and distillation operations on the pre-trained basic computing model through the cloud to obtain the deployable model corresponding to the basic computing model, the method further includes:

[0010] Predict the detection scenario parameters of the radar, and determine the computing requirement parameters of the device end corresponding to the radar according to the detection scenario parameters; the computing requirement parameters include computing volume requirement parameters and / or computing result requirement parameters;

[0011] Obtain the resource usage parameters of the device end, and determine the operating performance parameters of the device end according to the resource usage parameters; the operating performance parameters include response duration parameters and / or operating energy consumption parameters;

[0012] Determine the compression and distillation requirement parameters corresponding to the pre-trained basic computing model according to the computing requirement parameters and the operating performance parameters;

[0013] Among them, the operation of performing compression and distillation on the pre-trained basic computing model through the cloud to obtain the deployable model corresponding to the basic computing model includes:

[0014] Perform compression and distillation operations on the basic computing model through the cloud and the compression and distillation requirement parameters to obtain the deployable model corresponding to the basic computing model.

[0015] As an alternative implementation, in the first aspect of the present invention, before deploying the deployable model to the device end corresponding to the radar to obtain the deployed model corresponding to the device end, the method further includes:

[0016] Determine the model parameters of the deployable model; the model parameters include at least one of model size parameters, model computing volume parameters, model update method parameters, and framework compatibility requirement parameters;

[0017] Obtain the deployment environment parameters of the device end; the deployment environment parameters include deployment network parameters and / or deployment system parameters;

[0018] Determine the deployment parameters corresponding to the deployable model according to the model parameters, the deployment environment parameters, and the operating performance parameters; the deployment parameters include deployment location parameters and / or deployment time parameters;

[0019] Among them, the operation of deploying the deployable model to the device end corresponding to the radar to obtain the deployed model corresponding to the device end includes:

[0020] Deploy the model to be deployed to the device corresponding to the radar according to the deployment parameters, and obtain the deployed model corresponding to the device.

[0021] As an optional implementation manner, in the first aspect of the present invention, before determining the deployment parameters corresponding to the model to be deployed according to the model parameters, the deployment environment parameters, and the operating performance parameters, the method further includes:

[0022] Obtain the first radar parameters of the radar; the first radar parameters include at least one of a radar sampling rate parameter, a radar resolution parameter, and a radar data processing requirement parameter;

[0023] According to the first radar parameters, the deployment environment parameters, and the operating performance parameters, determine the deployment impact situation of the first radar parameters on the model to be deployed, and determine the deployment impact degree value corresponding to the deployment impact situation;

[0024] Judge whether the deployment impact degree value is greater than or equal to a preset deployment impact degree threshold;

[0025] When the judgment result is negative, trigger the operation of determining the deployment parameters corresponding to the model to be deployed according to the model parameters, the deployment environment parameters, and the operating performance parameters;

[0026] When the judgment result is positive, determine the deployment parameters corresponding to the model to be deployed according to the first radar parameters, the model parameters, the deployment environment parameters, and the operating performance parameters.

[0027] As an optional implementation manner, in the first aspect of the present invention, the target physical sign parameters at least include blood pressure parameters, and the body movement parameters include a body movement type parameter and a body movement amplitude parameter;

[0028] Among them, judging whether the target parameter meets the preset physical sign parameter calculation condition includes:

[0029] According to the type parameter of the target object, judge whether the type parameter is a preset type parameter;

[0030] When it is judged that the type parameter is the preset type parameter, judge whether the distance parameter is within a preset acquisition range according to the distance parameter of the target object;

[0031] When it is judged that the distance parameter is within the acquisition range, judge whether the body movement type parameter is a preset body movement type parameter according to the body movement type parameter of the target object;

[0032] When it is determined that the body movement type parameter is the preset body movement type parameter, according to the body movement amplitude parameter of the target object, determine whether the body movement amplitude parameter is within a preset body movement amplitude range;

[0033] When it is determined that the body movement amplitude parameter is within the body movement amplitude range, determine that the target parameter meets the preset physical sign parameter calculation condition.

[0034] As an optional implementation manner, in the first aspect of the present invention, before determining that the target parameter meets the preset physical sign parameter calculation condition, the method further includes:

[0035] Determine a first characteristic parameter and a second characteristic parameter corresponding to the basic physical sign parameter of the target object; the first characteristic parameter includes an acquisition duration parameter and / or an acquisition point number parameter, and the second characteristic parameter includes a waveform characteristic parameter;

[0036] Determine whether the first characteristic parameter is greater than or equal to a preset characteristic parameter threshold;

[0037] When it is determined that the first characteristic parameter is greater than or equal to the characteristic parameter threshold, according to the waveform characteristic parameter, determine the waveform integrity corresponding to the basic physical sign parameter, and determine whether the waveform integrity is greater than or equal to a preset waveform integrity threshold;

[0038] When it is determined that the waveform integrity is greater than or equal to the waveform integrity threshold, trigger the operation of determining that the target parameter meets the preset physical sign parameter calculation condition.

[0039] As an optional implementation manner, in the first aspect of the present invention, the characteristic parameter threshold includes an acquisition duration threshold and / or an acquisition point number threshold;

[0040] Wherein, the characteristic parameter threshold is determined by the following method:

[0041] Obtain a second radar parameter of the radar; the second radar parameter of the radar includes at least one of a radar sampling rate parameter, a radar resolution parameter, a radar data generation speed parameter, and a communication interface performance parameter of the radar;

[0042] According to the second radar parameter, determine the data transmission situation of the radar; the data transmission situation of the radar includes the data transmission speed situation and / or a transmission start frequency parameter of the radar;

[0043] According to the data transmission situation, determine the characteristic parameter threshold corresponding to the basic physical sign parameter of the target object.

[0044] The second aspect of the present invention discloses a radar detection physical sign parameter device based on local computing, and the device includes:

[0045] A compression and distillation module, configured to perform compression and distillation operations on a pre-trained basic calculation model through the cloud to obtain a to-be-deployed model corresponding to the basic calculation model;

[0046] A deployment module, configured to deploy the to-be-deployed model to the device side corresponding to the radar to obtain a deployed model corresponding to the device side;

[0047] A receiving module, configured to receive target parameters of a target object collected by the radar;

[0048] A judgment module, configured to judge whether the target parameters meet a preset physical sign parameter calculation condition;

[0049] A calculation module, configured to calculate target physical sign parameters of the target object through the deployed model and the target parameters when the judgment result of the judgment module is yes; the target parameters of the target object include type parameters, distance parameters, body movement parameters, and basic physical sign parameters of the target object.

