CN114142883A - Transmission device, reception device, transmission method, and reception method - Google Patents

Abstract

The disclosure provides a transmitting apparatus, a receiving apparatus, a transmitting method and a receiving method. The transmission apparatus includes: a processing unit for determining a learning reference signal sequence, wherein the learning reference signal sequence comprises signals relating to constellation points and non-constellation points in a constellation diagram; and a transmitting unit for transmitting information on the learning reference signal sequence.

Description

Transmission device, reception device, transmission method, and reception method

Technical Field

The present disclosure relates to the field of wireless communication, and more particularly, to a method and a device for transmitting and receiving a learned reference signal sequence and a corresponding transmitting device and receiving device.

Background

Deep neural network-based receivers (DNN-receivers) are an important research direction for B5G/6G. The receiver or the receiving device of the traditional deep neural network adopts an off-line training mode, so that the receiver of the traditional deep neural network can obtain better performance when the off-line training is the same as the transmission environment deployed on line, and the performance of the receiver of the traditional deep neural network is lost when the transmission environment deployed on line changes.

In order to improve the adaptability of the DNN to different environments, different training environments are introduced during offline training so as to improve the adaptability to the environment during online deployment. In particular, a larger data set may be introduced during offline training, however, this requires a large amount of data to be generated during training, which increases training complexity. In addition, multitasking or transfer learning may also be introduced for different training environments, however, the network may sacrifice performance for a single environment when migrating to multiple environments.

On the other hand, the online training of the model by using the online obtained samples is provided, so that the adaptability to different environments can be improved. The online training of the model using the online obtained samples may include incremental learning based online training, meta learning based online training. Specifically, the online training based on incremental learning refers to adding online obtained samples into a training set as incremental samples to update network weights, however, since the online samples are taken as increments, a larger number of online samples are required to realize adaptation to a new environment. On-line training based on meta-learning refers to that a network model is more easily adapted to a new task through network model design or optimization algorithm design, and meanwhile, an on-line sample is adopted to update a network during on-line deployment. Compared with other training modes, the online training based on the meta-learning considers the efficiency of the online training at the beginning of model design or optimization algorithm design, so the adaptability to new environment is strong, the demand on online samples is small, and the performance after online learning is better.

In the meta-Learning based online training, a receiving end learns a dynamic environment using an online training sample transmitted from a transmitting end, where the online training sample may also be referred to as a Learning Reference Signal (LRS) sequence. Currently, there is a lack of an LRS sequence design rule and a method for designing an LRS sequence under the rule, which results in that in the existing scheme, the LRS sequence is designed in a heuristic manner, in other words, the LRS sequence is formed by randomly selecting constellation points by a transmitting end. And the LRS formed by randomly selecting the constellation points cannot optimize the performance after online updating.

Disclosure of Invention

According to an aspect of the present disclosure, there is provided a transmission apparatus including: a processing unit for determining a learning reference signal sequence, wherein the learning reference signal sequence comprises signals relating to constellation points and non-constellation points in a constellation diagram; and a transmitting unit for transmitting information on the learning reference signal sequence.

According to another aspect of the present disclosure, there is provided a reception apparatus including: a receiving unit configured to receive information on a learning reference signal sequence, wherein the learning reference signal sequence includes signals on constellation points and non-constellation points in a constellation diagram; a processing unit, configured to update the network parameters of the receiving device according to the received information on the learned reference signal sequence, so as to perform data detection.

According to another aspect of the present disclosure, there is provided a transmission method including: determining a learning reference signal sequence, wherein the learning reference signal sequence comprises signals for constellation points and non-constellation points in a constellation diagram; and transmitting information on the learning reference signal sequence.

According to another aspect of the present disclosure, there is provided a reception method including: receiving information on a learning reference signal sequence, wherein the learning reference signal sequence comprises signals on constellation points and non-constellation points in a constellation diagram; and updating the network parameters of the receiving device according to the received information about the learning reference signal sequence so as to perform data detection.

