CN119030502B - Adaptive dual-frequency bandpass filter based on intelligent algorithm - Google Patents

Adaptive dual-frequency bandpass filter based on intelligent algorithm Download PDF

Info

Publication number
CN119030502B
CN119030502B CN202411505029.5A CN202411505029A CN119030502B CN 119030502 B CN119030502 B CN 119030502B CN 202411505029 A CN202411505029 A CN 202411505029A CN 119030502 B CN119030502 B CN 119030502B
Authority
CN
China
Prior art keywords
adaptive
sensor
passband
covariance
impedance
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.)
Active
Application number
CN202411505029.5A
Other languages
Chinese (zh)
Other versions
CN119030502A (en
Inventor
陆建国
徐军
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Electronic Science and Technology of China
Original Assignee
University of Electronic Science and Technology of China
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by University of Electronic Science and Technology of China filed Critical University of Electronic Science and Technology of China
Priority to CN202411505029.5A priority Critical patent/CN119030502B/en
Publication of CN119030502A publication Critical patent/CN119030502A/en
Application granted granted Critical
Publication of CN119030502B publication Critical patent/CN119030502B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01P—WAVEGUIDES; RESONATORS, LINES, OR OTHER DEVICES OF THE WAVEGUIDE TYPE
    • H01P1/00—Auxiliary devices
    • H01P1/20—Frequency-selective devices, e.g. filters
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/048—Activation functions
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03H—IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
    • H03H21/00—Adaptive networks
    • H03H21/0012—Digital adaptive filters
    • H03H21/0025—Particular filtering methods
    • H03H21/0029—Particular filtering methods based on statistics
    • H03H21/003—KALMAN filters
    • H—ELECTRICITY
    • H03—ELECTRONIC CIRCUITRY
    • H03H—IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
    • H03H21/00—Adaptive networks
    • H03H21/0012—Digital adaptive filters
    • H03H21/0043—Adaptive algorithms

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Molecular Biology (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Probability & Statistics with Applications (AREA)
  • Control Of Motors That Do Not Use Commutators (AREA)

Abstract

The invention relates to the technical field of filters, in particular to an intelligent algorithm-based self-adaptive dual-band pass filter, which comprises a main body frame, a plurality of sensors arranged in the sensor cabin, a core filter structure arranged in the main body frame, an impedance tuning module arranged at an input end and an output end of the main body frame and used for adjusting the characteristics of a first passband and a second passband, an impedance analyzer chip arranged in the main body frame, and a main control unit electrically connected with the sensor cabin, the core filter structure, the impedance tuning module and the impedance analyzer chip, wherein the main control unit is used for executing a multi-source data fusion algorithm, the core filter structure is adjusted through the impedance tuning module according to the result of the multi-source data fusion algorithm, the first passband and the second passband are self-adaptively adjusted, and the filter can dynamically adjust the performance according to environmental changes and signal characteristics through the multi-source data fusion algorithm and a real-time adjustment mechanism.