[0050] As an optional implementation manner, in the second aspect of the present invention, the device further includes:

[0051] A prediction module, configured to predict detection scene parameters of the radar before the compression and distillation module performs compression and distillation operations on a pre-trained basic calculation model through the cloud to obtain a to-be-deployed model corresponding to the basic calculation model;

[0052] A determination module, configured to determine calculation requirement parameters corresponding to the device side of the radar according to the detection scene parameters; the calculation requirement parameters include calculation amount requirement parameters and / or calculation result requirement parameters;

[0053] An acquisition module, configured to acquire resource usage parameters of the device side;

[0054] The determination module is further configured to determine operation performance parameters of the device side according to the resource usage parameters; the operation performance parameters include response duration parameters and / or operation energy consumption parameters; and determine compression and distillation requirement parameters corresponding to the pre-trained basic calculation model according to the calculation requirement parameters and the operation performance parameters;

[0055] Wherein, the manner in which the compression and distillation module performs compression and distillation operations on a pre-trained basic calculation model through the cloud to obtain a to-be-deployed model corresponding to the basic calculation model specifically includes:

[0056] Perform compression and distillation operations on the basic calculation model through the cloud and the compression and distillation requirement parameters to obtain the model to be deployed corresponding to the basic calculation model.

[0057] As an alternative implementation, in the second aspect of the present invention, the determination module is further configured to:

[0058] Before the deployment module deploys the model to be deployed to the device side corresponding to the radar to obtain the deployed model corresponding to the device side, determine the model parameters of the model to be deployed; the model parameters include at least one of a model size parameter, a model calculation amount parameter, a model update method parameter, and a framework compatibility requirement parameter;

[0059] The acquisition module is further configured to acquire the deployment environment parameters of the device side; the deployment environment parameters include deployment network parameters and / or deployment system parameters;

[0060] The determination module is further configured to determine the deployment parameters corresponding to the model to be deployed according to the model parameters, the deployment environment parameters, and the operation performance parameters; the deployment parameters include a deployment location parameter and / or a deployment time parameter;

[0061] Among them, the manner in which the deployment module deploys the model to be deployed to the device side corresponding to the radar to obtain the deployed model corresponding to the device side specifically includes:

[0062] Deploy the model to be deployed to the device side corresponding to the radar according to the deployment parameters to obtain the deployed model corresponding to the device side.

[0063] As an alternative implementation, in the second aspect of the present invention, the acquisition module is further configured to:

[0064] Before the determination module determines the deployment parameters corresponding to the model to be deployed according to the model parameters, the deployment environment parameters, and the operation performance parameters, acquire the first radar parameters of the radar; the first radar parameters include at least one of a radar sampling rate parameter, a radar resolution parameter, and a radar data processing requirement parameter;

[0065] The determination module is further configured to determine the deployment impact situation of the first radar parameters on the model to be deployed according to the first radar parameters, the deployment environment parameters, and the operation performance parameters, and determine the deployment impact degree value corresponding to the deployment impact situation;

[0066] The determination module is further configured to determine whether the deployment impact degree value is greater than or equal to a preset deployment impact degree threshold; when the determination result is negative, trigger the determination module to perform the operation of determining the deployment parameters corresponding to the to-be-deployed model according to the model parameters, the deployment environment parameters, and the operation performance parameters.

[0067] The determination module is further configured to, when the determination result of the determination module is positive, determine the deployment parameters corresponding to the to-be-deployed model according to the first radar parameter, the model parameter, the deployment environment parameter, and the operation performance parameter.

[0068] As an optional implementation manner, in the second aspect of the present invention, the target physical sign parameter at least includes a blood pressure parameter, and the body movement parameter includes a body movement type parameter and a body movement amplitude parameter.

[0069] Wherein, the manner in which the determination module determines whether the target parameter meets the preset physical sign parameter calculation condition specifically includes:

[0070] According to the type parameter of the target object, determine whether the type parameter is a preset type parameter.

[0071] When it is determined that the type parameter is the preset type parameter, according to the distance parameter of the target object, determine whether the distance parameter is within a preset acquisition range.

[0072] When it is determined that the distance parameter is within the acquisition range, according to the body movement type parameter of the target object, determine whether the body movement type parameter is a preset body movement type parameter.

[0073] When it is determined that the body movement type parameter is the preset body movement type parameter, according to the body movement amplitude parameter of the target object, determine whether the body movement amplitude parameter is within a preset body movement amplitude range.

[0074] When it is determined that the body movement amplitude parameter is within the body movement amplitude range, determine that the target parameter meets the preset physical sign parameter calculation condition.

[0075] As an optional implementation manner, in the second aspect of the present invention, the manner in which the determination module determines whether the target parameter meets the preset physical sign parameter calculation condition specifically further includes:

[0076] Before determining that the target parameter meets the preset physical sign parameter calculation condition, determine a first feature parameter and a second feature parameter corresponding to the basic physical sign parameter of the target object; the first feature parameter includes an acquisition duration parameter and / or an acquisition point number parameter, and the second feature parameter includes a waveform feature parameter.

[0077] Determine whether the first characteristic parameter is greater than or equal to a preset characteristic parameter threshold;

[0078] When it is determined that the first characteristic parameter is greater than or equal to the characteristic parameter threshold, determine the waveform integrity corresponding to the basic physical sign parameter according to the waveform characteristic parameter, and determine whether the waveform integrity is greater than or equal to a preset waveform integrity threshold;

[0079] When it is determined that the waveform integrity is greater than or equal to the waveform integrity threshold, trigger the operation of determining that the target parameter meets the preset physical sign parameter calculation condition.

[0080] As an optional implementation manner, in the second aspect of the present invention, the characteristic parameter threshold includes an acquisition duration threshold and / or an acquisition point number threshold;

[0081] Among them, the characteristic parameter threshold is determined by the following method:

[0082] Obtain the second radar parameter of the radar; the second radar parameter of the radar includes at least one of the radar sampling rate parameter, radar resolution parameter, radar data generation speed parameter, and communication interface performance parameter of the radar;

[0083] Determine the data transmission situation of the radar according to the second radar parameter; the data transmission situation of the radar includes the data transmission speed situation of the radar and / or the transmission start frequency parameter;

[0084] Determine the characteristic parameter threshold corresponding to the basic physical sign parameter of the target object according to the data transmission situation.