Drawings

The above and other objects, features and advantages of the present disclosure will become more apparent by describing in more detail embodiments of the present disclosure with reference to the attached drawings. The accompanying drawings are included to provide a further understanding of the embodiments of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the principles of the disclosure and not to limit the disclosure. In the drawings, like reference numbers generally represent like parts or steps.

Fig. 1 is a schematic block diagram illustrating a transmitting device according to one embodiment of the present disclosure.

Fig. 2 is a schematic diagram illustrating an LRS sequence according to one example of the present disclosure.

Fig. 3A is a diagram illustrating a candidate reference signal sequence table.

Fig. 3B is a diagram illustrating another candidate reference signal sequence table.

Fig. 4 is a schematic block diagram illustrating a receiving apparatus according to one embodiment of the present disclosure.

Fig. 5 is a flow diagram of a transmitting method according to one embodiment of the present disclosure.

Fig. 6 is a flowchart of a receiving method performed by a receiving device according to one embodiment of the present disclosure.

Fig. 7 is a schematic diagram of a hardware structure of an apparatus according to an embodiment of the present disclosure.

Detailed Description

In order to make the objects, technical solutions and advantages of the present disclosure more apparent, example embodiments according to the present disclosure will be described in detail below with reference to the accompanying drawings. In the drawings, like reference numerals refer to like elements throughout. It should be understood that: the embodiments described herein are merely illustrative and should not be construed as limiting the scope of the disclosure. Further, the terminal described herein may include various types of terminals, such as a User Equipment (UE), a mobile terminal (or referred to as a mobile station), or a fixed terminal, however, for convenience, the terminal and the UE are sometimes used interchangeably hereinafter. Further, in the following, a receiver and a receiving device are sometimes used interchangeably.

The meta-learning based online training may include metric-based meta-learning, model-based, and optimization-based meta-learning. Optimization-based meta-learning may be employed in embodiments consistent with the present disclosure. By improving the gradient descent algorithm of the neural network based on the optimized meta-learning, the online gradient descent algorithm can be rapidly converged under the condition of a small sample. The receiving device in the embodiments of the present application is a deep neural network-based receiving device (DNN-receiver). More specifically, the receiving apparatus in the embodiment of the present application may be an optimized meta learning based receiving apparatus.

A typical method of Meta-Learning based optimization is "Model-independent Meta-Learning (MAML)". The goal of offline training of MAML is to find the neural network parameter set such that the distance of the neural network parameter set to the optimal parameter manifold for tasks 1 and 2 is the shortest. For example, task 1 may be communication in an indoor environment, and task 2 may be communication in an outdoor environment, and the offline training of the deep neural network-based receiving device according to the embodiments of the present application aims to find the neural network parameter set such that the distance of the neural network parameter set to the optimal parameter manifold of the communication in the indoor environment and the communication in the outdoor environment is the shortest. Therefore, when the neural network parameter set is deployed on line, the neural network parameter set can be quickly updated to a parameter set suitable for a specific task in the face of the specific task.

Next, a transmission apparatus 100 according to an embodiment of the present disclosure will be described with reference to fig. 1. As shown in fig. 1, a transmitting device 100 according to one embodiment of the present disclosure may include a processing unit 110 and a transmitting unit 120. The transmission apparatus 100 may include other components in addition to the processing unit and the transmission unit, however, since these components are not related to the contents of the embodiments of the present disclosure, illustration and description thereof are omitted herein.

As shown in fig. 1, the processing unit 110 may determine a Learning Reference Signal (LRS) sequence, wherein the learning reference signal sequence includes signals regarding constellation points and non-constellation points in a constellation diagram. According to one example of the present disclosure, the constellation may be represented by a codebook, and the constellation points in the constellation may be represented by codewords in the codebook. Fig. 2 is a schematic diagram illustrating an LRS sequence according to one example of the present disclosure. The length of the LRS sequence is 2 in the example shown in fig. 2. As shown in fig. 2, a first LRS symbol 210 in the LRS sequence is a constellation point, and a second LRS symbol 220 in the LRS sequence is a non-constellation point. Different symbols have different importance when training a Deep Neural Network (DNN). For example, the symbols that are constellation points are simple samples (i.e., sharp samples) that are more important early in the network training, while the symbols that are non-constellation points are difficult samples (i.e., fuzzy samples) that are more important late in the network training.