Description

Self-adaptive dual-band-pass filter based on intelligent algorithm
Technical Field
The invention relates to the technical field of filters, in particular to an intelligent algorithm-based self-adaptive dual-band-pass filter.
Background
With the rapid development of modern radar technology, multi-band radar systems are widely used in the military and civil fields. These systems typically require processing signals in different frequency bands simultaneously, such as the GPS L1 band (1.575 GHz) for accurate positioning and the WiMAX band (3.5 GHz) for high speed data transmission. Thus, the high performance dual band pass filter (DBBPF) becomes a critical component in a multi-band radar system.
In recent years, researchers have proposed various methods of improving DBBPF performance. Among them, a common approach is to improve the performance of the filter by optimizing its physical structure. For example, a ladder impedance resonator, a mesh resonator, or a stub loading resonator is used. In addition, an intelligent algorithm is introduced to optimize the filter design process. For example, genetic algorithms or particle swarm optimization algorithms are used to find optimal filter parameters.
However, these prior art techniques still have some significant problems:
1. The lack of real-time adaptation capability-most existing DBBPF designs are static and cannot adjust the filter characteristics in real-time according to environmental changes and signal characteristics. In complex and varying electromagnetic environments, this may lead to a significant degradation of the filter performance.
2. Ambient sensing capabilities are inadequate-existing filter designs often lack the ability to sense the surrounding environment. They cannot acquire environmental parameters such as temperature, humidity, vibration, etc. in real time, and thus cannot be adjusted in time against environmental changes.
3. Multisource data fusion is inadequate-although some designs introduce intelligent algorithms, most algorithms are optimized based on only a single data source, ignoring complex interactions between multiple environmental factors and system parameters.
4. The feedback mechanism is imperfect, and the existing intelligent optimization method is off-line and lacks a real-time feedback mechanism. This results in the filter not being able to be dynamically adjusted according to the actual operating conditions.
5. The algorithm complexity is high, and although some intelligent optimization algorithms can improve the filter performance, the algorithm has higher calculation complexity and is difficult to run in real time in an embedded system with limited resources.
6. Reliability and stability problems-existing filters lack self-diagnosis and self-repair capabilities under long-term operation and severe environments, and are prone to performance degradation or failure.
Aiming at the problems, the invention provides an adaptive dual-band-pass filter based on an intelligent algorithm.
Disclosure of Invention
Aiming at the problems existing in the prior art, the invention provides an intelligent algorithm-based self-adaptive dual-band-pass filter, which is characterized in that the real-time self-adaptive adjustment of the filter characteristic is realized through a multi-source data fusion algorithm so as to adapt to complex and changeable electromagnetic environments.
The invention provides an intelligent algorithm-based self-adaptive dual-band pass filter, which comprises a main body frame, a plurality of sensors, a core filter structure, an impedance tuning module, an impedance analyzer chip and a main control unit, wherein independent sensor cabins are arranged in the main body frame, the plurality of sensors are arranged in the sensor cabins, the core filter structure is arranged in the main body frame and comprises two pairs of quarter-wavelength lambda/4 ladder impedance resonators and a branch loading three-mode resonator and is used for forming a first passband and a second passband, the impedance tuning module is arranged at an input end and an output end of the main body frame and is used for adjusting characteristics of the first passband and the second passband, the impedance analyzer chip is arranged in the main body frame, and the main control unit is electrically connected with the sensor cabins, the core filter structure, the impedance tuning module and the impedance analyzer chip, wherein the main control unit is used for executing a multi-source data fusion algorithm, and the core filter structure is adjusted through the impedance tuning module according to the result of the multi-source data fusion algorithm so as to achieve self-adaptive adjustment of the first passband and the second passband.
Specifically, the plurality of sensors include a temperature sensor array, a power detector, a standing wave ratio monitoring circuit, an ambient humidity sensor, and a vibration sensor, and are electrically connected with the main control unit through a high-speed digital interface.
Specifically, the impedance tuning module comprises a programmable waveform generator, a digital variable capacitor array and a variable inductor, and is electrically connected with the main control unit through an I2C bus.
Specifically, the impedance analyzer chip is electrically connected with the main control unit through an SPI interface.
The multi-source data fusion algorithm comprises the following steps of carrying out data preprocessing on raw data acquired by the plurality of sensors, carrying out sensor reliability evaluation based on the preprocessed data, carrying out adaptive Kalman filtering by utilizing the preprocessed data and the sensor reliability evaluation result, carrying out adaptive weight distribution according to the sensor reliability evaluation result, carrying out multi-source data fusion based on the adaptive Kalman filtering result and the adaptive weight, and outputting and feeding back the multi-source data fusion result to the impedance tuning module to adjust the characteristics of the first passband and the second passband.
The adaptive Kalman filtering step further comprises the substeps of carrying out non-Gaussian noise processing on the preprocessed data, carrying out adaptive gain adjustment based on the result of the non-Gaussian noise processing, and carrying out abnormal data detection and processing on the basis of the adaptive gain adjustment, wherein the result of the non-Gaussian noise processing is used for optimizing the follow-up substeps of adaptive gain adjustment and abnormal data detection and processing.
