WO2022151067A1

WO2022151067A1 - Wireless signal noise reduction method and apparatus, device, and storage medium

Info

Publication number: WO2022151067A1
Application number: PCT/CN2021/071545
Authority: WO
Inventors: 田文强; 肖寒; 刘文东
Original assignee: Oppo广东移动通信有限公司
Priority date: 2021-01-13
Filing date: 2021-01-13
Publication date: 2022-07-21
Also published as: CN116671024A

Abstract

The present application relates to the technical field of communications. Disclosed are a wireless signal noise reduction method and apparatus, a receiving end, and a storage medium. The method comprises: receiving a wireless signal sent by a sending end device to obtain receiving information; and processing the receiving information by means of a first noise reduction model to obtain receiving information subjected to noise reduction. In the solution, a machine learning model obtained by training by means of a receiving information sample and a sending information sample is added into the receiving end device, and before recovering source information of the receiving information, the receiving information is first subjected to noise reduction processing by means of the machine learning model, such that the influence of a channel environment and interference noise on the wireless signal is reduced, thereby improving a signal-to-noise ratio of a receiving signal and the receiving gain of a system, and improving the transmission performance of a wireless communication system.

Description

Wireless signal noise reduction method, device, device and storage medium

technical field

The present application relates to the field of communication technologies, and in particular, to a wireless signal noise reduction method, apparatus, device, and storage medium.

Background technique

In a wireless communication system, the channel environment and interference noise are one of the main problems that affect the wireless transmission performance between communication devices.

In the related art, in order to reduce the influence of the channel environment on wireless communication, the receiver usually performs channel estimation by means of pilot frequency, etc., and the receiver performs processing such as demodulation on the received signal according to the channel estimation result.

However, the channel estimation in the related art can only provide certain parameters for operations such as demodulation of the receiver, and cannot effectively eliminate the influence of the channel environment and interference noise on the wireless signal during transmission.

SUMMARY OF THE INVENTION

Embodiments of the present application provide a wireless signal noise reduction method, apparatus, device, and storage medium. The technical solution is as follows:

On the one hand, an embodiment of the present application provides a wireless signal noise reduction method, the method is performed by a receiving end device, and the method includes:

Receive the wireless signal sent by the sender device to obtain the received information;

Process the received information by using the first noise reduction model to obtain the received information after noise reduction;

The first noise reduction model is a machine learning model obtained by training the received information samples and the transmitted information samples; the received information samples are obtained by wireless signal reception of the transmitted information samples.

In a possible implementation manner, the received information is a wireless signal received by an antenna component of the receiving end device;

or,

The received information is information obtained by mathematically changing the wireless signal received by the antenna assembly of the receiving end device.

In a possible implementation, the method further includes:

Acquire channel measurement information, where the channel measurement information includes at least one of interference noise information and valid signal reception information; the interference noise information is used to indicate the interference noise situation in the environment where the receiving end device is located, and the valid signal The reception information is used to indicate the reception status of the valid signal sent by the sender device by the receiver device;

The first noise reduction model is selected from at least two candidate noise reduction models according to the channel measurement information.

In a possible implementation manner, when the channel measurement information includes the interference noise information, the acquiring the channel measurement information includes:

acquiring first information of a network configuration, where the first information is used to indicate a first resource;

The interference noise information is obtained by performing measurement on the first resource indicated by the first information.

In a possible implementation manner, the performing measurement on the first resource indicated by the first information to obtain the interference noise information includes:

Perform measurement on the first resource indicated by the first information to obtain first measurement information, where the first measurement information includes at least one of RSRP, RSRQ, RSSI, and SINR;

The interference noise information is acquired according to the first measurement information.

In one possible implementation,

When the receiving end device is a terminal, the first information is that the network side device transmits a broadcast message, a system information block (System Information Block, SIB), a radio resource control (Radio Resource Control, RRC) message, an RRC reconfiguration information command, downlink control information (Downlink Control Information, DCI), media access control layer (Media Access Control, MAC) layer control element (Control Element, CE), physical downlink control channel (Physical Downlink Control Channel, PDCCH) command ( order) at least one configuration information;

When the receiving end device is a network-side device, the first information is predefined information; or, the first information is determined by the network-side device according to unused resources.

In a possible implementation manner, when the channel measurement information includes the valid signal reception information, the acquiring the channel measurement information includes:

acquiring second information of the network configuration, where the second information is used to indicate a second resource;

On the second resource indicated by the second information, the designated signal sent by the receiving end device is measured to obtain the effective signal reception information.

In a possible implementation manner, measuring the designated signal sent by the receiving end device on the second resource indicated by the second information to obtain the valid signal receiving information includes:

Measure the designated signal sent by the receiving end device on the second resource indicated by the second information to obtain second measurement information, where the second measurement information includes RSRP, RSRQ, RSSI, SINR, packet loss at least one of a rate and a bit error rate;

The valid signal reception information is acquired according to the second measurement information.

In one possible implementation,

When the receiving end device is a terminal, the second information is information configured by the network side device through at least one of broadcast message, SIB, RRC message, RRC reconfiguration signaling, DCI, MAC CE, and PDCCH order;

When the receiving end device is a network-side device, the second information is predefined information; or, the second information is determined by the network-side device according to unused resources.

In one possible implementation,

When the receiving end device is a terminal, the designated signal is a signal predefined by a protocol, or the designated signal is a broadcast message, SIB, RRC message, RRC reconfiguration signaling, DCI, MAC A signal configured by at least one of CE and PDCCH order;

When the receiving end device is a network-side device, the designated signal is a signal predefined by a protocol, or the designated signal is a broadcast message, SIB, RRC message, or RRC reconfiguration signaling sent by the network-side device. , a signal that at least one of DCI, MAC CE, and PDCCH order is configured to the transmitting end device.

In a possible implementation manner, when the receiving end device is the network side device, the method further includes:

The second information is configured to the sending end device by at least one of broadcast message, SIB, RRC message, RRC reconfiguration signaling, DCI, MAC CE, and PDCCH order.

In a possible implementation manner, the first noise reduction model is a fully connected neural network model composed of N layers of fully connected layers, N≥1, and N is an integer.

In a possible implementation manner, the first noise reduction model is a convolutional neural network model composed of M layers of convolutional layers, where M≥1, and M is an integer.

In a possible implementation manner, the first noise reduction model sequentially includes a first fully connected layer, a first dimension adjustment layer, a first convolution layer, L common layers, a second convolution layer, and a second dimension An adjustment layer, and a second fully connected layer;

The common layer includes a third convolution layer, a normalization layer and an activation layer that are connected in sequence; the L common layers are connected in sequence; L≥1, and A is an integer.

In a possible implementation manner, a sampling operation is performed between the first dimension adjustment layer and the first convolution layer, and a stacking operation is performed between the second convolution layer and the second dimension adjustment layer ;

Wherein, the sampling operation is used to sample the output result of the first dimension adjustment layer, and output the sampling result to the first convolution layer and the superposition operation respectively; the superposition operation is used to The output result of the second convolution layer is superimposed with the output result of the sampling operation, and then output to the second dimension adjustment layer.

On the other hand, an embodiment of the present application provides a wireless signal noise reduction apparatus, the apparatus is used in a receiving end device, and the apparatus includes:

The receiving module is used to receive the wireless signal sent by the sending end device to obtain the received information;

a noise reduction processing module, configured to process the received information through the first noise reduction model to obtain the received information after noise reduction;

or,

In a possible implementation, the apparatus further includes:

A measurement information acquisition module, configured to acquire channel measurement information, where the channel measurement information includes at least one of interference noise information and valid signal reception information; the interference noise information is used to indicate interference in the environment where the receiving end device is located Noise condition, the valid signal reception information is used to indicate the reception condition of the valid signal sent by the sender device by the receiver device;

A model selection module, configured to select the first noise reduction model from at least two candidate noise reduction models according to the channel measurement information.

In a possible implementation manner, when the channel measurement information includes the interference noise information, the measurement information acquisition module is configured to:

In a possible implementation manner, the measurement information acquisition module is configured to:

In a possible implementation manner, when the receiving end device is a terminal, the first information is that the network side device transmits a broadcast message, SIB, RRC message, RRC reconfiguration signaling, DCI, MAC CE, PDCCH order Information about at least one of the configurations;

In a possible implementation manner, when the channel measurement information includes the effective signal reception information, the measurement information acquisition module is configured to:

In one possible implementation,

In a possible implementation manner, when the receiving end device is the network side device, the apparatus further includes:

A configuration module, configured to configure the second information to the sending end device through at least one of broadcast messages, SIBs, RRC messages, RRC reconfiguration signaling, DCI, MAC CE, and PDCCH order.

The common layer includes a third convolution layer, a normalization layer, and an activation layer that are connected in sequence; the L common layers are connected in sequence; L≥1, and L is an integer.

In another aspect, an embodiment of the present application provides a receiving end device, the receiving end device includes a processor, a memory and a transceiver, the memory stores a computer program, and the computer program is used by the processor/ The transceiver executes to implement the wireless signal noise reduction method described above.

In another aspect, an embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored in the storage medium, and the computer program is loaded and executed by a processor/transceiver to implement the above method for noise reduction of wireless signals .

In yet another aspect, the present application provides a computer program product or computer program comprising computer instructions stored in a computer-readable storage medium. The processor/transceiver of the computer device reads the computer instructions from the computer-readable storage medium, and the processor/transceiver executes the computer instructions, so that the computer device executes the wireless signal noise reduction method described above.

The technical solutions provided in the embodiments of the present application can bring the following beneficial effects:

By adding a machine learning model to the receiver device that is trained by receiving information samples and sending information samples, before recovering the source information of the received information, the machine learning model is used to perform noise reduction processing on the received information, thereby Reduce the influence of channel environment and interference noise on wireless signals, thereby improving the signal-to-noise ratio of received signals, as well as the receiving gain of the system, and improving the transmission performance of wireless communication systems.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following briefly introduces the drawings that are used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without creative effort.

1 is a schematic diagram of a network architecture provided by an embodiment of the present application;

FIG. 2 is a schematic diagram of the principle of a communication system provided by an embodiment of the present application;

3 is a schematic diagram of a neural network provided by an embodiment of the present application;

4 is a schematic diagram of a neural network provided by another embodiment of the present application;

5 is a flowchart of a wireless signal noise reduction method provided by an embodiment of the present application;

FIG. 6 is a schematic diagram of the principle of a communication system involved in the embodiment shown in FIG. 5;

7 is a flowchart of a wireless signal noise reduction method provided by an embodiment of the present application;

FIG. 8 is a schematic diagram of a noise reduction process on the terminal side involved in the embodiment shown in FIG. 7;

FIG. 9 is a schematic diagram of a noise reduction process on the terminal side involved in the embodiment shown in FIG. 7;

FIG. 10 is a schematic diagram of a noise reduction process on the network side involved in the embodiment shown in FIG. 7;

11 is a schematic structural diagram of a noise reduction model involved in the embodiment shown in FIG. 7;

12 is a schematic structural diagram of another noise reduction model involved in the embodiment shown in FIG. 7;

13 is a schematic structural diagram of another noise reduction model involved in the embodiment shown in FIG. 7;

14 is a block diagram of a wireless signal noise reduction apparatus provided by an embodiment of the present application;

FIG. 15 is a schematic structural diagram of a receiving end device provided by an embodiment of the present application.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

The network architecture and service scenarios described in the embodiments of the present application are for the purpose of illustrating the technical solutions of the embodiments of the present application more clearly, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. The evolution of new business scenarios and the emergence of new business scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

Please refer to FIG. 1 , which shows a schematic diagram of a network architecture of a communication system provided by an embodiment of the present application. The network architecture may include: terminal 10 and base station 20 .

The number of terminals 10 is usually multiple, and one or more terminals 10 may be distributed in a cell managed by each base station 20 . The terminal 10 may include various handheld devices with wireless communication functions, vehicle-mounted devices, wearable devices, computing devices or other processing devices connected to the wireless modem, as well as various forms of user equipment (User Equipment, UE), mobile stations ( Mobile Station, MS), terminal device, etc. For convenience of description, in the embodiments of the present application, the devices mentioned above are collectively referred to as terminals.

The base station 20 is a device deployed in the access network to provide the terminal 10 with a wireless communication function. The base station 20 may include various forms of macro base stations, micro base stations, relay stations, access points, and the like. In systems using different radio access technologies, the names of devices with base station functions may be different, for example, in the 5th generation (5th-Generation, 5G) NR system, they are called gNodeBs or gNBs. As communication technology evolves, the name "base station" may change. For convenience of description, in the embodiments of the present application, the above-mentioned apparatuses for providing wireless communication functions for the terminal 10 are collectively referred to as base stations.

Optionally, what is not shown in FIG. 1 is that the above-mentioned network architecture also includes other network devices, such as: a central control node (Central Network Control, CNC), an access and mobility management function (Access and Mobility Management Function, AMF) ) device, session management function (Session Management Function, SMF) or user plane function (User Plane Function, UPF) device, etc.

The "5G NR system" in the embodiments of the present disclosure may also be referred to as a 5G system or an NR system, but those skilled in the art can understand its meaning. The technical solutions described in the embodiments of the present disclosure may be applicable to the 4G system, the 5G NR system, and may also be applicable to the subsequent evolution system of the 5G NR system.

For ease of understanding, some related terms or background concepts involved in this application are introduced below:

1. Wireless communication.

Please refer to FIG. 2 , which shows a schematic schematic diagram of a communication system provided by an embodiment of the present application. As shown in FIG. 2 , in a wireless communication system, the basic workflow is that the transmitter performs coding, modulation, encryption and other operations on the information source at the transmitting end to form the transmission information to be transmitted. The transmitted information is transmitted to the receiving end through the wireless space, and the receiving end decodes, decrypts and demodulates the received received information, and finally restores the source information.

In the above process, the operations of encoding, modulation, encryption, decoding, demodulation, and decryption of the transmitting end and the receiving end are controllable, but the channel conditions and noise conditions in the space environment are uncontrollable, complex and changeable. of. At present, in order to recover the source information better, it is necessary to make a corresponding estimation of the channel environment in the space environment, so as to match the corresponding algorithm to obtain a better information receiving effect at the receiving end. For the interference noise in space, there is a corresponding lack of necessary processing solutions, and the source recovery conditions under different signal-to-noise ratios will show great differences.

Second, artificial intelligence.

In recent years, artificial intelligence research represented by neural networks has achieved great results in many fields, and it will also play an important role in people's production and life for a long time in the future.

Please refer to FIG. 3 , which shows a schematic diagram of a neural network provided by an embodiment of the present application. As shown in Figure 3, the basic structure of a simple neural network includes: input layer, hidden layer and output layer, as shown in Figure 2. The input layer is responsible for receiving data, the hidden layer processes the data, and the final result is generated in the output layer. Among them, each node represents a processing unit, which can be considered to simulate a neuron, and multiple neurons form a layer of neural network, and the multi-layer information transmission and processing constructs a whole neural network.

With the continuous development of neural network research, in recent years, neural network deep learning algorithms have been proposed, and more hidden layers have been introduced, and feature learning is carried out through multi-hidden layer neural network training layer by layer, which greatly improves the learning of neural networks. It is widely used in pattern recognition, signal processing, optimal combination, anomaly detection, etc.

Likewise, with the development of deep learning, convolutional neural networks have been further studied. Please refer to FIG. 4 , which shows a schematic diagram of a neural network provided by another embodiment of the present application. As shown in Figure 4, its basic structure includes: input layer, multiple convolution layers, multiple pooling layers, fully connected layers and output layers. The introduction of convolutional layers and pooling layers effectively controls the network The sharp increase of parameters limits the number of parameters and excavates the characteristics of local structures, which improves the robustness of the algorithm.

Combining the above-mentioned introduction to the principle of the basic wireless communication transceiver system, it can be known that the channel environment and noise are the biggest problems affecting the transmission scheme at present. For the channel environment, channel estimation schemes that consider pilots and other methods can alleviate the impact of channel uncertainty on communication performance to a certain extent. However, for spatial noise, the influence of this part is difficult to eliminate. It is difficult to process and eliminate the noise in the received signal in the traditional communication algorithm. Although some methods are beneficial to noise elimination, their algorithmic complexity It is also very large, and the effect is relatively limited.

In view of the above problems, this scheme considers the design of introducing a noise reduction network at the receiving end, and performs noise reduction processing on the received signal at the receiving end through a neural network, so as to obtain a better signal-to-noise ratio of the received signal and obtain a better receiving gain.

Please refer to FIG. 5 , which shows a flowchart of a wireless signal noise reduction method provided by an embodiment of the present application. The method can be executed by a receiving end device, wherein the receiving end device can be implemented as the network architecture shown in FIG. 1 . For a terminal or a base station in a corresponding communication system, the method may include the following steps:

Step 501: Receive the wireless signal sent by the sending end device to obtain received information.

In the embodiment of the present application, the receiving end device may receive the wireless signal sent by the transmitting end device through the antenna component, and obtain the original reception information. Due to the influence of the channel environment and interference noise, the received information usually carries certain noise information.

Step 502: Process the received information through the first noise reduction model to obtain the received information after noise reduction.

Please refer to 6, which shows a schematic schematic diagram of a communication system involved in an embodiment of the present application. As shown in FIG. 6 , in the embodiment of the present application, a noise reduction processing unit is added at the receiving end of the communication system. The noise reduction processing unit is a unit used to perform noise reduction processing on the received information, and one or more noise reduction models may be set therein. Optionally, after the received information is subjected to noise reduction processing, the original signal-to-noise ratio state can be improved, for example, from 10dB to 16dB.

The output of the above noise reduction model can be used for the receiving end device to perform corresponding processing to restore the source information, such as corresponding decoding, demodulation, decryption and other operations.

To sum up, in the solution shown in the embodiment of the present application, by adding a machine learning model obtained by training the received information samples and the transmitted information samples to the receiver device, before recovering the source information of the received information, firstly The machine learning model performs noise reduction processing on the received information, thereby reducing the influence of the channel environment and interference noise on the wireless signal, thereby improving the signal-to-noise ratio of the received signal, as well as the receiving gain of the system, and improving the transmission performance of the wireless communication system.

In practical applications, under the influence of factors such as different geographical locations and different time periods, the channel environment and interference noise will also be different, and the same set of noise reduction models often cannot effectively deal with all the channel environment and interference noise. Therefore, in the scheme shown in this application, the system can set up multiple sets of noise reduction models (for example, set multiple sets of model structures, each set of model structures corresponds to multiple sets of model parameters, and each model result is combined with a set of model parameters That is, a noise reduction model), the receiving end device can select a noise reduction model that matches the current channel environment and/or interference noise situation to perform noise reduction processing through the channel measurement result.

Please refer to FIG. 7 , which shows a flowchart of a wireless signal noise reduction method provided by an embodiment of the present application. The method may be executed by a receiving end device, wherein the receiving end device may be implemented as the network architecture shown in FIG. 1 . For a terminal or a base station in a corresponding communication system, the method may include the following steps:

Step 701: Receive the wireless signal sent by the sending end device to obtain received information.

Wherein, the received information is a wireless signal received by the antenna assembly of the receiving end device;

or,

In this embodiment of the present application, the input of the noise reduction model may include: the original wireless signal received by the receiving end device through the antenna component, or the receiving end device performs mathematical transformation processing on the original wireless signal (such as Fourier transform and information after at least one of mathematical processing such as singular decomposition). The output of the noise reduction network includes: the received information at the receiving end after noise reduction processing, the received information at the receiving end processed by mathematical transformation and then processed by the noise reduction network.

Step 702: Obtain channel measurement information, where the channel measurement information includes at least one of interference noise information and valid signal reception information.

Wherein, the interference noise information is used to indicate the interference noise in the environment where the receiver device is located, and the valid signal reception information is used to indicate the reception status of the valid signal sent by the receiver device to the sender device.

In this embodiment of the present application, the receiving end device can measure the channel to obtain the interference noise condition, or the reception condition of the valid signal, so as to accurately select the noise reduction model used for noise reduction processing on the received signal.

In a possible implementation solution, when the channel measurement information includes the interference noise information, the acquiring the channel measurement information includes:

acquiring first information of the network configuration, where the first information is used to indicate the first resource;

In this embodiment of the present application, the communication resource (ie, the above-mentioned first resource) used by the receiving end device to measure the interference and noise of the channel may be determined or indicated by the network.

For example, when the receiving end device is a UE, the network side configures the first information for the UE.

In a possible implementation manner, the first information is information configured by the network side device through at least one of broadcast messages, SIBs, RRC messages, RRC reconfiguration signaling, DCI, MAC CE, and PDCCH order.

That is to say, during the process of the UE accessing the network, or after accessing the network, the network-side device (such as the base station) transmits at least one of the broadcast message, SIB, RRC message, RRC reconfiguration signaling, DCI, MAC CE, and PDCCH order. One item is to configure the above-mentioned first information to the UE.

For example, please refer to FIG. 8 , which shows a schematic diagram of a noise reduction process on the terminal side involved in an embodiment of the present application. As shown in Figure 8, the noise reduction process on the terminal side is as follows:

S81, the network side indicates the first information to the UE;

S82, the UE performs measurement on the resource indicated by the first information, and estimates the current environmental noise situation, that is, obtains interference noise information;

S83, the UE determines a noise reduction processing unit (corresponding to a noise reduction model) based on the current environmental noise situation, and performs noise reduction processing on the received information.

For another example, when the receiving end device is a network-side device (such as a base station), the first information is predefined information; or, the first information is determined by the network-side device according to unused resources.

The manner in which the network-side device determines the first information may be in a predefined manner, including protocol pre-definition or operator pre-definition, so that single or multiple network devices can use a set of resources predefined by the protocol to estimate the corresponding Environmental noise situation.

Alternatively, the manner in which the network-side device determines the first information may also be determined by the resource allocation of the network-side device, for example, the network device may select resources that are not used for data transmission (for example, resources that are not allocated for uplink reception, and/or , without resources allocated for downlink transmission) to estimate the environmental noise situation.

In an exemplary solution, the resource information indicated by the foregoing first information may include time-domain resource information and frequency-domain resource information. For example, the first information may indicate a time domain start position, a time length, a time domain end position, period information, a frequency domain start position, a frequency domain range, an end position, mode information, and the like.

In an exemplary solution, the resource information indicated by the first information may be single-time resource information, multiple-time resource information, or periodic resource information.

In a possible implementation manner, the above-mentioned measuring on the first resource indicated by the first information to obtain the interference noise information includes:

Perform measurement on the first resource indicated by the first information to obtain first measurement information, where the first measurement information includes Reference Signal Receiving Power (RSRP), Reference Signal Receiving Quality (Reference Signal Receiving Quality, At least one of RSRQ), Received Signal Strength Indication (RSSI), and Signal to Interference plus Noise Ratio (SINR);

In this embodiment of the present application, when measuring interference noise, the receiving end device may measure and obtain parameters such as RSRP, RSRQ, RSSI, and SINR on the first resource indicated by the first information, and then determine the parameters used for the measurement based on the measured parameters. Interference noise information for model selection.

In an exemplary solution, the receiving end device may directly use at least one parameter of RSRP, RSRQ, RSSI and SINR obtained by measurement as the above-mentioned interference noise information.

In another exemplary solution, the receiving end device may perform preset fusion calculation or mapping according to the measured RSRP, RSRQ, RSSI, SINR and other information to obtain the interference noise information corresponding to the measured information, for example, according to the measured information The obtained information such as RSRP, RSRQ, RSSI, and SINR can obtain an interference noise level.

For example, the receiving end device may perform a weighted summation on the measured RSRP, RSRQ, RSSI, and SINR, and use the weighted summation result as the above-mentioned interference noise information, or use the weighted summation result corresponding to the interference noise level as the above-mentioned interference noise information.

On the second resource indicated by the second information, the designated signal sent by the receiving end device is measured to obtain the valid signal reception information.

In this embodiment of the present application, the communication resource (ie, the above-mentioned second resource) used by the receiving end device to measure the effective signal reception condition of the channel may be determined or indicated by the network.

For example, when the receiving end device is a UE, the network side configures the second information for the UE.

In a possible implementation manner, the second information is information configured by the network side device through at least one of broadcast messages, SIBs, RRC messages, RRC reconfiguration signaling, DCI, MAC CE, and PDCCH order.

That is to say, during the process of the UE accessing the network, or after accessing the network, the network-side device (such as the base station) transmits at least one of the broadcast message, SIB, RRC message, RRC reconfiguration signaling, DCI, MAC CE, and PDCCH order. One item is to configure the above-mentioned second information to the UE.

In a possible implementation manner, when the receiving end device is a terminal, the designated signal is a signal predefined by a protocol, or the designated signal is a broadcast message, SIB, RRC message, RRC reconfiguration by the network side device A signal configured by at least one of signaling, DCI, MAC CE, and PDCCH order.

The receiving end device and the transmitting end device need to predetermine the designated signal and transmit it on the second resource indicated by the second information, so that the receiving end device can measure the effective signal reception of the designated signal.

In an exemplary solution, when the receiving end device is a terminal, the specified signal is predefined by a protocol, so that the same specified signal can be determined by the predefined information of the protocol between the terminal and the network side device.

In another exemplary solution, when the receiving end device is a terminal, the designated signal is when the terminal accesses the network, or, after the terminal accesses the network, the network side device transmits a broadcast message, SIB, RRC message, RRC reset At least one of configuration signaling, DCI, MAC CE, and PDCCH order is configured for the terminal.

For example, please refer to FIG. 9 , which shows a schematic diagram of a noise reduction process on the terminal side involved in an embodiment of the present application. As shown in Figure 9, the noise reduction process on the terminal side is as follows:

S91, the network side indicates the second information to the UE;

S92, the UE performs measurement on the resource indicated by the second information, and estimates the current valid signal reception situation, that is, obtains the foregoing valid signal reception information;

S93, the UE determines a noise reduction processing unit (corresponding to a noise reduction model) based on the current valid signal reception, and performs noise reduction processing on the received information.

For another example, when the receiving end device is a network-side device (such as a base station), the second information is predefined information; or, the second information is determined by the network-side device according to unused resources.

The manner in which the network-side device determines the second information may be in a predefined manner, including protocol pre-definition or operator pre-definition, so that single or multiple network devices can use a set of resources predefined by the protocol to estimate the corresponding Valid signal reception.

Alternatively, the manner in which the network-side device determines the second information may also be determined by the resource allocation status of the network-side device, for example, the network device may select resources that are not used for data transmission (for example, resources that are not allocated for uplink reception, and/or , without resources allocated for downlink transmission) to estimate the effective signal reception.

Wherein, when the receiving end device is a network side device (such as a base station), the network side device also needs to instruct the transmitting end device (such as a UE) to send the above-mentioned specified signal on the resource corresponding to the second information, so that the network side device can understand the specified signal. Valid signal reception is measured. Therefore, in this embodiment of the present application, when the receiving end device is a network side device, the network side device also uses at least one of broadcast messages, SIBs, RRC messages, RRC reconfiguration signaling, DCI, MAC CE, and PDCCH order. , and configure the second information to the sending end device (such as a terminal).

Correspondingly, when the receiving end device is a network side device, the specified signal is a signal predefined by a protocol, or the specified signal is a broadcast message, SIB, RRC message, RRC reconfiguration signaling, The signal that at least one of DCI, MAC CE, and PDCCH order is configured to the transmitting end device.

For example, please refer to FIG. 10 , which shows a schematic diagram of a noise reduction process on the network side involved in an embodiment of the present application. As shown in Figure 10, the noise reduction process on the terminal side is as follows:

S1001, the network side indicates the second information to the UE;

S1002, the UE sends a designated signal on the resource indicated by the second information;

S1003, the network side measures on the resources indicated by the second information, and estimates the current valid signal reception situation, that is, obtains the valid signal reception information;

S1004, the network side determines a noise reduction processing unit (corresponding to a noise reduction model) based on the current valid signal reception, and performs noise reduction processing on the received information.

In a possible implementation manner, on the second resource indicated by the second information, the designated signal sent by the receiving end device is measured to obtain the valid signal reception information, including:

Measure the designated signal sent by the receiving end device on the second resource indicated by the second information to obtain second measurement information, where the second measurement information includes RSRP, RSRQ, RSSI, SINR, packet loss rate, and error at least one of the code rates;

According to the second measurement information, the valid signal reception information is acquired.

In the embodiment of the present application, when measuring the valid signal condition, the receiving end device may measure and obtain parameters such as RSRP, RSRQ, RSSI, SINR, packet loss rate, and bit error rate on the first resource indicated by the first information , and then determine the effective signal reception information for model selection based on the measured parameters.

In an exemplary solution, the receiving end device may directly use the measured at least one parameter of RSRP, RSRQ, RSSI, SINR, packet loss rate, and bit error rate as the above-mentioned effective signal reception information.

In another exemplary solution, the receiving end device may perform a preset fusion calculation or mapping according to the measured information such as RSRP, RSRQ, RSSI, SINR, packet loss rate, and bit error rate, and obtain the information corresponding to the measured information. For example, according to the measured information such as RSRP, RSRQ, RSSI, SINR, packet loss rate, and bit error rate, an effective signal reception level is obtained.

For example, the receiving end device may perform weighted summation on the measured RSRP, RSRQ, RSSI, SINR, packet loss rate, and bit error rate, and use the weighted summation result as the above-mentioned effective signal reception information, or The effective signal reception level corresponding to the result of the weighted summation is used as the above-mentioned effective signal reception information.

Step 703: Select the first noise reduction model from at least two candidate noise reduction models according to the channel measurement information.

In this embodiment of the present application, a plurality of noise reduction processing units may be preset in the receiving end device, and each noise reduction processing unit may correspond to a candidate noise reduction model; and, the receiving end device may also be preset with each noise reduction processing unit and Correspondence between channel measurement information; after acquiring the channel measurement information, the receiving end device can determine the noise reduction processing unit corresponding to the currently used first noise reduction model according to the preset correspondence.

Alternatively, in another exemplary solution, each noise reduction processing unit may correspond to a set of noise reduction model structures and multiple sets of model parameters, and the noise reduction model structure corresponding to each set of model parameters may constitute a noise reduction model; In addition, the model structure corresponding to each noise reduction processing unit may also be preset in the receiving end device, as well as the corresponding relationship between each set of model parameters corresponding to the model structure and the channel measurement information; the receiving end device obtains the above channel measurement information. After that, according to the preset correspondence, the corresponding noise reduction processing unit, the model parameters used by the model in the noise reduction processing unit, the model structure in the determined model noise reduction unit and the determined model parameters can be determined, That is, the above-mentioned first noise reduction model.

In this embodiment of the present application, the above-mentioned first information and second information may be configured independently, or may be configured in a unified manner.

The above-mentioned noise condition estimation based on the first information and the effective signal transmission condition estimation based on the second information can be used alone or in combination to judge the environmental noise and the effective signal reception condition.

That is to say, when the receiving end device determines the first noise reduction model, it can use the interference noise information alone to determine the first noise reduction model, or it can use the effective signal reception information to determine the first noise reduction model alone, or it can also combine the interference noise. The information and the valid signal reception information determine a first noise reduction model.

For example, when the first noise reduction model is determined by combining the interference noise information and the effective signal reception information, the receiving end device can fuse the interference noise information and the effective signal reception information (such as weighted summation) to obtain the fused information, and based on the The correspondence between the fused information and the noise reduction model determines the first noise reduction model.

In a possible implementation solution, the first noise reduction model is a fully connected neural network model composed of N layers of fully connected layers, N≥1, and N is an integer.

Please refer to FIG. 11 , which shows a schematic structural diagram of a noise reduction model involved in an embodiment of the present application. As shown in FIG. 11 , in an exemplary solution, the noise reduction model may use a fully connected network, that is, a noise reduction processing unit is formed by a fully connected network. The fully connected network here is composed of N fully connected layers, and the number of neurons in each fully connected layer is Cn. In FIG. 11, after the noise reduction information 1101 passes through the N-layer fully connected layer 1102, the noise reduction information 1103 is output.

In a possible implementation solution, the first noise reduction model is a convolutional neural network model composed of M layers of convolutional layers, M≥1, and M is an integer.

Please refer to FIG. 12 , which shows a schematic structural diagram of another noise reduction model involved in an embodiment of the present application. As shown in FIG. 12 , in an exemplary solution, the noise reduction processing unit is constituted by a convolutional neural network. The convolutional neural network here is composed of M layers of convolutional layers, the number of convolution kernels of the mth layer of convolutional layers is Km, the convolution kernel dimension Pm of the mth layer and the size of the convolution kernels are Dm=[D_1(m ), D_2(m), …D_Pm(m)]. In FIG. 12, after the noise reduction information 1201 passes through the M layers of convolution layers 1202, the noise reduction information 1203 is output.

In a possible implementation solution, the first noise reduction model sequentially includes a first fully connected layer, a first dimension adjustment layer, a first convolution layer, L common layers, a second convolution layer, and a second dimension adjustment layer. layer, and a second fully connected layer;

The common layer includes a third convolution layer, a normalization layer, and an activation layer connected in sequence; the L common layers are connected in sequence; L≥1, and A is an integer.

Wherein, the above-mentioned first fully connected layer includes a single fully connected layer, or is formed by connecting at least two fully connected layers in sequence, that is to say, the number of layers of the first fully connected layer may be 1 layer, or may be 2 layers or 2 floors or more.

Similarly, the above-mentioned first convolutional layer includes a single convolutional layer, or is formed by connecting at least two convolutional layers in sequence; the second convolutional layer includes a single convolutional layer, or is sequentially connected by at least two convolutional layers The third convolutional layer includes a single convolutional layer, or is formed by connecting at least two convolutional layers in sequence; the second fully-connected layer includes a single fully-connected layer, or is formed by connecting at least one fully-connected layer in sequence.

In a possible implementation manner, there is a sampling operation between the first dimension adjustment layer and the first convolution layer, and a stacking operation between the second convolution layer and the second dimension adjustment layer; the sampling operation It is used to sample the output result of the first dimension adjustment layer, and output the sampling results to the first convolution layer and the superposition operation respectively; the superposition operation is used for the output result of the second convolution layer to be combined with After the output results of the sampling operation are superimposed, they are output to the second dimension adjustment layer to form a Deep Residual Network (ResNet).

Please refer to FIG. 13 , which shows a schematic structural diagram of another noise reduction model involved in an embodiment of the present application. As shown in FIG. 13 , in an exemplary solution, the information 1301 to be denoised first passes through at least one fully connected layer (shown as one in FIG. 13 ), then undergoes dimension adjustment for sampling, and enters at least one volume The accumulation layer (shown as 1 in Figure 13), then enters L serially connected common layer modules, and then passes through at least one convolutional layer (shown as 1 in Figure 13) and the original sampling stack, and passes through After dimension adjustment, at least one fully connected layer (shown as one in FIG. 13 ) is used for post-processing, and finally denoised information 1302 is obtained and output. The common layer module here is composed of at least one convolutional layer (shown as one in FIG. 13 ), a normalization layer, and an activation function.

Wherein, the dimension adjustment operation of the first dimension adjustment layer and the dimension adjustment operation of the second dimension adjustment layer are the opposite operations. For example, the dimension adjustment operation of the first dimension adjustment layer adjusts the dimension of the input signal from 1×1024 to 32×32; the dimension adjustment operation of the second dimension adjustment layer adjusts the dimension of the input signal from 32×32 to 1×1024.

For example, the information to be denoised 1301 input in the above-mentioned first denoising model may be modulation symbols arranged in chronological order with a fixed length, for example, 1024 modulation symbols are input each time, and the input modulation symbols may be received by the antenna component The obtained modulation symbol, or the input modulation symbol may also be the result of performing mathematical transformation processing on the modulation symbol received by the antenna component. The noise reduction information 1301 first passes through one or more fully connected layers, and then obtains data with a dimension of 32×32 after dimension adjustment. It is superimposed with the sampled signal obtained by the original sampling processing, and then subjected to dimension adjustment and processing by one or more fully connected layers to obtain 1024 modulation symbols after noise reduction.

Step 704: Process the received information by using the first noise reduction model to obtain the received information after noise reduction.

Wherein, when the input of the noise reduction network includes: the wireless signal received by the receiving end device through the antenna assembly, the output of the first noise reduction model includes: the information after noise reduction processing is performed on the wireless signal.

When the input of the above noise reduction network includes: the wireless signal received by the receiving end device through the antenna assembly is processed by mathematical transformation, the output of the first noise reduction model includes: noise reduction processing is performed on the information after the mathematical transformation. later information.

The noise reduction model involved in the embodiments of the present application is obtained by training the received information samples and the transmitted information samples. It may be the information obtained by receiving the above-mentioned transmission information sample sent by the transmitting end device through the antenna assembly at the receiving end device side. The above-mentioned transmitted information samples may be modulation symbols obtained by encrypting, coding and modulating the information source by the transmitter of the transmitting end device, and the received information samples may be quasi-modulation symbols received by the receiving end device through an antenna.

In the embodiment of the present application, since multiple sets of candidate noise reduction models are set in the receiving end device, correspondingly, in the model training stage, multiple sample sets may be set respectively corresponding to different valid signal reception conditions and/or interference noise conditions , and according to the sample sets corresponding to the above-mentioned different valid signal reception conditions and/or interference noise conditions, candidate noise reduction models corresponding to different valid signal reception conditions and/or interference noise conditions are obtained by training.

Among them, in a possible training process of the noise reduction model, the received information samples can be input into a machine learning model with a pre-set model architecture (such as the model architecture shown in FIG. 13 ), and the output is processed by the machine learning model. The predicted denoised information, and the loss function value is calculated according to the denoised information and the corresponding sent information samples, and then the model parameters of the machine learning model are updated through the loss function value, and the above process is repeated until the model converges. Candidate noise reduction model.

For example, in one possible implementation, the training process of the noise reduction model is as follows:

1) Initialize the noise reduction model;

The training device corresponding to the noise reduction model initializes the weight parameters corresponding to the noise reduction model according to the model structure of the set noise reduction model, so as to obtain an initial untrained noise reduction model. Wherein, the initialization process may randomly assign each weight parameter of the noise reduction model, or input a preset initial value into the noise reduction model.

2) Input the received information samples into the noise reduction model;

Obtain a training sample set corresponding to the noise reduction model, where there are received information samples corresponding to the noise reduction model in the training sample set, input the received information samples into the noise reduction model, and obtain the output of the noise reduction model corresponding to the received information samples The predicted sample value; then the predicted sample value and the sent information sample corresponding to the received information sample are input into the loss function, and the loss function value corresponding to the received information sample is obtained.

3) Update the weight parameters of the noise reduction model;

After the loss function value corresponding to the received information sample is obtained according to the loss function, the noise reduction model can be gradient updated through the back propagation algorithm according to the loss function value. Wherein, the denoising model can be updated by gradient through a back-propagation algorithm according to one loss function value, or according to multiple loss function values (for example, through the sum of multiple loss function values or the mean of multiple loss function values), At the same time, gradient update of the noise reduction model is carried out through the back-propagation algorithm. Among them, the loss function can be a suitable loss function according to the type of the signal and the structure of the model, such as a cross entropy loss function, etc., and there is no limit here.

4) Obtain a trained noise reduction model;

The above process is repeated until the training meets the specified conditions, and the trained model is obtained as a trained candidate noise reduction model, so as to realize the noise reduction processing of the received information. Wherein, the specified condition may be that the number of training times reaches a training threshold, or the specified condition may be that the accuracy of the noise reduction model being verified through the validation set is greater than the validation threshold.

The above-mentioned model training process can be applied to the above-mentioned models of different structures. For example, the above-mentioned fully-connected neural network model, convolutional neural network model, deep residual network, etc. can all train the network model weights through the above-mentioned model training process.

In this solution, artificial intelligence (AI) technology is introduced into the wireless communication system to construct a noise reduction processing unit, which can perform corresponding noise reduction processing on the received signal and obtain a correspondingly better data recovery effect. The experimental test of this solution proves that after introducing the basic structure of the model shown in Figure 13 above, the noise reduction processing of the received signal can improve the signal-to-noise ratio by more than 6dB, and at the same time, a better data recovery effect can be obtained.

The reason for the above gain is that the noise reduction network can reduce the influence of noise in the received signal, and the reduction of this part of the influence is directly related to the subsequent data recovery effect. It is difficult for traditional methods to eliminate the environmental noise that is similar to white noise. However, the AI-based solution can reduce the above noise to a certain extent after a specific model structure is trained through a large number of training sets.

The following are apparatus embodiments of the present application, which can be used to execute the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

Please refer to FIG. 14 , which shows a block diagram of a wireless signal noise reduction apparatus provided by an embodiment of the present application. The apparatus may be in the receiving end device described above. As shown in Figure 14, the apparatus may include:

The receiving module 1401 is used for receiving the wireless signal sent by the sending end device to obtain the receiving information;

A noise reduction processing module 1402, configured to process the received information through the first noise reduction model to obtain the received information after noise reduction;

or,

In a possible implementation, the apparatus further includes:

In a possible implementation manner, when the receiving end device is a terminal, the second information is the network side device through broadcast messages, SIBs, RRC messages, RRC reconfiguration signaling, DCI, MAC CE, PDCCH order Information about at least one of the configurations;

In a possible implementation manner, when the receiving end device is a terminal, the specified signal is a signal predefined by a protocol, or the specified signal is a broadcast message, SIB, RRC message, A signal configured by at least one of RRC reconfiguration signaling, DCI, MAC CE, and PDCCH order;

It should be noted that, when the device provided in the above embodiment realizes its functions, only the division of the above functional modules is used as an example for illustration. In practical applications, the above functions can be allocated to different functional modules according to actual needs. That is, the content structure of the device is divided into different functional modules to complete all or part of the functions described above.

Regarding the apparatus in the above-mentioned embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be described in detail here.

Please refer to FIG. 15 , which shows a schematic structural diagram of a receiving end device 150 provided by an embodiment of the present application. The receiver device 150 may include: a processor 151 , a receiver 152 , a transmitter 153 , a memory 154 and a bus 155 .

The processor 151 includes one or more processing cores, and the processor 151 executes various functional applications and information processing by running software programs and modules.

The receiver 152 and the transmitter 153 may be implemented as a communication component, which may be a communication chip. The communication chip may also be referred to as a transceiver.

The memory 154 is connected to the processor 151 through the bus 155 .

The memory 154 can be used to store a computer program, and the processor 151 is used to execute the computer program, so as to implement each step performed by the receiving end device in the above method embodiments.

Additionally, memory 154 may be implemented by any type or combination of volatile or non-volatile storage devices including, but not limited to, magnetic or optical disks, electrically erasable programmable Read Only Memory, Erasable Programmable Read Only Memory, Static Anytime Access Memory, Read Only Memory, Magnetic Memory, Flash Memory, Programmable Read Only Memory.

In an exemplary embodiment, the receiver device includes a processor, a memory, and a transceiver (the transceiver may include a receiver and a transmitter, the receiver is used for receiving information, and the transmitter is used for transmitting information);

The transceiver is used to receive the wireless signal sent by the sending end device to obtain the received information;

the processor, configured to process the received information by using the first noise reduction model to obtain the received information after noise reduction;

For each method step performed by the receiving end device in the embodiment of the present application, reference may be made to all or part of the steps performed by the receiving end device in the embodiment shown in FIG. 5 or FIG. 7 , and details are not repeated here.

Embodiments of the present application further provide a computer-readable storage medium, where a computer program is stored in the storage medium, and the computer program is loaded and executed by a processor/transceiver to implement any of the above-mentioned ones shown in FIG. 5 or FIG. 7 . The various steps in the wireless signal noise reduction method.

The application also provides a computer program product or computer program, the computer program product or computer program comprising computer instructions stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes each step in the wireless signal noise reduction method shown in any one of FIG. 5 or FIG. 7 .

Those skilled in the art should realize that, in one or more of the above examples, the functions described in the embodiments of the present application may be implemented by hardware, software, firmware, or any combination thereof. When implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage medium can be any available medium that can be accessed by a general purpose or special purpose computer.

The above are only exemplary embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the protection of the present application. within the range.

Claims

A wireless signal noise reduction method, characterized in that the method is performed by a receiving end device, and the method includes:

Receive the wireless signal sent by the sender device to obtain the received information;

Process the received information by using the first noise reduction model to obtain the received information after noise reduction;

The first noise reduction model is a machine learning model obtained by training the received information samples and the transmitted information samples; the received information samples are obtained by wireless signal reception of the transmitted information samples.
The method of claim 1, wherein:

The received information is a wireless signal received by the antenna assembly of the receiving end device;

or,

The received information is information obtained by mathematically changing the wireless signal received by the antenna assembly of the receiving end device.
The method according to claim 1, wherein the method further comprises:

Obtain channel measurement information, where the channel measurement information includes at least one of interference noise information and valid signal reception information; the interference noise information is used to indicate the interference noise situation in the environment where the receiving end device is located, and the valid signal The reception information is used to indicate the reception status of the valid signal sent by the sender device by the receiver device;

The first noise reduction model is selected from at least two candidate noise reduction models according to the channel measurement information.
The method according to claim 3, wherein when the channel measurement information includes the interference noise information, the acquiring the channel measurement information comprises:

acquiring first information of a network configuration, where the first information is used to indicate a first resource;

The interference noise information is obtained by performing measurement on the first resource indicated by the first information.
The method according to claim 4, wherein the performing measurement on the first resource indicated by the first information to obtain the interference noise information comprises:

Perform measurement on the first resource indicated by the first information to obtain first measurement information, where the first measurement information includes reference signal received power RSRP, reference signal received quality RSRQ, received signal strength indication RSSI, and signal and at least one of the interference-plus-noise ratio SINR;

The interference noise information is acquired according to the first measurement information.
The method according to claim 4 or 5, wherein,

When the receiving end device is a terminal, the first information is that the network side device transmits broadcast messages, system information blocks SIB, radio resource control RRC messages, RRC reconfiguration signaling, downlink control information DCI, medium access control Information of at least one configuration in the layer control unit MAC CE and the physical downlink control channel order PDCCH order;

When the receiving end device is a network-side device, the first information is predefined information; or, the first information is determined by the network-side device according to unused resources.
The method according to claim 3, wherein when the channel measurement information includes the valid signal reception information, the acquiring the channel measurement information comprises:

acquiring second information of the network configuration, where the second information is used to indicate a second resource;

On the second resource indicated by the second information, the designated signal sent by the receiving end device is measured to obtain the effective signal reception information.
The method according to claim 7, wherein, on the second resource indicated by the second information, the designated signal sent by the receiving end device is measured to obtain the effective signal reception information ,include:

Measure the designated signal sent by the receiving end device on the second resource indicated by the second information to obtain second measurement information, where the second measurement information includes RSRP, RSRQ, RSSI, SINR, packet loss at least one of a rate and a bit error rate;

The valid signal reception information is acquired according to the second measurement information.
The method according to claim 7 or 8, characterized in that,

When the receiving end device is a terminal, the second information is the information configured by the network side device through at least one of broadcast message, SIB, RRC message, RRC reconfiguration signaling, DCI, MAC CE, PDCCH order;

When the receiving end device is a network-side device, the second information is predefined information; or, the second information is determined by the network-side device according to unused resources.
The method according to claim 7 or 8, wherein,

When the receiving end device is a terminal, the designated signal is a signal predefined by a protocol, or the designated signal is a broadcast message, SIB, RRC message, RRC reconfiguration signaling, DCI, MAC A signal configured by at least one of CE and PDCCH order;

When the receiving end device is a network-side device, the designated signal is a signal predefined by a protocol, or the designated signal is a broadcast message, SIB, RRC message, or RRC reconfiguration signaling sent by the network-side device. , a signal that at least one of DCI, MAC CE, and PDCCH order is configured to the transmitting end device.
The method according to claim 7 or 8, wherein when the receiving end device is the network side device, the method further comprises:

The second information is configured to the sending end device by at least one of broadcast message, SIB, RRC message, RRC reconfiguration signaling, DCI, MAC CE, and PDCCH order.
The method according to any one of claims 1 to 11, wherein the first noise reduction model is a fully connected neural network model composed of N layers of fully connected layers, N≥1, and N is an integer.
The method according to any one of claims 1 to 11, wherein the first noise reduction model is a convolutional neural network model composed of M layers of convolutional layers, where M≥1, and M is an integer.
The method according to any one of claims 1 to 11, wherein the first noise reduction model sequentially includes a first fully connected layer, a first dimension adjustment layer, a first convolution layer, L common layers, a first A second convolutional layer, a second dimension adjustment layer, and a second fully connected layer;

The common layer includes a third convolution layer, a normalization layer and an activation layer that are connected in sequence; the L common layers are connected in sequence; L≥1, and A is an integer.
The method of claim 14, wherein:

There is a sampling operation between the first dimension adjustment layer and the first convolution layer, and a stacking operation between the second convolution layer and the second dimension adjustment layer;

Wherein, the sampling operation is used to sample the output result of the first dimension adjustment layer, and output the sampling result to the first convolution layer and the superposition operation respectively; the superposition operation is used to The output result of the second convolution layer is superimposed with the output result of the sampling operation, and then output to the second dimension adjustment layer.
A wireless signal noise reduction device, characterized in that the device is used in a receiving end device, and the device comprises:

The receiving module is used to receive the wireless signal sent by the sending end device to obtain the received information;

a noise reduction processing module, configured to process the received information through the first noise reduction model to obtain the received information after noise reduction;

The first noise reduction model is a machine learning model obtained by training the received information samples and the transmitted information samples; the received information samples are obtained by wireless signal reception of the transmitted information samples.
The apparatus of claim 16, wherein:

The received information is a wireless signal received by the antenna assembly of the receiving end device;

or,

The received information is information obtained by mathematically changing the wireless signal received by the antenna assembly of the receiving end device.
The apparatus of claim 16, wherein the apparatus further comprises:

A measurement information acquisition module, configured to acquire channel measurement information, where the channel measurement information includes at least one of interference noise information and valid signal reception information; the interference noise information is used to indicate interference in the environment where the receiving end device is located Noise condition, the valid signal reception information is used to indicate the reception condition of the valid signal sent by the sender device by the receiver device;

A model selection module, configured to select the first noise reduction model from at least two candidate noise reduction models according to the channel measurement information.
The apparatus according to claim 18, wherein when the channel measurement information includes the interference noise information, the measurement information acquisition module is configured to:

acquiring first information of a network configuration, where the first information is used to indicate a first resource;

The interference noise information is obtained by performing measurement on the first resource indicated by the first information.
The device according to claim 19, wherein the measurement information acquisition module is configured to:

Perform measurement on the first resource indicated by the first information to obtain first measurement information, where the first measurement information includes at least one of RSRP, RSRQ, RSSI, and SINR;

The interference noise information is acquired according to the first measurement information.
The device according to claim 19 or 20, characterized in that,

When the receiving end device is a terminal, the first information is information configured by the network side device through at least one of broadcast message, SIB, RRC message, RRC reconfiguration signaling, DCI, MAC CE, and PDCCH order;

When the receiving end device is a network-side device, the first information is predefined information; or, the first information is determined by the network-side device according to unused resources.
The apparatus according to claim 18, wherein when the channel measurement information includes the effective signal reception information, the measurement information acquisition module is configured to:

acquiring second information of the network configuration, where the second information is used to indicate a second resource;

On the second resource indicated by the second information, the designated signal sent by the receiving end device is measured to obtain the effective signal reception information.
The device according to claim 22, wherein the measurement information acquisition module is configured to:

Measure the designated signal sent by the receiving end device on the second resource indicated by the second information to obtain second measurement information, where the second measurement information includes RSRP, RSRQ, RSSI, SINR, packet loss at least one of a rate and a bit error rate;

The valid signal reception information is acquired according to the second measurement information.
The device according to claim 22 or 23, characterized in that,

When the receiving end device is a terminal, the second information is information configured by the network side device through at least one of broadcast message, SIB, RRC message, RRC reconfiguration signaling, DCI, MAC CE, and PDCCH order;

When the receiving end device is a network-side device, the second information is predefined information; or, the second information is determined by the network-side device according to unused resources.
The device according to claim 22 or 23, characterized in that,

When the receiving end device is a terminal, the designated signal is a signal predefined by a protocol, or the designated signal is a network-side device through a broadcast message, SIB, RRC message, RRC reconfiguration signaling, DCI, MAC A signal configured by at least one of CE and PDCCH order;

When the receiving end device is a network-side device, the designated signal is a signal predefined by a protocol, or the designated signal is a broadcast message, SIB, RRC message, or RRC reconfiguration signaling sent by the network-side device. , a signal that at least one of DCI, MAC CE, and PDCCH order is configured to the transmitting end device.
The apparatus according to claim 22 or 23, wherein when the receiving end device is the network side device, the apparatus further comprises:

A configuration module, configured to configure the second information to the sending end device through at least one of broadcast message, SIB, RRC message, RRC reconfiguration signaling, DCI, MAC CE, and PDCCH order.
The apparatus according to any one of claims 16 to 26, wherein the first noise reduction model is a fully connected neural network model composed of N layers of fully connected layers, N≥1, and N is an integer.
The apparatus according to any one of claims 16 to 26, wherein the first noise reduction model is a convolutional neural network model composed of M layers of convolutional layers, where M≥1, and M is an integer.
The apparatus according to any one of claims 16 to 26, wherein the first noise reduction model sequentially comprises a first fully connected layer, a first dimension adjustment layer, a first convolution layer, L common layers, a first A second convolutional layer, a second dimension adjustment layer, and a second fully connected layer;

The common layer includes a third convolution layer, a normalization layer, and an activation layer that are connected in sequence; the L common layers are connected in sequence; L≥1, and L is an integer.
The apparatus of claim 29, wherein

There is a sampling operation between the first dimension adjustment layer and the first convolution layer, and a stacking operation between the second convolution layer and the second dimension adjustment layer;

Wherein, the sampling operation is used to sample the output result of the first dimension adjustment layer, and output the sampling results to the first convolution layer and the superposition operation respectively; the superposition operation is used to The output result of the second convolution layer is superimposed with the output result of the sampling operation, and then output to the second dimension adjustment layer.
A receiving end device, characterized in that the receiving end device includes a processor, a memory and a transceiver;

The transceiver is used to receive the wireless signal sent by the sending end device to obtain the received information;

the processor, configured to process the received information by using the first noise reduction model to obtain the received information after noise reduction;

The first noise reduction model is a machine learning model obtained by training the received information samples and the transmitted information samples; the received information samples are obtained by wireless signal reception of the transmitted information samples.
A computer-readable storage medium, characterized in that a computer program is stored in the storage medium, and the computer program is used to be executed by a processor/transceiver, so as to realize the method according to any one of claims 1 to 15. Wireless signal noise reduction method.