According to another example of the present disclosure, one or more candidate reference signal sequences may be generated according to a constellation diagram and a maximization loss function, where the loss function is a difference between a loss calculated based on training data received using a neural network parameter set before an update of the training reference signal sequence (hereinafter referred to as "loss before update") and a loss calculated based on training data received using a neural network parameter set after an update of the training reference signal sequence (hereinafter referred to as "loss after update"). And the processing unit 110 may determine the learning reference signal sequence from the generated one or more candidate reference signal sequences.

Specifically, one or more candidate reference signal sequences may be generated in advance by the first training device and the second training device. For example, the first training device and the second training device may be specific modules included in the base station. As another example, the first training device and the second training device may be dedicated devices deployed at the base station location. Furthermore, in order to obtain a true channel between the first training device and the second training device, the first training device may be arranged at the location of the base station, while the second device is arranged at a location where the user equipment may be present in the cell to which the base station belongs.

During off-line training, the first training device may generate a training reference signal sequence according to a constellation diagram and transmit the training reference signal sequence to the second training device, and the second training device may receive the training reference signal sequence and training data from the first training device. The second training device may detect training data using the initial set of neural network parameters and calculate a loss from the detected training data to obtain a pre-update loss. In addition, the second training device may further update the neural network parameter set according to the received training reference signal sequence, detect training data using the updated neural network parameter set, and calculate a loss according to the detected training data to obtain an updated loss. The second training device may then calculate a loss function, i.e. the difference between the loss before the update and the loss after the update and return to the first training device. The first training device may update the training reference signal sequence according to the constellation diagram and the loss function. And repeating the updating process of the training reference signal sequence until a preset condition is reached, and taking the obtained sequence as a candidate reference signal sequence.

In embodiments according to the present disclosure, the loss function may be arbitrary. For example, the loss may be cross entropy of data, Bit Error Rate (BER), or the like. Further, the loss function is not limited thereto in the embodiment according to the present disclosure, and the loss function may be adjusted according to actual needs or performance indexes.

According to another example of the present disclosure, the one or more candidate reference signal sequences may be generated using a reference signal sequence generation network. For example, the reference signal sequence generating network may be a single hidden layer, linear activation function. The processing unit 110 may perform network random initialization through the reference signal sequence generation network to generate multiple candidate reference signal sequences with close performance for the same constellation or codebook. For example, multiple candidate reference signal sequences with close performance for the same constellation or codebook may be generated by the reference signal sequence generation network using different initialization parameters. Furthermore, the processing unit 110 may also generate a plurality of candidate reference signal sequences respectively corresponding to a plurality of constellations or codebooks through a reference signal sequence generation network. For example, when performing offline training using the first training apparatus and the second training apparatus described above, the reference signal sequence generation network may be set in the first training apparatus. The first training device may generate a candidate reference signal sequence using a reference signal sequence generation network. The second training device may calculate the loss function based on candidate reference signal sequences generated by the first training device using the reference signal sequence generation network. In particular, the second training device may calculate the loss function based on the original candidate reference signal sequences generated by the first training device (i.e., candidate reference signal sequences that have not been transmitted over the channel) and the candidate reference signal sequences that have been transmitted over the channel. For another example, when online deployment is performed, a reference signal sequence generation network may be provided in the transmitting device.

