CN114510967A - Interference signal filtering method and device, electronic equipment and storage medium - Google Patents
Interference signal filtering method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN114510967A CN114510967A CN202210061661.XA CN202210061661A CN114510967A CN 114510967 A CN114510967 A CN 114510967A CN 202210061661 A CN202210061661 A CN 202210061661A CN 114510967 A CN114510967 A CN 114510967A
- Authority
- CN
- China
- Prior art keywords
- target
- sample
- motion
- signal
- doppler
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The application relates to an interference signal filtering method, an interference signal filtering device, electronic equipment and a storage medium, wherein the method comprises the following steps: acquiring an echo signal corresponding to a target environment; performing feature extraction on the echo signal to obtain a first motion feature and a second motion feature; inputting the first motion characteristic and the second motion characteristic into a target classification model, and determining a non-target interference signal in an echo signal; and filtering out non-target interference signals. This application is based on first motion characteristic and the second motion characteristic of object and is categorised the object in the target environment, can effectively discern less motion difference characteristic between non-target and the target, realizes the filtering to non-target interfering signal, and the filtering is effectual, can effectively improve equipment's interference killing feature.
Description
Technical Field
The present disclosure relates to the field of signal processing technologies, and in particular, to a method and an apparatus for filtering an interference signal, an electronic device, and a storage medium.
Background
The millimeter wave radar has the advantages of no privacy, all weather, non-contact, insensitivity to illumination temperature, weak signal detection capability, penetration capability to clothes, low radiation and the like, and is very suitable for being applied to indoor human body detection, such as human body vital sign detection, positioning and tracking. However, in practical applications, the characteristics of the millimeter wave radar determine that the millimeter wave radar is sensitive to all moving targets, including indoor non-human targets such as fans, air conditioners, curtains, green plants, pets and the like, and the difference between the motion of the targets and the motion of the human body is small, so that the millimeter wave radar can easily recognize the human body signal by mistake, and the final detection is invalid or wrong. Therefore, how to filter the interference signals to improve the interference resistance of the device is a technical problem which needs to be solved urgently in application.
Disclosure of Invention
Therefore, it is necessary to provide a method and an apparatus for filtering an interference signal, an electronic device, and a storage medium, which can filter the interference signal, have a good filtering effect, and effectively improve the anti-interference capability of the device.
An interference signal filtering method, comprising:
acquiring an echo signal corresponding to a target environment;
performing feature extraction on the echo signal to obtain a first motion feature and a second motion feature;
inputting the first motion characteristic and the second motion characteristic into a target classification model, and determining a non-target interference signal in the echo signal;
and filtering the non-target interference signal.
An interference signal filtering apparatus, comprising:
the signal acquisition module is used for acquiring an echo signal corresponding to a target environment;
the characteristic extraction module is used for extracting the characteristics of the echo signals to obtain a first motion characteristic and a second motion characteristic;
the classification module is used for inputting the first motion characteristic and the second motion characteristic into a target classification model and determining a non-target interference signal in the echo signal;
and the signal filtering module is used for filtering the non-target interference signal.
In one embodiment, the signal obtaining module includes:
an original echo signal acquiring unit, configured to acquire an original echo signal of the target environment;
the point cloud feature extraction unit is used for extracting features of the original echo signals to obtain point cloud features of the object in the target environment;
the area dividing unit is used for determining signals belonging to an invalid area in the original echo signals according to the point cloud characteristics;
and the signal filtering unit is used for filtering the signals belonging to the invalid region in the original echo signals to obtain the filtered echo signals of the target environment.
In one embodiment, the area dividing unit is configured to:
clustering point clouds of all objects in the target environment according to the point cloud characteristics;
and determining signals belonging to invalid regions in the original echo signals according to the clustering result.
In one embodiment, the first motion feature comprises a micro-doppler feature; the echo signals corresponding to the target environment are echo signals acquired in a first preset number of acquisition cycles, and each acquisition cycle comprises a second preset number of modulation cycles; the feature extraction module comprises:
the first transformation unit is used for performing fast Fourier transformation on the echo signals in each acquisition cycle in the echo signals to obtain a range-Doppler diagram corresponding to a second preset number of modulation cycles in each acquisition cycle;
the accumulation operation unit is used for carrying out accumulation operation on the Doppler information of the modulation cycles of the second preset number in the acquisition time of the first preset number according to the range Doppler image to obtain a micro Doppler image;
and the micro Doppler feature extraction unit is used for extracting micro Doppler features according to the micro Doppler image.
In one embodiment, the second motion characteristic comprises a range-doppler characteristic; the echo signals corresponding to the target environment are echo signals acquired in a first preset number of acquisition cycles, and each acquisition cycle comprises a second preset number of modulation cycles; the feature extraction module comprises:
the second transformation unit is used for performing fast Fourier transformation on the echo signals in each acquisition cycle in the echo signals to obtain a range-Doppler diagram corresponding to a second preset number of modulation cycles in each acquisition cycle;
the calculation unit is used for carrying out modular operation on the range-doppler diagram to obtain speed and distance and calculating the mean value and variance of the power of the acquisition cycles of a first preset number;
and the range-Doppler characteristic determining unit is used for determining the range-Doppler characteristic according to the mean value and the variance of the speed, the distance and the power.
In one embodiment, the apparatus further includes:
the system comprises a sample signal acquisition module, a signal processing module and a signal processing module, wherein the sample signal acquisition module is used for acquiring a sample target signal and a sample non-target interference signal;
the sample characteristic extraction module is used for respectively extracting a sample first motion characteristic and a sample second motion characteristic corresponding to the sample target signal and the sample non-target interference signal, and marking the sample first motion characteristic and the sample second motion characteristic with corresponding sample labels;
the prediction module is used for carrying out target classification on the first motion characteristics of the sample and the second motion characteristics of the sample through the initial classification model and determining a class prediction result corresponding to the first motion characteristics of the sample and the second motion characteristics of the sample;
and the training adjusting module is used for adjusting parameters of the initial classification model and continuing training according to the difference between the class prediction results corresponding to the first motion characteristics of the sample and the second motion characteristics of the sample and the sample labels of the first motion characteristics of the sample and the second motion characteristics of the sample, and stopping training until the training conditions are met to obtain the target classification model.
