Radio frequency front-end circuit, control method thereof and multi-band wireless communication system

Abstract

The invention discloses a radio frequency front-end circuit and a control method thereof, and a multi-band wireless communication system. A coupling module generates a coupling signal and an output signal according to an original signal of an antenna, sends the coupling signal to a control module, and sends the output signal to a filtering module; wherein the original signal comprises a multi-band electromagnetic wave signal; the control module obtains a tuning control signal according to an interference signal in the coupling signal, and sends the tuning control signal to the filtering module; and a filtering module filters the output signal under the control of the tuning control signal to obtain an updated output signal after filtering. The circuit provided by the invention is small in size, low in power consumption and good in compatibility, and can process a multi-band signal received by an antenna in a multi-band wireless communication system; the circuit provided by the invention is high in response speed, can dynamically generate the tuning control signal so as to suppress interference signals in different scenes, meets the requirements of different scenes, and is high in flexibility.

Description

Technical Field

[0001] This application relates to the field of radio frequency front-end circuit technology, specifically to a radio frequency front-end circuit and its control method, and a multi-band wireless communication system. Background Technology

[0002] Multi-band wireless communication refers to wireless communication technology that transmits data simultaneously across multiple frequency bands. By using multiple frequency bands, multi-band wireless communication systems can effectively utilize electromagnetic waves of different frequencies, thereby increasing data transmission rates, expanding network capacity, and improving signal coverage and interference resistance. Current mobile vehicles and IoT devices are standardly equipped with BeiDou positioning, 4G / 5G communication, Wi-Fi, and Bluetooth modules. However, electromagnetic compatibility (EMC) issues caused by the coexistence of multiple frequency bands are becoming increasingly prominent, impacting the performance of multi-band wireless communication systems. How to reduce signal interference in multi-band wireless communication systems and improve system performance is a pressing technical problem that needs to be solved. Summary of the Invention

[0003] This application provides a radio frequency front-end circuit and its control method, as well as a multi-band wireless communication system, which can dynamically suppress interference signals in the multi-band wireless communication system, offering high flexibility; it is also small in size and has low power consumption, thereby improving the performance of the multi-band wireless communication system.

[0004] In a first aspect, embodiments of this application provide a radio frequency front-end circuit, including: a coupling module, a control module, and a filtering module. The coupling module is connected to an antenna, the control module, and the filtering module, respectively. The control module is also connected to the filtering module. The coupling module is used to: generate a coupling signal and an output signal based on the original signal from the antenna, send the coupling signal to the control module, and send the output signal to the filtering module. The original signal includes multi-band electromagnetic wave signals. The control module is used to: obtain a tuning control signal based on interference signals in the coupling signal and send the tuning control signal to the filtering module. The filtering module is used to: filter the output signal under the control of the tuning control signal to obtain a filtered and updated output signal.

[0005] In some embodiments of this application, the control module includes: a prediction unit connected to the coupling module, and a signal generation unit connected to the prediction unit; wherein the prediction unit is used to: predict the interference frequency of the interference signal in the coupling signal according to a machine learning prediction model; and the signal generation unit is used to: generate the tuning control signal according to the interference frequency.

[0006] In some embodiments of this application, the prediction unit includes a first prediction subunit, which is configured to: determine the frequency domain signal of the coupled signal; determine the signal pattern of the interference signal based on first feature data of the frequency domain signal; and predict the interference frequency within a preset time period based on the signal pattern; wherein the first feature data includes: power data, peak data, amplitude data, and harmonic data.

[0007] In some embodiments of this application, the prediction unit further includes a second prediction subunit, which is configured to: predict the interference type corresponding to the interference signal based on the second feature data of the frequency domain signal, so as to predict the interference frequency in the first prediction subunit based on the interference type and the first feature data of the frequency domain signal; wherein, the second feature data includes: power data, peak data, bandwidth data, time domain feature data, and intermodulation data.

[0008] In some embodiments of this application, the tuning control signal includes a tuning voltage, and the filtering module includes a tunable filter; wherein, under the control of the tuning voltage, the tunable filter adjusts the filtering parameters of the tunable filter, and filters the output signal based on the tunable filter after the filtering parameters are adjusted.

[0009] In some embodiments of this application, the filtering parameters include the center frequency and bandwidth of the filtering module.

[0010] In some embodiments of this application, the coupling module is connected to the antenna via a signal input terminal, to the control module via a coupling output terminal, and to the filtering module via a signal output terminal; wherein, the coupling module performs coupling processing on the original signal according to a preset ratio to obtain the coupled signal, and sends the coupled signal to the control module via the coupling output terminal; the output signal is obtained based on the original signal and the coupled signal, and is sent to the filtering module via the signal output terminal.

[0011] Secondly, embodiments of this application provide a control method for a radio frequency front-end circuit, comprising: generating a coupling signal and an output signal based on an original signal from an antenna; wherein the original signal includes multi-band electromagnetic wave signals; generating a tuning control signal based on an interference signal in the coupling signal; and filtering the output signal under the control of the tuning control signal to obtain a filtered and updated output signal.

[0012] In some embodiments of this application, obtaining the tuning control signal based on the interference signal in the coupled signal includes: predicting the interference type of the interference signal in the coupled signal according to a first prediction model, predicting the interference frequency corresponding to the interference type according to a second prediction model, and generating the tuning control signal based on the interference frequency.

[0013] Thirdly, embodiments of this application provide a multi-band wireless communication system, including: a radio frequency module, and a radio frequency front-end circuit as described in the first aspect.