According to another example of the present disclosure, the transmitting device and the receiving device may include a storage unit to store the generated one or more candidate reference signal sequences. For example, the generated one or more candidate reference signal sequences may be stored in the form of a candidate reference signal sequence list, along with sequence indices corresponding to the respective sequences. Fig. 3A and 3B are schematic diagrams illustrating a candidate reference signal sequence table. In the example shown in fig. 3A, the candidate reference signal sequence table 310 includes all codebooks/constellations and one or more candidate reference signal sequences corresponding to each codebook/constellation. In the example shown in fig. 3B, the candidate reference signal sequence list comprises a plurality of sub-lists, such as a candidate reference signal sequence sub-list 321, a candidate reference signal sequence sub-list 322, and the like, wherein each candidate reference signal sequence sub-list is for one or more codebooks/constellations, and each candidate reference signal sequence sub-list may include one or more candidate reference signal sequences corresponding to the codebooks/constellations. The example shown in fig. 3B selects a reference signal sequence sub-table for a codebook/constellation construction, which improves the flexibility of indication and reduces the signaling overhead.

Returning to fig. 1, the transmitting unit 120 may transmit information about the learned reference signal sequence determined by the processing unit 110. The receiving device may update the neural network parameter set of the receiving device according to the information of the learned reference signal sequence transmitted by the transmitting unit 120 to perform data detection.

According to an example of the present disclosure, the information on the learning reference signal sequence includes a sequence index of the determined learning reference signal sequence and at least one of the determined learning reference signal sequences. For example, as described above, there may be multiple candidate reference signal sequences. In this case, the transmitting unit 120 may transmit a sequence index to indicate the learning reference signal sequence determined by the processing unit 110. And the transmitting unit 120 may also transmit the learning reference signal sequence corresponding to the sequence index.

According to another example of the present disclosure, the learning reference signal sequence may also be pre-configured. In this case, the sequence index does not need to be sent, and the information of the learned reference signal sequence may include the learned reference signal sequence determined by the processing unit 110. For example, in case of uplink transmission, that is, the transmitting apparatus is a user equipment and the receiving apparatus is a base station, the base station may configure a learning reference signal sequence which it should use for the user equipment in advance, so that the user equipment may transmit the learning reference signal sequence in advance according to the configuration. According to another example of the present disclosure, the transmitting unit 120 may transmit the learning reference signal sequence with data. For example, the transmitting unit 120 may transmit the learning reference signal sequence first and then transmit data next when data is to be transmitted. Alternatively, the transmitting unit 120 may transmit information on the learning reference signal sequence according to a trigger signal. For example, when a receiving device or a transmitting device detects that a channel environment changes or is not as expected, transmission of a learning reference signal sequence may be triggered. For example, the receiving device may calculate a loss from the received data and trigger the transmitting device to transmit the sequence of learning reference signals when the loss satisfies a predetermined condition. Specifically, when the loss is a Bit Error Rate (BER), if the loss is greater than or equal to a predetermined threshold, transmission of the learning reference signal sequence is triggered. When the loss is cross entropy, if the loss is less than or equal to a predetermined threshold, then triggering transmission of a learned reference signal sequence.

Further, in an example according to the invention, the learned reference signal sequence transmitted by the transmitting unit may be used to replace a reference signal in an existing system, and the receiving apparatus may perform channel estimation based on the received learned reference signal sequence. For example, the learned reference signal sequence replaces a Cell Reference Signal (CRS), a demodulation reference signal (DMRS), a channel state reference signal (CSI-RS), etc. in the existing system. For another example, the learned reference signal sequence may also replace the reference, broadcast channel, etc. of the SSB, SRS, etc. in the existing system.

Next, a reception apparatus 400 according to an embodiment of the present disclosure will be described with reference to fig. 4. The receiving device 400 may be a deep neural network-based receiver (DNN-receiver). As shown in fig. 4, a receiving apparatus 400 according to one embodiment of the present disclosure may include a receiving unit 410 and a processing unit 420. The receiving apparatus 400 may include other components in addition to the processing unit and the receiving unit, however, since these components are not related to the contents of the embodiments of the present disclosure, illustration and description thereof are omitted herein.