In one embodiment, the sample signal acquiring module is configured to:
acquiring sample target signals acquired under different motion states of a sample target and sample non-target interference signals acquired under different motion states of a plurality of non-sample targets respectively;
the sample feature extraction module comprises:
the characteristic extraction unit is used for respectively extracting the characteristics of the sample target signals and the sample non-target interference signals corresponding to different motion states to obtain sample first motion characteristics corresponding to the different motion states of the sample target and sample second motion characteristics corresponding to the different motion states of the sample non-target;
and the sample label marking unit is used for marking the corresponding sample labels for the sample first motion characteristics and the sample second motion characteristics.
In one embodiment, the apparatus further includes:
the target information determining module is used for determining a target detection result according to the target signal obtained by filtering;
and the control module is used for sending a corresponding control instruction to the target equipment to control the target equipment to execute the control instruction if the target detection result meets the equipment control condition.
An electronic device comprising a processor, a memory and a computer program stored on the memory and executable on the processor, the computer program, when executed by the processor, implementing the steps of the interference signal filtering method as described above.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the interference signal filtering method as described above.
The interference signal filtering method, the interference signal filtering device, the electronic equipment and the storage medium acquire echo signals corresponding to a target environment; performing feature extraction on the echo signal to obtain a first motion feature and a second motion feature; inputting the first motion characteristic and the second motion characteristic into a target classification model, and determining a non-target interference signal in an echo signal; and filtering out non-target interference signals. This application is based on first motion characteristic and the second motion characteristic of object and is categorised the object in the target environment, can effectively discern less motion difference characteristic between non-target and the target, realizes the filtering to non-target interfering signal, and the filtering is effectual, can effectively improve equipment's interference killing feature.
Drawings
FIG. 1 is a diagram of an exemplary embodiment of an interference signal filtering method;
FIG. 2 is a flowchart illustrating a method for filtering an interference signal according to an embodiment;
FIG. 3 is a schematic diagram of echo signal acquisition logic in an embodiment;
FIG. 4 is a schematic diagram illustrating a process for obtaining a micro-Doppler map according to an embodiment;
FIG. 5 is a flowchart illustrating a method for filtering interference signals according to another embodiment;
FIG. 6 is a flowchart illustrating a method for filtering an interference signal according to another embodiment;
FIG. 7 is a flowchart illustrating a method for filtering interference signals according to another embodiment;
FIG. 8 is a block diagram of an interference signal filtering apparatus according to an embodiment;
FIG. 9 is a block diagram of an interference signal filtering apparatus according to another embodiment;
FIG. 10 is a block diagram of an interference signal filtering apparatus according to another embodiment;
FIG. 11 is a diagram of the internal structure of an electronic device in one embodiment;
fig. 12 is an internal structural view of an electronic apparatus in another embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The interference signal filtering method provided by the application can be applied to the application environment shown in fig. 1. The terminal 102 communicates with the server 104 through a network, the server 104 communicates with the gateway 106 through the network, the internet of things device accesses the gateway 106 through a local area network, the internet of things device includes the millimeter wave radar 101, and may further include devices such as the intelligent desk lamp 103 and the intelligent air conditioner 105, which are not limited herein. The millimeter wave radar 101 is used for emitting detection signals and collecting echo signals of a target environment, the terminal 102 or the server 104 or the millimeter wave radar 101 can extract features of the echo signals to obtain first motion features and second motion features, the first motion features and the second motion features are input into a target classification model to determine non-target interference signals in the echo signals, the non-target interference signals are filtered, and then the filtered target signals can be used for carrying out relevant operations such as human body posture monitoring, motion monitoring or position monitoring, so that the terminal 102 or the server 104 can generate corresponding control instructions of the internet of things equipment based on monitoring results.
The terminal 102 may send a control instruction to the gateway 106 through the network and the server 104, so as to implement remote control on the internet of things device. In practical implementation, after the terminal 102 accesses the local area network where the gateway 106 is located, the terminal may also directly send a control instruction to the gateway 106 through the local area network, so as to control the internet of things device. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers.
As shown in fig. 2, there is provided an interference signal filtering method, including the following steps:
the target environment is a designated space environment in which a non-target signal needs to be filtered to detect a target, and may specifically be an environment in which the millimeter wave radar detection device is installed and applied, for example, an indoor space such as a room, a living room, and a balcony. One or more millimeter wave radars may be installed in the target environment to enable the millimeter wave radars to fully monitor objects in the space. The object includes a target and a non-target, wherein the target refers to an object needing monitoring, such as a human body, and the non-target refers to an object which interferes with the monitoring of the target, such as a fan, an air conditioner, a curtain, a green plant, a pet and the like in the environment. The echo signal corresponding to the target environment includes an echo signal reflected by a non-target in the target environment, that is, includes a non-target interference signal. When there is a person or other object in the target environment, the echo signal will include the echo signal reflected by the object in the target environment, that is, the target signal.
the first motion characteristic and the second motion characteristic refer to characteristics capable of representing two different motion attributes expressed by the echo signal; for example, the first motion characteristic includes characteristics capable of reflecting the motion intensity, the motion frequency, the motion period, the offset and the like of the object, and the second motion characteristic includes characteristics capable of reflecting the reflected signal power, the object distance, the object speed and the like of the object.
And performing feature extraction on the echo signals to obtain a first motion feature and a second motion feature. By combining various motion characteristics, the motion difference characteristics between the non-target and the target can be effectively identified.
In an embodiment, the first motion feature includes a micro-doppler feature, the second motion feature includes a range-doppler feature, the electronic device performs fast fourier transform on the echo signal, then performs an accumulation operation on a result of the fast fourier transform to extract the micro-doppler feature corresponding to the echo signal, and performs a modulo operation on a result of the fast fourier transform to extract the range-doppler feature corresponding to the echo signal.