Specifically, the non-gaussian noise processing includes the steps of:
firstly, observing a data sequence to perform nuclear density estimation:
,
Wherein, As a function of the estimated probability density,For the number of samples to be taken,As a parameter of the bandwidth it is,As a function of the kernel,Is the firstA number of sample points are taken,Is the point to be estimated;
Then based on the estimated noise distribution, calculating the variance of the equivalent gaussian noise:
,
Wherein, As the covariance of the equivalent gaussian noise,In order to observe the value of the value,In order to observe the matrix,Is a state vector;
Finally, the said Substituting the standard Kalman filtering formula.
Specifically, the adaptive gain adjustment includes the steps of:
first, the innovation sequence covariance is calculated:
,
Wherein, In order to innovate the sequence covariance,In order to achieve a sliding window size,In order to create a sequence of innovations,,In order to observe the value of the value,In order to observe the matrix,Estimating for a priori state;
The theoretical innovation sequence covariance is then calculated:
,
Wherein, For the theoretical innovation of the sequence covariance,For the a priori estimation of the error covariance,To observe the noise covariance, k is the current time instant,Is an observation matrix;
Then defining an adaptive factor:
,
Wherein, Is an adaptive factor, tr [ ] is the trace of the matrix;
Finally, the Kalman gain is adjusted:
,
Wherein, Is the adjusted kalman gain.
Specifically, the abnormal data detection and processing includes the steps of:
Firstly, calculating the mahalanobis distance between an observed value and a predicted value:
,
Wherein, Is the square of the mahalanobis distance,For the theoretical innovation of the sequence covariance,Superscript (-1) indicates matrix inversion,For an a priori state estimate at time k,For the a priori estimation of the error covariance,Is the covariance of the equivalent gaussian noise, k is the current time instant,In order to observe the matrix,Is thatAn observation value of time;
then set a threshold value When (when)When the data is abnormal, judging the data as abnormal data;
then for outlier data, adjust the observed noise covariance:
,
Finally using the adjusted And carrying out Kalman filtering updating.
Specifically, the adaptive weight allocation step includes:
based on the sensor reliability evaluation result, calculating the weight of each sensor by using a Softmax function:
,
Wherein, Is the firstThe weight of the individual sensors is determined,Is the firstThe reliability score of the individual sensors,As a function of the natural index of refraction,For sum symbols, representing the sum of the values for all sensorsSumming; for indexing of sensors, the range is from 1 to the total number of sensors.
The self-adaptive dual-band-pass filter based on the intelligent algorithm has the following remarkable beneficial effects:
1. and the real-time self-adaption capability is that the filter can dynamically adjust the performance of the filter according to environmental change and signal characteristics through a multi-source data fusion algorithm and a real-time adjustment mechanism, so that the optimal working state is maintained.
2. The environment perception is enhanced, a plurality of sensors in the independent sensor cabin can monitor environmental parameters such as temperature, humidity, vibration and the like in real time, and comprehensive data support is provided for the self-adaptive algorithm.
3. The multi-source data fusion algorithm executed by the main control unit can comprehensively consider various environmental factors and system parameters, and achieves more accurate performance optimization.
4. And the complete feedback mechanism is that the filter can analyze the working state of the filter in real time and correspondingly adjust the working state through the impedance analyzer chip and the impedance tuning module to form closed loop control.
5. And the low-complexity algorithm adopts an optimized multi-source data fusion algorithm, reduces the calculation complexity while ensuring the performance, and is suitable for running in an embedded system.
6. The reliability and stability are improved, namely, the filter can keep stable performance under long-term working and severe environments through continuous self-monitoring and adjustment, and has certain self-diagnosis and repair capability.
7. The design method is not only suitable for the dual-band pass filter, but also can be extended to other types of adjustable filters, and has wide application prospect.
In a word, the self-adaptive dual-band-pass filter effectively solves the key problems of real-time self-adaptation, environment sensing, data fusion and the like in the prior art by introducing an intelligent algorithm and a multi-source data fusion technology. The innovative design provides a highly intelligent and high-performance filtering solution for the modern multi-band radar system, and is expected to play an important role in various advanced radar applications in the military and civil fields.
Drawings
FIG. 1 is a schematic diagram of a framework of an intelligent algorithm-based adaptive dual-band-pass filter of the present invention;
FIG. 2 is a flow chart of the multi-source data fusion algorithm of the present invention;
FIG. 3 is a circuit block diagram of a core filter structure of the present invention including geometric parameters;
FIG. 4 is a coupling topology of the core filter structure of the present invention, with solid and dashed lines representing direct coupling and cross coupling, respectively;
FIG. 5 is a block diagram of a branch-loaded three-mode resonator of the core filter structure of the present invention;
FIG. 6 is an odd mode equivalent circuit of a branch loaded three-mode resonator of the core filter structure of the present invention;
Fig. 7 is an even mode equivalent circuit of a branch loaded three-mode resonator of the core filter structure of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description of the specific implementation, structure, characteristics and effects thereof is given with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