[0085] The third aspect of the present invention discloses another radar detection physical sign parameter device based on local calculation, and the device includes:

[0086] A memory storing executable program code;

[0087] A processor coupled to the memory;

[0088] The processor calls the executable program code stored in the memory and executes the radar detection physical sign parameter method based on local calculation disclosed in the first aspect of the present invention.

[0089] The fourth aspect of the present invention discloses a computer storage medium, and the computer storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the radar detection physical sign parameter method based on local calculation disclosed in the first aspect of the present invention.

[0090] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0091] In the embodiments of the present invention, the basic computing model is compressed and distilled through the cloud to obtain a to-be-deployed model corresponding to the basic computing model, and the to-be-deployed model is deployed to the device side corresponding to the radar to obtain a deployed model corresponding to the device side; the target parameters of the target object collected by the radar are received, and it is determined whether the target parameters meet the physical sign parameter calculation condition. If so, the target physical sign parameters of the target object are calculated through the deployed model and the target parameters. It can be seen that implementing the present invention can calculate the target physical sign parameters of the target object through the deployed model deployed on the device side and the target parameters collected by the radar. In this way, when the radar detects the physical sign parameters, the offline use of the device side is realized, thereby improving the real-time performance of data transmission and the convenience of data detection, and thus improving the detection efficiency of the radar for the physical sign parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0093] Figure 1 It is a schematic diagram of a technical framework for a radar to detect physical sign parameters based on local computing disclosed in the embodiments of the present invention;

[0094] Figure 2 It is a schematic flowchart of a method for a radar to detect physical sign parameters based on local computing disclosed in the embodiments of the present invention;

[0095] Figure 3 It is a schematic flowchart of another method for a radar to detect physical sign parameters based on local computing disclosed in the embodiments of the present invention;

[0096] Figure 4 It is a schematic structural diagram of a device for a radar to detect physical sign parameters based on local computing disclosed in the embodiments of the present invention;

[0097] Figure 5 It is a schematic structural diagram of another device for a radar to detect physical sign parameters based on local computing disclosed in the embodiments of the present invention;

[0098] Figure 6 It is a schematic structural diagram of yet another device for a radar to detect physical sign parameters based on local computing disclosed in the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0099] To enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0100] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, device, product or terminal including a series of steps or units is not limited to the listed steps or units, but optionally further includes steps or units not listed, or optionally further includes other steps or units inherent to these processes, methods, products or terminals.

[0101] Referring to "embodiment" herein means that a specific feature, structure or characteristic described in connection with the embodiment may be included in at least one embodiment of the present invention. The phrase appears in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0102] The present invention discloses a method and device for detecting physical sign parameters by radar based on local computing. When detecting physical sign parameters by radar, offline use of the device end is realized, thereby improving the real-time performance of data transmission and the convenience of data detection, and thus improving the detection efficiency of the radar for physical sign parameters.

[0103] Embodiment 1

[0104] Please refer to Figure 2 , Figure 2 is a schematic flowchart of a method for detecting physical sign parameters by radar based on local computing disclosed in an embodiment of the present invention. Among them, Figure 2 The described method for detecting physical sign parameters by radar based on local computing can be applied to detecting target physical sign parameters (such as blood pressure parameters - systolic / diastolic blood pressure, heart rate parameters, respiratory parameters, etc.) of a target human body or a target animal by radar. The embodiments of the present invention do not limit the detection target object and the type of target physical sign parameters. Optionally, the method can be implemented by a physical sign parameter detection device, which can be integrated in a physical sign parameter detection device (such as a smart computer, a smart phone), or can be a local server for processing the physical sign parameter detection process, etc. The embodiments of the present invention do not limit it. Such as Figure 2As shown, the method for detecting physical sign parameters by radar based on local computing may include the following operations:

[0105] 101. Perform compression and distillation operations on the pre-trained basic computing model through the cloud to obtain the model to be deployed corresponding to the basic computing model.

[0106] In an embodiment of the present invention, optionally, the compression and distillation operations include at least one of distillation, quantization, pruning, and low-rank decomposition operations.

[0107] 102. Deploy the model to be deployed to the device side corresponding to the radar to obtain the deployed model corresponding to the device side.

[0108] In an embodiment of the present invention, as Figure 1 (a) shown, after training the basic computing model (cloud model) in the cloud, compression and distillation operations can be performed on the basic computing model, so that the obtained model to be deployed can be deployed to the device side corresponding to the radar and become the deployed model, that is, the local model. In this way, even in an offline state, the calculation process of the target parameters of the target object collected by the radar can be completed through the deployed model.

[0109] 103. Receive the target parameters of the target object collected by the radar, and determine whether the target parameters meet the preset physical sign parameter calculation conditions. If so, calculate the target physical sign parameters of the target object through the deployed model and the target parameters.

[0110] In an embodiment of the present invention, optionally, the target parameters of the target object include the type parameter, distance parameter, body movement parameter, and basic physical sign parameters of the target object (such as waveform data of heartbeat, breathing, etc. collected by the radar). Further optionally, the target physical sign parameters may include blood pressure parameters, and may also include heartbeat parameters, breathing parameters, etc., and the body movement parameters include body movement type parameters (such as hands hanging vertically in a sitting state, hands held flat in a standing state, etc.) and body movement amplitude parameters.

[0111] It should be noted that, as Figure 1 (b) shown, it can be understood that after the radar main control software is started, the model file on the device side is loaded, the radar is initialized to enter the detection mode, and then the present invention can combine the sensed millimeter-wave radar signal with the local AI algorithm framework to complete functions such as target recognition and human / animal physical sign detection (breathing, heartbeat, blood pressure, etc.) in an offline state, that is, a hardware product solution with local target detection function is realized.

[0112] Further, after calculating the target physical sign parameters of the target object, they can be stored on the device side, or the target physical sign parameters can be displayed to users in need (when the target physical sign parameters are abnormal, timely feedback can be provided), and they can also be reported to the cloud through IoT technology.

[0113] It can be seen that implementing the embodiments of the present invention can calculate the target physical sign parameters of the target object through the post-deployment model deployed on the device side and the target parameters collected by the radar. In this way, when the radar detects the physical sign parameters, offline use on the device side is achieved, thereby improving the real-time performance of data transmission and the convenience of data detection, and thus improving the detection efficiency of the radar for physical sign parameters; at the same time, additional communication costs and latency overhead are also reduced.

[0114] In an optional embodiment, before performing the compression and distillation operation on the pre-trained basic calculation model through the cloud in step 101 above to obtain the model to be deployed corresponding to the basic calculation model, the method further includes:

[0115] Predict the detection scenario parameters of the radar, and determine the calculation requirement parameters of the device side corresponding to the radar according to the detection scenario parameters;

[0116] Obtain the resource usage parameters of the device side, and determine the operating performance parameters of the device side according to the resource usage parameters;

[0117] Determine the compression and distillation requirement parameters corresponding to the pre-trained basic calculation model according to the calculation requirement parameters and the operating performance parameters.