The Micro Doppler characteristic is a characteristic related to Micro Doppler of an object, Doppler information representing Micro movement of a component can be represented in an echo signal by Micro movement of an internal component of a non-rigid system or a rigid system, namely Micro Doppler (Micro-Doppler), and due to the difference of the Micro movement of the component among different types of objects, the Micro Doppler among the different types of objects has a remarkable difference and can be used as an effective characteristic for object classification. The Range-Doppler characteristic refers to a characteristic associated with a Range-Doppler Map (RDM) of an object, one dimension of which is a Range and the other dimension is a Doppler frequency, which represents a velocity of the object. By combining the micro-Doppler feature and the range-Doppler feature, the small motion difference feature between the non-target and the target can be effectively identified.
the target classification model is used for identifying and classifying the non-target interference signals and the target signals, and may be obtained by training a classifier in advance based on first motion features and second motion features of the non-target interference signals and the target signals corresponding to a plurality of motion states, where the classifier may be, for example, KNN (K-Nearest Neighbors) or svm (support Vector machines).
After determining the non-target interference signals in the echo signals, filtering the non-target interference signals, so as to extract signals belonging to the target. Taking the signal of extracting the human body target as an example, because the characteristic of the millimeter wave radar determines that the millimeter wave radar is sensitive to all moving objects, and besides people, many other moving objects such as fans, curtains and the like exist indoors, the echo signals of the objects cause interference to the detection of the radar, the interference is filtered by the method, the anti-interference capability of the radar can be improved, the detection of the human body by the millimeter wave radar is really realized, the information of the existence of the human body, the position of the human body and the like is further obtained, the monitoring accuracy is high, and the false alarm rate is greatly reduced.
In the interference signal filtering method, an echo signal corresponding to a target environment is obtained; performing feature extraction on the echo signal to obtain a first motion feature and a second motion feature; inputting the first motion characteristic and the second motion characteristic into a target classification model, and determining a non-target interference signal in an echo signal; and filtering out non-target interference signals. This application is based on first motion characteristic and the second motion characteristic of object and is categorised the object in the target environment, can effectively discern less motion difference characteristic between non-target and the target, realizes the filtering to non-target interfering signal, and the filtering is effectual, can effectively improve equipment's interference killing feature.
In one embodiment, the process of training the target classification model includes the following steps:
obtaining a sample target signal and a sample non-target interference signal;
respectively extracting a sample first motion characteristic and a sample second motion characteristic corresponding to the sample target signal and the sample non-target interference signal, and marking the sample first motion characteristic and the sample second motion characteristic with corresponding sample labels;
performing target classification on the sample first motion characteristics and the sample second motion characteristics through an initial classification model, and determining class prediction results corresponding to the sample first motion characteristics and the sample second motion characteristics;
and adjusting parameters of the initial classification model and continuing training according to the difference between the class prediction results corresponding to the first motion characteristics of the sample and the class prediction results corresponding to the second motion characteristics of the sample and the sample labels of the first motion characteristics of the sample and the second motion characteristics of the sample, and stopping training until the training conditions are met to obtain the target classification model.
In one embodiment, obtaining a sample target signal and a sample non-target interfering signal comprises:
acquiring sample target signals acquired under different motion states of a sample target and sample non-target interference signals acquired under different motion states of a plurality of non-sample targets respectively.
In an embodiment, extracting a sample first motion feature and a sample second motion feature corresponding to a sample target signal and a sample non-target interference signal, respectively, and labeling the sample first motion feature and the sample second motion feature with corresponding sample labels includes:
respectively extracting the characteristics of a sample target signal and a sample non-target interference signal corresponding to different motion states to obtain a sample first motion characteristic corresponding to the different motion states of the sample target and a sample second motion characteristic corresponding to the different motion states of the sample non-target;
and marking the corresponding sample label for the sample first motion characteristic and the sample second motion characteristic. The sample first motion features include sample micro-doppler features and the sample second motion features include sample range-doppler features. The sample micro Doppler characteristics comprise micro Doppler characteristics of a sample target in different motion states and micro Doppler characteristics of a sample non-target in different motion states, and the sample range Doppler characteristics comprise range Doppler characteristics of the sample target in different motion states and range Doppler characteristics of the sample non-target in different motion states. The different motion states of the sample target can be various walking, standing and sitting postures, the different motion states of the sample non-target can be different working states, swinging states and the like, and taking the fan as an example, the different motion states can be states of various angles, rotating speeds, swinging heads and the like of the fan.
The echo signal acquisition logic is shown in fig. 3, and when a sample target signal and a sample non-target interference signal are acquired, a first preset number of acquisition cycles are respectively acquired when the sample target and the sample non-target are respectively in different motion states, each acquisition cycle includes a second preset number of modulation cycles, and each modulation cycle includes a third preset number of sampling points. During collection, a space environment with only an object needing to collect signals at present can be created as a test environment, for example, only one fan is placed in a room, so that independent sampling of the object is realized, and a data processing process of a sample target signal and a sample non-target interference signal can be simplified.
After a sample target signal and a sample non-target interference signal are collected, extracting micro Doppler characteristics, wherein the extraction process is as follows:
performing fast Fourier transform on the signal in each acquisition period to obtain a range-Doppler image corresponding to a second preset number of modulation periods in each acquisition period;
according to the range-Doppler image, performing accumulation operation on the Doppler information of the second preset number of modulation cycles in the acquisition time of the first preset number of acquisition cycles to obtain a micro-Doppler image;
and extracting micro Doppler features according to the micro Doppler image.