With the rapid development of modern radar technology, multi-band radar systems are widely used in the military and civil fields. These systems typically require processing signals in different frequency bands simultaneously, such as the GPS L1 band for accurate positioning and the WiMAX band for high speed data transmission. Thus, the high performance dual band pass filter (DBBPF) becomes a critical component in a multi-band radar system.
The invention provides an intelligent algorithm-based self-adaptive dual-band pass filter, which is characterized in that the real-time self-adaptive adjustment of the filter characteristic is realized through a multi-source data fusion algorithm so as to adapt to complex and changeable electromagnetic environments.
As shown in fig. 1, the adaptive dual-band-pass filter based on the intelligent algorithm provided by the invention comprises:
A main body frame 10, wherein an independent sensor compartment 11 is arranged in the main body frame 10;
a plurality of sensors disposed within the sensor pod 11;
a core filter structure 12 disposed within the body frame 10, the core filter structure 12 for forming a first passband and a second passband;
An impedance tuning module 13 disposed at an input/output end of the main body frame 10, the impedance tuning module 13 being configured to adjust characteristics of the first passband and the second passband;
An impedance analyzer chip 14 disposed within the main body frame 10, and a main control unit 15 electrically connected to the sensor compartment 11, the core filter structure 12, the impedance tuning module 13, and the impedance analyzer chip 14.
Preferably, as shown in FIGS. 3-7, the core filter structure 12 of the present invention includes two pairs of quarter wavelength lambda/4 stepped impedance resonators (SSIR) and one stub loaded three-mode resonator (STR). Wherein, two pairs of lambda/4 SSIR are used to form a first passband (GPS L1 band, 1.575 GHz), STR is used to form a second passband (WiMAX band, 3.5 GHz). Specifically, the two pairs lambda/4 SSIR include a first resonator R1 and a fourth resonator R4, which constitute a first pair of resonators, and a second resonator R2 and a third resonator R3, which constitute a second pair of resonators. The coupling between R1 and R4, and R2 and R3 forms a first passband of the GPS frequency band.
Further, the STR comprises an odd mode o1 and two even modes e1 and e2, satisfying the frequency relationship of fe1 < fo1 < fe2. This design ensures that the WiMAX band has proper bandwidth and steep edges, and high-selectivity filtering is realized. It is worth noting that R1 and R4 not only participate in GPS band formation, but also provide a feed structure for WiMAX band coupling with source load, realizing multi-functionalization of resonators, further reducing circuit size.
Preferably, the present invention uses a Rogers5880 substrate design core filter structure 12, which has a thickness of 0.508mm and a relative dielectric constant of 2.2. The normalized circuit size of the filter is 0.08λg×0.32λg, where λg is the waveguide wavelength at the center frequency of the GPS band (1.575 GHz). The design not only realizes miniaturization of the filter, but also ensures excellent high-frequency performance and temperature stability.
The core filter structure 12 of the present invention also includes the following key geometric parameters (unit: mm):
L1 = 4.44, L2 = 4.52, L3 = 21.82, L4 = 11.44, L5 = 6.29, L6 = 4.92, L7 = 32.03, L8 = 0.77, L9 = 1.7, L10 = 5.5, L11 = 5.9, W0 = 1.55, W1 = 0.4, W2 = 1.23, W3 = 0.5, W4 = 3.1, W5 = 1.2, W6 = 0.5, W7 = 1.49, W8 = 1.2, S1 = 0.29, S2= 0.40, S3 = 0.12, S4 = 0.13, d1 = 2.22, d2 = 0.93.
For the first pair of The length of the high impedance section of the stepped impedance resonator,For the second pair ofThe length of the high impedance section of the stepped impedance resonator,For the first pair ofThe low impedance section length of the stepped impedance resonator,For the second pair ofThe low impedance section length of the stepped impedance resonator,The dendrites are loaded with the main transmission line length of the three-mode resonator,The branch is loaded with the auxiliary transmission line length of the three-mode resonator,Loading the open stub length of the three-mode resonator for the stub,Loading the short-circuit branch length of the three-mode resonator for the branch,-Loading open branch detail sizes of the three-mode resonator for the branches; For the input/output port width to be the same, Is thatThe width of the high impedance section of the stepped impedance resonator,Is thatThe width of the low impedance section of the stepped impedance resonator,-The transmission line width of the tri-mode resonator is loaded for the stub,-The length of the branch loading three-mode resonator is W8 represents the width of the short-circuit branch in the branch loading three-mode resonator,AndIs thatThe coupling gap between the stepped impedance resonators,AndLoading a tri-mode resonator for a stubThe coupling gap between the stepped impedance resonators,AndIs the position parameter of the through hole;
for the transmission line electrical length; the admittance is input for the even mode, Is the transmission line characteristic admittance; Inputting admittance for the odd mode;
Accurate control of these parameters is critical to achieving high performance of the filter. For example, L3 and L4 affect the center frequency and bandwidth of the GPS band, L7 affects the center frequency of the WiMAX band, S1 and S2 control the coupling strength of the GPS band, and S3 and S4 control the coupling strength of the WiMAX band. In particular, the larger width of W4 (3.1 mm) increases the power handling capability of the filter, making it suitable for high power radar systems.
Preferably, in one embodiment of the present invention, the main body frame 10 is made of an aluminum alloy 6061-T6 material, which has good thermal conductivity and corrosion resistance. The main body frame 10 has dimensions of 200mm×150mm×50mm (length×width×height), and has an independent sensor compartment 11 provided therein, and the sensor compartment 11 has dimensions of 100mm×80mm×30mm. This compact design facilitates the integration of the filter in a spatially limited radar system.