[0014] The technical solution provided in this application embodiment involves a coupling module generating a coupling signal and an output signal based on the original signal from the antenna, sending the coupling signal to a control module, and sending the output signal to a filtering module. The original signal includes multi-band electromagnetic wave signals. The control module obtains a tuning control signal based on interference signals in the coupling signal and sends this tuning control signal to the filtering module. Under the control of the tuning control signal, the filtering module filters the output signal to obtain a filtered and updated output signal. The circuit in this application embodiment is small in size, has low power consumption, and good compatibility. Through the RF front-end circuit of this application embodiment, multi-band signals received by the antenna in a multi-band wireless communication system can be processed, filtering out interference signals between multi-band signals. This application embodiment has a fast response speed and can dynamically generate tuning control signals to suppress interference signals in different scenarios, meeting the needs of different scenarios and offering high flexibility. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a structural diagram of a radio frequency front-end circuit provided in an embodiment of this application;

[0017] Figure 2 This is a schematic diagram of a coupling module provided in an embodiment of this application;

[0018] Figure 3 This is a structural diagram of another radio frequency front-end circuit provided in an embodiment of this application;

[0019] Figure 4 This is a flowchart of a control method for a radio frequency front-end circuit provided in an embodiment of this application;

[0020] Figure 5 This is a flowchart of another control method for a radio frequency front-end circuit provided in an embodiment of this application.

[0021] Explanation of reference numerals in the attached figures:

[0022] 100. Coupling module; 200. Control module; 300. Filtering module. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] In the following description, specific embodiments of the invention will be illustrated with reference to steps and symbols performed by one or more computers, unless otherwise stated. Therefore, these steps and operations will be referred to several times as being performed by a computer, and computer execution as referred to herein includes operations by a computer processing unit representing electronic signals of data in a structured format. This operation transforms the data or maintains it at a location in the computer's memory system, which can be reconfigured or otherwise altered in a manner well known to those skilled in the art. The data structure maintained by the data is the physical location of the memory, which has specific characteristics defined by the data format. However, the principles of the invention described above are not intended to be limiting, and those skilled in the art will understand that many of the steps and operations described below can also be implemented in hardware.

[0025] The terms "module" or "unit" as used herein can be considered as software objects executing on the computing system. The various components, modules, engines, and services described herein can be considered as implementations on the computing system. While the apparatus and methods described herein are preferably implemented in software, they can also be implemented in hardware, both of which are within the scope of this invention.

[0026] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this specification means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when an element is “connected” or “coupled” to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein may include wireless connections or wireless coupling. The term “and / or” as used herein includes all or any units and all combinations of one or more associated listed items.

[0027] Multi-band wireless communication systems can transmit data simultaneously across multiple frequency bands. With the rapid development of the Internet of Things (IoT) and intelligent transportation technologies, the demand for integrated multi-band wireless communication devices has surged. For example, new electric bicycles are now standardly equipped with BeiDou positioning, 4G / 5G communication, Wi-Fi, and Bluetooth modules. As wireless communication technology advances, 5G / 6G technology requires direct connection between terminal devices and satellites, making multi-band wireless communication systems increasingly complex. In multi-band wireless communication systems, electromagnetic compatibility (EMC) issues caused by the coexistence of multiple frequency band signals are becoming increasingly prominent, affecting the performance of communication equipment.

[0028] Intermodulation interference in multi-band wireless communication systems refers to new frequency components generated when two or more frequency signals pass through nonlinear devices such as power amplifiers. For example, the third-order intermodulation product between the BeiDou B1 band (frequency 1561.098MHz) and the Wi-Fi 2.4GHz band is expressed as: 2fWi-Fi-fB1=2×2400MHz-1561.098MHz=3238.902MHz. Intermodulation interference signals may fall into other communication frequency bands, such as the 5G Sub-6GHz band, causing a decrease in the receiving sensitivity of other communication bands and potentially leading to malfunctions in IoT devices. Furthermore, the typical signal strength of the BeiDou B1 band is -130dBm, while the transmit power of Wi-Fi signals can reach 20dBm. Currently, for signals with low strength, such as the BeiDou B1 band, gain antennas are used for signal amplification, but this amplification introduces additional noise.

[0029] Currently, the most common approach is to use a fixed-frequency bandpass filter to filter out interference signals. Alternatively, separate antennas through spatial layout to achieve physical isolation. However, fixed-frequency bandpass filters cannot dynamically adapt to sudden interference, and physical isolation increases the size of the RF module, resulting in high hardware redundancy.

[0030] In view of this, embodiments of this application provide a radio frequency front-end circuit and its control method, as well as a multi-band wireless communication system, which can dynamically suppress interference signals in the multi-band wireless communication system, offering high flexibility; and is small in size and low in power consumption, thereby improving the performance of the multi-band wireless communication system.

[0031] Please see Figure 1 , Figure 1 This is a structural diagram of a radio frequency front-end circuit provided in an embodiment of this application. For example... Figure 1 As shown, the radio frequency front-end circuit includes: a coupling module 100, a control module 200, and a filtering module 300. The coupling module 100 is connected to the antenna, the control module 200, and the filtering module 300, respectively. The control module 200 is also connected to the filtering module 300.

[0032] like Figure 1 As shown, the coupling module 100 is connected to the antenna. The antenna receives electromagnetic wave signals and sends the received electromagnetic wave signals as the raw signal to the coupling module 100. The coupling module 100 generates a coupling signal and an output signal based on the raw signal from the antenna, sends the coupling signal to the control module 200, and sends the output signal to the filtering module 300. In the RF front-end circuit of this application, the raw signal received by the antenna includes multi-band electromagnetic wave signals, which can be applied to multi-band wireless communication systems.

[0033] The control module 200 obtains the tuning control signal based on the interference signal in the coupled signal and sends the tuning control signal to the filtering module 300. Under the control of the tuning control signal, the filtering module 300 performs filtering processing on the output signal to obtain the filtered and updated output signal.

[0034] The coupled signal is the signal obtained by coupling the original signal through coupling module 100, and can be understood as a part of the original signal. Therefore, the interference signal in the coupled signal has the same characteristics as the interference signal in the original signal. In this embodiment, the coupled signal output by coupling module 100 is analyzed and processed to determine the interference signal in the coupled signal. Then, a tuning control signal is generated based on the interference signal in the coupled signal. This tuning control signal is used to control the control filtering module 300, enabling the filtering module 300 to filter the interference signal in the output signal under the control of the tuning control signal, thereby removing the interference signal and obtaining a filtered and updated output signal.

[0035] The control module 200 receives the coupling signal output by the coupling module 100. The coupling module 100 is directly connected to the antenna and can couple and process the original signal received by the antenna to obtain the coupling signal, which is then sent to the control module 200. This allows the control module 200 to directly process and analyze the coupling signal generated from the original signal to obtain the corresponding interference signal in the original signal. Then, based on the predicted interference signal, a tuning control signal is generated to control the filtering module 300 to filter the interference signal in the output signal, resulting in a clean output signal after filtering.