[0118] In this optional embodiment, further, the operation of performing the compression and distillation operation on the pre-trained basic calculation model through the cloud in step 101 above to obtain the model to be deployed corresponding to the basic calculation model includes:

[0119] Perform a compression and distillation operation on the basic calculation model through the cloud and the compression and distillation requirement parameters to obtain the model to be deployed corresponding to the basic calculation model.

[0120] In this optional embodiment, optionally, the calculation requirement parameters include calculation amount requirement parameters and / or calculation result requirement parameters. Further optionally, the operating performance parameters include response duration parameters and / or operating energy consumption parameters.

[0121] For example, if it is necessary to detect the blood pressure of family members in a home scenario, at this time, it can be determined that the computational requirement of the device end corresponding to the radar is relatively small. Then, in combination with the resource usage parameters of the device end, such as CPU occupancy rate, memory occupancy parameter, storage read and write speed, etc., the response duration, power consumption parameter, heat generation parameter, etc. of the device end can be determined, so as to determine the compression and distillation requirement parameters corresponding to the basic calculation model, such as removing some neurons in the neural network, needing to maintain a high floating-point parameter, needing to decompose the weight matrix into the product of 3 low-rank matrices, etc., so as to implement the compression and distillation operation of the basic calculation model.

[0122] It can be seen that this optional embodiment can determine the computational requirement parameters of the device end corresponding to the radar according to the detection scenario parameters of the radar, and determine the operating performance parameters of the device end according to the resource usage parameters of the device end. Then, according to the computational requirement parameters and the operating performance parameters, the compression and distillation requirement parameters corresponding to the basic calculation model can be determined, so as to perform compression and distillation on the basic calculation model according to the compression and distillation requirement parameters. In this way, the execution reliability and accuracy of the compression and distillation operation of the basic calculation model can be improved, and then the normal operation of the device during subsequent operations of the device end can be ensured, so that the target physical sign parameters of the target object can be accurately calculated.

[0123] In another optional embodiment, before deploying the model to be deployed to the device end corresponding to the radar in step 102 above to obtain the deployed model corresponding to the device end, the method further includes:

[0124] Determine the model parameters of the model to be deployed;

[0125] Obtain the deployment environment parameters of the device end;

[0126] Determine the deployment parameters corresponding to the model to be deployed according to the model parameters, deployment environment parameters, and operating performance parameters.

[0127] In this optional embodiment, further, deploying the model to be deployed to the device end corresponding to the radar in step 102 above to obtain the deployed model corresponding to the device end includes:

[0128] Deploy the model to be deployed to the device end corresponding to the radar according to the deployment parameters to obtain the deployed model corresponding to the device end.

[0129] In this optional embodiment, optionally, the model parameters include at least one of a model size parameter, a model computation amount parameter, a model update method parameter (such as update frequency, update time, etc.), and a framework compatibility requirement parameter, and the deployment environment parameters include deployment network parameters and / or deployment system parameters (such as the operating systems supported by the device side, such as Linux, Windows, or an embedded system, etc., or the hardware configuration of the device side, such as CPU, GPU, memory, storage, etc.). Further optionally, the deployment parameters include a deployment location parameter and / or a deployment time parameter (such as deploying when the device is idle or at low load).

[0130] It can be seen that this optional embodiment can determine the deployment parameters corresponding to the model to be deployed according to the model parameters of the model to be deployed, the deployment environment parameters of the device side, and in combination with the running performance parameters of the device side, and then implement the deployment process of the model to be deployed. In this way, the deployment reliability and accuracy of the model to be deployed can be improved, ensuring the normal operation of the device side after the model is deployed, and thus facilitating the improvement of the calculation reliability, accuracy, and effectiveness of the target physical sign parameters of the target object.

[0131] In another optional embodiment, before determining the deployment parameters corresponding to the model to be deployed according to the model parameters, deployment environment parameters, and running performance parameters in the above steps, the method further includes:

[0132] Obtain the first radar parameters of the radar;

[0133] According to the first radar parameters, deployment environment parameters, and running performance parameters, determine the deployment impact situation caused by the first radar parameters on the model to be deployed, and determine the deployment impact degree value corresponding to the deployment impact situation;

[0134] Judge whether the deployment impact degree value is greater than or equal to a preset deployment impact degree threshold;

[0135] When the judgment result is negative, trigger the operation of determining the deployment parameters corresponding to the model to be deployed according to the model parameters, deployment environment parameters, and running performance parameters;

[0136] When the judgment result is positive, determine the deployment parameters corresponding to the model to be deployed according to the first radar parameters, model parameters, deployment environment parameters, and running performance parameters.

[0137] In this optional embodiment, optionally, the first radar parameters include at least one of a radar sampling rate parameter (such as the number of samples per second), a radar resolution parameter (such as range resolution, angular resolution), and a radar data processing requirement parameter (such as data filtering, pulse interference rejection, differential calculation, etc.).

[0138] For example, in a millimeter-wave radar system, if the radar sampling rate is 1 MHz, the resolution is 0.1 m, and the data processing requirements include pulse interference rejection, by combining the deployment environment parameters and operating performance parameters of the device side, it can be analyzed that the high sampling rate and high resolution result in a large amount of input data, which requires high storage and computing capabilities on the device side (that is, if the model deployment location is inappropriate, it may increase the computing burden on the device side and the real-time response of the model). Then, it can be determined that the deployment impact degree value is greater than the preset deployment impact degree threshold. At this time, it is necessary to further determine the deployment parameters corresponding to the model to be deployed based on the first radar parameter to reasonably deploy the model (such as deploying it to an edge server, adjusting the CPU and memory allocation ratio of the device side, etc.).

[0139] It can be seen that this optional embodiment can further determine the deployment impact of the first radar parameter on the model to be deployed according to the first radar parameter of the radar, the deployment environment parameter of the device side, and the operating performance parameter. And when the deployment impact is relatively significant, based on the first radar parameter, determine the deployment parameters corresponding to the model to be deployed. In this way, it can ensure that the deployment of the model on the device side highly matches the actual operation requirements of the radar, thereby improving the overall adaptability of the system; at the same time, it can optimize the deployment time and location of the model on the device side, thereby improving the computing real-time performance and reliability of the system for target physical sign parameters.