Further, the sensor pod is modular in design, equipped with 5 standardized sensor slots (50-pin ZIF interface). Each slot is connected to the main control unit 15 via a high-speed digital interface (LVDS, max 2.5 Gbps). The design scheme not only improves the flexibility and maintainability of the system, but also ensures high-speed and stable data transmission.
Specifically, the plurality of sensors 12 includes a temperature sensor array, a power detector, a standing wave ratio monitoring circuit, an ambient humidity sensor, and a vibration sensor;
The temperature sensor array adopts 4X Texas Instruments TMP to 117 with the precision of +/-0.1 ℃, the model of a power detector is Analog DEVICES ADL5519, the dynamic range is 60dB, the standing wave ratio monitoring circuit adopts Analog DEVICES AD to 8302, the precision of +/-0.5 dB, the ambient humidity sensor adopts sensor SHT85 with the precision of +/-1.5 percent RH, and the vibration sensor adopts INVENSENSE MPU-6050,3 shaft accelerometers and gyroscopes.
The configuration of the high-precision sensors enables the filter to comprehensively sense the working environment of the filter, and accurate input data is provided for a subsequent intelligent algorithm.
Preferably, in another embodiment of the present invention, the impedance tuning module 13 comprises an Analog DEVICES AD5930 programmable waveform generator, a PEREGRINE SEMICONDUCTOR PE64904 integrated digital variable capacitor array, and a Coilcraft 0603CS series variable inductor. The module communicates with the main control unit 15 via an I2C bus with a clock frequency of 400kHz. This configuration allows the filter to achieve fast and flexible impedance adjustment while maintaining high accuracy.
Further, the impedance analyzer chip 14 adopts Analog DEVICES AD5933, the frequency range is 1kHz-100MHz, the impedance measurement error is <0.5%, and the phase measurement error is <0.5 °. This high precision impedance analysis capability provides reliable basis data for adaptive tuning of the filter.
Next, the main control unit 15 includes a processor, a memory and an interface, the processor adopts an ARM Cortex-M7 kernel, an STM32H753VIT6, a main frequency of 480MHz, the memory is 1MB Flash,1MB SRAM, the interface includes 5 LVDS interfaces for sensor communication, an I2C interface for tuning module control, an SPI interface for impedance analysis, and a USB 2.0 interface for external communication.
This high performance main control unit 15 configuration ensures efficient execution of complex algorithms and seamless collaboration of parts of the system.
Referring to fig. 2, based on the above hardware structure, the core of the present invention is a multi-source data fusion algorithm executed by the main control unit 15. The algorithm comprises the following steps sequentially executed:
First, data preprocessing. In this step, a moving average filter is applied to the raw data for each sensor, with a window size of 5. Then, using the Min-Max normalization method:
,
Wherein, AndHistorical minimum and maximum values for each sensor.
The preprocessing method not only can effectively remove high-frequency noise, but also can unify the data of different sensors to the same scale, thereby laying a foundation for subsequent processing.
And secondly, evaluating the reliability of the sensor. This step calculates the signal-to-noise ratio (SNR) and data stability, and then synthesizes the scores:
,
,
,
Wherein, As the root mean square value of the signal,Is the root mean square value of the noise,As a result of the data standard deviation,. These weights are optimized through a large number of experiments, and can be well balanced in most application scenarios.
And thirdly, adaptive Kalman filtering. This step is the core innovation of the present invention and comprises the following three sub-steps:
a) Non-gaussian noise processing:
First, a kernel density estimation is performed on an observation data sequence:
,
Wherein, As a function of the estimated probability density,For the number of samples to be taken,As a parameter of the bandwidth it is,As a function of the kernel,Is the firstA number of sample points are taken,Is the point to be estimated.
Preferably, the present invention employs an adaptive bandwidth selection algorithm:
Wherein the method comprises the steps of IQR is the standard deviation and IQR is the quartile range. The method can automatically select the optimal bandwidth according to the characteristics of data distribution, and improves the accuracy of estimation.
Then, based on the estimated noise distribution, the variance of the equivalent gaussian noise is calculated:
,
Wherein, As the covariance of the equivalent gaussian noise,In order to observe the value of the value,In order to observe the matrix,Is a state vector.
B) Adaptive gain adjustment:
first, the innovation sequence covariance is calculated:
,
Wherein, In order to innovate the sequence covariance,For the sliding window size (20 in the present invention),Is an innovative sequence.
Next, the theoretical innovation sequence covariance is calculated:
,
Wherein, For the theoretical innovation of the sequence covariance,For a priori estimating the error covariance, k is the current time instant,Is an observation matrix. Then, an adaptation factor is defined:
,
Wherein, For the adaptation factor, tr [ ] is the trace of the matrix.
Finally, the kalman gain is adjusted:
,
Wherein, Is the adjusted kalman gain.
C) Abnormal data detection and processing:
First, the mahalanobis distance between the observed value and the predicted value is calculated:
,
Wherein, Is the square of the mahalanobis distance,For the theoretical innovation of the sequence covariance,Superscript (-1) indicates matrix inversion,For an a priori state estimate at time k,For the a priori estimation of the error covariance,Is the covariance of the equivalent gaussian noise, k is the current time instant,In order to observe the matrix,Is thatThe observed value of the moment is then set as a threshold value。
Preferably, the present invention employs an adaptive threshold:
,
Wherein the method comprises the steps of (Corresponding to a confidence level of 0.99),,. Such an adaptive threshold allows for a higher tolerance in the initial stages of the filter, tightening gradually over time.
When (when)And if the observed noise covariance is determined to be abnormal data, adjusting the observed noise covariance:
Finally, use the adjusted And carrying out Kalman filtering updating.
And fourthly, self-adaptive weight distribution. Based on the sensor reliability assessment results, weights for each sensor are calculated using a modified Softmax function:
,
Wherein, Is the firstThe weight of the individual sensors is determined,Is the firstThe reliability score of the individual sensors,As a function of the natural index of refraction,For sum symbols, representing the sum of the values for all sensorsSumming; for indexing of the sensors, the range from 1 to the total number of sensors, Is a temperature compensation coefficient. The temperature compensation coefficient is introduced to further improve the accuracy of weight distribution by taking into account the influence of temperature on the performance of different sensors.
And fifthly, multi-source data fusion. Data fusion is carried out by adopting a covariance cross-validation method:
,
,
Wherein the method comprises the steps of As a covariance matrix after the fusion,For the state estimation after the fusion,Covariance matrix estimated for each sensor,The filtering result of each sensor.
And sixthly, outputting and feeding back a result. The fusion result is output to an external system through a USB 2.0 interface (12 Mbps). Meanwhile, an adaptive PID control algorithm is realized based on the fusion result:
,
Wherein the method comprises the steps of 、、And the PID parameters are adjusted in real time by a fuzzy logic controller.
Preferably, the present invention employs the following parameter ranges:
;
these parameter ranges are obtained through a large number of simulations and experiments, and can achieve good control effects in most application scenarios.
In order to further improve the system performance, the invention also adopts the following optimization measures:
a) FPGA co-processing adopts Xilinx Artix-7 XC7A35T-1FTG256C FPGA to realize data preprocessing and Fast Fourier Transform (FFT) acceleration. The FPGA communicates with the main control unit 15 through an AXI4 interface, and the data transmission rate reaches 1GB/s.
B) And the real-time operating system adopts FreeRTOS to ensure the real-time performance and reliability of task scheduling. The system realizes the dynamic adjustment of task priority and optimizes the response performance of the system.
C) Electromagnetic compatibility (EMC) design, the PCB adopts 4-layer design, and the inner layer is a complete ground plane. The key signal lines adopt differential pair wiring, so that electromagnetic interference is reduced. EMI filters, such as TDK MPZ2012S102AT000, are added around the power supply inlet and the sensitive module.
D) And the heat dissipation design is adopted, and a Fischer Elektronik ICK BGA x 14 x 10 heat radiator is adopted by the main processor and the FPGA. A built-in temperature control fan is of the model Sunon MF40101V1-1000U-A99, and the rotating speed is automatically adjusted according to the temperature.
E) Software optimization, optimizing algorithm performance using ARM CMSIS-DSP libraries. To further optimize the system operating efficiency, we implement circular buffers and zero copy techniques to minimize data transfer overhead. Meanwhile, a dynamic memory allocation strategy is adopted, so that the memory use efficiency is optimized, and the stable performance of the system is ensured to be maintained in long-time operation.
It is noted that the adaptive dual band pass filter of the present invention is excellent in practical applications. The filter exhibits excellent performance in the GPS L1 band (1.575 GHz) and WiMAX band (3.5 GHz) tests. Specifically, in the GPS L1 band, the measured center frequency is 1.55GHz, the 3dB relative bandwidth is 13.4%, the insertion loss is 0.74dB, and the return loss is 17.4dB. In the WiMAX frequency band, the measured center frequency is 3.49GHz, the 3dB relative bandwidth is 7.8%, the insertion loss is 1.33dB, and the return loss is 12dB. These data fully demonstrate the advantages of the present invention in achieving high selectivity and low insertion loss.
More importantly, the filter of the invention implements five transmission zeros distributed at 1.28, 1.92, 2.68, 3.65 and 4.1GHz. The carefully designed transmission zeros significantly enhance the frequency selectivity and isolation of the two pass bands and effectively suppress interference in adjacent frequency bands. In addition, the 14-dB rejection degree stop band bandwidth of the filter can be extended from 3.63GHz to 9.83GHz, so that a wide stop band of 1.79f02 is realized, which is far superior to the prior art, and high-frequency harmonic interference is effectively inhibited.
In practical applications, the filter of the present invention exhibits excellent environmental suitability. The filter is capable of maintaining stable performance over a wide temperature range of-40 ℃ to 85 ℃ through temperature compensation and adaptive algorithms. Meanwhile, due to the adoption of the Rogers5880 substrate (the thickness is 0.508mm, and the relative dielectric constant is 2.2), the filter has excellent temperature stability and low loss characteristics, and is particularly suitable for being used in radar systems in severe environments.
In addition, the filter of the invention has excellent power processing capability. By reasonably designing the transmission line width (e.g., w4=3.1 mm), the filter can withstand input power up to 50W, meeting the needs of many high power radar systems. At the same time, the normalized circuit size of the filter is only 0.08λg×0.32λg (where λg is the waveguide wavelength at the center frequency of the GPS band), which makes it well suited for space-constrained radar systems.
In a word, the self-adaptive dual-band-pass filter based on the intelligent algorithm realizes a filtering solution with high integration, high performance and high reliability through an innovative hardware design and an advanced multi-source data fusion algorithm. The filter not only can process signals of the GPS L1 frequency band and the WiMAX frequency band simultaneously, but also can adapt to complex and changeable electromagnetic environments in real time, and provides key signal processing support for a modern multi-frequency band radar system.
It should be noted that the above-mentioned embodiments are only preferred embodiments of the present invention, and are not intended to limit the present invention, and any modifications, equivalent substitutions, improvements, etc. within the principle of the present invention should be included in the protection scope of the present invention.