As shown in (a) - (d) of fig. 4, a schematic diagram of a process of obtaining a micro doppler map is shown, in which a first preset number is denoted as N, a second preset number is denoted as M, and a third preset number is denoted as W, and all of the numbers are integers greater than 0. Firstly, performing transverse one-dimensional FFT (fast fourier transform) on W sampling points of each modulation period, repeating M modulation periods, and performing two-dimensional FFT on each column to obtain a range-doppler plot corresponding to M modulation periods for each acquisition period, as shown in fig. 4 (a).
Next, the FFT data of W sampling points of the same modulation period are accumulated to obtain the superposition of doppler information at all distances, that is, the doppler information corresponding to M modulation periods for each acquisition period, as shown in fig. 4 (b).
Then, the doppler information of each acquisition cycle is accumulated over the acquisition time of N acquisition cycles, and assuming that each acquisition cycle is 0.01T seconds, the T seconds are accumulated to obtain a micro doppler map, specifically, M pieces of doppler information of the latest acquisition cycle are obtained each time, a T seconds window is slid, new data is inserted into the window, and the result shown in fig. 4 (c) is obtained.
Finally, as shown in fig. 4 (d), the maximum value is taken for each column of data to form a waveform aa in the graph, which is used to represent the main frequency variation of the object, an upper envelope waveform bb is formed for the peak of the waveform aa, and a lower envelope waveform cc is formed for the valley of the waveform aa. For the waveform aa, the waveform is represented as S1(n), 0.01T seconds obtain one data, T seconds accumulate 100 points of data, and perform FFT operation thereon
Taking the intensity Pc1 and the corresponding frequency Fc1 corresponding to the maximum value of the signal, and taking the intensity of the next largest peak as Pc2 and frequency Fc2, then obtaining the motion period delta Fc of the object as | Fc1-Fc2 |; taking the Doppler frequency Fua corresponding to the peak and the Doppler frequency Fub corresponding to the trough of the upper envelope waveform bb, taking the Doppler frequency Fda corresponding to the peak and the Doppler frequency Fdb corresponding to the trough of the lower envelope waveform cc, and calculating the total Doppler signal of the obtained objectThe bandwidth is BW ═ Fua-Fdb |, and the Offset of the total doppler of the available objects is calculated as | Fua-Fdb |/2. As such, the micro-doppler signature includes: the intensity Pc1 and frequency Fc1 of the dominant motion doppler, the motion period Δ Fc, the total doppler bandwidth BW, the total doppler Offset.
After the sample target signal and the sample non-target interference signal are collected, the range-Doppler characteristic is extracted, and the extraction process is as follows:
performing fast Fourier transform on the signal in each acquisition period to obtain a range-Doppler image corresponding to a second preset number of modulation periods in each acquisition period;
performing a modulo operation on the range-doppler plot to obtain a speed and a distance, and calculating a mean and a variance of the power of a first preset number of acquisition cycles;
and determining the range-Doppler characteristics according to the mean value and variance of the speed, the range and the power.
As described above, the signal of each acquisition period in the signal is subjected to fast fourier transform to obtain a range-doppler plot corresponding to M modulation periods for each acquisition period, and then the range-doppler plot is subjected to modulo calculation, wherein in the modulo result, the horizontal direction represents the distance direction, the higher the frequency is, the farther the distance is, the longitudinal direction represents the speed, and the higher the frequency is, the faster the speed is. In addition, based on the modulo result, the power corresponding to M modulation periods for each acquisition period may also be calculated, and the mean and variance of the power may be obtained by performing an accumulation operation on the power corresponding to M modulation periods for each acquisition period over the acquisition time of N acquisition periods, for example, if the acquisition period is 0.01T seconds, the T seconds time is accumulated, and the mean and variance of 100 acquisitions may be calculated. The formula used in the above operation process is as follows:
wherein F (M, N) represents a matrix of size M x N, where M is 0,1,2, M-1, N is 0,1,2, N-1, F (u, v) represents the fourier transform of F (M, N);
the spectral amplitude of the fourier transform is represented as: if (u, v) | R2(u，v)+I2(u，v)]1/2Wherein R represents a real part of Fourier transform, and I represents an imaginary part of Fourier transform;
the power level is expressed as:
P(u，v)＝|F(u，v)|2＝[R2(u，v)+I2(u，v)]；
the mean power of 100 acquisition cycles is expressed as:
wherein, P (u, v) is the power of an object, and k represents the accumulated times;
the power square difference for 100 acquisition cycles is:
where P (u, v) is the power level of an object, and k represents the number of accumulations.
Thus, the range-doppler signature includes: mean value of reflected signal power
Variance s of power2Distance R, velocity magnitude v.
After the first motion characteristic of the sample and the second motion characteristic of the sample are obtained, corresponding sample labels are marked on the first motion characteristic of the sample and the second motion characteristic of the sample to obtain training data, and the sample labels are used for distinguishing the characteristics which are corresponding to the target from the characteristics which are not corresponding to the target. And then, dividing the training data into a test set and a training set, and training the classifier through the test set and the training set to obtain a target classification model. Taking a training KNN classifier as an example, firstly setting a KNN classifier, inputting the feature vectors and sample labels of all training sets into the classifier, calculating the distance between each sample of a test set and all samples of the training set, outputting and adjusting to obtain a proper K value, and finishing the training of the classifier.
By the method, when the target classification model is trained, the micro Doppler characteristic and the range Doppler characteristic of the target and the non-targets in different motion states are adopted, so that the small motion difference characteristic between the non-targets and the target can be effectively identified, and the target classification model has the capability of accurately identifying non-target interference signals.
Before the target classification model is applied to the classification in step 206, the echo signals in the target environment need to be acquired in step 202, and when the signals are acquired, a first preset number of acquisition cycles are acquired respectively, each acquisition cycle includes a second preset number of modulation cycles, and each modulation cycle includes a third preset number of sampling points. Next, feature extraction is performed on the echo signal of the target environment through step 204 to obtain a first motion feature and a second motion feature. When the first motion feature includes a micro doppler feature, the step 204 performs feature extraction on the echo signal to obtain a first motion feature and a second motion feature, including:
performing fast Fourier transform on the signal of each acquisition period in the echo signals to obtain a range-Doppler image of each acquisition period, which corresponds to a second preset number of modulation periods;
according to the range-Doppler diagram, performing accumulation operation on the Doppler information of the modulation cycles of the second preset number on the acquisition time of the acquisition cycles of the first preset number to obtain a micro-Doppler diagram;
and extracting micro Doppler features according to the micro Doppler image.