[0140] Embodiment 2

[0141] Please refer to Figure 3 , Figure 3 which is a schematic flowchart of another method for detecting physical sign parameters of a radar based on local computing disclosed in an embodiment of the present invention. Among them, Figure 3 the method for detecting physical sign parameters of a radar based on local computing described can be applied to detecting the target physical sign parameters (such as blood pressure parameters - systolic / diastolic blood pressure) of a target human body or a target animal through a radar. The embodiment of the present invention does not limit the detection target object and the type of target physical sign parameters. Optionally, this method can be implemented by a physical sign parameter detection device, which can be integrated in a physical sign parameter detection device (such as a smart computer, a smart phone), or can be a local server for processing the physical sign parameter detection process, etc. The embodiment of the present invention does not make a limitation.

[0142] As Figure 3 shown, the method for detecting physical sign parameters of a radar based on local computing may include the following operations:

[0143] 201. Perform compression and distillation operations on the pre-trained basic calculation model through the cloud to obtain the model to be deployed corresponding to the basic calculation model.

[0144] 202. Deploy the model to be deployed to the device corresponding to the radar to obtain the deployed model corresponding to the device end.

[0145] 203. Receive the target parameters of the target object collected by the radar, and determine whether the type parameter is a preset type parameter according to the type parameter of the target object.

[0146] 204. When it is determined that the type parameter is a preset type parameter, determine whether the distance parameter is within the preset acquisition range according to the distance parameter of the target object.

[0147] 205. When it is determined that the distance parameter is within the acquisition range, determine whether the body movement type parameter is a preset body movement type parameter according to the body movement type parameter of the target object.

[0148] 206. When it is determined that the body movement type parameter is a preset body movement type parameter, determine whether the body movement amplitude parameter is within the preset body movement amplitude range according to the body movement amplitude parameter of the target object.

[0149] 207. When it is determined that the body movement amplitude parameter is within the body movement amplitude range, determine that the target parameter meets the preset physical sign parameter calculation condition.

[0150] In the embodiment of the present invention, for example, the judgment process of steps 203-207 can be understood as follows: first judge whether the presence of a human body is true; if so, judge whether the distance between this human body and the radar is between 0.1-1 m; if so, judge whether this human body is in a sitting state; if so, judge whether the body movement amplitude of this human body is less than the preset body movement amplitude limit value 20 (the greater the body movement amplitude, the greater the movement of this human body); if so, it can be determined that the target parameters collected by the radar meet the preset physical sign parameter calculation conditions. Further, if any judgment result in the judgment process of steps 203-207 is negative, it can be determined that the target parameters collected by the radar do not meet the preset physical sign parameter calculation conditions.

[0151] 208. When it is determined that the target parameter meets the preset physical sign parameter calculation condition, calculate the target physical sign parameter of the target object through the deployed model and the target parameter.

[0152] In the embodiment of the present invention, for other descriptions of steps 201, 202 and 208, please refer to the detailed descriptions of steps 101-103 in Embodiment 1, and the embodiment of the present invention will not be repeated here.

[0153] It can be seen that implementing the embodiments of the present invention can perform multiple parameter validity judgments on the target parameters of the target object collected by the radar. In this way, it is possible to effectively reduce the processing of irrelevant targets or interference data, thereby improving the computing efficiency and accuracy of the system. At the same time, it is also beneficial to improve the adaptability and reliability of the system in complex scenarios and enhance the user experience of the system.

[0154] In an optional embodiment, before determining that the target parameters meet the preset physical sign parameter calculation conditions in step 207 above, the method further includes:

[0155] Determine the first characteristic parameter and the second characteristic parameter corresponding to the basic physical sign parameter of the target object; the second characteristic parameter includes a waveform characteristic parameter;

[0156] Judge whether the first characteristic parameter is greater than or equal to a preset characteristic parameter threshold;

[0157] When it is judged that the first characteristic parameter is greater than or equal to the characteristic parameter threshold, determine the waveform integrity corresponding to the basic physical sign parameter according to the waveform characteristic parameter, and judge whether the waveform integrity is greater than or equal to a preset waveform integrity threshold;

[0158] When it is judged that the waveform integrity is greater than or equal to the waveform integrity threshold, trigger the operation of determining that the target parameters meet the preset physical sign parameter calculation conditions.

[0159] In this optional embodiment, optionally, the first characteristic parameter includes an acquisition duration parameter and / or an acquisition point number parameter. Further optionally, the characteristic parameter threshold includes an acquisition duration threshold and / or an acquisition point number threshold. For example, it can be understood as judging whether the basic physical sign parameter of the target object collected by the radar has accumulated to a certain duration (such as continuous acquisition for 5 seconds) and / or the acquisition sample size; if so, further judge whether the waveform of the basic physical sign parameter (the parameter feedback by the radar is a waveform signal, such as the waveform data group of heartbeat and respiration within a preset time period) is complete enough; if so, then it can be determined that the target parameters meet the preset physical sign parameter calculation conditions.

[0160] Further, the characteristic parameter threshold is determined by the following method:

[0161] Obtain the second radar parameter of the radar;

[0162] Determine the data transmission situation of the radar according to the second radar parameter;

[0163] Determine the characteristic parameter threshold corresponding to the basic physical sign parameter of the target object according to the data transmission situation.

[0164] In this optional embodiment, optionally, the second radar parameter of the radar includes at least one of the radar sampling rate parameter, the radar resolution parameter, the radar data generation speed parameter, and the communication interface performance parameter of the radar. Further optionally, the data transmission situation of the radar includes the data transmission speed situation and / or the transmission start frequency parameter of the radar, so that it can be ensured that the data will not be congested in the radar firmware, nor will it cause the radar to start transmission too frequently.

[0165] It can be seen that this optional embodiment can effectively screen out high-quality data through the first characteristic parameter judgment of the basic physical sign parameters of the target object and the waveform integrity analysis, reduce misjudgment or calculation errors caused by incomplete data or insufficient collection, so as to improve the calculation reliability and accuracy of the target physical sign parameters; at the same time, by dynamically determining the characteristic parameter threshold through the second radar parameter of the radar, the adjustment flexibility of the radar data collection requirements can be improved, and then it can be ensured that the system can operate efficiently on different radar devices, thereby enhancing the adaptability and versatility of the system.