Claims (4)

1.基于智能算法的自适应双频带通滤波器，其特征在于，包括：主体框架，所述主体框架内设有独立传感器舱；设置在所述传感器舱内的多个传感器；设置在所述主体框架内的核心滤波器结构，所述核心滤波器结构包括两对四分之一波长λ/4阶梯阻抗谐振器和一个枝节加载三模谐振器，用于形成第一通带和第二通带；设置在所述主体框架输入输出端的阻抗调谐模块，所述阻抗调谐模块用于调节所述第一通带和所述第二通带的特性；设置在所述主体框架内的阻抗分析仪芯片；以及与所述传感器舱、所述核心滤波器结构、所述阻抗调谐模块和所述阻抗分析仪芯片电连接的主控制单元；其中，所述主控制单元用于执行多源数据融合算法，根据所述多源数据融合算法的结果通过所述阻抗调谐模块调整所述核心滤波器结构，以实现所述第一通带和所述第二通带的自适应调节；1. An adaptive dual-frequency bandpass filter based on an intelligent algorithm, characterized in that it comprises: a main frame, wherein an independent sensor cabin is arranged in the main frame; a plurality of sensors arranged in the sensor cabin; a core filter structure arranged in the main frame, wherein the core filter structure comprises two pairs of quarter-wavelength λ/4 ladder impedance resonators and a branch-loaded three-mode resonator, which are used to form a first passband and a second passband; an impedance tuning module arranged at the input and output ends of the main frame, wherein the impedance tuning module is used to adjust the characteristics of the first passband and the second passband; an impedance analyzer chip arranged in the main frame; and a main control unit electrically connected to the sensor cabin, the core filter structure, the impedance tuning module and the impedance analyzer chip; wherein the main control unit is used to execute a multi-source data fusion algorithm, and adjust the core filter structure through the impedance tuning module according to the result of the multi-source data fusion algorithm to achieve adaptive adjustment of the first passband and the second passband; 所述多源数据融合算法包括以下依次执行的步骤：对所述多个传感器采集的原始数据进行数据预处理；基于所述预处理后的数据，进行传感器可靠性评估；利用所述预处理后的数据和所述传感器可靠性评估结果，执行自适应卡尔曼滤波；根据所述传感器可靠性评估结果，进行自适应权重分配；基于所述自适应卡尔曼滤波结果和所述自适应权重，执行多源数据融合；将所述多源数据融合的结果输出并反馈至所述阻抗调谐模块，以调整所述第一通带和所述第二通带的特性；The multi-source data fusion algorithm includes the following steps that are performed in sequence: preprocessing the raw data collected by the multiple sensors; performing sensor reliability assessment based on the preprocessed data; performing adaptive Kalman filtering using the preprocessed data and the sensor reliability assessment result; performing adaptive weight allocation based on the sensor reliability assessment result; performing multi-source data fusion based on the adaptive Kalman filtering result and the adaptive weight; outputting the result of the multi-source data fusion and feeding it back to the impedance tuning module to adjust the characteristics of the first passband and the second passband; 所述自适应卡尔曼滤波步骤进一步包括以下依次执行的子步骤：对所述预处理后的数据进行非高斯噪声处理；基于所述非高斯噪声处理的结果，执行自适应增益调整；在所述自适应增益调整的基础上，进行异常数据检测与处理；其中，所述非高斯噪声处理的结果用于优化后续的自适应增益调整和异常数据检测与处理子步骤；The adaptive Kalman filtering step further includes the following sub-steps executed in sequence: performing non-Gaussian noise processing on the pre-processed data; performing adaptive gain adjustment based on the result of the non-Gaussian noise processing; performing abnormal data detection and processing based on the adaptive gain adjustment; wherein the result of the non-Gaussian noise processing is used to optimize the subsequent adaptive gain adjustment and abnormal data detection and processing sub-steps; 所述非高斯噪声处理包括以下步骤：The non-Gaussian noise processing comprises the following steps: 首先观测数据序列进行核密度估计：First, observe the data sequence for kernel density estimation: 其中，f(y)为估计的概率密度函数，n为样本数量，h为带宽参数，K为核函数，yi为第i个样本点，y为待估计点；Where f(y) is the estimated probability density function, n is the number of samples, h is the bandwidth parameter, K is the kernel function, yi is the i-th sample point, and y is the point to be estimated; 然后基于估计的噪声分布，计算等效高斯噪声的方差：Then based on the estimated noise distribution, the variance of the equivalent Gaussian noise is calculated: Rk＝E[(yk-Hkxk)2]，R k =E[(y k -H k x k ) 2 ], 其中，Rk为等效高斯噪声的协方差，yk为观测值，Hk为观测矩阵，xk为状态向量；Among them, R k is the covariance of the equivalent Gaussian noise, y k is the observation value, H k is the observation matrix, and x k is the state vector; 最后将所述Rk代入标准卡尔曼滤波公式；Finally, substitute the R k into the standard Kalman filter formula; 所述自适应增益调整包括以下步骤：The adaptive gain adjustment comprises the following steps: 首先计算创新序列协方差： First calculate the innovation series covariance: 其中，Cv(k)为创新序列协方差，M为滑动窗口大小，v(i)为创新序列，y(i)为观测值，Hi为观测矩阵，为先验状态估计；Where C v (k) is the innovation sequence covariance, M is the sliding window size, v(i) is the innovation sequence, y(i) is the observation value, Hi is the observation matrix, is the prior state estimate; 然后计算理论创新序列协方差：Then calculate the theoretical innovation series covariance: 其中，Sk为理论创新序列协方差，为先验估计误差协方差，Rk为等效高斯噪声的协方差，k为当前时刻，Hk为观测矩阵；Among them, Sk is the covariance of the theoretical innovation sequence, is the prior estimation error covariance, R k is the covariance of the equivalent Gaussian noise, k is the current moment, and H k is the observation matrix; 接着定义自适应因子：Next, define the adaptive factor: 其中，λ(k)为自适应因子，tr[]为矩阵的迹；Among them, λ(k) is the adaptive factor, tr[] is the trace of the matrix; 最后调整卡尔曼增益：Finally adjust the Kalman gain: 其中，Kk为调整后的卡尔曼增益；Where K k is the adjusted Kalman gain; 所述异常数据检测与处理包括以下步骤：The abnormal data detection and processing comprises the following steps: 首先计算观测值与预测值之间的马氏距离：First, calculate the Mahalanobis distance between the observed and predicted values: 其中，为马氏距离的平方，Sk为理论创新序列协方差，上标(-1)表示矩阵求逆，为k时刻的先验状态估计，为先验估计误差协方差，Rk为等效高斯噪声的协方差，k为当前时刻，Hk为观测矩阵，yk为k时刻的观测值；in, is the square of Mahalanobis distance, S k is the covariance of theoretical innovation sequence, The superscript (-1) indicates matrix inversion. is the prior state estimate at time k, is the prior estimation error covariance, R k is the covariance of the equivalent Gaussian noise, k is the current moment, H k is the observation matrix, and y k is the observation value at moment k; 然后设定阈值γ，当时，判定为异常数据；Then set the threshold γ, when When , it is judged as abnormal data; 接着对于异常数据，调整观测噪声协方差：Then, for abnormal data, adjust the observation noise covariance: 最后使用调整后的进行卡尔曼滤波更新；Finally, use the adjusted Perform Kalman filter update; 所述自适应权重分配步骤包括：The adaptive weight allocation step comprises: 基于所述传感器可靠性评估结果，使用Softmax函数计算各传感器的权重：Based on the sensor reliability evaluation results, the weight of each sensor is calculated using the Softmax function: 其中，wi为第i个传感器的权重，ri为第i个传感器的可靠性评分，exp( )为自然指数函数，∑为求和符号，表示对所有传感器的exp(rj)进行求和；j为传感器的索引，范围从1到传感器总数。Where wi is the weight of the ith sensor, ri is the reliability score of the ith sensor, exp( ) is the natural exponential function, ∑ is the summation symbol, indicating the summation of exp( rj ) of all sensors; j is the index of the sensor, ranging from 1 to the total number of sensors. 2.根据权利要求1所述的滤波器，其特征在于，所述多个传感器包括温度传感器阵列、功率检测器、驻波比监测电路、环境湿度传感器和振动传感器，所述多个传感器通过高速数字接口与所述主控制单元电连接。2. The filter according to claim 1 is characterized in that the multiple sensors include a temperature sensor array, a power detector, a standing wave ratio monitoring circuit, an ambient humidity sensor and a vibration sensor, and the multiple sensors are electrically connected to the main control unit through a high-speed digital interface. 3.根据权利要求1所述的滤波器，其特征在于，所述阻抗调谐模块包括可编程波形发生器、数字可变电容器阵列和可变电感器，所述阻抗调谐模块通过I2C总线与所述主控制单元电连接。3. The filter according to claim 1 is characterized in that the impedance tuning module comprises a programmable waveform generator, a digital variable capacitor array and a variable inductor, and the impedance tuning module is electrically connected to the main control unit through an I2C bus. 4.根据权利要求1所述的滤波器，其特征在于，所述阻抗分析仪芯片通过SPI接口与所述主控制单元电连接。4. The filter according to claim 1, characterized in that the impedance analyzer chip is electrically connected to the main control unit through an SPI interface.
CN202411505029.5A 2024-10-28 2024-10-28 Adaptive dual-frequency bandpass filter based on intelligent algorithm Active CN119030502B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202411505029.5A CN119030502B (en) 2024-10-28 2024-10-28 Adaptive dual-frequency bandpass filter based on intelligent algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202411505029.5A CN119030502B (en) 2024-10-28 2024-10-28 Adaptive dual-frequency bandpass filter based on intelligent algorithm