In one embodiment, when the second motion feature includes a range-doppler feature, the step 204 performs feature extraction on the echo signal to obtain a first motion feature and a second motion feature, including:
performing fast Fourier transform on the signal of each acquisition period in the echo signals to obtain a range-Doppler image corresponding to a second preset number of modulation periods in each acquisition period;
performing modular operation on the range-doppler diagram to obtain the speed and the distance of the target, and calculating the mean value and the variance of the power of the first preset number of acquisition cycles;
and determining the range-Doppler characteristics according to the mean value and variance of the speed, the range and the power.
The above process of extracting the micro doppler feature and the range-doppler feature of the echo signal in the target environment is the same as the process of extracting the sample micro doppler feature and the sample range-doppler feature of the sample target signal and the sample non-target interference signal during training, and is not described herein again. The micro Doppler characteristics and the range Doppler characteristics of echo signals of a target environment are used as samples to be detected to be input into a classifier for prediction, K target categories with the closest distance to all the characteristics are selected, one category with the most target categories is output, and then the signals which are identified as non-target categories can be filtered.
Fig. 5 is a flowchart illustrating an interference signal filtering method according to another embodiment. As shown in fig. 5, the step 202 of acquiring the echo signal corresponding to the target environment may include the following steps:
the target environment is an environment in which the millimeter wave radar is installed and applied, and is an indoor space such as a room, a living room, a balcony and the like. One or more millimeter wave radars may be installed in the target environment to enable the millimeter wave radars to fully monitor objects in the space. The object includes a target and a non-target, wherein the target refers to an object needing monitoring, such as a human body, and the non-target refers to an object which interferes with the monitoring of the target, such as a fan, an air conditioner, a curtain, a green plant, a pet and the like in the environment. The original echo signal of the target environment comprises an echo signal reflected by a non-target in the target environment, i.e. comprises a non-target interfering signal. When there is a person or other target in the target environment, the original echo signal will simultaneously include the echo signal reflected by the target in the target environment, that is, the target signal.
the original echo signals adopt the same acquisition logic as the above embodiment, and the fast fourier transform is performed on the echo signals of each acquisition cycle in the original echo signals, so as to obtain a range-doppler diagram of the object in the target environment in each acquisition cycle, which corresponds to a second preset number of modulation cycles. And then, carrying out angle estimation on a plurality of equivalent echo signals contained in the radar antenna array, wherein the angle estimation can be a phase time difference method such as three-dimensional FFT (fast Fourier transform), a beam forming method such as Capon, Music and the like, and point cloud characteristics of all objects in the space, including distance, speed, horizontal angle and pitch angle, are obtained.
Specifically, based on the same algorithm of the above embodiment, a two-dimensional range-doppler plot FFT result F (u, v) of a single antenna can be obtained, meanwhile, the power magnitude P (u, v) of the object is obtained, the power response threshold TH1 is set, and the distance R and the velocity v corresponding to the object exceeding TH1 are extracted. Then, for the millimeter wave radar antenna array, the number of antennas in the horizontal direction is P, the number of antennas in the vertical direction is Q, and the three-dimensional FFT is calculated for the signal in the horizontal direction as follows:
the corresponding power levels are expressed as follows:
PA(u，v)＝|A(u，v，l)|2＝[R2(u，v，l)+I2(u，v，l)}；
the vertical direction signal computes a three-dimensional FFT as follows:
the corresponding power levels are expressed as follows:
PE(u，v)＝|E(u，v，l)|2＝[R2(u，v，l)+I2(u，v，l)}。
a threshold TH2 is set for the response to the power spectrum, and the peak of the spectrum that exceeds the threshold corresponds to the index, i.e., the angular information IE of the horizontal and vertical directions IA corresponding to the object.
Thus, the point cloud features of the object include: distance R, velocity magnitude v, horizontal angle IA, vertical angle IE, power magnitudes PA and PE.
in one embodiment, determining a signal belonging to an invalid region in an original echo signal according to a point cloud feature includes:
clustering the point cloud of each object in the target environment according to the point cloud characteristics;
and determining signals belonging to invalid regions in the original echo signals according to the clustering result.
The method comprises the steps of clustering point clouds of objects in a target environment according to point cloud characteristics, extracting objects with large and dense point cloud data by using a clustering method such as KMeans and DBSCAN, classifying the object point clouds with large pieces of linear arrangement into walls, classifying objects with high and low pitch angle height information into the ground or ceilings, such as pets, sweeping robots, hanging air conditioners and the like, and classifying coordinates of positions of the objects as point cloud invalid areas so as to determine signals belonging to the invalid areas in original echo signals.
And step 508, filtering out signals belonging to the invalid region in the original echo signals to obtain filtered echo signals of the target environment.
The echo signal of the target environment after filtering, that is, the echo signal of the target environment used in the above embodiment, is obtained after filtering the signal belonging to the invalid region in the original echo signal in the target environment, so that the data processing amount of the echo signal can be reduced, and the response speed can be improved.
Referring to fig. 6, in the present embodiment, the steps 502 to 508 are performed to perform the preliminary filtering on the original echo signal in the target environment, and then the steps 204 to 208 are performed to perform the secondary filtering, so as to reduce the data processing amount on the echo signal and improve the response speed. The specific limitations of steps 204 to 208 and the training process of the target classification model may refer to the limitations of the interference signal filtering method in the above embodiments, which are not described herein again.