[0166] Embodiment III

[0167] Please refer to Figure 4 , Figure 4 which is a schematic structural diagram of a radar detection physical sign parameter device based on local computing disclosed in an embodiment of the present invention. As Figure 4 shown, the radar detection physical sign parameter device based on local computing may include:

[0168] A compression and distillation module 301, configured to perform a compression and distillation operation on a pre-trained basic calculation model through the cloud to obtain a to-be-deployed model corresponding to the basic calculation model;

[0169] A deployment module 302, configured to deploy the to-be-deployed model to the device side corresponding to the radar to obtain a deployed model corresponding to the device side;

[0170] A receiving module 303, configured to receive the target parameters of the target object collected by the radar;

[0171] A judgment module 304, configured to judge whether the target parameters meet the preset physical sign parameter calculation conditions;

[0172] A calculation module 305, configured to calculate the target physical sign parameters of the target object through the deployed model and the target parameters when the judgment result of the judgment module 304 is yes.

[0173] In the embodiment of the present invention, the target parameters of the target object include the type parameter, the distance parameter, the body movement parameter, and the basic physical sign parameter of the target object.

[0174] It can be seen that the implementation Figure 4The described radar detection vital sign parameter device based on local computing can calculate the target vital sign parameters of the target object through the post-deployment model deployed on the device side and the target parameters collected by the radar. In this way, when detecting the vital sign parameters by the radar, offline use on the device side is achieved, thereby improving the real-time performance of data transmission and the convenience of data detection, and thus improving the detection efficiency of the radar for vital sign parameters; at the same time, additional communication costs and latency overhead are also reduced.

[0175] In an optional embodiment, the device further includes:

[0176] A prediction module 306, configured to predict the detection scenario parameters of the radar before the compression and distillation module 301 performs a compression and distillation operation on the pre-trained basic calculation model through the cloud to obtain a to-be-deployed model corresponding to the basic calculation model;

[0177] A determination module 307, configured to determine the calculation requirement parameters of the device side corresponding to the radar according to the detection scenario parameters;

[0178] An acquisition module 308, configured to acquire the resource usage parameters of the device side;

[0179] The determination module 307 is further configured to determine the operating performance parameters of the device side according to the resource usage parameters; and determine the compression and distillation requirement parameters corresponding to the pre-trained basic calculation model according to the calculation requirement parameters and the operating performance parameters;

[0180] Among them, the manner in which the compression and distillation module 301 performs a compression and distillation operation on the pre-trained basic calculation model through the cloud to obtain a to-be-deployed model corresponding to the basic calculation model specifically includes:

[0181] Performing a compression and distillation operation on the basic calculation model through the cloud and the compression and distillation requirement parameters to obtain a to-be-deployed model corresponding to the basic calculation model.

[0182] In this optional embodiment, the calculation requirement parameters include calculation amount requirement parameters and / or calculation result requirement parameters; the operating performance parameters include response duration parameters and / or operating energy consumption parameters.

[0183] It can be seen that implementing Figure 5The described radar detection physical sign parameter device based on local computing can determine the computing requirement parameters of the device end corresponding to the radar according to the detection scene parameters of the radar, and determine the operating performance parameters of the device end according to the resource usage parameters of the device end. Then, according to the computing requirement parameters and the operating performance parameters, the compression and distillation requirement parameters corresponding to the basic computing model are determined, so as to perform compression and distillation on the basic computing model according to the compression and distillation requirement parameters. In this way, the execution reliability and accuracy of the compression and distillation operation of the basic computing model can be improved, and then the normal operation of the device during subsequent operations of the device end can be ensured, so that the target physical sign parameters of the target object can be accurately calculated.

[0184] In another alternative embodiment, the determining module 307 is further configured to:

[0185] Before the deployment module 302 deploys the model to be deployed to the device end corresponding to the radar to obtain the deployed model corresponding to the device end, determine the model parameters of the model to be deployed;

[0186] The obtaining module 308 is further configured to obtain the deployment environment parameters of the device end;

[0187] The determining module 307 is further configured to determine the deployment parameters corresponding to the model to be deployed according to the model parameters, the deployment environment parameters, and the operating performance parameters;

[0188] Among them, the manner in which the deployment module 302 deploys the model to be deployed to the device end corresponding to the radar to obtain the deployed model corresponding to the device end specifically includes:

[0189] Deploy the model to be deployed to the device end corresponding to the radar according to the deployment parameters to obtain the deployed model corresponding to the device end.

[0190] In this alternative embodiment, the model parameters include at least one of the model size parameter, the model computation amount parameter, the model update method parameter, and the framework compatibility requirement parameter; the deployment environment parameters include the deployment network parameter and / or the deployment system parameter; the deployment parameters include the deployment location parameter and / or the deployment time parameter.

[0191] It can be seen that implementing Figure 5 The described radar detection physical sign parameter device based on local computing can determine the deployment parameters corresponding to the model to be deployed according to the model parameters of the model to be deployed and the deployment environment parameters of the device end, and in combination with the operating performance parameters of the device end, and then realize the deployment process of the model to be deployed. In this way, the deployment reliability and accuracy of the model to be deployed can be improved, ensuring the normal operation of the device end after the model is deployed, which is conducive to improving the computing reliability, accuracy, and effectiveness of the target physical sign parameters of the target object.

[0192] In yet another alternative embodiment, the obtaining module 308 is further configured to:

[0193] Before the determining module 307 determines the deployment parameters corresponding to the model to be deployed according to the model parameters, the deployment environment parameters, and the running performance parameters, obtain the first radar parameters of the radar;

[0194] The determining module 307 is further configured to determine the deployment impact situation of the first radar parameters on the model to be deployed according to the first radar parameters, the deployment environment parameters, and the running performance parameters, and determine the deployment impact degree value corresponding to the deployment impact situation;

[0195] The judging module 304 is further configured to judge whether the deployment impact degree value is greater than or equal to a preset deployment impact degree threshold; when the judgment result is negative, trigger the operation of the determining module 307 to determine the deployment parameters corresponding to the model to be deployed according to the model parameters, the deployment environment parameters, and the running performance parameters;

[0196] The determining module 307 is further configured to, when the judgment result of the judging module 304 is positive, determine the deployment parameters corresponding to the model to be deployed according to the first radar parameters, the model parameters, the deployment environment parameters, and the running performance parameters.

[0197] In this alternative embodiment, the first radar parameters include at least one of a radar sampling rate parameter, a radar resolution parameter, and a radar data processing requirement parameter.