As shown in fig. 4, the receiving unit 410 may receive information on a learning reference signal sequence including signals on constellation points and non-constellation points in a constellation diagram. The learning reference signal sequence is described in detail above with reference to fig. 2, and therefore is not described herein again.

The processing unit 420 updates the neural network parameter set of the receiving apparatus 400 according to the information on the learning reference signal sequence for data detection. According to an example of the present disclosure, the processing unit 420 may obtain a learning reference signal sequence that is not transmitted through the channel according to the information on the learning reference signal sequence, and update the neural network parameter set of the receiving device according to the learning reference signal sequence that is transmitted through the channel and the learning reference signal sequence that is not transmitted through the channel.

Specifically, as described above, the information of the learning reference signal sequence may include at least one of a sequence index of the learning reference signal sequence and the learning reference signal sequence. The processing unit 420 may obtain a learned reference signal sequence that is not transmitted through a channel from one or more candidate reference signal sequences stored in advance according to the sequence index. Alternatively, in the case where the learning reference signal sequence has been previously configured, the previously configured learning reference signal sequence may be obtained as a learning reference signal sequence that is not transmitted through the channel. Further, the processing unit 420 may take the learning reference signal sequence transmitted from the transmitting apparatus received by the accepting unit 410 as the learning reference signal sequence after being transmitted through the channel. The processing unit 420 may update the neural network parameter set of the receiving device according to the learning reference signal sequence transmitted through the channel and the learning reference signal sequence not transmitted through the channel. Further, according to an example of the present disclosure, after data detection using the updated set of neural network parameters, the processing unit 420 may also determine whether a loss calculated from the detected data satisfies a predetermined condition. The receiving device may further include a transmitting unit. When the processing unit 420 determines that the loss satisfies the predetermined condition, the transmission unit may transmit the learning reference signal sequence trigger signal to the transmission apparatus. For example, when the loss is a Bit Error Rate (BER), if the loss is greater than or equal to a predetermined threshold, the processing unit 420 may determine that the loss satisfies a predetermined condition, and the transmitting unit may transmit the learning reference signal sequence trigger signal to the transmitting apparatus. For another example, when the loss is cross entropy, if the loss is less than or equal to a predetermined threshold, the processing unit 420 may determine that the loss satisfies a predetermined condition, and the transmission unit may transmit the learning reference signal sequence trigger signal to the transmission apparatus.

In the transmitting apparatus and the receiving apparatus according to the embodiments of the present disclosure, training of neural network parameters for different scenarios and channel environments may be better adapted by generating and transmitting a learning reference signal sequence including signals regarding constellation points and non-constellation points in a constellation diagram to a receiver. In addition, by generating a candidate reference signal sequence based on a difference between a loss calculated based on training data received using a neural network parameter set before the training reference signal sequence is updated and a loss calculated based on training data received using a neural network parameter set after the training reference signal sequence is updated, and further obtaining a learning reference signal sequence, it is possible to maximally improve the accuracy of data detection after the receiving device updates the network parameters online.

In embodiments according to the present disclosure, the transmitting device may be a base station and the receiving device may be a user equipment, or vice versa. For example, in the case of downlink transmission, the transmitting device is a base station and the receiving device is a User Equipment (UE). A reference signal sequence generating network may be deployed at a base station. In particular, the base station may deploy a variety of networks with different structures, parameters, and coefficients to cope with different scenarios and channel environments. The base station may decide which network to use and determine to learn the reference signal sequence according to the frequency band, bandwidth, etc. of the system configuration and/or the UE receiving signals (e.g., RACH signals). The base station can configure the learning reference signal sequence used by the UE through signaling so that the UE can perform online update of neural network parameters. Accordingly, the UE inputs the received learned reference signal sequence into the meta-training network and updates the DNN receiver parameters. And then the UE inputs the received downlink data into the updated DNN receiver for data detection.