Fig. 7 is a flowchart illustrating a method for filtering an interference signal according to another embodiment. As shown in fig. 7, after filtering out the non-target interference signal, the method of the present application further includes:
after the non-target interference signals are filtered, the target signal can be obtained, and then the target detection result can be obtained according to the target signal. The target detection result includes, but is not limited to, at least one of a target position, a target posture and a target action.
After determining the target detection result based on the filtered target signal, it may be further determined whether the target detection result satisfies a device control condition, for example, whether a control trigger condition in a device linkage control scheme or an automation control scheme is satisfied. If the target detection result meets the requirement, a control instruction of the corresponding target equipment is generated according to the target detection result, and the control of the corresponding target equipment is realized, so that equipment linkage control or automatic control based on the target detection result is realized, and the control accuracy is improved
The application also provides an application scenario of the smart home, and the application scenario applies the interference signal filtering method of the embodiment. Specifically, the application of the interference signal filtering method in the application scenario is as follows:
the target is a human body, and the non-target is an object other than a human body, such as a fan, a curtain, an animal, and the like. The method comprises the steps of detecting echo signals of non-targets and targets through a millimeter wave radar, carrying out region division and feature extraction on the echo signals to obtain micro Doppler features and range Doppler features of the non-targets and the targets, inputting the micro Doppler features and the range Doppler features into a trained target classification model, determining non-target interference signals in the echo signals and filtering the non-target interference signals, and further obtaining information such as existence of a human body, position of the human body and the like. Then, after the accurate position of the human body is detected, the application such as scene linkage control or automatic control can be carried out according to the current position of the human body, for example, the current position of the human body is at a sofa, the floor lamp beside the sofa can be automatically controlled to be turned on, and the sofa is more intelligent, convenient and accurate.
Fig. 8 is a block diagram of an interference signal filtering apparatus according to an embodiment. As shown in fig. 8, the present application also provides an interference signal filtering apparatus, including:
a signal obtaining module 802, configured to obtain an echo signal corresponding to a target environment;
a feature obtaining module 804, configured to perform feature extraction on the echo signal to obtain a first motion feature and a second motion feature;
the signal classification module 806 inputs the first motion characteristic and the second motion characteristic into a target classification model, and determines a non-target interference signal in the echo signal;
and a signal filtering module 808 for filtering out non-target interference signals.
In one embodiment, the first motion characteristic comprises a micro-doppler characteristic and the second motion characteristic comprises a range-doppler characteristic; the echo signals corresponding to the target environment are echo signals acquired in a first preset number of acquisition cycles, and each acquisition cycle comprises a second preset number of modulation cycles; a feature extraction module 804, comprising:
the first transformation unit is used for performing fast Fourier transformation on the echo signals in each acquisition cycle in the echo signals to obtain a range-Doppler diagram corresponding to a second preset number of modulation cycles in each acquisition cycle;
the accumulation operation unit is used for carrying out accumulation operation on the Doppler information of the modulation cycles of the second preset number in the acquisition time of the first preset number according to the range Doppler image to obtain a micro Doppler image;
and the micro Doppler feature extraction unit is used for extracting micro Doppler features according to the micro Doppler image.
In one embodiment, the first motion characteristic comprises a micro-doppler characteristic and the second motion characteristic comprises a range-doppler characteristic; the echo signals corresponding to the target environment are echo signals acquired in a first preset number of acquisition cycles, and each acquisition cycle comprises a second preset number of modulation cycles; a feature extraction module 804, comprising:
the second transformation unit is used for performing fast Fourier transformation on the echo signals in each acquisition cycle in the echo signals to obtain a range-Doppler diagram corresponding to a second preset number of modulation cycles in each acquisition cycle;
the calculation unit is used for carrying out modular operation on the range-doppler diagram, acquiring speed and distance and calculating the mean value and variance of the power of the acquisition cycles of the first preset number;
and the range-Doppler characteristic determining unit is used for determining the range-Doppler characteristic according to the mean value and the variance of the speed, the distance and the power.
In an embodiment, the apparatus further includes:
the system comprises a sample signal acquisition module, a signal processing module and a signal processing module, wherein the sample signal acquisition module is used for acquiring a sample target signal and a sample non-target interference signal;
the sample characteristic extraction module is used for respectively extracting a sample first motion characteristic and a sample second motion characteristic corresponding to the sample target signal and the sample non-target interference signal, and marking the sample first motion characteristic and the sample second motion characteristic with corresponding sample labels;
the prediction module is used for carrying out target classification on the first motion characteristics of the sample and the second motion characteristics of the sample through the initial classification model and determining a class prediction result corresponding to the first motion characteristics of the sample and the second motion characteristics of the sample;
and the training adjusting module is used for adjusting parameters of the initial classification model and continuing training according to the difference between the class prediction results corresponding to the first motion characteristics of the sample and the second motion characteristics of the sample and the sample labels of the first motion characteristics of the sample and the second motion characteristics of the sample, and stopping training until the training conditions are met to obtain the target classification model.
In one embodiment, the sample signal acquisition module is configured to:
acquiring sample target signals acquired under different motion states of a sample target and sample non-target interference signals acquired under different motion states of a plurality of non-sample targets respectively;
a sample feature extraction module comprising:
the characteristic extraction unit is used for respectively extracting the characteristics of the sample target signals and the sample non-target interference signals corresponding to different motion states to obtain sample first motion characteristics corresponding to the different motion states of the sample target and sample second motion characteristics corresponding to the different motion states of the sample non-target;
and the sample label marking unit is used for marking the corresponding sample labels for the sample first motion characteristics and the sample second motion characteristics.
For the specific definition of the interference signal filtering device, reference may be made to the definition of the interference signal filtering method in the above embodiments, which is not described herein again. All or part of the modules in the interference signal filtering device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
Fig. 9 is a block diagram of an interference signal filtering apparatus in another embodiment, as shown in fig. 9, the signal obtaining module 802 may include:
a first signal acquiring unit 902, configured to acquire an original echo signal of a target environment;
a point cloud feature obtaining unit 904, configured to perform feature extraction on the original echo signal to obtain a point cloud feature of an object in a target environment;
a signal clustering unit 906, configured to determine, according to the point cloud feature, a signal belonging to an invalid region in the original echo signal;
the signal filtering unit 908 is configured to filter signals belonging to an invalid region in the original echo signal, so as to obtain an echo signal of the filtered target environment.