Publications (2)

Publication Number Publication Date
CN119030502A CN119030502A (en) 2024-11-26
CN119030502B true CN119030502B (en) 2025-03-18

Family

ID=93531030

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202411505029.5A Active CN119030502B (en) 2024-10-28 2024-10-28 Adaptive dual-frequency bandpass filter based on intelligent algorithm

Country Status (1)

Country Link
CN (1) CN119030502B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119363072B (en) * 2024-11-27 2025-09-12 深圳市诺信博通讯有限公司 A waveguide bandpass filter and its optimized control system
CN119253222B (en) * 2024-12-02 2025-03-04 深圳市诺信博通讯有限公司 Intelligent tunable 5G circulator and system

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107124158A (en) * 2017-03-29 2017-09-01 同济大学 Wireless sensor network filtering information processing system and method based on logarithmic quantization
US12061277B1 (en) * 2023-08-10 2024-08-13 Beihang University Airborne positioning method in aviation navigation network based on multi-source information fusion

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9014790B2 (en) * 2013-06-03 2015-04-21 Fitbit, Inc. Heart rate data collection
CN114609912B (en) * 2022-03-18 2023-10-03 电子科技大学 Angle measurement target tracking method based on pseudo-linear maximum correlation entropy Kalman filtering
CN118763370A (en) * 2024-09-06 2024-10-11 电子科技大学 A radar filter assembly

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107124158A (en) * 2017-03-29 2017-09-01 同济大学 Wireless sensor network filtering information processing system and method based on logarithmic quantization
US12061277B1 (en) * 2023-08-10 2024-08-13 Beihang University Airborne positioning method in aviation navigation network based on multi-source information fusion

Also Published As

Publication number Publication date
CN119030502A (en) 2024-11-26

Similar Documents

Publication Publication Date Title
CN119030502B (en) Adaptive dual-frequency bandpass filter based on intelligent algorithm
CN108647372A (en) The active phase array antenna Electrical Analysis method and system of integrated temperature knowledge
Alessandri et al. New efficient full wave optimization of microwave circuits by the adjoint network method
CN106503359B (en) A Rapid Optimal Design Method of Microwave Window Based on NSGA-Ⅱ
CN116663476A (en) A High-Order High-Temperature Superconducting Filter and Its Circuit Topology Rapid Design Method
CN113191036B (en) Coplanar waveguide equivalent circuit structure and parameter extraction method thereof
Kretschmer et al. An improved algorithm for adaptive processing
CN109856964A (en) Moving target control method and system based on miss distance processing
Stošić et al. Stochastic signal propagation in five-section wideband branch-line hybrid
CN112039496B (en) Rapid Gaussian particle filter data fusion method based on artificial fish swarm optimization
Aghanim et al. Q-Learning Empowered Cavity Filter Tuning with Epsilon Decay Strategy.
Duan et al. Noise reduction analysis of deformation data based on CEEMD-PE-SVD modeling
CN114759361B (en) A phased array antenna based on interdigital capacitor loading and RCS reduction method
CN222691702U (en) An E-band high-gain millimeter-wave low-noise amplifier
CN120012622B (en) Optimization method, system and storage medium for microstrip converter and interface
CN115345105A (en) An Analog Circuit Optimization Algorithm Based on Multi-task Modeling Bayesian Optimization
Zhou et al. A novel metamaterial power divider optimized by artificial neural network
CN111965589B (en) Multi-Bernoulli Filter Multi-Source DOA Tracking Method Based on Spatial Smoothing
Ding et al. Multiobjective optimization of microwave circuits with many structural parameters and objectives
Mapare et al. Employment of artificial neural network in manipulating design constraints of rectangular micro strip patch antenna
CN120850689B (en) Unmanned aerial vehicle electronic compatibility evaluation method and system based on ADI-FDTD
CN221552125U (en) Microstrip filter phase shifter
Yu et al. Research on Converting and Connecting Process of Coaxial Connector to Microstrip Line for Microwave and Millimeter-Wave Modules
Koziel et al. Constrained Surrogates and Dimensionality Reduction for Low-Cost Multi-Objective Optimization of Compact Microwave Components
Zhang et al. Multi-Objective Cavity Filter Design Based on SR-BP Neural Network Surrogate Model Optimized by Slime Mould Algorithm

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
GR01 Patent grant
GR01 Patent grant