For another example, in the case of uplink transmission, the transmitting device is a User Equipment (UE) and the receiving device is a base station. A generated network of LRSs may be deployed at a UE. Specifically, the UE may deploy a variety of LRS generating networks with different structures, parameters, and coefficients, some or all of which may be configured by the base station through signaling, to cope with different scenarios and channel environments. Accordingly, a DNN receiver may be deployed at the base station device, and the DNN receiver parameters may be updated according to the received learned reference signal sequence for data detection.

Further, in an embodiment according to the present disclosure, the learned reference signal sequence may be for a wireless transmission environment of a single user equipment, and may also be for a wireless transmission environment of a plurality of users.

Furthermore, in embodiments according to the present disclosure, the learned reference signal sequence may be cell-specific, i.e. user equipments within the same cell use the same learned reference signal sequence. Alternatively, the learning reference signal sequence may also be specific to a particular user equipment. That is, different users use different learning reference signal sequences according to their own environmental changes.

Furthermore, the receiver DNN according to the embodiments of the present disclosure described above may be deployed in a CU or a DU under a CU (Centralized Unit) -DU (distributed Unit) network architecture. When online adjustment of DNN network parameters is performed based on the learned reference sequence, different scenarios may be considered.

In particular, when the receiver DNN is deployed in a CU, the DU may transmit the received information about the learned reference sequence signal to the CU, which completes the online update of the DNN parameters. On the other hand, when the receiver DNN is deployed in a DU, the DU may adjust DNN network parameters according to the received information about the learned reference signals, and the DU may transmit the adjusted DNN parameters to the CU for distribution by the CU to other DUs that need parameter updates.

Next, a transmission method according to an embodiment of the present disclosure is described with reference to fig. 5. Fig. 5 is a flow diagram of a transmission method 500 according to one embodiment of the present disclosure. Since the steps of the transmission method 500 correspond to the operation of the transmission apparatus 100 described above with reference to the drawings, a detailed description of the same is omitted here for the sake of simplicity.

As shown in fig. 5, in step S501, a learning reference signal sequence is determined, wherein the learning reference signal sequence includes signals about constellation points and non-constellation points in a constellation diagram. Different symbols have different importance when training a Deep Neural Network (DNN). For example, the symbols that are constellation points are simple samples (i.e., sharp samples) that are more important early in the network training, while the symbols that are non-constellation points are difficult samples (i.e., fuzzy samples) that are more important late in the network training.

According to another example of the present disclosure, one or more candidate reference signal sequences may be generated according to a constellation diagram and a maximization loss function, where the loss function is a difference between a loss calculated based on training data received using a neural network parameter set before an update of the training reference signal sequence (hereinafter referred to as "loss before update") and a loss calculated based on training data received using a neural network parameter set after an update of the training reference signal sequence (hereinafter referred to as "loss after update"). And the learning reference signal sequence may be determined from the generated one or more candidate reference signal sequences in step S501.

Specifically, the method shown in fig. 5 may further include generating one or more candidate reference signal sequences by the first training device and the second training device in advance. During off-line training, the first training device may generate a training reference signal sequence according to a constellation diagram and transmit the training reference signal sequence to the second training device, and the second training device may receive the training reference signal sequence and training data from the first training device. The second training device may detect training data using the initial set of neural network parameters and calculate a loss from the detected training data to obtain a pre-update loss. In addition, the second training device may further update the neural network parameter set according to the received training reference signal sequence, detect training data using the updated neural network parameter set, and calculate a loss according to the detected training data to obtain an updated loss. The second training device may then calculate a loss function, i.e. the difference between the loss before the update and the loss after the update and return to the first training device. The first training device may update the training reference signal sequence according to a loss function. And repeating the updating process of the training reference signal sequence until a preset condition is reached, and taking the obtained sequence as a candidate reference signal sequence.