In an embodiment, the signal clustering unit 908 is configured to:
clustering the point cloud of each object in the target environment according to the point cloud characteristics;
and determining signals belonging to invalid regions in the original echo signals according to the clustering result.
For the specific definition of the interference signal filtering device, reference may be made to the definition of the above method embodiment, which is not described herein again. All or part of the modules in the interference signal filtering device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent of a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
Fig. 10 is a block diagram of a structure of an interference signal filtering apparatus in another embodiment, as shown in fig. 10, the apparatus of the present application includes:
a target information determining module 1002, configured to determine a target detection result according to the filtered target signal;
the control module 1004 is configured to send a corresponding control instruction to the target device to control the target device to execute the control instruction if the target detection result meets the device control condition.
For the specific definition of the interference signal filtering device, reference may be made to the definition of the above method embodiment, which is not described herein again. All or part of the modules in the interference signal filtering device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, an electronic device is provided, and the electronic device may be a server, and the internal structure thereof may be as shown in fig. 11. The electronic device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the electronic device is configured to provide computing and control capabilities. The memory of the electronic equipment comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the electronic device is used for storing interference signal filtering data. The network interface of the electronic device is used for connecting and communicating with an external terminal through a network. The computer program is executed by a processor to implement a method of interference signal filtering.
Those skilled in the art will appreciate that the architecture shown in fig. 11 is a block diagram of only a portion of the architecture associated with the subject application, and does not constitute a limitation on the electronic devices to which the subject application may be applied, and that a particular electronic device may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, an electronic device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 12. The electronic device comprises a processor, a memory, a network interface, a display screen and an input device which are connected through a system bus. Wherein the processor of the electronic device is configured to provide computing and control capabilities. The memory of the electronic equipment comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The network interface of the electronic device is used for connecting and communicating with an external terminal through a network. The computer program is executed by a processor to implement a method of interference signal filtering. The display screen of the electronic equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the electronic equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the electronic equipment, an external keyboard, a touch pad or a mouse and the like.
Those skilled in the art will appreciate that the structure shown in fig. 12 is a block diagram of only a portion of the structure relevant to the present disclosure, and does not constitute a limitation on the electronic device to which the present disclosure may be applied, and that a particular electronic device may include more or less components than those shown, or combine certain components, or have a different arrangement of components.
In one embodiment, an electronic device is provided, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method embodiments when executing the computer program. For specific limitations of the electronic device, reference may be made to the limitations of the interference signal filtering method in the above embodiments, and details are not repeated here.
In one embodiment, an electronic device is provided, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method embodiments when executing the computer program. :
for specific limitations of the electronic device, reference may be made to the limitations of the interference signal filtering method in the above embodiments, and details are not repeated here.
In one embodiment, a computer program product or computer program is provided that includes computer instructions stored in a computer-readable storage medium. The computer instructions are read by a processor of a computer device from a computer-readable storage medium, and the computer instructions are executed by the processor to cause the computer device to perform the steps in the above-mentioned method embodiments.
In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which computer program, when being executed by a processor, carries out the steps of the above-mentioned method embodiments. For specific limitations of the steps, reference may be made to limitations of interference signal filtering methods in various method embodiments, and details are not described herein.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (11)
1. A method for filtering an interference signal, the method comprising:
acquiring an echo signal corresponding to a target environment;
performing feature extraction on the echo signal to obtain a first motion feature and a second motion feature;
inputting the first motion characteristic and the second motion characteristic into a target classification model, and determining a non-target interference signal in the echo signal;
and filtering the non-target interference signal.
2. The method of claim 1, wherein the acquiring echo signals of a target environment comprises:
acquiring an original echo signal of the target environment;
extracting the characteristics of the original echo signals to obtain point cloud characteristics of the object in the target environment;
determining signals belonging to an invalid region in the original echo signals according to the point cloud characteristics;
and filtering signals belonging to an invalid region in the original echo signals to obtain filtered echo signals of the target environment.
3. The method of claim 2, wherein determining the signal corresponding to the null region in the raw echo signal from the point cloud features comprises:
clustering point clouds of all objects in the target environment according to the point cloud characteristics;
and determining signals belonging to invalid regions in the original echo signals according to the clustering result.
4. The method of claim 1 or 2, wherein the first motion characteristic comprises a micro-doppler characteristic; the echo signals corresponding to the target environment are echo signals acquired in a first preset number of acquisition cycles, and each acquisition cycle comprises a second preset number of modulation cycles; the extracting the characteristics of the echo signals to obtain a first motion characteristic and a second motion characteristic comprises the following steps:
performing fast Fourier transform on the echo signals of each acquisition period in the echo signals to obtain a range-Doppler diagram of each acquisition period, wherein the range-Doppler diagram corresponds to a second preset number of modulation periods;
according to the range-Doppler image, performing accumulation operation on the Doppler information of a second preset number of modulation cycles in the first preset number of acquisition time to obtain a micro-Doppler image;
and extracting micro Doppler features according to the micro Doppler image.
5. The method of claim 1 or 2, wherein the second motion characteristic comprises a range-doppler characteristic; the echo signals corresponding to the target environment are echo signals acquired in a first preset number of acquisition cycles, and each acquisition cycle comprises a second preset number of modulation cycles; the extracting the characteristics of the echo signals to obtain a first motion characteristic and a second motion characteristic comprises the following steps:
performing fast Fourier transform on the echo signal of each acquisition cycle in the echo signals to obtain a range-Doppler image of each acquisition cycle, which corresponds to a second preset number of modulation cycles;
performing a modulo operation on the range-doppler plot to obtain a speed and a distance, and calculating a mean and a variance of powers of a first preset number of acquisition cycles;
and determining the range-Doppler characteristics according to the mean value and the variance of the speed, the distance and the power.