[0036] This embodiment ensures that the control module 200 analyzes the original signal and predicts the corresponding interference signal. The control module 200 generates a tuning control signal based on the predicted interference signal from the coupled signal. When the coupled signal changes, the control module 200 can dynamically adjust to obtain a new tuning control signal, offering high flexibility. This embodiment can respond quickly even when the communication scenario changes, meeting the needs of dynamic scenarios and exhibiting strong anti-interference capabilities.

[0037] Please see Figure 2 , Figure 2 This is a schematic diagram of a coupling module 100 provided in an embodiment of this application. In some embodiments, such as Figure 2 As shown, the coupling module 100 is a directional coupler, including a signal input terminal, a coupling output terminal, an isolation terminal, and a signal output terminal. The coupling module 100 is connected to the antenna via the signal input terminal and to the filter module 300 via the signal output terminal, forming the main signal transmission channel. Furthermore, the coupling module 100 is also connected to the control module 200 via the coupling output terminal, outputting a coupling signal to the control module 200. The isolation terminal is used for signal isolation, reducing signal reflection and crosstalk, minimizing mutual interference between ports, and improving the quality of the output signal from the coupling module 100. The isolation terminal typically does not output a signal.

[0038] The coupling module 100 allocates power between the coupling output terminal and the signal output terminal according to the coupling degree, so as to extract a small amount of signal from the original signal as a coupling signal and send it to the control module 200 while minimizing the signal loss of the main channel. The coupling module 100 couples the original signal according to a preset ratio to obtain the coupling signal, and sends the coupling signal to the control module 200 through the coupling output terminal; then, it obtains the output signal based on the original signal and the coupling signal, and sends the output signal to the filtering module 300 through the signal output terminal. The coupling degree can be expressed as:

[0039] Coupling degree = 10log 10 (P-input / P-coupling);

[0040] The unit of coupling is dB. In the coupling degree calculation formula, Pinput represents the power of the original signal at the signal input terminal, and Pcoupling represents the power of the coupled signal output at the coupling output terminal. The power of the coupled signal is calculated based on the coupling degree, and then the power of the output signal can be determined based on the power of the original signal and the power of the coupled signal. Taking a coupling degree of 20 dB as an example, the power of the coupled signal is determined to be 1% of the power of the original signal, and correspondingly, the power of the output signal is 99% of the power of the original signal.

[0041] This application embodiment determines a preset ratio between the coupled signal and the original signal based on the coupling degree, thereby obtaining the coupled signal and the output signal. This application embodiment can also determine a tuning control signal for the interference frequency based on the coupled signal, and then filter the output signal according to the tuning control signal to remove interference signals from the output signal, thereby improving the performance of the wireless communication system.

[0042] Please see Figure 3 , Figure 3 This is a structural diagram of another radio frequency front-end circuit provided in an embodiment of this application. In some embodiments, such as Figure 3 As shown, the control module 200 includes a prediction unit and a signal generation unit; the prediction unit is connected to the coupling module 100 and is used to predict the interference frequency of the interference signal in the coupling signal according to the machine learning prediction model; the signal generation unit is connected to the prediction unit and is used to generate a tuning control signal according to the interference frequency.

[0043] Multi-band wireless communication systems involve electromagnetic wave signals across multiple frequency bands. Depending on the time and spatial dimensions of the signals, nonlinear intermodulation mechanisms, and the characteristics of the devices and protocols within the system, different forms of interference signals may exist. In other words, different application scenarios, different signal acquisition times, and different communication device states may all correspond to different interference signals. This embodiment incorporates a prediction unit in the control module 200, which uses a machine learning prediction model to predict the interference frequency of the interference signal, thereby improving the prediction efficiency and accuracy. After predicting the interference frequency, this embodiment uses a signal generation unit to generate a tuning control signal based on the real-time predicted interference frequency. Under the control of this tuning control signal, the filtering module 300 filters the interference signal, obtaining a clean, filtered electromagnetic wave signal as the updated output signal, thus improving the performance of the multi-band wireless communication system.

[0044] In some embodiments, the prediction unit includes a first prediction subunit, which is configured to: determine the frequency domain signal of the coupled signal; determine the signal pattern of the interference signal based on the first feature data of the frequency domain signal; and predict the interference frequency within a preset time period based on the signal pattern; wherein the first feature data includes: power data, peak data, amplitude data, and harmonic data.

[0045] Frequency domain signals are the representation of coupled signals in the frequency domain, and can be obtained by performing a Fourier transform on a time domain signal. Unlike time domain signals, which describe the continuous variation of a signal over time, frequency domain signals describe the energy distribution, amplitude, and phase characteristics of a signal at different frequencies. By analyzing frequency domain signals, the main frequency components and interfering frequencies in the signal can be identified, allowing for the design of filters to filter signals at specific frequencies.

[0046] Feature data represents data extracted from a frequency domain signal that characterizes the signal's features. Feature data can be used to determine the frequency components, energy distribution, and phase relationships of a frequency domain signal. Among the feature data of a frequency domain signal, power data can be the power distribution of the signal at different frequencies. The power spectrum can be obtained by squaring the amplitude spectrum. Power data can be used to determine the energy distribution of the frequency domain signal and identify the main frequency components and noise. Peak data can be the frequency corresponding to the maximum value in the frequency domain signal. Peak data can be used to determine the center frequency of the frequency domain signal. Amplitude data can be the magnitude of the energy of the frequency domain signal at different frequencies. Amplitude data can be used to determine the main frequency components and energy distribution of the frequency domain signal. Harmonic data can be the distribution of the fundamental frequency and its integer multiples in the frequency domain signal. Harmonic data can be used to analyze the periodicity of the frequency domain signal and identify harmonic distortion in the frequency domain signal.

[0047] As shown in the aforementioned embodiments, different interference signals exist depending on the time dimension, spatial dimension, nonlinear intermodulation mechanism, and device and protocol characteristics of the multi-band signals. The following provides a detailed explanation of each dimension.

[0048] In the time dimension, user behavior cycles, protocol timing mechanisms, or device operating modes can lead to regular interference signals. The interference signal pattern corresponding to user behavior cycles can manifest as increased interference signal density during peak Wi-Fi 2.4GHz usage periods in urban street scenarios. For example, the increased duty cycle of Wi-Fi 2.4GHz signals between 18:00 and 22:00 may increase the probability of third-order intermodulation products. The interference signal pattern corresponding to protocol timing mechanisms can manifest as periodic pulse interference caused by signaling. For example, the 1ms uplink scheduling cycle of Long Term Evolution (LTE) causes 800μs burst interference every 1ms in the 1560MHz band. The interference signal pattern corresponding to device operating modes, such as intermittent wake-up of low-power devices, can cause interference. For instance, in an underground parking garage, Bluetooth Low Energy (BLE) beacons broadcasting for 30ms every two seconds can cause a transient spike at 2402MHz, generating interference signals.

[0049] In the spatial dimension, building structure, electromagnetic environment complexity, or geographical topology can lead to regular interference signals. The interference signal pattern corresponding to building structure can manifest as prolonged frequency dwell time due to multipath reflection; for example, the concrete walls of an underground parking garage attenuate the 2.4GHz signal by up to 20dB, but the multipath effect extends the interference on channel 6 (2437MHz) to 200ms. The interference signal pattern corresponding to electromagnetic environment complexity can manifest as the superposition of intermodulation products in areas with high-density equipment; for example, in urban centers where multiple access points (APs) coexist, the third-order intermodulation products of 2.4GHz form a continuous noise floor in the 3238-3250MHz range. The interference signal pattern corresponding to geographical topology can manifest as differences in the interference spectrum between open spaces and enclosed scenarios; for example, the LTE uplink interference bandwidth at street intersections extends to 5MHz, while the interference bandwidth in indoor scenarios is typically 1MHz.

[0050] In nonlinear intermodulation mechanisms, third-order intermodulation, fifth-order intermodulation, and harmonic interference can lead to regular interference signals. The interference signal corresponding to third-order intermodulation can be expressed as: fIMD3 = 2f1 - f2 or fIMD3 = f1 + f2 - f3; for example, interference may occur between the Band 3 (1800MHz) signal of automotive LTE and the 2.4GHz Wi-Fi signal at a range of 2 × 1800 - 2400 = 1200MHz. The interference signal corresponding to fifth-order intermodulation can be expressed as: fIMD5 = 3f1 - 2f2; for example, the combination of 3.5GHz of 5G New Radio (NR) and Bluetooth 2.48GHz may cause interference at a range of 3 × 3500 - 2 × 2480 = 6540MHz, affecting the C-band. Harmonic interference can be expressed in relation to the fundamental frequency as: fharmonic = n × ffundamental, (n = 2, 3, ...); where fharmonic represents the harmonic frequency and ffundamental represents the fundamental frequency, which is the lowest frequency in the periodic waveform. The fundamental frequency can be considered as the dominant frequency in the signal. For example, the second harmonics of the B13 and B14 bands of LTE will generate interference signals in the Global Navigation Satellite System (GNSS) band.

[0051] Among the characteristics of equipment and protocols, transmit power level, modulation bandwidth, and multiple access mechanisms can lead to regular interference signals. The interference signal pattern corresponding to transmit power level can manifest as high-power equipment dominating and generating intermodulation signals, causing interference. For example, LTE Band 3 (1800MHz) in vehicles and Wi-Fi 2.4GHz signals can generate 2×1800-2400=1200MHz of interference. The interference signal pattern corresponding to modulation bandwidth can manifest as wideband signals causing spectrum leakage, leading to interference. For example, a 5G NR 100MHz carrier can cause out-of-band spurious signals in the adjacent BeiDou B3 band to rise by 15dB at 1268MHz, generating interference. The interference signal pattern corresponding to multiple access mechanisms can manifest as interference caused by channel occupancy conflicts due to contention protocols. For example, when Wi-Fi CSMA / CA backoff fails, channels 1-11 in the 2.4GHz band may experience continuous interference with an 80% duty cycle.

[0052] In summary, the occurrence of interference frequencies between multi-band signals exhibits certain patterns across time, space, nonlinear intermodulation mechanisms, and device and protocol characteristics. The first prediction subunit can analyze and process historical frequency domain signals to predict the signal patterns of interference signals within the multi-band signals. Based on these patterns, it can predict the interference frequencies that may occur within a preset time period, and determine the tuning control signal according to the predicted interference frequencies.

[0053] In some embodiments, the prediction unit further includes a second prediction subunit, which is configured to: predict the interference type corresponding to the interference signal based on the second feature data of the frequency domain signal, so as to predict the interference frequency in the first prediction subunit based on the interference type and the first feature data of the frequency domain signal; wherein the second feature data includes: power data, peak data, bandwidth data, time domain feature data, and intermodulation data.

[0054] In multi-band wireless communication systems, interference between signals is related to the operating frequency band, modulation technique, time-domain characteristics, and intermodulation interference modes of communication technologies such as Wi-Fi, Bluetooth, and cellular. In this application embodiment, before predicting the interference frequency through the first prediction subunit, a second prediction subunit can be used to predict the interference type corresponding to the interference signal, and then predict the corresponding interference frequency based on the interference type, thereby improving prediction efficiency.

[0055] First, let's introduce the operating frequency bands of different signals. Wi-Fi primarily operates in the 2.4GHz and 5GHz bands. Channel allocation can divide the 2.4GHz band into 14 channels and the 5GHz band into 25 channels. Channel overlap may lead to adjacent channel leakage, resulting in wideband continuous interference. Bluetooth primarily operates in the 2.4GHz band, corresponding to 79 1MHz channels. It may experience fast-hopping narrowband transient interference. Cellular networks primarily operate in the 700-2600MHz band for 4G LTE, the Sub-6GHz (3.3-4.2GHz) band for 5G NR, and millimeter wave. Channel allocation for LTE can be based on 1.4 / 3 / 5 / 10 / 20MHz.

[0056] To achieve variable bandwidth allocation, 5G signals use 100-400MHz carrier aggregation, which may lead to wideband blocking interference, especially during uplink burst transmissions.

[0057] Next, the modulation methods and waveform characteristics of different signals are introduced. Wi-Fi can be modulated based on Orthogonal Frequency Division Multiplexing (OFDM) technology. The signal has a peak-to-average power ratio (PAPR), such as a PAPR > 8dB. The PAPR affects the transmission efficiency and quality of the wireless signal, especially when there are bursts of transmission, the waveform may become more unstable, leading to signal distortion, increased bit error rate, and other problems. Bluetooth can be modulated using Gaussian Frequency Shift Keying (GFSK) or π / 4-Shifted Differential Quadrature Phase Shift Keying (π / 4-DQPSK). Bluetooth signals have a low duty cycle and are prone to producing brief transient spikes. Cellular networks can use quadrature phase shift keying (QPSK), 16QAM (quadrature amplitude modulation), or 64QAM quadrature amplitude modulation to transmit information through changes in phase or amplitude. Cellular networks employ a continuous frame structure to ensure stable communication.

[0058] The temporal characteristics of different signals are described below. Wi-Fi uses Carrier Sense Multiple Access with Collision Avoidance (CSMA / CA) transmission mode, avoiding collisions by monitoring channel conditions. Data packets are spaced at microsecond intervals, resulting in a high burst frequency characteristic and a packet duration of 1 to 4 milliseconds. Bluetooth uses fixed-slot frequency hopping transmission mode, resulting in a low Bluetooth signal duty cycle and a connection event interval of 7.5 milliseconds to 4 seconds. Cellular transmission uses timed, synchronized, continuous transmission, resulting in a high cellular signal duty cycle. The uplink scheduling cycle is related to Hybrid Automatic Repeat Request (HARQ) retransmissions.

[0059] Finally, the intermodulation interference modes between signals are explained. The third-order intermodulation product between Wi-Fi and BeiDou B1 can be expressed as 2f1-f2=2×2400-1561=3239MHz. Due to the high power of Wi-Fi devices and the saturation characteristics of nonlinear devices, Wi-Fi devices have high intermodulation sensitivity and are prone to interference in the 5GSub-6 GHz frequency band, such as 3.3-3.8GHz. Bluetooth is easily affected by neighboring Bluetooth channels, and its third-order intermodulation product can be expressed as 2f1-f2=2×2402-2405=2399MHz. Due to the low power of Bluetooth devices and the frequency hopping energy dissipation characteristics, Bluetooth devices have low intermodulation sensitivity. The third-order intermodulation product of cellular signals can be expressed as 2f1-f2=2×1800-900=2700MHz. Due to the high power of the base station during cellular communication, the receiver may be blocked, resulting in moderate intermodulation sensitivity of cellular equipment, which is prone to interference in the LTE Band 7 frequency band, such as 2500-2570MHz.

[0060] In summary, based on the analysis of the operating frequency bands, modulation techniques, time-domain characteristics, and intermodulation interference modes of different communication technologies, different interference types exist between multi-band signals. When predicting the interference frequency of an interfering signal, the signal type of the interfering signal can be predicted first, and then the interference frequency can be predicted based on the predicted interference type.

[0061] For example, the first prediction subunit can predict the interference frequency of the interference signal in the coupled signal using an LSTM model, while the second prediction subunit can predict the interference type in the interference signal using a random forest model. Combining the first and second prediction subunits, through their collaborative work, improves the prediction efficiency and reliability of interference prediction.

[0062] In some embodiments, the tuning control signal includes a tuning voltage, and the filtering module 300 includes a tunable filter; wherein, under the control of the tuning voltage, the tunable filter adjusts its filtering parameters, and filters the output signal based on the tunable filter with adjusted filtering parameters.

[0063] After predicting the interference frequency, the control module 200 converts the signal using a digital-to-analog converter to obtain a tuning control signal, and then controls the filtering module 300 to perform filtering based on the tuning control signal. The filtering module 300 uses the tuning control signal to filter out the interference signal corresponding to the interference frequency in the output signal, obtaining a clean output signal.

[0064] The filtering module 300 in this embodiment is a voltage-controlled tunable filter (VCF). The tuning control signal is the tuning voltage. The VCF can adjust the filtering parameters according to the applied tuning voltage so that the VCF can filter the predicted interference frequency.

[0065] In some embodiments, the filtering parameters include the center frequency and bandwidth of the filtering module 300.

[0066] The center frequency represents the dominant frequency of the VCF response, and the bandwidth represents the frequency range that can pass through the VCF near the center frequency. The center frequency and bandwidth can be adjusted by tuning the voltage. This allows the VCF, with its adjusted filtering parameters, to filter interference signals corresponding to the interference frequencies in the output signal, resulting in a clean, filtered output signal, which is then output to the RF module.

[0067] The circuit embodiments of this application are small in size, have low power consumption, and good compatibility. Using the RF front-end circuit of this application, multi-band signals received by the antenna in a multi-band wireless communication system can be processed, filtering out interference signals between the multi-band signals. The embodiments of this application have fast response speed and can dynamically generate tuning control signals to suppress interference signals in different scenarios, meeting the needs of various scenarios and offering high flexibility.

[0068] Please see Figure 4 , Figure 4 This is a flowchart illustrating a control method for a radio frequency front-end circuit provided in an embodiment of this application. Figure 4 As shown, a control method for an RF front-end circuit based on the RF front-end circuit provided in the above embodiments is provided, including the following steps S410 to S430.

[0069] Step S410: Generate a coupling signal and an output signal based on the original signal from the antenna; wherein the original signal includes multi-band electromagnetic wave signals;

[0070] Step S420: Generate a tuning control signal based on the interference frequency of the coupled signal;

[0071] Step S430: Under the control of the tuning control signal, the output signal is filtered to obtain the updated output signal after filtering.

[0072] The RF front-end circuit of this application processes the multi-band raw signals received by the antenna, splitting the raw signals into two outputs: a coupling signal and an output signal. Through processing and analysis of the coupling signal, the interference frequency in the coupling signal is predicted, and a tuning control signal is obtained based on the predicted interference frequency. Then, based on the tuning control signal, a filtering module is used to filter out the signal in the output signal corresponding to the interference frequency, resulting in a filtered and updated output signal. This updated, clean output signal is then output to the RF module. In embodiments of this application, when the signal received by the antenna changes, the interference frequency can be dynamically predicted and the corresponding interference signal can be filtered out, improving the performance of the multi-band wireless communication system.

[0073] In some embodiments, obtaining a tuning control signal based on an interference signal in a coupled signal includes: predicting the interference type of the interference signal in the coupled signal based on a first prediction model, predicting the interference frequency corresponding to the interference type based on a second prediction model, and then generating a tuning control signal based on the interference frequency.

[0074] This application embodiment sets up a first prediction model and a second prediction model working together. The first prediction model predicts the interference type of the interference signal, and the second prediction model predicts the interference frequency of the interference signal. This can improve the prediction efficiency, reliability, and accuracy of the interference frequency. In some implementations, the interference frequency can also be predicted directly using the second prediction model, and then a tuning control signal can be generated based on the interference frequency.

[0075] The first prediction model can predict the type of interference signal based on the characteristic data of the frequency domain signal corresponding to the coupled signal. The second prediction model can predict the interference signal within a preset time period based on the signal pattern of the interference signal in the coupled signal. For example, the first prediction model can be a random forest prediction model, and the second prediction model can be a long short-term memory (LSTM) prediction model.

[0076] After obtaining the interference frequency by combining the first and second prediction models, a tuning control signal is obtained through digital-to-analog conversion based on the interference frequency. The filtering parameters of the filtering module are then adjusted according to the generated tuning control signal to filter out the interference signal.

[0077] The following example illustrates the prediction methods of the first and second prediction models.

[0078] Before predicting the interference frequency, the coupled signal is first acquired and preprocessed. In the RF front-end circuit, the control module preprocesses the coupled signal output by the coupling module through the coupling output terminal.

[0079] For example, the coupling signal can be a frequency band from 0.6 GHz to 6 GHz. The coupling signal is the same as the original signal, both being multi-frequency electromagnetic signals. The sampling rate is set to 100 MS / s, and the ADC quantization bit depth is 12 bits. It can be understood that the frequency band of the coupling signal is determined according to the application scenario; for example, in satellite communication, the frequency band of the coupling signal can be extended to the 6 GHz to 8 GHz band.

[0080] A Fourier transform is performed on the sampled signal to obtain the frequency domain signal of the coupled signal. For example, one frame of spectral data is generated every 10 milliseconds, meaning the resolution of the frequency domain signal is 1 MHz. The 10-millisecond processing time is based on the hardware response time and algorithm inference time. Reducing the time typically increases the system's computational load and energy consumption. For high-frequency signals in the 5 GHz to 6 GHz band, path loss is significant, requiring longer integration times to improve the signal-to-noise ratio. For low-frequency signals in the 0.6 GHz to 1 GHz band, the processing time can be shortened.

[0081] In some embodiments, the second prediction model is an LSTM model; predicting the interference frequency using the second prediction model may include the following steps:

[0082] (A1) Extract the first feature data of the frequency domain signal corresponding to the coupled signal.

[0083] The first feature data is extracted from the frequency domain signal after the coupled signal undergoes Fourier transform. This first feature data includes power data, peak value data, amplitude data, and harmonic data. As described in the previous embodiments, interference between signals is related to the operating frequency band, modulation technique, time domain characteristics, and intermodulation interference modes of communication technologies such as Wi-Fi, Bluetooth, and cellular networks. Therefore, the frequency domain signal is labeled, for example, by simulating application scenarios such as city streets and underground parking garages in a laboratory environment, to label different interference types and interference frequencies in order to construct a training dataset for the LSTM model.

[0084] (A2) Construct and train the LSTM prediction model.

[0085] The LSTM prediction model uses a lightweight, single-layer LSTM network. The input layer takes time-series spectral data; for example, a time step of T=10 represents 10 consecutive frames of spectral data. The hidden layer has 128 neurons, capturing long-term dependencies in the time series to help predict future frequency changes. The output layer shows the predicted interference frequencies within the next 10ms and their corresponding confidence levels.

[0086] In LSTM prediction models, frequency point prediction is performed using the mean squared error (MSE) loss function. The mathematical expression for the MSE loss function is:

[0087]

[0088] Among them, y i Indicates the actual interference frequency. This represents the predicted frequency point, and N represents the batch size, such as 32.

[0089] MSE has the advantage of convexity, which helps LSTM models converge to the global optimum. Furthermore, the derivative of MSE is simple to calculate, making it... Suitable for implementing backpropagation in low-power MCUs.

[0090] The embodiments of this application can also weight the mean square error. For example, for key frequency bands, such as 1561MHz of Beidou B1, a higher weight is assigned through weighting, and the above formula is adjusted to:

[0091]

[0092] Where, ω i This represents the protection weighting coefficient. For weak signals with low frequencies, such as Beidou B1, protection can be achieved through weighting.

[0093] In the LSTM prediction model, confidence classification prediction is also performed using cross-entropy loss. Cross-entropy loss measures the difference between the predicted probability distribution and the actual distribution. The mathematical expression for cross-entropy loss is:

[0094]

[0095] Where C represents the number of categories, such as the interference type being Wi-Fi, cellular, Bluetooth, or others. i,c Encoded data representing the true category, This represents the predicted probability output by the softmax function. The cross-entropy function can directly measure the difference between the predicted probability distribution and the true distribution, making it more suitable for classification tasks than MSE.

[0096] For scenarios with imbalanced interference types, adjustments can be made using category weights. For example, in applications like underground parking lots where cellular interference samples are relatively small, the weights can be set as follows:

[0097]

[0098] in, Inversely proportional to the category frequency, y c Indicates the true category label, This represents the predicted probability.

[0099] The total loss function is expressed as:

[0100]

[0101] Here, α represents the weighting coefficient. In actual deployment, the weights can be dynamically optimized based on the interference intensity distribution of the target scenario. For example, α can be increased for densely populated cellular network areas.

[0102] The LSTM prediction model also incorporates the Adam optimizer to automatically adjust the learning rate during training, making model training more efficient.

[0103] First, configure the core parameters of the Adam optimizer. Set the learning rate to an empirical value of 0.001, suitable for most computer vision (CV) or natural language processing (NLP) tasks, but it needs adjustment based on the characteristics of the spectral data. If the loss fluctuates wildly in the early stages of training, the learning rate can be reduced to 0.0005. If convergence is too slow, cosine annealing can be used, such as changing it from 0.001 to 0.0001. The momentum parameters are β1 = 0.9 and β2 = 0.999. β1 is the first-order momentum, used to balance the current gradient with the historical gradient direction. β2 is the second-order momentum, used to adaptively adjust the learning rate of each parameter. In predicting sudden disturbances, the momentum mechanism can mitigate the impact of temporal abrupt noise in the spectral data.

[0104] For example, during Adam optimizer training, the batch size is set to 32 to balance memory constraints and gradient stability. The training epoch is set to 100. Early stopping is used for training; if the validation set loss does not decrease for 10 consecutive epochs, training is terminated. Analysis of the learning curve shows that under normal convergence conditions, the training loss tends to stabilize between epochs 30 and 50. If the training loss continues to decrease after 100 epochs, there may be anomalies such as underfitting, requiring an increase in model capacity.

[0105] (A3) Predict the interference frequency based on the LSTM prediction model.

[0106] The constructed LSTM model is deployed on the processor, and the trained model is quantized using the TensorFlow Lite machine learning inference framework, reducing the model's precision to 8 bits. This significantly reduces the model size, improves inference speed, and keeps the inference time within 2ms.

[0107] Normalize the frequency domain signal, for example, with an input dimension of (Batch_size, T, 6000), corresponding to the 0.6GHz to 6GHz frequency band, with one data point per MHz. Normalizing the frequency points can prevent gradient explosion caused by large values.

[0108] The normalized frequency domain signal is input into the LSTM prediction model. The LSTM prediction model outputs the interference frequency prediction result every 10ms. The interference frequency domain prediction result can be expressed as: [1.561GHz, 2.437GHz], confidence level = [0.89, 0.76]). That is, the confidence level for the predicted interference frequency of 1.561GHz is 0.89, and the confidence level for the predicted interference frequency of 2.437GHz is 0.76.

[0109] In some embodiments, the first prediction model is a random forest prediction model; predicting the type of disturbance using the first prediction model may include the following steps:

[0110] (B1) Extract the second feature data of the frequency domain signal corresponding to the coupled signal.

[0111] The second feature data is extracted from the frequency domain signal after the coupled signal undergoes Fourier transform. This second feature data includes power data, peak data, bandwidth data, time-domain feature data, and intermodulation data. Examples include the frequency and power data corresponding to the peak value of the frequency domain signal, signal bandwidth, the duty cycle of the signal in the time domain (e.g., the burst transmission characteristics of Wi-Fi), and third-order intermodulation data between signals.

[0112] (B2) Construct and train a random forest prediction model.

[0113] For example, the random forest prediction model has 50 decision trees with a maximum depth of 10 layers. The purity of the dataset is measured using Gini impurity. The output categories of the random forest prediction model include: Wi-Fi, Bluetooth, cellular networks, and other interference.

[0114] The training dataset for the random forest prediction model can be 100,000 labeled spectrum data collected in the laboratory. The training dataset should cover typical interference scenarios, such as Wi-Fi channel conflict, Bluetooth frequency hopping, and LTE uplink burst interference.

[0115] During model training, the Scikit-learn library was used to standardize the feature data. The cross-validation accuracy was greater than 95%, and the confusion matrix showed that the Wi-Fi classification accuracy was 98% and the Bluetooth classification accuracy was 97%.

[0116] (B3) Predict the type of interference signal based on the random forest prediction model.

[0117] The feature vector of each frame of spectrum data is input into the random forest prediction model, and the input data dimension can be 20. The random forest outputs the interference type label of the interference signal, as well as the probability distribution of the corresponding interference type. For example, Wi-Fi: 0.95, Bluetooth: 0.03, indicating that the probability of the interference type being Wi-Fi is 95% and the probability of the interference type being Bluetooth is 3%.

[0118] (B4) Combine the random forest prediction model with the LSTM prediction model to predict the interference frequency.

[0119] This application combines a random forest prediction model and an LSTM prediction model. The LSTM prediction model uses the temporal characteristics of the frequency domain signal, such as the power at each frequency point and the temporal evolution of harmonic distribution, to determine the power fluctuation period and frequency correlation, thus predicting the interference frequency within a preset time period. The random forest prediction model is used for decision support, determining the interference type (Wi-Fi or Bluetooth) based on the spectral characteristics of the frequency domain signal to optimize the prediction results of the LSTM prediction model. For example, if the random forest prediction model determines the interference type is Wi-Fi, the LSTM prediction model prioritizes predicting Wi-Fi-related signals; if the random forest prediction model determines the interference type is Bluetooth frequency hopping, the time window can be shortened by controlling the LSTM prediction model due to the high dynamics of Bluetooth frequency hopping.

[0120] The following specific example illustrates the radio frequency front-end circuit control method of this application.

[0121] Please see Figure 5 , Figure 5 This is a flowchart of another control method for a radio frequency front-end circuit provided in an embodiment of this application. For example... Figure 5 As shown, the control method for the radio frequency front-end circuit includes the following steps S510 to S560.

[0122] Step S510: The multi-band electromagnetic wave signal received by the antenna is used as the original signal.

[0123] For example, the multi-band electromagnetic wave signals received by the antenna include GNSS, Bluetooth, Wi-Fi, cellular signals, etc., with different signals corresponding to different frequency bands.

[0124] In step S520, the coupling module divides the original signal into a coupling signal and an output signal, sends the coupling signal to the control module, and sends the output signal to the filtering module.

[0125] In step S530, the control module predicts the interference frequency based on the coupling signal.

[0126] The control module first samples and performs Fourier transform on the coupled signal to obtain its frequency domain signal. It then performs feature extraction and classification on the frequency domain signal, predicts the interference type of the interfering signal in the coupled signal using a random forest prediction model, and predicts the corresponding interference frequency using an LSTM prediction model.

[0127] Step S540: Generate a harmonic control signal based on the interference frequency.

[0128] Step S550: Adjust the filtering parameters of the filtering module according to the harmonic control signal.

[0129] Step S560: The filter module, after adjusting the filter parameters, processes the output signal to filter out interference signals in the output signal.

[0130] After the filtering module removes interference signals from the output signal, it obtains an updated and clean output signal, which is then output to the RF module.

[0131] For example, if a high-power burst signal is found in the 2.4GHz band based on spectral characteristics, the interference type is Wi-Fi channel 6. The LSTM prediction model predicts that the third-order intermodulation product of Wi-Fi channel 6 and BeiDou B1 at 1.561GHz is: 2×1.561-2.437=0.685GHz, and a harmonic control signal is generated to control the VCF to filter out signals near 0.685GHz.

[0132] This application embodiment uses dynamic hardware tuning combined with a predictive model to predict interference frequencies, which can suppress interference signals in multi-frequency band coexistence application scenarios. The circuit structure is simple and small in size, and has both high compatibility and low power consumption characteristics, and can be widely used in intelligent transportation, Internet of Things and other fields.

[0133] Thirdly, embodiments of this application provide a multi-band wireless communication system, including: a radio frequency module, and a radio frequency front-end circuit as provided in the above embodiments.

[0134] Please continue reading. Figure 1 After filtering the output signal, the filtering module 300 of the RF front-end circuit outputs the filtered and cleaned output signal to the RF module. The RF module performs demodulation and other processing on the filtered and cleaned output signal to improve the information transmitted in the output signal.

[0135] The multi-band wireless communication system of this application filters interference signals in multi-band signals through radio frequency front-end circuits to obtain a clean output signal after filtering and output it to the radio frequency module. It can suppress interference signals in different scenarios, meet the needs of different scenarios, and has high flexibility.

[0136] It should be noted that those skilled in the art can choose to combine prediction models according to actual needs. For example, when predicting the type of interference, in addition to the random forest prediction model proposed in the above embodiments, one or more of the following can be used: Support Vector Machine (SVM), lightweight Convolutional Neural Network (CNN), or Gradient Boosting Decision Tree (GBDT). When predicting the frequency of interference, in addition to the LSTM model proposed in the above embodiments, one or more of the following can be used: Gated Recurrent Unit (GRU), Temporal Convolutional Network (TCN), or Transformer encoder, to achieve similar temporal prediction functions.

[0137] The radio frequency front-end circuit and its control method, as well as the multi-band wireless communication system provided in the embodiments of this application, have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A radio frequency front-end circuit, characterized in that, include: The system includes a coupling module (100), a control module (200), and a filtering module (300). The coupling module (100) is connected to the antenna, the control module (200), and the filtering module (300), respectively. The control module (200) is also connected to the filtering module (300). The coupling module (100) is used to: generate a coupling signal and an output signal based on the original signal of the antenna, send the coupling signal to the control module (200), and send the output signal to the filtering module (300); wherein the original signal includes multi-band electromagnetic wave signals; The control module (200) is used to: obtain a tuning control signal based on the interference signal in the coupling signal, and send the tuning control signal to the filtering module (300); The filtering module (300) is used to: filter the output signal under the control of the tuning control signal to obtain a filtered and updated output signal.

2. The circuit according to claim 1, characterized in that, The control module (200) includes: a prediction unit connected to the coupling module (100), and a signal generation unit connected to the prediction unit; wherein, The prediction unit is used to: predict the interference frequency of the interference signal in the coupled signal according to a machine learning prediction model; The signal generation unit is used to generate the tuning control signal according to the interference frequency.

3. The circuit according to claim 2, characterized in that, The prediction unit includes a first prediction subunit, which is used for: Determine the frequency domain signal of the coupled signal; The signal pattern of the interference signal is determined based on the first feature data of the frequency domain signal, and the interference frequency within a preset time period is predicted based on the signal pattern. The first feature data includes: power data, peak data, amplitude data, and harmonic data.

4. The circuit according to claim 3, characterized in that, The prediction unit further includes a second prediction subunit, the second prediction subunit being used for: The interference type corresponding to the interference signal is predicted based on the second feature data of the frequency domain signal, so that the interference frequency is predicted in the first prediction subunit based on the interference type and the first feature data of the frequency domain signal. The second feature data includes: power data, peak data, bandwidth data, time-domain feature data, and intermodulation data.

5. The circuit according to claim 1, characterized in that, The tuning control signal includes a tuning voltage, and the filtering module (300) includes a tunable filter; wherein, The tunable filter adjusts its filtering parameters under the control of the tuning voltage, and filters the output signal based on the adjusted tunable filter parameters.

6. The circuit according to claim 5, characterized in that, The filtering parameters include the center frequency and bandwidth of the filtering module (300).

7. The circuit according to claim 1, characterized in that, The coupling module (100) is connected to the antenna via a signal input terminal, to the control module (200) via a coupling output terminal, and to the filtering module (300) via a signal output terminal; wherein, The coupling module (100) performs coupling processing on the original signal according to a preset ratio to obtain the coupled signal, and sends the coupled signal to the control module (200) through the coupling output terminal; The output signal is obtained based on the original signal and the coupling signal, and the output signal is sent to the filtering module (300) through the signal output terminal.

8. A control method for a radio frequency front-end circuit, characterized in that, include: A coupling signal and an output signal are generated based on the original signal from the antenna; wherein the original signal includes multi-band electromagnetic wave signals; The tuning control signal is obtained based on the interference signal in the coupling signal; Under the control of the tuning control signal, the output signal is filtered to obtain a filtered and updated output signal.

9. The method according to claim 8, characterized in that, The step of obtaining the tuning control signal based on the interference signal in the coupled signal includes: The interference type in the coupled signal is predicted according to the first prediction model, and the interference frequency corresponding to the interference type is predicted according to the second prediction model. The tuning control signal is generated based on the interference frequency.

10. A multi-band wireless communication system, characterized in that, include: The radio frequency module, and the radio frequency front-end circuit as described in any one of claims 1 to 7.

Images