[0198] It can be seen that the Figure 5 described radar detection physical sign parameter device based on local computing can further determine the deployment impact situation of the first radar parameters on the model to be deployed according to the first radar parameters of the radar, the deployment environment parameters of the device end, and the running performance parameters, and when the deployment impact situation is relatively significant, determine the deployment parameters corresponding to the model to be deployed based on the first radar parameters, so that the deployment of the model on the device end can be ensured to highly match the actual operation requirements of the radar, thereby improving the overall adaptability of the system; at the same time, the deployment time and location of the model on the device end can be optimized, thereby improving the real-time performance and reliability of the system's calculation of the target physical sign parameters.

[0199] In yet another alternative embodiment, the target physical sign parameters at least include blood pressure parameters, and the body movement parameters include a body movement type parameter and a body movement amplitude parameter;

[0200] Among them, the specific manner in which the judging module 304 judges whether the target parameters meet the preset physical sign parameter calculation conditions includes:

[0201] According to the type parameter of the target object, judge whether the type parameter is a preset type parameter;

[0202] When it is determined that the type parameter is a preset type parameter, according to the distance parameter of the target object, it is judged whether the distance parameter is within a preset acquisition range;

[0203] When it is judged that the distance parameter is within the acquisition range, according to the body movement type parameter of the target object, it is judged whether the body movement type parameter is a preset body movement type parameter;

[0204] When it is judged that the body movement type parameter is a preset body movement type parameter, according to the body movement amplitude parameter of the target object, it is judged whether the body movement amplitude parameter is within a preset body movement amplitude range;

[0205] When it is judged that the body movement amplitude parameter is within the body movement amplitude range, it is determined that the target parameter meets the preset physical sign parameter calculation condition.

[0206] It can be seen that implementing Figure 5 The described radar detection physical sign parameter device based on local calculation can perform multiple parameter validity judgments on the target parameters of the target object collected by the radar. In this way, it can effectively reduce the processing of irrelevant targets or interference data, and thus can improve the calculation efficiency and accuracy of the system; at the same time, it is also beneficial to improve the adaptability and reliability of the system in complex scenarios and enhance the user experience of the system.

[0207] In another optional embodiment, the manner in which the judgment module 304 judges whether the target parameter meets the preset physical sign parameter calculation condition specifically further includes:

[0208] Before determining that the target parameter meets the preset physical sign parameter calculation condition, determine the first characteristic parameter and the second characteristic parameter corresponding to the basic physical sign parameter of the target object; the second characteristic parameter includes a waveform characteristic parameter;

[0209] Judge whether the first characteristic parameter is greater than or equal to a preset characteristic parameter threshold;

[0210] When it is judged that the first characteristic parameter is greater than or equal to the characteristic parameter threshold, according to the waveform characteristic parameter, determine the waveform integrity corresponding to the basic physical sign parameter, and judge whether the waveform integrity is greater than or equal to a preset waveform integrity threshold;

[0211] When it is judged that the waveform integrity is greater than or equal to the waveform integrity threshold, trigger the operation of determining that the target parameter meets the preset physical sign parameter calculation condition.

[0212] In this optional embodiment, the first characteristic parameter includes an acquisition duration parameter and / or an acquisition point number parameter; the characteristic parameter threshold includes an acquisition duration threshold and / or an acquisition point number threshold;

[0213] In this optional embodiment, further, the characteristic parameter threshold is determined by the following method:

[0214] Obtain the second radar parameter of the radar;

[0215] Determine the data transmission situation of the radar according to the second radar parameter;

[0216] Determine the characteristic parameter threshold corresponding to the basic physical sign parameter of the target object according to the data transmission situation.

[0217] In this optional embodiment, optionally, the second radar parameter of the radar includes at least one of the radar sampling rate parameter, the radar resolution parameter, the radar data generation speed parameter, and the communication interface performance parameter; the data transmission situation of the radar includes the data transmission speed situation of the radar and / or the transmission start frequency parameter.

[0218] It can be seen that implementing Figure 5 The described radar detection physical sign parameter device based on local calculation can effectively screen out high-quality data through the first characteristic parameter judgment of the basic physical sign parameter of the target object and the waveform integrity analysis, reduce misjudgment or calculation errors caused by incomplete data or insufficient collection, thereby improving the calculation reliability and accuracy of the target physical sign parameter; at the same time, by dynamically determining the characteristic parameter threshold through the second radar parameter of the radar, the adjustment flexibility of the radar data acquisition requirements can be improved, and then it can be ensured that the system can operate efficiently on different radar devices, thereby enhancing the adaptability and versatility of the system.

[0219] Embodiment Four

[0220] Please refer to Figure 6 , Figure 6 which is a schematic structural diagram of another radar detection physical sign parameter device based on local calculation disclosed in the embodiments of the present invention. As Figure 6 shown, the radar detection physical sign parameter device based on local calculation may include:

[0221] A memory 401 storing executable program code;

[0222] A processor 402 coupled to the memory 401;

[0223] The processor 402 calls the executable program code stored in the memory 401 and executes the steps in the method for detecting physical sign parameters by radar based on local calculation described in Embodiment One or Embodiment Two of the present invention.

[0224] Embodiment Five

[0225] An embodiment of the present invention discloses a computer storage medium. The computer storage medium stores computer instructions, which, when called, are used to execute the steps in the method for detecting physical sign parameters by radar based on local computing described in Embodiment 1 or Embodiment 2 of the present invention.

[0226] Embodiment 6

[0227] An embodiment of the present invention discloses a computer program product. The computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute the steps in the method for detecting physical sign parameters by radar based on local computing described in Embodiment 1 or Embodiment 2.

[0228] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative labor.

[0229] Through the specific descriptions of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, and the storage medium includes read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc memories, magnetic disk memories, tape memories, or any other computer-readable medium capable of carrying or storing data.

[0230] Finally, it should be noted that: The method and device for detecting physical sign parameters based on local computing disclosed in the embodiments of the present invention only disclose the preferred embodiments of the present invention, and are only used to illustrate the technical solutions of the present invention, rather than limiting them; Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; And these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting vital sign parameters of radar based on local calculation, characterized in that: The method comprises: Performing compression and distillation operations on the pre-trained basic computing model through the cloud to obtain a model to be deployed corresponding to the basic computing model; Deploy the model to be deployed to the device end corresponding to the radar, and obtain a deployed model corresponding to the device end; Receive the target parameters of the target object collected by the radar, and determine whether the target parameters meet the preset vital sign parameter calculation conditions. If so, calculate the target vital sign parameters of the target object through the post-deployment model and the target parameters; the target parameters of the target object include the type parameters, distance parameters, body motion parameters and basic vital sign parameters of the target object.