6. The method of claim 1, further comprising:
obtaining a sample target signal and a sample non-target interference signal;
respectively extracting a sample first motion characteristic and a sample second motion characteristic corresponding to the sample target signal and the sample non-target interference signal, and marking the sample first motion characteristic and the sample second motion characteristic with corresponding sample labels;
performing target classification on the first motion characteristics of the sample and the second motion characteristics of the sample through an initial classification model, and determining a class prediction result corresponding to the first motion characteristics of the sample and the second motion characteristics of the sample;
and adjusting parameters of the initial classification model and continuing training according to the difference between the class prediction results corresponding to the first motion characteristics of the sample and the class prediction results corresponding to the second motion characteristics of the sample and the sample labels of the first motion characteristics of the sample and the second motion characteristics of the sample, and stopping training until the training conditions are met to obtain the target classification model.
7. The method of claim 6, wherein obtaining the sample target signal and the sample non-target interfering signal comprises:
acquiring sample target signals acquired under different motion states of a sample target and sample non-target interference signals acquired under different motion states of a plurality of non-sample targets respectively;
the extracting a sample first motion feature and a sample second motion feature corresponding to the sample target signal and the sample non-target interference signal, respectively, and labeling the sample first motion feature and the sample second motion feature with corresponding sample labels includes:
respectively extracting the characteristics of sample target signals and sample non-target interference signals corresponding to different motion states to obtain sample first motion characteristics corresponding to the different motion states of the sample target and sample second motion characteristics corresponding to the different motion states of the sample non-target;
and marking the corresponding sample label for the sample first motion characteristic and the sample second motion characteristic.
8. The method of claim 1, wherein after filtering the non-target interfering signal, the method further comprises:
determining a target detection result according to the target signal obtained by filtering;
and if the target detection result meets the equipment control condition, sending a corresponding control instruction to the target equipment to control the target equipment to execute the control instruction.
9. An interference signal filtering apparatus, comprising:
the signal acquisition module is used for acquiring an echo signal corresponding to a target environment;
the characteristic extraction module is used for extracting the characteristics of the echo signals to obtain a first motion characteristic and a second motion characteristic;
the classification module is used for inputting the first motion characteristic and the second motion characteristic into a target classification model and determining a non-target interference signal in the echo signal;
and the signal filtering module is used for filtering the non-target interference signal.
10. An electronic device comprising a processor, a memory and a computer program stored on the memory and executable on the processor, the computer program, when executed by the processor, implementing the steps of the interference signal filtering method according to any one of claims 1 to 8.
11. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the interference signal filtering method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210061661.XA CN114510967A (en) | 2022-01-19 | 2022-01-19 | Interference signal filtering method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210061661.XA CN114510967A (en) | 2022-01-19 | 2022-01-19 | Interference signal filtering method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114510967A true CN114510967A (en) | 2022-05-17 |
Family
ID=81549929
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210061661.XA Pending CN114510967A (en) | 2022-01-19 | 2022-01-19 | Interference signal filtering method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114510967A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115980676A (en) * | 2023-01-10 | 2023-04-18 | 扬州宇安电子科技有限公司 | Radar signal data analysis system and method based on big data |
-
2022
- 2022-01-19 CN CN202210061661.XA patent/CN114510967A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115980676A (en) * | 2023-01-10 | 2023-04-18 | 扬州宇安电子科技有限公司 | Radar signal data analysis system and method based on big data |
| CN115980676B (en) * | 2023-01-10 | 2023-09-19 | 扬州宇安电子科技有限公司 | Radar signal data analysis system and method based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wu et al. | Lameness detection of dairy cows based on the YOLOv3 deep learning algorithm and a relative step size characteristic vector | |
| Wagner et al. | Radar signal processing for jointly estimating tracks and micro-Doppler signatures | |
| CN112698322A (en) | Integration of tracking and classifier in MMWAVE radar | |
| CN115345908B (en) | Human body posture recognition method based on millimeter wave radar | |
| US8098186B2 (en) | Through-the-obstacle radar system and method of operation | |
| Rajagopalan et al. | Higher order statistical learning for vehicle detection in images | |
| CN111178331B (en) | Radar image recognition system, method, apparatus, and computer-readable storage medium | |
| Ni et al. | Gait-based person identification and intruder detection using mm-wave sensing in multi-person scenario | |
| WO2021248472A1 (en) | Target tracking method and apparatus based on ultra-wideband radar, device, and storage medium | |
| CN106842165A (en) | One kind is based on different distance angular resolution radar centralization asynchronous fusion method | |
| Li et al. | Human behavior recognition using range-velocity-time points | |
| CN111142102A (en) | Respiratory data calculation method and related equipment | |
| CN114814775A (en) | Radar fall detection method and equipment based on ResNet network | |
| CN114510967A (en) | Interference signal filtering method and device, electronic equipment and storage medium | |
| CN114814832A (en) | Millimeter wave radar-based real-time monitoring system and method for human body falling behavior | |
| CN109581343A (en) | Multi-radar network device and multi-direction object detection method | |
| Alujaim et al. | Human motion detection using planar array FMCW Radar through 3D point clouds | |
| CN111025241A (en) | Boundary area detection method and device, electronic equipment and storage medium | |
| Xie et al. | Lightweight midrange arm-gesture recognition system from mmwave radar point clouds | |
| CN107578036A (en) | A kind of depth image tumble recognizer based on wavelet moment | |
| Raeis et al. | InARMS: Individual activity recognition of multiple subjects with FMCW radar | |
| CN110398735A (en) | A kind of perception data processing method and system based on more radars | |
| CN109766737B (en) | Image processing-based indoor human body posture preliminary classification method | |
| TWI809241B (en) | Life form radar system, identification method, and database establishment method | |
| WO2022101390A2 (en) | Optimization of dataset |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |