WO2023016231A1 - 射频增益控制方法、装置及通信设备 - Google Patents

射频增益控制方法、装置及通信设备 Download PDF

Info

Publication number
WO2023016231A1
WO2023016231A1 PCT/CN2022/107390 CN2022107390W WO2023016231A1 WO 2023016231 A1 WO2023016231 A1 WO 2023016231A1 CN 2022107390 W CN2022107390 W CN 2022107390W WO 2023016231 A1 WO2023016231 A1 WO 2023016231A1
Authority
WO
WIPO (PCT)
Prior art keywords
radio frequency
gain control
target
gain
parameter
Prior art date
Application number
PCT/CN2022/107390
Other languages
English (en)
French (fr)
Inventor
李大国
Original Assignee
Oppo广东移动通信有限公司
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 Oppo广东移动通信有限公司 filed Critical Oppo广东移动通信有限公司
Publication of WO2023016231A1 publication Critical patent/WO2023016231A1/zh

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04BTRANSMISSION
    • H04B1/00Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/02Transmitters
    • H04B1/04Circuits
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology
    • G06N3/045Combinations of networks
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/08Learning methods
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04BTRANSMISSION
    • H04B1/00Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
    • H04B1/02Transmitters
    • H04B1/04Circuits
    • H04B2001/0408Circuits with power amplifiers
    • H04B2001/0416Circuits with power amplifiers having gain or transmission power control

Definitions

  • Embodiments of the present invention relate to the field of communication technologies, and in particular, to a radio frequency gain control method, device, and communication equipment.
  • the 3rd Generation Partnership Project (3gpp) has strict specification requirements for RF gain parameters (including: transmit power, transmit power consumption, etc.) of different communication systems (such as 2G, 3G, 4G, 5G), so Gain control is one of the important indicators to measure the performance of terminal equipment. For example, through radio frequency power control, the range and quality of radio frequency signals sent by the terminal equipment can be determined by controlling the transmit power.
  • the current RF power control scheme of the RF system is to control and adjust the power of each component of the RF system through two stages of laboratory debugging and factory debugging, so as to ensure that the transmit power in the RF system of the terminal equipment meets the requirements of 3GPP specifications, and will pass through the above
  • the power control parameters (which may include power control factors and adjustment weights of each factor) obtained after debugging are stored in the terminal device for subsequent use.
  • the above solution requires manual debugging by chip practitioners in the laboratory stage, so the reliability after debugging is low, and the cost of debugging equipment in the factory debugging stage is high. Therefore, it is urgent to achieve low-cost, high-reliability RF gain control. problem to be solved.
  • Embodiments of the present invention provide a radio frequency gain control method, device, and communication equipment, so as to realize low-cost, high-reliability radio frequency gain control.
  • a radio frequency gain control method including:
  • the target connection parameter corresponding to the target gain parameter is obtained after training the first gain control model based on the sample values of the plurality of gain control factors, and the first gain control model is for the first gain control model A neural network model for gain control established by a radio frequency component;
  • a radio frequency gain control device including:
  • a radio frequency gain control device including:
  • An acquisition module configured to acquire a target connection parameter corresponding to a target gain parameter, where the target connection parameter is obtained after training a first gain control model based on sample values of the plurality of gain control factors, and the first gain
  • the control model is a neural network model for gain control established for the first radio frequency component
  • a configuration module configured to configure the plurality of gain control factors of the first radio frequency component according to the target connection parameter, so as to perform gain control on the first radio frequency component.
  • a communication device including: the radio frequency gain control device according to the second aspect or the third aspect, and a first radio frequency component.
  • a computer-readable storage medium where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the radio frequency gain control method according to the first aspect is implemented.
  • a computer program product stores a computer program, and when the computer program is executed by a processor, the radio frequency gain control method according to the first aspect is implemented.
  • FIG. 1 is a schematic structural diagram of a radio frequency system solution provided by an embodiment of the present invention
  • FIG. 2A is a schematic structural diagram of a radio frequency system provided by an embodiment of the present invention.
  • FIG. 2B is a first schematic flow diagram of a radio frequency gain control method provided by an embodiment of the present invention.
  • FIG. 3 is a schematic flow diagram II of a radio frequency gain control method provided by an embodiment of the present invention.
  • FIG. 4 is a schematic flow diagram 3 of a radio frequency gain control method provided by an embodiment of the present invention.
  • FIG. 5 is a structural schematic diagram 1 of a radio frequency gain control device according to an embodiment of the present invention.
  • FIG. 6 is a schematic structural diagram II of a radio frequency gain control device provided by an embodiment of the present invention.
  • FIG. 7 is a schematic structural diagram of a communication device provided by an embodiment of the present invention.
  • FIG. 1 it is a schematic diagram of the architecture of a wireless radio frequency system solution, which mainly includes a baseband chip, a radio frequency transceiver chip (Radio Frequency Integrated Circuit, RFIC), a radio frequency front end (Radio Frequency Front End, RFFE) chip and an antenna (antenna) consists of several parts.
  • RFIC Radio Frequency Integrated Circuit
  • RFFE Radio Frequency Front End
  • antenna antenna
  • NR New radio
  • FR1 Frequency range 1
  • FR2 Frequency range 2
  • Table 1 the frequency domain ranges included in FR1 and FR2 are shown in Table 1.
  • the embodiments of the present invention may be applied to frequency bands FR1 and FR2, and may also be applied to other frequency bands, for example, frequency bands from 52.6 GHz to 71 GHz, or frequency bands from 71 GHz to 100 GHz, which are not limited in this application.
  • Frequency band definition Corresponding frequency range FR1 410MHz–7.125GHz FR2 24.25GHz–52.6GHz
  • the RF front-end is provided with a receiving (Rx) path and a transmitting (Tx) path, as well as a switch connecting the receiving path or the transmitting path and the antenna; among them, in the transmitting path
  • a power amplifier (Power Amplifier, PA) matrix is provided.
  • the antenna and other RF front-end devices have been integrated into a millimeter-wave antenna system integrated chip, which mainly includes a phase shifter matrix, a PA matrix and an antenna matrix.
  • radio frequency power control is a measure of terminal equipment performance.
  • the radio frequency power control in the wireless communication system is also mainly controlled and implemented through the above four modules of baseband chip, radio frequency transceiver chip, radio frequency front-end chip and antenna as shown in Figure 1.
  • the specific power control scheme is mainly implemented through the following two aspects :
  • Chip optimization and debugging in the laboratory requires high experience requirements for practitioners. At the same time, debugging takes a long time, human factors are uncontrollable, and the power control efficiency of each module cannot be well optimized and combined;
  • the antenna and other RF front-end devices such as PA are already integrated in one chip.
  • the conduction line loss is relatively large when using the traditional LTE method for verification testing.
  • the embodiment of the present invention provides a radio frequency gain control method, device and communication equipment, which can obtain the target connection parameters corresponding to the target gain parameters, and the target connection parameters are based on the sample values of multiple gain control factors for the first gain control model Obtained after training, the first gain control model is a neural network model for gain control established for the first radio frequency component; according to the target connection parameters, configure multiple gain control factors of the first radio frequency component to control the first radio frequency component for gain control.
  • the radio frequency gain control method provided by the embodiment of the present invention can be applied to a radio frequency gain control device or a communication device, and the radio frequency gain control device can be a functional module or a functional entity in the communication device for implementing the radio frequency gain control method.
  • the communication device involved may be a network device or a terminal device.
  • the above terminal equipment may be referred to as user equipment (user equipment, UE), access terminal, subscriber unit, subscriber station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device, etc.
  • user equipment user equipment, UE
  • access terminal subscriber unit, subscriber station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device, etc.
  • the terminal device can be a station (STAION, ST) in the WLAN, a cellular phone, a cordless phone, a Session Initiation Protocol (Session Initiation Protocol, SIP) phone, a wireless local loop (Wireless Local Loop, WLL) station, a personal digital processing (Personal 40Digital Assistant, PDA) devices, handheld devices with wireless communication functions, computing devices or other processing devices connected to wireless modems, vehicle-mounted devices, wearable devices, next-generation communication systems such as terminal devices in NR networks, or future The terminal equipment in the evolved public land mobile network (Public Land Mobile Network, PLMN) network, etc.
  • PLMN Public Land Mobile Network
  • the terminal device can be deployed on land, including indoor or outdoor, handheld, wearable or vehicle-mounted; it can also be deployed on water (such as ships, etc.); it can also be deployed in the air (such as aircraft, balloons and satellites) superior).
  • the terminal device can also be a mobile phone (Mobile Phone), a tablet computer (Pad), a computer with a wireless transceiver function, a virtual reality (Virtual Reality, VR) terminal device, an augmented reality (Augmented Reality, AR) terminal device, an industrial control (industrial control), wireless terminal devices in self driving, wireless terminal devices in remote medical, wireless terminal devices in smart grid, transportation safety ), wireless terminal devices in a smart city, or wireless terminal devices in a smart home.
  • a mobile phone Mobile Phone
  • a tablet computer Pad
  • a computer with a wireless transceiver function a virtual reality (Virtual Reality, VR) terminal device, an augmented reality (Augmented Reality, AR) terminal device, an industrial control (industrial control), wireless terminal devices in self driving, wireless terminal devices in remote medical, wireless terminal devices in smart grid, transportation safety ), wireless terminal devices in a smart city, or wireless terminal devices in a smart home.
  • a virtual reality Virtual Reality, VR
  • AR Augmented Reality
  • industrial control industrial control
  • the network device involved in this embodiment of the present invention may be an access network device.
  • the access network device may be a long-term evolution (long-term evolution, LTE) system, a next-generation (mobile communication system) (next radio, NR) system or an authorized auxiliary access long-term evolution (LAA- Evolved base station (evolutional node B, abbreviated as eNB or e-NodeB) macro base station, micro base station (also called “small base station”), pico base station, access point (access point, AP), Transmission point (transmission point, TP) or new generation base station (new generation Node B, gNodeB), etc.
  • LTE long-term evolution
  • NR next-generation
  • LAA- Evolved base station evolutional node B, abbreviated as eNB or e-NodeB
  • eNB next-generation
  • NR next-generation
  • LAA- Evolved base station evolutional node B, abbreviated as
  • the network device can be a device for communicating with the mobile device, and the network device can be an access point (Access Point, AP) in WLAN, a base station (Base Transceiver Station, BTS) in GSM or CDMA , or a base station (NodeB, NB) in WCDMA, or an evolved base station (Evolutional Node B, eNB or eNodeB) in LTE, or a relay station or access point, or a vehicle-mounted device, a wearable device, and an NR network
  • the network equipment (gNB) in the network or the network equipment in the future evolved PLMN network or the network equipment in the NTN network, etc.
  • the network device may provide services for a cell, and the terminal device communicates with the network device through the transmission resources (for example, frequency domain resources, or spectrum resources) used by the cell, and the cell may be a network device ( For example, a cell corresponding to a base station), the cell may belong to a macro base station, or may belong to a base station corresponding to a small cell (Small cell), and the small cell here may include: a metro cell (Metro cell), a micro cell (Micro cell), a pico cell ( Pico cell), Femto cell, etc. These small cells have the characteristics of small coverage and low transmission power, and are suitable for providing high-speed data transmission services.
  • the transmission resources for example, frequency domain resources, or spectrum resources
  • the cell may be a network device (
  • the cell may belong to a macro base station, or may belong to a base station corresponding to a small cell (Small cell)
  • the small cell here may include: a metro cell (Metro cell), a micro cell (Micro
  • the technical solution of the embodiment of the present invention can be applied to various communication systems, such as: Global System of Mobile communication (Global System of Mobile communication, GSM) system, code division multiple access (Code Division Multiple Access, CDMA) system, broadband code division multiple access (Wideband Code Division Multiple Access, WCDMA) system, General Packet Radio Service (GPRS), Long Term Evolution (LTE) system, Advanced long term evolution (LTE-A) system , New Radio (NR) system, evolution system of NR system, LTE (LTE-based access to unlicensed spectrum, LTE-U) system on unlicensed spectrum, NR (NR-based access to unlicensed spectrum) on unlicensed spectrum unlicensed spectrum (NR-U) system, Non-Terrestrial Networks (NTN) system, Universal Mobile Telecommunications System (UMTS), Wireless Local Area Networks (WLAN), Wireless Fidelity (Wireless Fidelity, WiFi), fifth-generation communication (5th-Generation, 5G) system or other communication systems, etc.
  • GSM Global System of Mobile
  • the radio frequency system includes: a baseband chip, a radio frequency transceiver chip, a radio frequency front-end chip, an antenna and a neural network chip, wherein the neural network
  • the chip can be a microprocessor (Graphics Processing Unit, GPU) or the neural network chip can be a central processing unit (Central Processing Unit, CPU).
  • the provided radio frequency gain control method is used as an example for illustration.
  • FIG. 2B it is a schematic flow diagram of a radio frequency gain control method provided by an embodiment of the present invention, and the method flow includes:
  • the first gain control model is a neural network model for gain control established for the first radio frequency component, and the target connection parameter is obtained after training the first gain control model based on sample values of multiple gain control factors .
  • multiple sets of correspondences between gain parameters and connection parameters may be stored in advance in the communication device.
  • Table 2 is in the form of a correspondence table, taking n groups of gain parameters and connection parameters stored as an example for exemplary description.
  • the correspondence between the target gain parameter and the target connection parameter may be pre-stored in the communication device.
  • the correspondence between the target gain parameter and the target connection parameter may be one of the n groups of correspondences stored above, where n is greater than or equal to 2.
  • connection parameters corresponding to the gain parameters are the same for different gain parameters and connection parameters, the following uses the specific process of obtaining the target connection parameters corresponding to the target gain parameters as an example for exemplary description.
  • the communication device may first train the first gain control model based on the sample values of multiple gain control factors of the first radio frequency component to obtain the target gain control model, and then store the target gain parameters corresponding to the multiple
  • the target connection parameters of a gain control factor, the target connection parameters include: the connection weight and bias value between any two adjacent hidden layers of the target gain control model.
  • the error between the first gain parameter output by the target gain control model and the target gain parameter is less than or equal to the first error threshold
  • the first gain control model is a neural network model established for gain control of the first radio frequency component.
  • one or more training processes may be experienced. Specifically, the following training steps of a, b and c can be executed cyclically until the error between the gain parameter output by the adjusted first gain control model and the target transmission gain is less than or equal to the first error threshold, and the adjusted first gain The control model serves as the target gain control model.
  • connection parameters include: connection weights and bias values between any two adjacent hidden layers of the first gain control model.
  • the number of times the above training steps a, b and c are cyclically executed can be one or more times, as long as the error between the gain parameter output by the first gain control model and the target gain parameter is less than or equal to the first error threshold.
  • the first error threshold may be set according to the actual training accuracy of the first gain control model, which is not limited in the embodiment of the present invention.
  • the foregoing first radio frequency component may include one or more devices in a radio frequency system.
  • the first radio frequency component includes at least one of the following:
  • the first gain control model is a neural network model established for the baseband chip for gain control.
  • the first gain control model is a neural network model established for the baseband chip and the radio frequency transceiver chip for gain control.
  • the first gain control model is a neural network model established for the baseband chip, the radio frequency transceiver chip and the radio frequency front end for gain control.
  • the first gain control model is a neural network model established for the baseband chip, radio frequency transceiver chip, radio frequency front end and antenna for gain control.
  • the first gain control model is a neural network model established for the radio frequency transceiver chip and the radio frequency front end for gain control.
  • the first gain control model is a neural network model established for the radio frequency transceiver chip, the radio frequency front end and the antenna for gain control.
  • the first gain control model is a neural network model established for the radio frequency front end and the antenna for gain control.
  • the first gain control model is a neural network model established for gain control of the antenna.
  • the gain control involved in this embodiment of the present invention includes but is not limited to at least one of the following:
  • Transmission power control that is, control of the transmission power of the first radio frequency component
  • Transmission power consumption control that is, control of the transmission power consumption of the first radio frequency component
  • Received signal strength control that is, to control the received signal strength of the first radio frequency component.
  • the transmission power P is mainly determined by the baseband chip, the radio frequency transceiver chip, the radio frequency front-end chip and the antenna.
  • RF power control is a typical multi-factor control problem.
  • factors that affect power control for the baseband chip, radio frequency transceiver chip, radio frequency front end, and antenna are as follows:
  • the power control factors of the baseband chip mainly include: signal strength, frequency, temperature, device process level, etc.
  • the power control factors of the RF transceiver chip mainly include: power amplifiers at all levels, channels, temperature, voltage, device technology level, etc.
  • the power control factors of the radio frequency front-end chip mainly include: power amplifier, voltage, temperature, device process level, device printed circuit board (Printed Circuit Board, PCB) layout method, and the like.
  • the power control factors of the antenna mainly include: antenna shape, operating frequency range, temperature, PCB layout, etc.
  • radio frequency power control in wireless communication is a typical multi-factor control problem
  • the power control process of the components in the radio frequency system based on the neural network can be divided into: 1) the self-learning training phase and 2) the training result use phase, two phases, wherein, 1) the self-learning training phase
  • the specific implementation process is as follows:
  • Self-learning training stage which uses multi-layer neural network (Multi-layer neural network, MLP) algorithm for learning and training, this stage is mainly to statically analyze and learn the influence of power control factors of each component in the radio frequency system on power,
  • MLP multi-layer neural network
  • the MLP neural network algorithm is the most effective multi-layer neural network learning method. Its main feature is that the signal is transmitted forward, and the error is propagated backwards. By continuously adjusting the connection parameters (including weight values and bias values) in the neural network , so that the final output of the neural network is as close as possible to the desired output.
  • a multilayer neural network consists of L layers of neurons, and L is greater than and equal to 2. The first layer is called the input layer, and the Lth layer is called the output layer.
  • the neural network model constructed based on the first radio frequency component composed of a baseband chip and a radio frequency transceiver chip is taken as an example
  • the algorithm process is described, and the corresponding neural network model (ie, the first gain control model) can be defined as follows:
  • H (l) represents the neuron output of the lth hidden layer
  • l is greater than or equal to 1, less than or equal to L, s is the number of neurons in layer l, represent the outputs of different s neurons respectively;
  • connection weight is the connection weight from the rth neuron of layer l-1 to the tth neuron of layer l, Indicates the bias value of the tth neuron in layer l, and is the connection parameter, then And for the first layer in the hidden layer, Among them, f() is the activation function of the neuron.
  • the neural network model from the power control factor to the output transmission power can be established, and the connection parameters corresponding to each hidden layer can be updated by repeated training k times, namely and Make the error between the final calculated transmit power and the preset target transmit power less than or equal to the first error threshold, and consider the trained neural network model as the target power control model (ie, corresponding to the target gain control model).
  • the target power control model can be obtained and and will and Stored in the communication device for subsequent use in power control.
  • NVM Non-Volatile Memory
  • the target connection parameters include: the connection weight and bias value between any two adjacent hidden layers of the target gain control model.
  • configuring the multiple gain control factors of the first radio frequency component according to the target connection parameters includes: calculating target values of the multiple gain control factors according to the target connection parameters and sample values of the multiple gain control factors; The target value of each gain control factor is configured to configure multiple gain control factors of the first radio frequency component to perform gain control on the first radio frequency component.
  • the calculation of the target values of the multiple gain control factors according to the target connection parameters and the sample values of the multiple gain control factors includes: performing the first operation according to the target connection parameters and the sample values of the multiple gain control factors , to obtain target values of multiple gain control factors; the first operation is an inverse operation for the trained first gain control model.
  • the trained first gain control model is the above-mentioned target gain control model, that is, the first operation is an inverse operation for the target gain control model.
  • the radio frequency gain control method provided by the embodiment of the present invention can obtain the target connection parameter corresponding to the target gain parameter, and the target connection parameter is obtained after training the first gain control model based on the sample values of multiple gain control factors.
  • the gain control model is a neural network model established for gain control of the first radio frequency component; according to the target connection parameters, multiple gain control factors of the first radio frequency component are configured to perform gain control on the first radio frequency component.
  • the first gain control model can be trained for the corresponding multiple gain control factors, and the target connection parameters corresponding to the target gain parameters are obtained, and the multiple gain control factors of the first radio frequency component are configured through the target connection parameters In this way, the gain control of the first radio frequency component can be performed, so that the first radio frequency component outputs a radio frequency signal according to the target gain parameter, which improves the reliability compared with the manual debugging in the prior art, and does not require complicated debugging equipment, so low cost-effective, high-reliability RF gain control.
  • FIG. 3 it is a schematic flow diagram of another radio frequency gain control method provided in the embodiment of the present invention.
  • the method flow includes:
  • the above-mentioned 301 needs to further introduce an input gain parameter as a training An input to the gain control model. That is, obtaining the target gain control model may include: training the first gain control model based on multiple gain control factors of the first radio frequency component and input gain parameters, so as to obtain the target gain control model.
  • the input gain parameter is a gain parameter of the second radio frequency component
  • the second radio frequency component inputs a radio frequency signal to the first radio frequency component
  • the input gain parameter is a gain parameter of the second radio frequency component after gain control.
  • the second radio frequency component may be a baseband chip.
  • the first radio frequency component is (F) a radio frequency transceiver chip, a radio frequency front end, and an antenna; then the second radio frequency component may be a baseband chip.
  • the first radio frequency component is (G) a radio frequency front end and an antenna; then the second radio frequency component may be a radio frequency transceiver chip and a baseband chip, or the second radio frequency component may be a radio frequency transceiver chip.
  • the second radio frequency component may be a radio frequency front end, a radio frequency transceiver chip, and a baseband chip, or the second radio frequency component may be a radio frequency front end.
  • the target connection parameters include: the connection weight and bias value between any two adjacent hidden layers of the target gain control model.
  • the first gain control model can be trained for the corresponding multiple gain control factors, and the obtained first gain parameter is close to the target gain control model of the target gain parameter, and the target gain parameter and the The target connection parameters corresponding to the target gain control model are saved, so that the corresponding saved target connection data can be used as parameters for power control when the target gain parameters need to be output later.
  • the target connection parameter is obtained after training the first gain control model based on the sample values of multiple gain control factors, that is, obtained through the above-mentioned 301 and 302 .
  • the above 303 and 304 correspond to the above 2) training result use stage, in this stage, the target transmission power corresponding to the first radio frequency component can be determined first, and the previous training results obtained from the communication device match the stored target with the target transmission power connection parameters, and according to the target connection parameters, configure multiple power control factors of the first radio frequency component to perform power control on the first radio frequency component.
  • the current application scenario of the communication device can be identified first, and the first radio frequency corresponding to the current application scenario can be determined.
  • the component's target gain parameter can be determined.
  • the communication device applied by this method as a terminal device as an example, it is also possible to determine the current application scenario of the terminal device according to the usage data of the terminal device, and then determine the current application scenario.
  • a target gain parameter (such as a target transmit power) of the first radio frequency component corresponding to the scene.
  • the usage data of the terminal equipment includes at least one of the following:
  • the transmission power needs to be reduced in this scenario to save power consumption of the terminal device.
  • the transmit power when it is determined according to the temperature of the terminal device that the terminal device is currently in a high-temperature scene, in order to avoid overheating of the terminal device in this scene, the transmit power may be reduced to reduce the Terminal equipment heating problem.
  • the transmit power when it is determined according to the holding mode of the terminal device that the terminal device is not in use, the transmit power may be reduced; when the terminal device is in use, the transmit power may be increased to avoid Due to the limitation of transmission power in the process of using the terminal equipment, the transmission distance of the terminal equipment is limited.
  • the corresponding transmission power range configured by the serving base station for the current scene of the terminal device may be determined according to the serving base station of the terminal device, and the transmission power of the terminal device may be determined within the transmission power range.
  • other factors for example, the remaining power of the terminal device, the temperature of the terminal device, etc. may also be considered within the range of the transmit power to select the transmit power of the terminal device.
  • the usage scenario of the terminal device may also be determined by using parameters of other terminal devices.
  • a neural network model for identifying usage scenarios of terminal devices may also be established, and the neural network model may be trained according to usage parameters of terminal devices in different scenarios to obtain a scenario recognition model.
  • the usage parameters of the terminal equipment obtained in real time can be used as the input of the scene recognition model, and the current application scene corresponding to the output of the scene recognition model can be obtained, and the target of the first radio frequency component corresponding to the current application scene can be determined
  • the transmission power can then be determined according to the target transmission power as a reference, and the corresponding target connection parameters stored in the terminal device.
  • gain control as an example of transmit power control.
  • the implementation of processes such as transmit power consumption and received signal strength is similar to the implementation of the above transmit power control process.
  • the embodiment of the present invention will not be described in detail.
  • a self-learning training flag may be used to identify whether a trained target connection parameter for the first radio frequency component is stored in the communication device.
  • the terminal device If there is a self-learning training flag in the terminal device, it means that the corresponding target connection parameters of the first radio frequency component already exist in the terminal device, and subsequent gain control can be performed; if there is no self-learning training flag in the terminal device, it means that the terminal device has There is no target connection parameter saved for the first radio frequency component, and the target connection parameter needs to be obtained through neural network training.
  • the target connection parameter is the connection parameter obtained from the target gain control model
  • the error between the first gain parameter output by the target gain control model and the target gain parameter is less than or equal to the first error threshold
  • the first gain control model is for the first The neural network model built by the radio frequency component for gain control.
  • the radio frequency gain control method provided by the embodiment of the present invention can obtain the target connection parameter corresponding to the target gain parameter, and the target connection parameter is obtained after training the first gain control model based on the sample values of multiple gain control factors.
  • the gain control model is a neural network model established for gain control of the first radio frequency component; according to the target connection parameters, multiple gain control factors of the first radio frequency component are configured to perform gain control on the first radio frequency component.
  • the first gain control model can be trained for the corresponding multiple gain control factors, and the target connection parameters corresponding to the target gain parameters are obtained, and the multiple gain control factors of the first radio frequency component are configured through the target connection parameters In this way, the gain control of the first radio frequency component can be performed, so that the first radio frequency component outputs a radio frequency signal according to the target gain parameter, which improves the reliability compared with the manual debugging in the prior art, and does not require complicated debugging equipment, so low cost-effective, high-reliability RF gain control.
  • the embodiment of the present invention also provides a gain control method, which includes:
  • the target connection parameter is the target connection parameter obtained by the terminal device through 301 and 302 in the above-mentioned embodiment as shown in FIG. 3 , and the target connection parameter and the target gain parameter are correspondingly stored in the communication device.
  • the actual gain parameter and the target gain parameter are corresponding physical quantities.
  • the actual gain parameter when the target gain parameter indicates the physical quantity of transmit power, the actual gain parameter also indicates the physical quantity of transmit power;
  • the actual gain parameter when the parameter indicates the physical quantity of transmission power consumption, the actual gain parameter also indicates the physical quantity of transmission power consumption.
  • the above-mentioned actual gain parameter may be the transmission power actually detected in the radio frequency system.
  • the transmit power in the radio frequency system will change nonlinearly with the above-mentioned power control factors, and a saturation range may appear.
  • a power feedback point may be set in the radio frequency system to detect the actual transmit power after power control, so as to evaluate the effect of power control.
  • M is a power feedback point located between the RF transceiver chip and the RF front-end.
  • the actual transmission power at this point can be detected by a power sensor, and the actual transmission power can be used to further optimize the target gain. Control the model to get the updated target connection parameters.
  • point M is mainly to evaluate the transmission power of the first radio frequency component composed of the baseband chip and the radio frequency transceiver chip. If there is a large error between the actual transmission power detected at point M and the target transmission power during the previous training, Then, the training may continue on the first radio frequency component composed of the baseband chip and the radio frequency transceiver chip.
  • the actual gain parameter is the actual transmission power detected at point M
  • the target gain parameter is the above target transmission power
  • the error between the actual transmission power and the target transmission power is determined.
  • the relationship between the error between the actual transmit power and the target transmit power and the second error threshold can be judged. If the error between the actual gain parameter and the target gain parameter is greater than the second error threshold, it means Corresponding to the saved target connection parameters, after the power control of the baseband chip and the RF transceiver chip, there is a large deviation between the actual transmission power obtained and the expected target transmission power, then it may be necessary to update the saved target connection parameters to Make the matching degree in subsequent power control.
  • the second error threshold may be the same as or different from the first error threshold. Specifically, it may be set according to the actual training accuracy of the first gain control model and the target gain control model. Here No limit.
  • the third error threshold is smaller than the first error threshold.
  • the target connection parameter used in the current power control does not meet the actual power control requirements, so it can be considered to further improve the current target power control model Training
  • the control range of the error can be narrowed, that is, the first error threshold is changed to a third error threshold with a smaller value, so that the error between the first gain parameter and the target gain parameter is less than or equal to the third
  • the error threshold is used as the training cut-off condition
  • the trained target gain control model that is more in line with the actual power control requirements can be obtained, and the target connection parameters saved corresponding to the target gain parameters can be updated with the trained target gain control model. It is guaranteed that the re-stored target connection parameters can be closer to the actual power control requirements.
  • the power feedback point is set in the actual radio frequency system, so that after the power control is performed according to the training result in the embodiment of the present invention, the actual transmission power and the target transmission power can be compared, so that the error between the two is relatively large
  • the previously saved training results that is, the connection parameters
  • the target gain control model of the previous training can be trained again based on the actual transmission power to obtain the optimized gain control model, and extract the target connection parameters from the optimized gain control model, and update the target connection parameters saved corresponding to the target gain parameters.
  • the power feedback point can also be set at other positions. Specifically, the position of the power feedback point is set to be the same as that of the first radio frequency component defined in the embodiment of the present invention. match.
  • a power feedback point can also be set at point N, and the actual transmission power can be detected through a power sensor.
  • the first radio frequency component can be in the following situations:
  • the power feedback point N is located at the antenna port, and mainly evaluates the final transmission power of the terminal device.
  • the performance of the radio frequency front end and the antenna can be analyzed. If there is a large error between the actual transmission power detected at point N and the target transmission power during previous training, then the first radio frequency component composed of the radio frequency front end and the antenna can be continued to be trained, and the communication system can be reconfigured , other radio frequency components on the radio frequency front-end and the upper end of the antenna are trained, for example, the radio frequency components composed of the baseband chip and the radio frequency transceiver chip are trained.
  • the radio frequency gain control is performed on the radio frequency component composed of the baseband chip and the radio frequency transceiver chip according to the method in this embodiment, and the radio frequency gain control is performed on the radio frequency component composed of the radio frequency front end and the antenna.
  • the training process can be similar to the above-mentioned training process for radio frequency components composed of baseband chips and radio frequency transceiver chips.
  • the difference is that for the input layer, the neural network
  • Z represents the set of sample values of the power control factors of the RF front-end chip
  • V represents the set of sample values of the power control factors of the antenna, z 1 , z 2 , z 3 ...
  • z m represent the sample values of m different power control factors of the RF front-end chip, and m is greater than or equal to 2; v 1 , v 2 , v 3 ...v n represent sample values of n power control factors of the antenna.
  • the radio frequency transceiver chip may also use the gain parameter (for example, transmission power) after gain control as the input of the neural network to train the neural network.
  • the gain parameter for example, transmission power
  • each module in the radio frequency system needs to be in the linear working range to avoid being in the saturated working range, and at the same time, the transmission power is required to show continuous linear changes, so as to avoid the saturated range and discontinuous changes in power control.
  • an embodiment of the present invention provides a radio frequency gain control device, which is characterized in that it includes:
  • the obtaining module 501 is used to obtain the target connection parameter corresponding to the target gain parameter.
  • the target connection parameter is obtained after training the first gain control model based on the sample values of multiple gain control factors.
  • the first gain control model is for the first gain control model.
  • the configuration module 502 is configured to configure a plurality of gain control factors of the first radio frequency component according to the target connection parameter, so as to perform gain control on the first radio frequency component.
  • the configuration module 502 is specifically configured to perform the first operation according to the target connection parameters and the sample values of the multiple gain control factors to obtain the target values of the multiple gain control factors; the first calculation is for the trained first an inverse operation of the gain control model;
  • the multiple gain control factors of the first radio frequency component are configured to perform gain control on the first radio frequency component.
  • the radio frequency gain control device also includes:
  • the training module 503 is used to train the first gain control model based on the sample values of multiple gain control factors of the first radio frequency component before the acquisition module 501 acquires the target connection parameter corresponding to the target gain parameter, so as to obtain the target gain In the control model, the error between the first gain parameter output by the target gain control model and the target gain parameter is less than or equal to the first error threshold;
  • the saving module 504 is used for correspondingly saving the target gain parameter and the target connection parameter for multiple gain control factors, the target connection parameter includes: the connection weight and bias value between any two adjacent hidden layers of the target gain control model .
  • the training module 503 is specifically used for:
  • the adjusted first gain control model is used as the target gain control model.
  • the first radio frequency component includes any of the following:
  • Baseband chips Baseband chips, RF transceiver chips, RF front-ends and antennas;
  • RF transceiver chips RF front-ends and antennas
  • the first radio frequency component includes any of the following;
  • RF transceiver chips RF front-ends and antennas
  • the training module 503 is specifically used for:
  • the input gain parameter is a gain parameter of the second radio frequency component
  • the second radio frequency component is a radio frequency component that inputs a radio frequency signal to the first radio frequency component
  • the training module 503 is further configured to: acquire an actual gain parameter output by the first radio frequency component;
  • the target gain control model is trained based on the sample values of multiple gain control factors of the first radio frequency component, and the first gain parameter and the target gain are known.
  • the parameter error is less than or equal to the third error threshold, and the target connection parameters saved corresponding to the target gain parameters are updated according to the target gain control model obtained after training, wherein the third error threshold is smaller than the first error threshold.
  • the training module 503 is further configured to: configure multiple gain control factors of the first radio frequency component according to the target connection parameters, so as to determine the terminal device according to the usage data of the terminal device before performing gain control on the first radio frequency component. current application scenarios;
  • usage data includes at least one of the following:
  • the gain control includes at least one of the following:
  • the embodiment of the present invention also provides a radio frequency gain control device, the radio frequency control device includes: a memory 601, a processor 602, and a computer stored in the memory 601 and operable on the processor 602. program, when the computer program is executed by a processor, the radio frequency gain control method in the foregoing method embodiment can be implemented.
  • the radio frequency gain control device may be a radio frequency system as shown in FIG. 2A, or, the radio frequency gain control device may include the radio frequency system as shown in FIG. 2A, and the function of the processor 602 in FIG. 6 may be determined by
  • the neural network chip (such as GPU) in the radio frequency system as shown in Figure 2A realizes, or, the function of processor 602 in Figure 6 can be realized by the neural network chip and the baseband chip in the radio frequency system as shown in Figure 2A, Among them, the neural network chip can be used to mainly realize the training process of the gain control model.
  • the baseband chip can be used to mainly realize the training process and apply the training process.
  • the connection parameters (such as the target connection parameters) of the gain control model are used for the process of power control.
  • an embodiment of the present invention provides a communication device, and the communication device includes: a radio frequency gain control apparatus 701 and a first radio frequency component 702 .
  • the radio frequency gain control device 701 in the communication device shown in FIG. 7 may be as shown in 6, or, as shown in FIG. Then it can be realized by any of the following in the above Figure 2A:
  • Baseband chips Baseband chips, RF transceiver chips, RF front-ends and antennas;
  • RF transceiver chips RF front-ends and antennas
  • An embodiment of the present invention provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium.
  • the computer program is executed by a processor, the radio frequency gain control method provided in the above method embodiment is implemented, and the same Technical effects, in order to avoid repetition, will not be repeated here.
  • the computer-readable storage medium may be a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a magnetic disk or an optical disk, and the like.
  • ROM Read-Only Memory
  • RAM Random Access Memory
  • magnetic disk or an optical disk and the like.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (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)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Transmitters (AREA)
  • Transceivers (AREA)

Abstract

本发明实施例公开了一种射频增益控制方法、装置及通信设备,应用于通信技术领域,可以实现低成本、高可靠性的射频增益控制。包括:获取目标增益参数对应的目标连接参数,目标连接参数为基于多个增益控制因素的样本取值对第一增益控制模型进行训练后得到的,第一增益控制模型为针对第一射频组件建立的用于进行增益控制的神经网络模型;根据目标连接参数,配置第一射频组件的多个增益控制因素,以对第一射频组件进行增益控制。

Description

射频增益控制方法、装置及通信设备
本申请要求于2021年8月13日提交、申请号为202110931606.7、发明名称为“射频增益控制方法、装置及通信设备”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本发明实施例涉及通信技术领域,尤其涉及一种射频增益控制方法、装置及通信设备。
背景技术
第三代合作计划(3rd Generation Partnership Project,3gpp)对不同通信制式(例如2G、3G、4G、5G)的射频增益参数(包括:发射功率、发射功耗等)有着严格的规范要求,因此射频增益控制是衡量终端设备性能的重要指标之一,例如,通过射频功率控制可以通过控制发射功率来决定终端设备发送射频信号的范围和质量。目前射频系统的射频功率控制方案,是通过实验室调试和工厂调试两个阶段对射频系统各组件进行功率控制调整,以保证终端设备的射频系统中的发射功率满足3gpp规范要求,并将经过上述调试后得到的功率控制参数(可以包括功率控制因素,及各个因素的调整权值),保存在终端设备中,以供后续应用。但是上述方案,由于实验室阶段需要芯片从业人员人为调试,因此调试后的可靠性较低,并且在工厂调试阶段调试设备的成本较高,因此实现低成本、高可靠性的射频增益控制是亟需解决的问题。
发明内容
本发明实施例提供一种射频增益控制方法、装置及通信设备,用以实现低成本、高可靠性的射频增益控制。
为了解决上述技术问题,本发明实施例是这样实现的:
第一方面,提供一种射频增益控制方法,包括:
获取目标增益参数对应的目标连接参数,所述目标连接参数为基于所述多个增益控制因素的样本取值对第一增益控制模型进行训练后得到的,所述第一增益控制模型为针对第一射频组件建立的用于进行增益控制的神经网络模型;
根据所述目标连接参数,配置所述第一射频组件的所述多个增益控制因素,以对所述第一射频组件进行增益控制。
第二方面,提供一种射频增益控制装置,包括:
处理器,存储器及存储在所述存储器上并可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如第一方面所述的射频增益控制方法。
第三方面,提供一种射频增益控制装置,包括:
获取模块,用于获取目标增益参数对应的目标连接参数,所述目标连接参数为基于所述多个增益控制因素的样本取值对第一增益控制模型进行训练后得到的,所述第一增益控制模型为针对第一射频组件建立的用于进行增益控制的神经网络模型;
配置模块,用于根据所述目标连接参数,配置所述第一射频组件的所述多个增益控制因素,以对所述第一射频组件进行增益控制。
第四方面,提供一种通信设备,包括:如第二方面或第三方面所述的射频增益控制装置,以及第一射频组件。
第五方面,提供一种计算机可读存储介质,计算机可读存储介质上存储计算机程序, 计算机程序被处理器执行时实现如第一方面的射频增益控制方法。
第六方面,提供一种计算程序产品,该计算机程序产品存储有计算机程序,所述计算机程序被处理器执行时实现如第一方面的射频增益控制方法。
本申请的一个或多个实施例的细节在下面的附图和描述中提出。本申请的其它特征和有益效果将从说明书、附图以及权利要求书中体现。
附图说明
为了更清楚地说明本申请实施例中的技术方案,下面将对实施例中所需要使用的附图进行简单的介绍,显而易见地,下面描述中的附图仅仅是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1为本发明实施例提供的一种无线射频系统方案的架构示意图;
图2A为本发明实施例提供的一种射频系统的架构示意图;
图2B为本发明实施例提供的一种射频增益控制方法的流程示意图一;
图3为本发明实施例提供的一种射频增益控制方法的流程示意图二;
图4为本发明实施例提供一种射频增益控制方法的流程示意图三;
图5为本发明实施例提供一种射频增益控制装置的结构示意图一;
图6为本发明实施例提供的一种射频增益控制装置的结构示意图二;
图7为本发明实施例提供的一种通信设备的结构示意图。
具体实施方式
下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。
需要说明的是,本发明实施例中,“示例性的”或者“例如”等词用于表示作例子、例证或说明。本发明实施例中被描述为“示例性的”或者“例如”的任何实施例或设计方案不应被解释为比其它实施例或设计方案更优选或更具优势。确切而言,使用“示例性的”或者“例如”等词旨在以具体方式呈现相关概念。
首先对本发明实施例的相关技术内容进行介绍:
如图1所示,为一种无线射频系统方案的架构示意图,主要包含基带(baseband)芯片,射频收发芯片(Radio Frequency Integrated Circuit,RFIC)、射频前端(Radio Frequency Front End,RFFE)芯片和天线(antenna)几部分组成。
新无线(New radio,NR)系统的研究目前主要考虑两个频段,频段FR1(Frequency range 1)和频段FR2(Frequency range 2),其中,FR1和FR2包括的频域范围如表1所示。应理解,本发明实施例可以应用于FR1和FR2频段,也可以应用于其他频段,例如52.6GHz到71GHz的频段,或71GHz到100GHz的频段等,本申请对此并不限定。
频段定义 对应频段范围
FR1 410MHz–7.125GHz
FR2 24.25GHz–52.6GHz
表1
针对2G、3G、4G,以及NR中的FR1,射频前端中设置有接收(Rx)通路和发送(Tx)通路,以及连接接收通路或发送通路与天线之间的开关;其中,在发送通路中设置有功率放大器 (Power Amplifier,PA)矩阵。
针对NR中的FR2,天线与其他射频前端器件已经集成为毫米波天线系统集成芯片,其中,主要包括移相器矩阵、PA矩阵和天线矩阵。
3gpp对不同通信制式(2G、3G、4G、5G)的射频发射功率有着严格的规范要求,终端设备需要严格按照网络设备要求的发射功率进行射频信号发射,因此射频功率控制是衡量终端设备性能的重要指标之一。无线通信系统中射频功率控制也主要是通过如图1中所示的以上基带芯片、射频收发芯片、射频前端芯片和天线四个模块进行控制实施,具体的功率控制方案主要是通过以下两方面实施:
(1)实验室中芯片工程师,根据芯片特性和3gpp规范要求,完成部分终端样本的功率控制方案调试,对上述各模块的功率控制调整,保证实验终端设备天线的发射功率满足3gpp规范要求,并将功率控制方案和功率控制数据应用于所有终端设备;
(2)基于第一步优化调试之后,所有终端设备都可以初步正常发射,但因终端设备内部器件的个体差异的影响,可能会导致天线的发射功率不一定满足3gpp规范要求,故需要对所有终端设备进行工厂校准,可以基于第一步功率控制方案和功率控制数据进行校准,保证每一终端设备的天线口功率都符合3gpp规范。
通过上述功率方案调试,能够实现功率控制,但存在以下几点问题:
1)实验室中芯片优化调试,需要对从业人员经验要求较高,同时调试耗时较长,人为因素不可控,对各模块功率控制效率不能很好优化组合;
2)目前方案中射频功率控制复杂,为了做好射频功率控制,会引入多种功率控制方式,比如数字预失真技术(Digital Pre-Distortion,DPD)、包络跟踪(Envelope tracking,ET)等方式,导致功率控制变复杂和芯片成本变高;
3)随着通信制式增加,通信控制方案变复杂,功率控制方案也越来越复杂,其次会增加工厂生产校准时间,严重影响工厂产率,同时需要昂贵的生产设备,以及占用大量的生产资源,导致终端成本越来越高;
4)针对NR中的RF2,天线与PA等其他射频前端器件已经是集成于一颗芯片,同时因通信频率较高,使用传统LTE方式进行验证测试时,传导线损较大。
为了解决上述问题,实现低成本、高可靠性的射频增益控制。本发明实施例提供了一种射频增益控制方法、装置及通信设备,可以获取目标增益参数对应的目标连接参数,目标连接参数为基于多个增益控制因素的样本取值对第一增益控制模型进行训练后得到的,第一增益控制模型为针对第一射频组件建立的用于进行增益控制的神经网络模型;根据目标连接参数,配置第一射频组件的多个增益控制因素,以对第一射频组件进行增益控制。
本发明实施例提供的射频增益控制方法,可以应用于射频增益控制装置或者或者通信设备,该射频增益控制装置可以为通信设备中用于实现该射频增益控制方法的功能模块,或者功能实体。
本发明实施例中,所涉及的通信设备可以为网络设备或者终端设备。
上述终端设备可以称之为用户设备(user equipment,UE)、接入终端、用户单元、用户站、移动站、移动台、远方站、远程终端、移动设备、用户终端、终端、无线通信设备、用户代理或用户装置等。
终端设备可以是WLAN中的站点(STAION,ST),可以是蜂窝电话、无绳电话、会话启动协议(Session Initiation Protocol,SIP)电话、无线本地环路(Wireless Local Loop,WLL)站、个人数字处理(Personal 40Digital Assistant,PDA)设备、具有无线通信功能的手持设备、计算设备或连接到无线调制解调器的其它处理设备、车载设备、可穿戴设备、下一代通信系统例如NR网络中的终端设备,或者未来演进的公共陆地移动网络(Public  Land Mobile Network,PLMN)网络中的终端设备等。在本发明实施例中,终端设备可以部署在陆地上,包括室内或室外、手持、穿戴或车载;也可以部署在水面上(如轮船等);还可以部署在空中(例如飞机、气球和卫星上等)。
终端设备还可以是手机(Mobile Phone)、平板电脑(Pad)、带无线收发功能的电脑、虚拟现实(Virtual Reality,VR)终端设备、增强现实(Augmented Reality,AR)终端设备、工业控制(industrial control)中的无线终端设备、无人驾驶(self driving)中的无线终端设备、远程医疗(remote medical)中的无线终端设备、智能电网(smart grid)中的无线终端设备、运输安全(transportation safety)中的无线终端设备、智慧城市(smart city)中的无线终端设备或智慧家庭(smart home)中的无线终端设备等。
本发明实施例涉及的网络设备可以为接入网设备。接入网设备可以是长期演进(long-term evolution,LTE)系统、下一代(移动通信系统)(next radio,NR)系统或者授权辅助接入长期演进(authorized auxiliary access long-term evolution,LAA-LTE)系统中的演进型基站(evolutional node B,简称可以为eNB或e-NodeB)宏基站、微基站(也称为“小基站”)、微微基站、接入站点(access point,AP)、传输站点(transmission point,TP)或新一代基站(new generation Node B,gNodeB)等。在本发明实施例中,网络设备可以是用于与移动设备通信的设备,网络设备可以是WLAN中的接入点(Access Point,AP),GSM或CDMA中的基站(Base Transceiver Station,BTS),也可以是WCDMA中的基站(NodeB,NB),还可以是LTE中的演进型基站(Evolutional Node B,eNB或eNodeB),或者中继站或接入点,或者车载设备、可穿戴设备以及NR网络中的网络设备(gNB)或者未来演进的PLMN网络中的网络设备或者NTN网络中的网络设备等。在本发明实施例中,网络设备可以为小区提供服务,终端设备通过该小区使用的传输资源(例如,频域资源,或者说,频谱资源)与网络设备进行通信,该小区可以是网络设备(例如基站)对应的小区,小区可以属于宏基站,也可以属于小小区(Small cell)对应的基站,这里的小小区可以包括:城市小区(Metro cell)、微小区(Micro cell)、微微小区(Pico cell)、毫微微小区(Femto cell)等,这些小小区具有覆盖范围小、发射功率低的特点,适用于提供高速率的数据传输服务。
本发明实施例的技术方案可以应用于各种通信系统,例如:全球移动通讯(Global System of Mobile communication,GSM)系统、码分多址(Code Division Multiple Access,CDMA)系统、宽带码分多址(Wideband Code Division Multiple Access,WCDMA)系统、通用分组无线业务(General Packet Radio Service,GPRS)、长期演进(Long Term Evolution,LTE)系统、先进的长期演进(Advanced long term evolution,LTE-A)系统、新无线(New Radio,NR)系统、NR系统的演进系统、非授权频谱上的LTE(LTE-based access to unlicensed spectrum,LTE-U)系统、非授权频谱上的NR(NR-based access to unlicensed spectrum,NR-U)系统、非地面通信网络(Non-Terrestrial Networks,NTN)系统、通用移动通信系统(Universal Mobile Telecommunication System,UMTS)、无线局域网(Wireless Local Area Networks,WLAN)、无线保真(Wireless Fidelity,WiFi)、第五代通信(5th-Generation,5G)系统或其他通信系统等。
如图2A所示,为本发明实施例中提供的一种射频系统的架构示意图,该射频系统中包括:基带芯片、射频收发芯片、射频前端芯片、天线和神经网络芯片,其中,该神经网络芯片可以为微型处理器(Graphics Processing Unit,GPU)或者,该神经网络芯片可以为中央处理器(Central Processing Unit,CPU)下面将结合如图2A所示的射频系统的架构示意图,对本发明实施例提供的射频增益控制方法进行示例性的说明。
如图2B所示,为本发明实施例提供的一种射频增益控制方法的流程示意图,该方法流 程包括:
201、获取目标增益参数对应的目标连接参数。
其中,第一增益控制模型为针对第一射频组件建立的用于进行增益控制的神经网络模型,目标连接参数为基于多个增益控制因素的样本取值对第一增益控制模型进行训练后得到的。
可选的,在通信设备中可以预先保存有多组增益参数与连接参数的对应关系。示例性的,下述表2以对应关系表的形式,以保存有n组增益参数与连接参数为例,进行示例性的说明。
增益参数1 连接参数1
增益参数2 连接参数2
增益参数3 连接参数3
…… ……
增益参数n 连接参数n
表2
可选的,通信设备中可以预先保存有目标增益参数与目标连接参数的对应关系,例如,目标增益参数与目标连接参数的对应关系可以为上述保存的n组对应关系中的一组,n大于或等于2。
由于针对不同的增益参数和连接参数,获取与增益参数对应的连接参数的方式均相同,因此下面以获取目标增益参数对应的目标连接参数的具体过程为例示例性的说明。
在上述201之前,通信设备可以先基于第一射频组件的多个增益控制因素的样本取值,对第一增益控制模型进行训练,以得到目标增益控制模型,然后对应保存目标增益参数与针对多个增益控制因素的目标连接参数,该目标连接参数包括:目标增益控制模型的任意两个相邻隐含层之间的连接权重和偏置值。
其中,目标增益控制模型输出的第一增益参数与目标增益参数的误差小于或等于第一误差门限,第一增益控制模型为针对第一射频组件建立的用于进行增益控制的神经网络模型。
在基于第一射频组件的多个增益控制因素,对第一增益控制模型进行训练,以得到目标增益控制模型的过程中,可能经历了一次或多次的训练过程。具体的,可以循环执行以下a、b和c的训练步骤,直到调整后的第一增益控制模型输出的增益参数与目标发射增益的误差小于或等于第一误差门限,将调整后的第一增益控制模型作为目标增益控制模型。
a、将第一射频组件的多个增益控制因素,输入至第一增益控制模型。
b、获取第一增益控制模型输出的增益参数。
c、根据第一增益控制模型输出的增益参数与目标增益参数的误差,调整第一增益控制模型中的连接参数。
其中,该连接参数包括:第一增益控制模型的任意两个相邻隐含层之间的连接权重和偏置值。
其中,循环执行以上a、b和c的训练步骤的次数可以为一次或多次,以能够使得第一增益控制模型输出的增益参数与目标增益参数的误差小于或等于第一误差门限为准。
需要说明的是,本发明实施例中,第一误差门限可以根据实际针对第一增益控制模型的训练精度进行设置,本发明实施例不做限定。
本发明实施例中,上述第一射频组件可以包括射频系统中的一个或多个器件。可选的,以上述图2A所示的射频系统为例,第一射频组件包括以下至少一项:
(A)基带芯片;
相应的,第一增益控制模型为针对基带芯片建立的用于进行增益控制的神经网络模型。
(B)基带芯片和射频收发芯片;
相应的,第一增益控制模型为针对基带芯片和射频收发芯片建立的用于进行增益控制的神经网络模型。
(C)基带芯片、射频收发芯片和射频前端;
相应的,第一增益控制模型为针对基带芯片、射频收发芯片和射频前端建立的用于进行增益控制的神经网络模型。
(D)基带芯片、射频收发芯片、射频前端和天线;
相应的,第一增益控制模型为针对基带芯片、射频收发芯片、射频前端和天线建立的用于进行增益控制的神经网络模型。
(E)射频收发芯片和射频前端;
相应的,第一增益控制模型为针对射频收发芯片和射频前端建立的用于进行增益控制的神经网络模型。
(F)射频收发芯片、射频前端和天线;
相应的,第一增益控制模型为针对射频收发芯片、射频前端和天线建立的用于进行增益控制的神经网络模型。
(G)射频前端和天线;
相应的,第一增益控制模型为针对射频前端和天线建立的用于进行增益控制的神经网络模型。
(H)天线。
相应的,第一增益控制模型为针对天线建立的用于进行增益控制的神经网络模型。
可选的,本发明实施例中所涉及的增益控制包括当不限于以下至少一种:
1、发射功率控制,即针对第一射频组件的发射功率的控制;
2、发射功耗控制,即针对第一射频组件的发射功耗的控制;
3、接收信号强度控制,即针对第一射频组件的接收信号强度的控制。
下面将以针对增益控制为发射功率控制作为示例,对本发明实施例提供的射频增益控制方法进行详细的说明:
在如图2A所示的射频系统中,发射功率P主要是由基带芯片、射频收发芯片、射频前端芯片和天线决定。
在通信系统中,射频功率控制是典型的多因素控制问题。可选的,对于基带芯片、射频收发芯片、射频前端和天线影响功率控制的因素分别如下所示:
基带芯片的功率控制因素主要包括:信号强度、频率、温度、器件工艺水平等。本发明实施例中X表示基带芯片的功率控制因素的样本取值的集合,X=[x 1,x 2,x 3...x i],其中,x 1,x 2,x 3...x i表示基带芯片的i个不同的功率控制因素的样本取值,i大于或等于2;
射频收发芯片的功率控制因素主要包括:各级功率放大器、信道、温度、电压、器件工艺水平等。本发明实施例中Y表示射频收发芯片的功率控制因素的样本取值的集合,Y=[y 1,y 2,y 3...y j],其中,y 1,y 2,y 3...y j表示射频收发芯片的j个不同的功率控制因素的样本取值,j大于或等于2;
射频前端芯片的功率控制因素主要包括:功率放大器、电压、温度、器件工艺水平、 器件印制电路板(Printed Circuit Board,PCB)布局方式等。本发明实施例中Z表示射频前端芯片的功率控制因素样本取值的集合,Z=[z 1,z 2,z 3...z m],其中,z 1,z 2,z 3...z m表示射频前端芯片的m个不同的功率控制因素的样本取值,m大于或等于2;
天线的功率控制因素主要包括:天线形状、工作频率范围、温度、PCB布局方式等。本发明实施例中V表示天线的功率控制因素样本取值的集合,V=[v 1,v 2,v 3...v n],其中,v 1,v 2,v 3...v n表示天线的n个功率控制因素的样本取值。
由于无线通信中射频功率控制是典型的多因素控制问题,因此在本发明实施例中提出采用神经网络控制单元对各射频组件性能进行自学习训练,实现各射频组件动态基于功率控制因素的功率动态分配,确保终端的发射功率处于最优状态。
本发明实施例中,可以基于神经网络针对射频系统中组件的功率控制过程可以分为:1)自学习训练阶段和2)训练结果使用阶段,两个阶段,其中,1)自学习训练阶段的具体实现过程如下:
1)自学习训练阶段,该阶段采用多层神经网络(Multi-layer neural network,MLP)算法进行学习训练,该阶段主要是静态分析和学习射频系统中各个组件的功率控制因素对功率的影响,具体的算法实现如下:
MLP神经网络算法是一种最有效的多层神经网络学习方法,其主要特点是信号前向传递,而误差后向传播,通过不断调节神经网络中的连接参数(包括权重值和偏置值),使得神经网络的最终输出与期望输出尽可能接近,通常多层神经网络由L层神经元组成,L大于且等于2,其中第1层称为输入层,第L层被称为输出层,其余各层均被称为隐含层;结合本发明实施例中的图2A所示的射频系统,以基于基带芯片和射频收发芯片所组成的第一射频组件,所构建的神经网络模型为例对算法过程进行说明,对应的神经网络模型(即第一增益控制模型)可以定义如下:
输入层:神经网络的输入为XY=[x 1,x 2,x 3...x i,y 1,y 2,y 3...y j];其中,X表示基带芯片的功率控制因素的样本取值的集合,Y表示射频收发芯片的功率控制因素的样本取值的集合,x 1,x 2,x 3...x i表示基带芯片的i个不同的功率控制因素的样本取值,i大于或等于2,y 1,y 2,y 3...y j表示射频收发芯片的j个不同的功率控制因素的样本取值,j大于或等于2。
输出层:P=[p 1,p 2,p 3...p k],其中,P表示训练过程中训练得到的发射功率的集合,p 1,p 2,p 3...p k为进行k次训练过程中,每次训练后所输出的发射功率,其中,k大于或等于1。
隐含层:H (l)表示第l隐含层的神经元输出,
Figure PCTCN2022107390-appb-000001
其中,l大于或等于1,小于或等于L,s为第l层神经元的个数,
Figure PCTCN2022107390-appb-000002
分别表示不同的s个神经元的输出;
假设
Figure PCTCN2022107390-appb-000003
表示从l-1层第r个神经元到l层的第t个神经元之间的连接权重,
Figure PCTCN2022107390-appb-000004
表示l层第t个神经元的偏置值,
Figure PCTCN2022107390-appb-000005
Figure PCTCN2022107390-appb-000006
即为连接参数,那么
Figure PCTCN2022107390-appb-000007
并且针对隐含层中的第1层来说,
Figure PCTCN2022107390-appb-000008
其中,f()为神经元的激活函数。
根据上述函数式可以建立起从功率控制因素到输出的发射功率之间的神经网络模型,可以通过反复训练k次更新各隐含层对应的连接参数,即
Figure PCTCN2022107390-appb-000009
Figure PCTCN2022107390-appb-000010
以使得最终计算出的发射功率与预设的目标发射功率的误差小于或等于第一误差门限,并认为此时训练后的神经网络模型为目标功率控制模型(即对应于目标增益控制模型)。
进一步的,可以获取该目标功率控制模型中的
Figure PCTCN2022107390-appb-000011
Figure PCTCN2022107390-appb-000012
并将
Figure PCTCN2022107390-appb-000013
Figure PCTCN2022107390-appb-000014
保存在通信设备中,以备后续使用进行功率控制时使用。
可选的,可以将学习好的
Figure PCTCN2022107390-appb-000015
Figure PCTCN2022107390-appb-000016
保存在非易失性存储器(Non-Volatile Memory,NVM)。
202、根据目标连接参数,配置第一射频组件的多个增益控制因素,以对第一射频组件进行增益控制。
其中,目标连接参数包括:目标增益控制模型的任意两个相邻隐含层之间的连接权重和偏置值。
可选的,根据目标连接参数,配置第一射频组件的多个增益控制因素包括:根据目标连接参数和多个增益控制因素的样本取值,计算多个增益控制因素的目标取值;根据多个增益控制因素的目标取值,配置第一射频组件的多个增益控制因素,以对第一射频组件进行增益控制。
进一步的,上述根据目标连接参数和多个增益控制因素的样本取值,计算多个增益控制因素的目标取值,包括:根据目标连接参数和多个增益控制因素的样本取值进行第一运算,得到多个增益控制因素的目标取值;第一运算为针对训练后的第一增益控制模型的逆运算。这里训练后的第一增益控制模型为上述目标增益控制模型,即第一运算为针对目标增益控制模型的逆运算。
本发明实施例提供的射频增益控制方法,可以获取目标增益参数对应的目标连接参数,目标连接参数为基于多个增益控制因素的样本取值对第一增益控制模型进行训练后得到的,第一增益控制模型为针对第一射频组件建立的用于进行增益控制的神经网络模型;根据目标连接参数,配置第一射频组件的多个增益控制因素,以对第一射频组件进行增益控制。通过该方案,可以针对对应的多个增益控制因素对第一增益控制模型进行训练,得到与目标增益参数对应目标连接参数,并通过该目标连接参数,配置第一射频组件的多个增益控制因素,如此可以对第一射频组件进行增益控制,使得第一射频组件按照目标增益参数输出射频信号,相比于现有技术中人为调试提高了可靠性,并且无需复杂的调试设备,因此可以实现低成本、高可靠性的射频增益控制。
如图3所示,为本发明实施例中提供的另一种射频增益控制方法的流程示意图,该方法流程包括:
301、基于第一射频组件的多个增益控制因素的样本取值,对第一增益控制模型进行训练,以得到目标增益控制模型。
可选的,针对第一射频组件包括上述(E)、(F)、(G)和(H)中的至少一项的实施例,上述301在实现时还需要进一步引入输入增益参数作为训练第一增益控制模型的输入。也就是说,得到目标增益控制模型可以包括:基于第一射频组件的多个增益控制因素,以及输入增益参数,对第一增益控制模型进行训练,以得到目标增益控制模型。
其中,输入增益参数为第二射频组件的增益参数,第二射频组件向第一射频组件输入射频信号的射频组件。
可选的,输入增益参数为第二射频组件经过增益控制之后的增益参数。
示例性的,假设第一射频组件为(E)射频收发芯片和射频前端,那么第二射频组件可以为基带芯片。
示例性的,假设第一射频组件为(F)射频收发芯片、射频前端和天线;那么第二射频组件可以为基带芯片。
示例性的,假设第一射频组件为(G)射频前端和天线;那么第二射频组件可以为射频收发芯片和基带芯片,或者,第二射频组件可以为射频收发芯片。
示例性的,假设第一射频组件为(H)天线,那么第二射频组件可以为射频前端、射频收发芯片和基带芯片,或者,第二射频组件可以为射频前端。
302、对应保存目标增益参数与针对多个增益控制因素的目标连接参数。
其中,目标连接参数包括:目标增益控制模型的任意两个相邻隐含层之间的连接权重和偏置值。
通过上述301和302,可以针对对应的多个增益控制因素对第一增益控制模型进行训练,得到出的第一增益参数接近于目标增益参数的目标增益控制模型,并将该目标增益参数与该目标增益控制模型对应的目标连接参数进行保存,如此对应保存的该目标连接数据可以作为在后续需要输出目标增益参数时,进行功率控制的参数。
303、获取目标增益参数对应的目标连接参数。
其中,目标连接参数为基于多个增益控制因素的样本取值对第一增益控制模型进行训练后得到的,即通过上述301和302得到的。
304、根据目标连接参数,配置第一射频组件的多个增益控制因素,以对第一射频组件进行增益控制。
上述303和304对应于以上2)训练结果使用阶段,在该阶段可以先确定第一射频组件对应的目标发射功率,并从通信设备中获取之前的训练结果中,与目标发射功率匹配存储的目标连接参数,并根据该目标连接参数,配置第一射频组件的多个功率控制因素,对第一射频组件进行功率控制。
在一些可选的实现方式中,由于通信设备在不同应用场景中对于射频增益参数的需求有所不同,该阶段可以先识别通信设备的当前应用场景,并确定与当前应用场景对应的第一射频组件的目标增益参数。
也就是说,本发明实施例中,在上述301之前,以该方法应用的通信设备是终端设备为例,还可以根据终端设备的使用数据,确定终端设备的当前应用场景,然后确定与当前应用场景对应的第一射频组件的目标增益参数(如目标发射功率)。
其中,终端设备的使用数据包括以下至少一种:
(1)终端设备的剩余电量;
(2)终端设备的温度;
(3)终端设备的握持方式;
(4)终端设备的服务基站。
一种可能的实现方式中,在根据终端设备的剩余电量确定终端设备当前处于低电量场景中时,该场景下需要将发射功率降低,以节省终端设备的耗电量。
一种可能的实现方式中,在根据终端设备的温度确定终端设备当前处于高温场景中时,该场景下为了避免终端设备过热,可以将发射功率降低,以降低终端设备因为较高发射功率导致的终端设备发热问题。
另一种可能的实现方式中,在根据终端设备的握持方式确定终端设备处于非使用状态时,可以将发射功率降低;在终端设备处于使用状态时,可以将发射功率升高,以避免在使用终端设备过程中由于发射功率的限制,导致终端设备的传输距离受限。
还一种可能的实现方式中,可以根据终端设备的服务基站,确定该服务基站为终端设备当前场景所配置的相应发射功率范围,并在该发射功率范围内,确定终端设备的发射功率。可选的,还可以在该发射功率范围内考虑其他因素(例如,终端设备的剩余电量、终端设备的温度等),选择终端设备的发射功率。
需要说明的是,除了上述(1)(2)(3)和(4)所示的几种终端设备的使用参数之外,还可以通过其他终端设备的使用参数来确定终端设备的使用场景。
可选的,本发明实施例中还可以建立用于识别终端设备使用场景的神经网络模型,并根据不同场景下的终端设备的使用参数,对该神经网络模型进行训练,以得到场景识别模型。在后续使用中,可以根据实时获取的终端设备的使用参数作为该场景识别模型的输入,并获取该场景识别模型对应输出的当前应用场景,并确定与当前应用场景对应的第一射频组件的目标发射功率,然后可以按照目标发射功率作为基准,确定对应保存在终端设备中的目标连接参数。
需要说明的是,上述是以增益控制为发射功率控制为例进行说明的,针对其他增益控制,例如,发射功耗、接收信号强度等过程的实现方式,与上述发射功率控制过程的实现类似,本发明实施例不再赘述。
在一些可能的实现方式中,可以采用自学习训练标志来标识通信设备中是否保存有已经训练过的针对第一射频组件的目标连接参数。
若终端设备中存在自学习训练标志,说明终端设备中已经存在第一射频组件的对应的目标连接参数,可以进行后续的增益控制;若终端设备中不存在自学习训练标志,则说明终端设备中不存在针对第一射频组件保存的目标连接参数,需要通过神经网络训练获取目标连接参数。
其中,目标连接参数为从目标增益控制模型中获取的连接参数,目标增益控制模型输出的第一增益参数与目标增益参数的误差小于或等于第一误差门限,第一增益控制模型为针对第一射频组件建立的用于进行增益控制的神经网络模型。
本发明实施例提供的射频增益控制方法,可以获取目标增益参数对应的目标连接参数,目标连接参数为基于多个增益控制因素的样本取值对第一增益控制模型进行训练后得到的,第一增益控制模型为针对第一射频组件建立的用于进行增益控制的神经网络模型;根据目标连接参数,配置第一射频组件的多个增益控制因素,以对第一射频组件进行增益控制。通过该方案,可以针对对应的多个增益控制因素对第一增益控制模型进行训练,得到与目标增益参数对应目标连接参数,并通过该目标连接参数,配置第一射频组件的多个增益控制因素,如此可以对第一射频组件进行增益控制,使得第一射频组件按照目标增益参数输出射频信号,相比于现有技术中人为调试提高了可靠性,并且无需复杂的调试设备,因此可以实现低成本、高可靠性的射频增益控制。
可选的,如图4所示,本发明实施例还提供一种增益控制方法,该方法包括:
401、获取目标增益参数对应的目标连接参数。
402、根据目标连接参数,配置第一射频组件的多个增益控制因素,以对第一射频组件进行增益控制。
其中,该目标连接参数为终端设备通过上述如图3所示的实施例中301和302得到的目标连接参数,该目标连接参数与目标增益参数对应保存在通信设备中。
403、获取第一射频组件输出的实际增益参数。
可选的,本发明实施例中实际增益参数与目标增益参数为相对应的物理量,例如,在目标增益参数指示发射功率这一物理量时,实际增益参数也指示发射功率这一物理量;在目标增益参数指示发射功耗这一物理量时,实际增益参数也指示发射功耗这一物理量。
其中,在上述目标增益参数指示发射功率这一物理量时,上述实际增益参数可以为实际在射频系统中检测出的发射功率。
射频系统中的发射功率,会随着上述功率控制因素非线性变化,可能会出现饱和区间。本发明实施例中,可以在射频系统中设置功率反馈点,来检测功率控制后的实际发射功率,以此来评估功率控制的效果。
如图2A中所示,M为位于射频收发芯片与射频前端之间的一个功率反馈点,在该点可以通过功率感应器检测该点的实际发射功率,并利用实际发射功率去进一步优化目标增益控制模型,得到更新后的目标连接参数。
在图2A中,M点主要是评估基带芯片和射频收发芯片所组成的第一射频组件的发射功率,若根据M点的检测的实际发射功率与之前训练时的目标发射功率存在较大误差,那么可以继续对基带芯片和射频收发芯片所组成的第一射频组件进行训练。
404、确定实际增益参数与目标增益参数的误差。
针对功率控制,实际增益参数为M点检测的实际发射功率,目标增益参数为上述目标发射功率,确定实际发射功率与目标发射功率之间的误差。
405、若实际增益参数与目标增益参数的误差大于第二误差门限,则基于第一射频组件的多个增益控制因素,对目标增益控制模型进行训练。
其中,可以判断实际发射功率与目标发射功率之间的误差与第二误差门限的大小关系,在实际增益参数与目标增益参数的误差大于第二误差门限的情况下,说明根据之前与目标发射功率对应保存的目标连接参数,对基带芯片和射频收发芯片进行功率控制之后,得到的实际发射功率与预期的目标发射功率,存在较大的偏差,那么可能需要更新之间保存的目标连接参数,以使得后续进行功率控制时的匹配度。
需要说明的是,本发明实施例中,第二误差门限与第一误差门限可以相同也可以不同,具体可以根据实际针对第一增益控制模型,以及目标增益控制模型的训练精度进行设置,在此不做限定。
406、直到第一增益参数与目标增益参数的误差小于或等于第三误差门限,根据训练后的目标增益控制模型,更新与目标增益参数对应保存的目标连接参数。
其中,第三误差门限小于第一误差门限。
在实际增益参数与目标增益参数的误差大于第二误差门限时,可以说明当前进行功率控制时使用的目标连接参数并未达到实际功率控制要求,以此可以考虑对当前的目标功率控制模型进一步进行训练,在再次进行训练时,可以缩小误差的控制范围,即将第一误差门限更改为取值更小的第三误差门限,如此在将第一增益参数与目标增益参数的误差小于或等于第三误差门限作为,训练截止的条件时,可以得到更加符合实际功率控制要求的训练后的目标增益控制模型,并以此训练后的目标增益控制模型更新与目标增益参数对应保存的目标连接参数,可以保证重新存储的目标连接参数可以更加接近实际功率控制要求。
上述实施例中,在实际射频系统中设置功率反馈点,这样可以在根据本发明实施例中训练结果进行功率控制之后,根据实际发射功率和目标发射功率进行对比,这样在两者的误差较大时,可以说明之前保存的训练结果(即连接参数)在实际应用中并不匹配,如此可以继续基于实际发射功率再次对之间训练的目标增益控制模型再次进行训练,以得到优化后的增益控制模型,并从该优化后的增益控制模型中提取目标连接参数,更新与目标增益参数对应保存的目标连接参数。进一步的,还可以在图2A所示的射频系统中,还可以在其他位置设置功率反馈点,具体的,功率反馈点的位置的设置与本发明实施例中所定义的第一射频组件的相匹配。
示例性的,如图2A所示,还可以在N点设置功率反馈点,通过功率感应器检测实际发射功率,相应的,此时第一射频组件可以为以下几种情况:
(D)基带芯片、射频收发芯片、射频前端和天线;
(F)射频收发芯片、射频前端和天线;
(G)射频前端和天线;
(H)天线。
结合图2A所示,功率反馈点N位于天线端口,主要评估终端设备的最终发射功率,在本发明实施例中上述第一射频组件由射频前端和天线组成时,可以分析射频前端和天线性能,若根据N点的检测的实际发射功率与之前训练时的目标发射功率存在较大误差,那么可以继续对由射频前端和天线组成所组成的第一射频组件进行训练,以及可以重新对通信系统中,处于射频前端和天线上端的其他射频组件进行训练,例如,针对基带芯片和射频收发芯片所组成的射频组件进行训练。
由于射频前端和天线容易受终端设备中的器件布局和其他环境因素影响,并且针对不同产品可能会有不同的影响因素,而基带芯片和射频收发芯片相比之下受到外界影响较小,因此可以针对基带芯片和射频收发芯片组成的射频组件按照本实施例中方法进行射频增益控制,以及针对射频前端和天线组成的射频组件进行射频增益控制。
在针对射频前端芯片和天线组成的射频组件进行神经网络训练时,可以与上述针对基带芯片和射频收发芯片所组成的射频组件的训练过程类似的训练过程,其中,不同在于针对输入层,神经网络的输入为ZV=[z 1,z 2,z 3...z m,v 1,v 2,v 3...v n]其中,Z表示射频前端芯片的功率控制因素样本取值的集合,V表示天线的功率控制因素样本取值的集合,z 1,z 2,z 3...z m表示射频前端芯片的m个不同的功率控制因素的样本取值,m大于或等于2;v 1,v 2,v 3...v n表示天线的n个功率控制因素的样本取值。
可选的,还可以将射频收发芯片按照增益控制后的增益参数(例如,发射功率)也作为神经网络的输入,对神经网络进行训练。
进一步的,在进行功率控制时,需要射频系统中的各模块处于线性工作区间,避免处于饱和工作区间,同时要求发射功率呈现连续线性变化,避免功率控制出现饱和区间和非连续性变化。
如图5所示,本发明实施例提供一种射频增益控制装置,其特征在于,包括:
获取模块501,用于获取目标增益参数对应的目标连接参数,目标连接参数为基于多个增益控制因素的样本取值对第一增益控制模型进行训练后得到的,第一增益控制模型为针对第一射频组件建立的用于进行增益控制的神经网络模型;
配置模块502,用于根据目标连接参数,配置第一射频组件的多个增益控制因素,以对第一射频组件进行增益控制。
可选的,配置模块502,具体用于根据目标连接参数和多个增益控制因素的样本取值进行第一运算,得到多个增益控制因素的目标取值;第一运算为针对训练后的第一增益控制模型的逆运算;
根据多个增益控制因素的目标取值,配置第一射频组件的多个增益控制因素,以对第一射频组件进行增益控制。
可选的,射频增益控制装置还包括:
训练模块503,用于在获取模块501获取目标增益参数对应的目标连接参数之前,基于第一射频组件的多个增益控制因素的样本取值,对第一增益控制模型进行训练,以得到目标增益控制模型,目标增益控制模型输出的第一增益参数与目标增益参数的误差小于或等于第一误差门限;
保存模块504,用于对应保存目标增益参数与针对多个增益控制因素的目标连接参数,目标连接参数包括:目标增益控制模型的任意两个相邻隐含层之间的连接权重和偏置值。
可选的,训练模块503具体用于:
循环执行以下训练步骤:
将第一射频组件的多个增益控制因素,输入至第一增益控制模型;
获取第一增益控制模型输出的增益参数;
根据第一增益控制模型输出的增益参数与目标增益参数的误差,调整第一增益控制模型中的连接参数;
直到调整后的第一增益控制模型输出的增益参数与目标发射增益的误差小于或等于第一误差门限,将调整后的第一增益控制模型作为目标增益控制模型。
可选的,第一射频组件包括以下任一项:
基带芯片;
基带芯片和射频收发芯片;
基带芯片、射频收发芯片和射频前端;
基带芯片、射频收发芯片、射频前端和天线;
射频收发芯片和射频前端;
射频收发芯片、射频前端和天线;
射频前端和天线;
天线。
可选的,第一射频组件包括以下任一项;
射频收发芯片和射频前端;
射频收发芯片、射频前端和天线;
射频前端和天线;
天线;
训练模块503,具体用于:
基于第一射频组件的多个增益控制因素的样本取值,以及输入增益参数,对第一增益控制模型进行训练,以得到目标增益控制模型;
其中,输入增益参数为第二射频组件的增益参数,第二射频组件为向第一射频组件输入射频信号的射频组件。
可选的,训练模块503,还用于:获取第一射频组件输出的实际增益参数;
确定实际增益参数与目标增益参数的误差;
若实际增益参数与目标增益参数的误差或等于第二误差门限,则基于第一射频组件的多个增益控制因素的样本取值,对目标增益控制模型进行训练,知道第一增益参数与目标增益参数的误差小于或等于第三误差门限,根据训练后得到的目标增益控制模型,更新与目标增益参数对应保存的目标连接参数,其中,第三误差门限小于第一误差门限。
可选的,训练模块503,还用于:根据目标连接参数,配置第一射频组件的多个增益控制因素,以对第一射频组件进行增益控制之前,根据终端设备的使用数据,确定终端设备的当前应用场景;
确定与当前应用场景对应的第一射频组件的目标增益参数;
其中,使用数据包括以下至少一种:
剩余电量、温度、握持方式、服务基站。
可选的,增益控制包括以下至少一种:
发射功率控制;
发射功耗控制;
接收信号强度控制。
如图6所示,本发明实施例还提供一种射频增益控制装置,该射频控制设备包括:存储器601和处理器602,以及存储在存储器601、上并可在处理器602、上运行的计算机程序,该计算机程序被处理器执行时可以实现上述方法实施例中的射频增益控制方法。
可选的,该射频增益控制装置可以为如图2A所示的射频系统,或者,该射频增益控制装置可以包括如图2A所示的该射频系统,图6中的处理器602的功能可以由如图2A所示的射频系统中的神经网络芯片(如GPU)实现,或者,图6中的处理器602的功能可以由如图2A所示的射频系统中的神经网络芯片和基带芯片实现,其中,神经网络芯片可以用于主要实现对增益控制模型的训练过程,例如,针对第一增益控制模型,以及目标增益控制模型的训练过程,基带芯片则可以用于主要实现训练后,应用训练后的增益控制模型的连接参数(如目标连接参数)进行功率控制的过程。
如图7,本发明实施例提供一种通信设备,该通信设备包括:射频增益控制装置701和第一射频组件702。
可选的,图7中所示的该通信设备中的射频增益控制装置701可以为如6所示,或者,如图5中所示的射频增益控制装置,图7中的第一射频组件702则可以通过上述图2A中的以下任一项实现:
基带芯片;
基带芯片和射频收发芯片;
基带芯片、射频收发芯片和射频前端;
基带芯片、射频收发芯片、射频前端和天线;
射频收发芯片和射频前端;
射频收发芯片、射频前端和天线;
射频前端和天线;
天线。
本发明实施例提供一种计算机可读存储介质,该计算机可读存储介质上存储计算机程序,该计算机程序被处理器执行时实现上述方法实施例中提供的射频增益控制方法,且能达到相同的技术效果,为避免重复,这里不再赘述。
其中,该计算机可读存储介质可以为只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、磁碟或者光盘等。
需要说明的是,在本文中,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他 性的包含,从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。
通过以上的实施方式的描述,本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现,当然也可以通过硬件,但很多情况下前者是更佳的实施方式。基于这样的理解,本发明的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中,包括若干指令用以使得一台终端设备(可以是手机,计算机,服务器,空调器,或者网络设备等)执行本发明各个实施例的方法。
上面结合附图对本发明的实施例进行了描述,但是本发明并不局限于上述的具体实施方式,上述的具体实施方式仅仅是示意性的,而不是限制性的,本领域的普通技术人员在本发明的启示下,在不脱离本发明宗旨和权利要求所保护的范围情况下,还可做出很多形式,均属于本发明的保护之内。

Claims (21)

  1. 一种射频增益控制方法,其特征在于,包括:
    获取目标增益参数对应的目标连接参数,所述目标连接参数为基于所述多个增益控制因素的样本取值对第一增益控制模型进行训练后得到的,所述第一增益控制模型为针对第一射频组件建立的用于进行增益控制的神经网络模型;
    根据所述目标连接参数,配置所述第一射频组件的所述多个增益控制因素,以对所述第一射频组件进行增益控制。
  2. 根据权利要求1所述的方法,其特征在于,所述根据所述目标连接参数,配置第一射频组件对应的多个增益控制因素,包括:
    根据所述目标连接参数和所述多个增益控制因素的样本取值进行第一运算,得到所述多个增益控制因素的目标取值;所述第一运算为针对训练后的所述第一增益控制模型的逆运算;
    根据所述多个增益控制因素的目标取值,配置所述第一射频组件的所述多个增益控制因素,以对所述第一射频组件进行增益控制。
  3. 根据权利要求1所述的方法,其特征在于,所述获取目标增益参数对应的目标连接参数之前,所述方法还包括:
    基于所述第一射频组件的所述多个增益控制因素的样本取值,对第一增益控制模型进行训练,以得到目标增益控制模型,所述目标增益控制模型输出的第一增益参数与所述目标增益参数的误差小于或等于第一误差门限;
    对应保存所述目标增益参数与针对所述多个增益控制因素的目标连接参数,所述目标连接参数包括:所述目标增益控制模型的任意两个相邻隐含层之间的连接权重和偏置值。
  4. 根据权利要求3所述的方法,其特征在于,所述基于所述第一射频组件的所述多个增益控制因素的样本取值,对第一增益控制模型进行训练,以得到目标增益控制模型,包括:
    循环执行以下训练步骤:
    将所述第一射频组件的多个增益控制因素的样本取值,输入至所述第一增益控制模型;
    获取所述第一增益控制模型输出的增益参数;
    根据所述第一增益控制模型输出的增益参数与所述目标增益参数的误差,调整所述第一增益控制模型中的连接参数;
    直到调整后的所述第一增益控制模型输出的增益参数与所述目标发射增益的误差小于或等于第一误差门限,将所述调整后的所述第一增益控制模型作为所述目标增益控制模型。
  5. 根据权利要求1至4任一项所述的方法,其特征在于,所述第一射频组件包括以下任一项:
    基带芯片;
    所述基带芯片和射频收发芯片;
    所述基带芯片、所述射频收发芯片和射频前端;
    所述基带芯片、所述射频收发芯片、所述射频前端和天线;
    所述射频收发芯片和所述射频前端;
    所述射频收发芯片、所述射频前端和所述天线;
    所述射频前端和所述天线;
    所述天线。
  6. 根据权利要求3所述的方法,其特征在于,所述第一射频组件包括以下任一项;
    所述射频收发芯片和所述射频前端;
    所述射频收发芯片、所述射频前端和所述天线;
    所述射频前端和所述天线;
    所述天线;
    所述基于所述第一射频组件的所述多个增益控制因素的样本取值,对第一增益控制模型进行训练,以得到目标增益控制模型,包括:
    基于所述第一射频组件的多个增益控制因素的样本取值,以及输入增益参数,对所述第一增益控制模型进行训练,以得到目标增益控制模型;
    其中,所述输入增益参数为第二射频组件的增益参数,所述第二射频组件为向所述第一射频组件输入射频信号的射频组件。
  7. 根据权利要求3所述的方法,其特征在于,所述根据所述目标连接参数,配置所述第一射频组件的所述多个增益控制因素,以对所述第一射频组件进行增益控制之后,还包括:
    获取所述第一射频组件输出的实际增益参数;
    确定所述实际增益参数与所述目标增益参数的误差;
    若所述实际增益参数与所述目标增益参数的误差大于第二误差门限,则基于所述多个增益控制因素的样本取值,对所述目标增益控制模型进行训练,直到所述第一增益参数与目标增益参数的误差小于或等于第三误差门限,根据训练后的所述目标增益控制模型,更新与所述目标增益参数对应保存的所述目标连接参数,所述第三误差门限小于所述第一误差门限。
  8. 根据权利要求1所述的方法,其特征在于,所述根据所述目标连接参数,配置所述第一射频组件的所述多个增益控制因素,以对所述第一射频组件进行增益控制之前,所述方法还包括:
    根据终端设备的使用数据,确定所述终端设备的当前应用场景;
    确定与所述当前应用场景对应的第一射频组件的所述目标增益参数;
    其中,所述使用数据包括以下至少一种:
    剩余电量、温度、握持方式、服务基站。
  9. 根据权利要求1所述的方法,其特征在于,所述增益控制包括以下至少一种:
    发射功率控制;
    发射功耗控制;
    接收信号强度控制。
  10. 一种射频增益控制装置,其特征在于,包括:
    处理器,存储器及存储在所述存储器上并可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如权利要求1至9中任一项所述的射频增益控制方法。
  11. 一种射频增益控制装置,其特征在于,包括:
    获取模块,用于获取目标增益参数对应的目标连接参数,所述目标连接参数为基于所述多个增益控制因素的样本取值对第一增益控制模型进行训练后得到的,所述第一增益控制模型为针对第一射频组件建立的用于进行增益控制的神经网络模型;
    配置模块,用于根据所述目标连接参数,配置所述第一射频组件的所述多个增益控制因素,以对所述第一射频组件进行增益控制。
  12. 根据权利要求11所述的射频增益控制装置,其特征在于,所述配置模块在用于根据所述目标连接参数,配置所述第一射频组件的所述多个增益控制因素时,包括:
    根据所述目标连接参数和所述多个增益控制因素的样本取值进行第一运算,得到所述多个增益控制因素的目标取值;所述第一运算为针对训练后的所述第一增益控制模型的逆运算;
    根据所述多个增益控制因素的目标取值,配置所述第一射频组件的所述多个增益控制因素,以对所述第一射频组件进行增益控制。
  13. 根据权利要求11所述的射频增益控制装置,其特征在于,所述射频增益控制装置还包括:
    训练模块,用于基于所述第一射频组件的所述多个增益控制因素的样本取值,对第一增益控制模型进行训练,以得到目标增益控制模型,所述目标增益控制模型输出的第一增益参数与所述目标增益参数的误差小于或等于第一误差门限;
    保存模块,用于对应保存所述目标增益参数与针对所述多个增益控制因素的目标连接参数,所述目标连接参数包括:所述目标增益控制模型的任意两个相邻隐含层之间的连接权重和偏置值。
  14. 根据权利要求13所述的射频增益控制装置,其特征在于,所述训练模块在用于基于所述第一射频组件的所述多个增益控制因素的样本取值,对第一增益控制模型进行训练,以得到目标增益控制模型时,包括:
    循环执行以下训练步骤:
    将所述第一射频组件的多个增益控制因素的样本取值,输入至所述第一增益控制模型;
    获取所述第一增益控制模型输出的增益参数;
    根据所述第一增益控制模型输出的增益参数与所述目标增益参数的误差,调整所述第一增益控制模型中的连接参数;
    直到调整后的所述第一增益控制模型输出的增益参数与所述目标发射增益的误差小于或等于第一误差门限,将所述调整后的所述第一增益控制模型作为所述目标增益控制模型。
  15. 根据权利要求11至14任一项所述的射频增益控制装置,其特征在于,所述第一射频组件包括以下任一项:
    基带芯片;
    所述基带芯片和射频收发芯片;
    所述基带芯片、所述射频收发芯片和射频前端;
    所述基带芯片、所述射频收发芯片、所述射频前端和天线;
    所述射频收发芯片和所述射频前端;
    所述射频收发芯片、所述射频前端和所述天线;
    所述射频前端和所述天线;
    所述天线。
  16. 根据权利要求13所述的射频增益控制装置,其特征在于,所述第一射频组件包括以下任一项;
    所述射频收发芯片和所述射频前端;
    所述射频收发芯片、所述射频前端和所述天线;
    所述射频前端和所述天线;
    所述天线;
    所述训练模块在用于基于所述第一射频组件的所述多个增益控制因素的样本取值,对第一增益控制模型进行训练,以得到目标增益控制模型时,包括:
    基于所述第一射频组件的多个增益控制因素的样本取值,以及输入增益参数,对所述第一增益控制模型进行训练,以得到目标增益控制模型;
    其中,所述输入增益参数为第二射频组件的增益参数,所述第二射频组件为向所述第一射频组件输入射频信号的射频组件。
  17. 根据权利要求13所述的射频增益控制装置,其特征在于,所述训练模块,还用于在所述配置模块根据所述目标连接参数,配置所述第一射频组件的所述多个增益控制因素,以对所述第一射频组件进行增益控制之后,获取所述第一射频组件输出的实际增益参数;以及,
    确定所述实际增益参数与所述目标增益参数的误差;以及,
    若所述实际增益参数与所述目标增益参数的误差大于第二误差门限,则基于所述多个增益控制因素的样本取值,对所述目标增益控制模型进行训练,直到所述第一增益参数与目标增益参数的误差小于或等于第三误差门限,根据训练后的所述目标增益控制模型,更新与所述目标增益参数对应保存的所述目标连接参数,所述第三误差门限小于所述第一误差门限。
  18. 根据权利要求11所述的射频增益控制装置,其特征在于,所述训练模块,还用于在所述配置模块根据所述目标连接参数,配置所述第一射频组件的所述多个增益控制因素,以对所述第一射频组件进行增益控制之前,根据终端设备的使用数据,确定所述终端设备的当前应用场景;以及,
    确定与所述当前应用场景对应的第一射频组件的所述目标增益参数;
    其中,所述使用数据包括以下至少一种:
    剩余电量、温度、握持方式、服务基站。
  19. 根据权利要求11所述的射频增益控制装置,其特征在于,所述增益控制包括以下至少一种:
    发射功率控制;
    发射功耗控制;
    接收信号强度控制。
  20. 一种通信设备,其特征在于,包括:如权利要求10或11至19任一项所述的射频增益控制装置,以及所述第一射频组件。
  21. 一种计算机可读存储介质,其特征在于,包括:计算机可读存储介质上存储计算机程序,计算机程序被处理器执行时实现如权利要求1至9任一项所述的射频增益控制方法。
PCT/CN2022/107390 2021-08-13 2022-07-22 射频增益控制方法、装置及通信设备 WO2023016231A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110931606.7A CN113708781B (zh) 2021-08-13 2021-08-13 射频增益控制方法、装置及通信设备
CN202110931606.7 2021-08-13

Publications (1)

Publication Number Publication Date
WO2023016231A1 true WO2023016231A1 (zh) 2023-02-16

Family

ID=78652674

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/107390 WO2023016231A1 (zh) 2021-08-13 2022-07-22 射频增益控制方法、装置及通信设备

Country Status (2)

Country Link
CN (1) CN113708781B (zh)
WO (1) WO2023016231A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113708781B (zh) * 2021-08-13 2023-02-17 Oppo广东移动通信有限公司 射频增益控制方法、装置及通信设备
CN113917881B (zh) * 2021-12-13 2022-08-05 深圳市华杰智通科技有限公司 基于fpga的射频参数自动调节系统及方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006279490A (ja) * 2005-03-29 2006-10-12 Ricoh Co Ltd 無線lanチップおよび受信ゲイン制御方法
CN109100975A (zh) * 2018-09-03 2018-12-28 深圳市智物联网络有限公司 一种参数优化方法及系统
CN112580720A (zh) * 2020-12-18 2021-03-30 华为技术有限公司 一种模型训练方法及装置
CN112770398A (zh) * 2020-12-18 2021-05-07 北京科技大学 一种基于卷积神经网络的远端射频端功率控制方法
CN113708781A (zh) * 2021-08-13 2021-11-26 Oppo广东移动通信有限公司 射频增益控制方法、装置及通信设备

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101782743A (zh) * 2010-02-11 2010-07-21 浙江大学 神经网络建模方法及系统
CN109284551B (zh) * 2018-09-12 2023-12-08 天津工业大学 一种基于神经网络空间映射的超高频rfid天线增益建模方法
CN110012529B (zh) * 2019-03-13 2022-03-15 维沃移动通信有限公司 一种增益模式切换方法和移动终端
CN112769504B (zh) * 2021-01-25 2023-04-21 展讯通信(上海)有限公司 发射功率校准方法、电子设备和存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006279490A (ja) * 2005-03-29 2006-10-12 Ricoh Co Ltd 無線lanチップおよび受信ゲイン制御方法
CN109100975A (zh) * 2018-09-03 2018-12-28 深圳市智物联网络有限公司 一种参数优化方法及系统
CN112580720A (zh) * 2020-12-18 2021-03-30 华为技术有限公司 一种模型训练方法及装置
CN112770398A (zh) * 2020-12-18 2021-05-07 北京科技大学 一种基于卷积神经网络的远端射频端功率控制方法
CN113708781A (zh) * 2021-08-13 2021-11-26 Oppo广东移动通信有限公司 射频增益控制方法、装置及通信设备

Also Published As

Publication number Publication date
CN113708781B (zh) 2023-02-17
CN113708781A (zh) 2021-11-26

Similar Documents

Publication Publication Date Title
WO2023016231A1 (zh) 射频增益控制方法、装置及通信设备
TWI738067B (zh) 準同位框架之增強方法及使用者設備
US11985607B2 (en) Dynamic transmit power adjustment
US20160029225A1 (en) Active Antenna System And Method With Full Antenna Ports Dimension
US10903888B2 (en) Configuration method and configuration device for reference signal and communication node
RU2748706C1 (ru) Способ передачи данных и оконечное устройство
CN106464280B (zh) 用于控制无线电传输的方法和无线电节点
CN108476499B (zh) 无线网络中相同链路方向的子帧部分之间的保护时段
CN109792304A (zh) 用于基于互易性的协调多点通信的enb间空中校准
WO2017092705A1 (zh) 多模功率放大器模组、芯片及通信终端
CN108092698A (zh) 一种波束训练方法及装置
CN106803747A (zh) 多模功率放大器模组、芯片及通信终端
JP7128820B2 (ja) 端末及び通信方法
US20160028451A1 (en) Active Antenna System And Method With Full Antenna Ports Dimension
US20230354217A1 (en) Electronic device transmitting uplink signal and operating method thereof
EP3554151A1 (en) Power control method and device
CN116076025A (zh) 电子装置和用于通过电子装置设置发送信号的路径的方法
CN102378201B (zh) 一种载波聚合环境下的小区的管理方法及管理装置
CN109769293A (zh) Rru定标方法、装置、计算机设备和存储介质
CN111106864B (zh) 上行波束训练方法、终端设备和网络侧设备
CN111245374A (zh) 用于控制频谱再生的方法和装置
US20180198507A1 (en) Method and apparatus for determining direction for transmission to establish wireless connections
CN115603671A (zh) 用于测量间隙的干扰管理
CN113273092B (zh) 阵列天线控制装置和方法
CN107547146A (zh) 天线校正方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22855214

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22855214

Country of ref document: EP

Kind code of ref document: A1