2. The method for detecting vital sign parameters of radar based on local calculation according to claim 1, characterized in that: Before performing a compression and distillation operation on the pre-trained basic computing model through the cloud to obtain a to-be-deployed model corresponding to the basic computing model, the method further includes: Predicting detection scene parameters of the radar, and determining computing requirement parameters of a device corresponding to the radar according to the detection scene parameters; the computing requirement parameters include computing amount requirement parameters and / or computing result requirement parameters; Acquiring resource usage parameters of the device end, and determining operation performance parameters of the device end according to the resource usage parameters; the operation performance parameters include response time parameters and / or operation energy consumption parameters; Determining compression distillation requirement parameters corresponding to a pre-trained basic computing model according to the computing requirement parameters and the operating performance parameters; The method of performing compression and distillation operations on the pre-trained basic computing model through the cloud to obtain a model to be deployed corresponding to the basic computing model includes: Through the cloud and the compression and distillation requirement parameters, the basic computing model is compressed and distilled to obtain a model to be deployed corresponding to the basic computing model.

3. The method for detecting vital sign parameters of radar based on local calculation according to claim 2, characterized in that: Before deploying the to-be-deployed model to the device end corresponding to the radar to obtain the deployed model corresponding to the device end, the method further includes: Determine the model parameters of the model to be deployed; the model parameters include at least one of a model size parameter, a model calculation amount parameter, a model update method parameter, and a framework compatibility requirement parameter; Acquire deployment environment parameters of the device end; the deployment environment parameters include deployment network parameters and / or deployment system parameters; Determine deployment parameters corresponding to the to-be-deployed model according to the model parameters, the deployment environment parameters, and the operating performance parameters; the deployment parameters include deployment location parameters and / or deployment time parameters; The step of deploying the model to be deployed to a device end corresponding to the radar to obtain a deployed model corresponding to the device end includes: According to the deployment parameters, the model to be deployed is deployed to the device end corresponding to the radar to obtain a deployed model corresponding to the device end.

4. The method for detecting vital sign parameters of radar based on local calculation according to claim 3, characterized in that: Before determining the deployment parameters corresponding to the to-be-deployed model according to the model parameters, the deployment environment parameters, and the operating performance parameters, the method further includes: Acquire a first radar parameter of the radar; the first radar parameter includes at least one of a radar sampling rate parameter, a radar resolution parameter, and a radar data processing requirement parameter; Determine, according to the first radar parameter, the deployment environment parameter, and the operation performance parameter, a deployment impact situation caused by the first radar parameter on the model to be deployed, and determine a deployment impact degree value corresponding to the deployment impact situation; Determining whether the deployment impact degree value is greater than or equal to a preset deployment impact degree threshold; When the judgment result is no, triggering the execution of the operation of determining the deployment parameters corresponding to the to-be-deployed model according to the model parameters, the deployment environment parameters and the operating performance parameters; When the judgment result is yes, the deployment parameters corresponding to the model to be deployed are determined according to the first radar parameters, the model parameters, the deployment environment parameters and the operating performance parameters.

5. The radar detection vital sign parameter method based on local calculation according to any one of claims 1 to 4, characterized in that: The target vital sign parameters at least include blood pressure parameters, and the body movement parameters include body movement type parameters and body movement amplitude parameters; Wherein, the step of judging whether the target parameter satisfies a preset vital sign parameter calculation condition includes: According to the type parameter of the target object, determining whether the type parameter is a preset type parameter; When it is determined that the type parameter is the preset type parameter, judging whether the distance parameter is within a preset acquisition range according to the distance parameter of the target object; When it is determined that the distance parameter is within the acquisition range, determining whether the body motion type parameter is a preset body motion type parameter according to the body motion type parameter of the target object; When it is determined that the body movement type parameter is the preset body movement type parameter, judging whether the body movement amplitude parameter is within a preset body movement amplitude range according to the body movement amplitude parameter of the target object; When it is determined that the body movement amplitude parameter is within the body movement amplitude range, it is determined that the target parameter meets a preset physical sign parameter calculation condition.

6. The method for detecting vital sign parameters of radar based on local calculation according to claim 5, characterized in that: Before determining that the target parameter satisfies a preset vital sign parameter calculation condition, the method further includes: Determine a first characteristic parameter and a second characteristic parameter corresponding to the basic vital sign parameter of the target object; the first characteristic parameter includes a collection time parameter and / or a collection point quantity parameter, and the second characteristic parameter includes a waveform characteristic parameter; Determining whether the first characteristic parameter is greater than or equal to a preset characteristic parameter threshold; When it is determined that the first characteristic parameter is greater than or equal to the characteristic parameter threshold, determining the waveform integrity corresponding to the basic vital sign parameter according to the waveform characteristic parameter, and determining whether the waveform integrity is greater than or equal to a preset waveform integrity threshold; When it is determined that the waveform integrity is greater than or equal to the waveform integrity threshold, the operation of determining whether the target parameter satisfies the preset vital sign parameter calculation condition is triggered.

7. The method for detecting vital sign parameters of radar based on local calculation according to claim 6, characterized in that: The characteristic parameter threshold includes a collection time threshold and / or a collection point quantity threshold; The characteristic parameter threshold is determined in the following way: Acquire a second radar parameter of the radar; the second radar parameter of the radar includes at least one of a radar sampling rate parameter, a radar resolution parameter, a radar data generation speed parameter, and a communication interface performance parameter of the radar; Determining the data transmission status of the radar according to the second radar parameter; the data transmission status of the radar includes the data transmission speed of the radar and / or the transmission start frequency parameter; According to the data transmission situation, a characteristic parameter threshold corresponding to the basic vital sign parameter of the target object is determined.

8. A radar detection vital sign parameter device based on local calculation, characterized in that: The device comprises: A compression and distillation module is used to perform compression and distillation operations on the pre-trained basic computing model through the cloud to obtain a model to be deployed corresponding to the basic computing model; A deployment module, used to deploy the model to be deployed to the device end corresponding to the radar, and obtain a deployed model corresponding to the device end; A receiving module, used for receiving target parameters of the target object collected by the radar; A judgment module, used to judge whether the target parameter meets the preset vital sign parameter calculation conditions; A calculation module is used to calculate the target vital sign parameters of the target object through the post-deployment model and the target parameters when the judgment result of the judgment module is yes; the target parameters of the target object include the type parameter, distance parameter, body motion parameter and basic vital sign parameter of the target object.

9. A radar detection vital sign parameter device based on local calculation, characterized in that: The device comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the radar vital sign parameter detection method based on local calculation as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: The computer storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the radar vital sign parameter detection method based on local calculation as described in any one of claims 1 to 7.

Images