According to another example of the present disclosure, the one or more candidate reference signal sequences may be generated using a reference signal sequence generation network. For example, the reference signal sequence generating network may be a single hidden layer, linear activation function. The network random initialization may be performed by a reference signal sequence generation network to generate multiple candidate reference signal sequences with close performance for the same constellation or codebook. For example, multiple candidate reference signal sequences with close performance for the same constellation or codebook generation may be generated by the reference signal sequence generation network using different initialization parameters. In addition, a plurality of candidate reference signal sequences respectively corresponding to a plurality of constellations or codebooks may also be generated by the reference signal sequence generation network. For example, when performing offline training using the first training apparatus and the second training apparatus described above, the reference signal sequence generation network may be set in the first apparatus. The first training device may generate a candidate reference signal sequence using a reference signal sequence generation network. The second training device may calculate the loss function based on candidate reference signal sequences generated by the first training device using the reference signal sequence generation network. In particular, the second training device may calculate the loss function based on the original candidate reference signal sequences generated by the first training device (i.e., candidate reference signal sequences that have not been transmitted over the channel) and the candidate reference signal sequences that have been transmitted over the channel. For another example, when online deployment is performed, a reference signal sequence generation network may be provided in the transmitting device.

According to another example of the present disclosure, the method illustrated in fig. 5 may further include storing the generated one or more candidate reference signal sequences. For example, the generated one or more candidate reference signal sequences may be stored in the form of a candidate reference signal sequence list, along with sequence indices corresponding to the respective sequences.

In step S502, information on the learning reference signal sequence is transmitted. According to an example of the present disclosure, the information on the learning reference signal sequence includes a sequence index of the determined learning reference signal sequence and at least one of the determined learning reference signal sequences. For example, as described above, there may be multiple candidate reference signal sequences. In this case, a sequence index may be transmitted in step S502 to indicate the learning reference signal sequence determined in step S501. And a learning reference signal sequence corresponding to the sequence index may also be transmitted in step S502.

According to another example of the present disclosure, the learning reference signal sequence may also be pre-configured. In this case, it is not necessary to send a sequence index, and the information of the learned reference signal sequence may include the determined learned reference signal sequence. For example, in case of uplink transmission, that is, the transmitting apparatus is a user equipment and the receiving apparatus is a base station, the base station may configure a learning reference signal sequence which it should use for the user equipment in advance, so that the user equipment may transmit the learning reference signal sequence in advance according to the configuration.

According to another example of the present disclosure, a learning reference signal sequence may be transmitted with data in step S502. For example, when data is to be transmitted in step S502, the learning reference signal sequence may be transmitted first, and then the data may be transmitted next. Alternatively, information on the learning reference signal sequence may be transmitted according to a trigger signal in step S502. For example, when a receiving device or a transmitting device detects that a channel environment changes or is not as expected, transmission of a learning reference signal sequence may be triggered. For example, the receiving device may calculate a loss from the received data and trigger the transmitting device to transmit the sequence of learning reference signals when the loss satisfies a predetermined condition.

Further, in an example according to the invention, the learned reference signal sequence transmitted by the transmitting unit may be used to replace a reference signal in an existing system, and the receiving apparatus may perform channel estimation based on the received learned reference signal sequence. For example, the learned reference signal sequence replaces a Cell Reference Signal (CRS), a demodulation reference signal (DMRS), a channel state reference signal (CSI-RS), etc. in the existing system. For another example, the learned reference signal sequence may also replace the reference, broadcast channel, etc. of the SSB, SRS, etc. in the existing system.

Next, a reception method 600 according to an embodiment of the present disclosure will be described with reference to fig. 6. Fig. 6 is a flowchart of a receiving method 600 performed by a receiving device, wherein the receiving device may be a deep neural network-based receiver (DNN-receiver), according to an embodiment of the present disclosure. Since the steps of the receiving method 600 correspond to the operation of the receiving apparatus 400 described above with reference to fig. 4, a detailed description of the same is omitted here for the sake of simplicity.

As shown in fig. 6, in step S601, information on a learning reference signal sequence including signals on constellation points and non-constellation points in a constellation diagram is received. Then, in step S602, the neural network parameter set of the receiving device is updated according to the received information on the learning reference signal sequence to perform data detection. According to an example of the present disclosure, in step S602, a learning reference signal sequence that is not transmitted through a channel may be obtained according to information about the learning reference signal sequence, and a neural network parameter set of the receiving device may be updated according to the learning reference signal sequence that is transmitted through the channel and the learning reference signal sequence that is not transmitted through the channel.

Specifically, as described above, the information of the learning reference signal sequence may include at least one of a sequence index of the learning reference signal sequence and the learning reference signal sequence. In step S602, a learned reference signal sequence that is not transmitted through a channel may be obtained from one or more candidate reference signal sequences stored in advance according to a sequence index. Alternatively, in the case where the learning reference signal sequence has been previously configured, the previously configured learning reference signal sequence may be obtained as a learning reference signal sequence that is not transmitted through the channel. Further, the learning reference signal sequence transmitted from the transmitting apparatus received at step S601 may be taken as the learning reference signal sequence after being transmitted through the channel in step S602. In step S602, the neural network parameter set of the receiving device may be updated according to the learning reference signal sequence transmitted through the channel and the learning reference signal sequence not transmitted through the channel.

Further, according to an example of the present disclosure, the receiving method 600 may further include determining whether a loss calculated from the detected data satisfies a predetermined condition after data detection using the updated neural network parameter set. And when the loss is determined to meet the predetermined condition, sending a learning reference signal sequence trigger signal to the sending equipment. For example, when the loss is a Bit Error Rate (BER), if the loss is greater than or equal to a predetermined threshold, the processing unit 420 may determine that the loss satisfies a predetermined condition, and the transmitting unit may transmit the learning reference signal sequence trigger signal to the transmitting apparatus. For another example, when the loss is cross entropy, if the loss is less than or equal to a predetermined threshold, the processing unit 420 may determine that the loss satisfies a predetermined condition, and the transmission unit may transmit the learning reference signal sequence trigger signal to the transmission apparatus.

In the transmitting method and the receiving method according to the embodiments of the present disclosure, training of neural network parameters for different scenes and channel environments may be better adapted by generating and transmitting a learning reference signal sequence including signals about constellation points and non-constellation points in a constellation diagram to a receiver. In addition, by generating a candidate reference signal sequence based on a difference between a loss calculated based on training data received using a neural network parameter set before the training reference signal sequence is updated and a loss calculated based on training data received using a neural network parameter set after the training reference signal sequence is updated, and further obtaining a learning reference signal sequence, it is possible to maximally improve the accuracy of data detection after the receiving device updates the network parameters online.

< hardware Structure >

The block diagrams used in the description of the above embodiments show blocks in units of functions. These functional blocks (structural units) are implemented by any combination of hardware and/or software. Note that the means for implementing each functional block is not particularly limited. That is, each functional block may be implemented by one apparatus which is physically and/or logically combined, or may be implemented by a plurality of apparatuses which are directly and/or indirectly (for example, by wire and/or wirelessly) connected by two or more apparatuses which are physically and/or logically separated.

For example, a device (such as a transmitting device, a receiving device, or the like) of one embodiment of the present disclosure may function as a computer that performs processing of the wireless communication method of the present disclosure. Fig. 7 is a schematic diagram of a hardware structure of an apparatus 700 (base station or terminal) according to an embodiment of the present disclosure. The apparatus 700 (base station or terminal) may be physically configured as a computer device including a processor 710, a memory 720, a storage 730, a communication device 740, an input device 750, an output device 760, a bus 770, and the like.

In the following description, the words "device" or the like may be replaced with circuits, devices, units, or the like. The hardware configuration of the user terminal and the base station may include one or more of the devices shown in the figures, or may not include some of the devices.

For example, processor 710 is shown as only one, but may be multiple processors. The processing may be executed by one processor, or may be executed by one or more processors at the same time, sequentially, or by other methods. In addition, the processor 710 may be mounted by more than one chip.

The functions of the device 700 are implemented, for example, as follows: by reading predetermined software (program) into hardware such as the processor 710 and the memory 720, the processor 710 performs an operation to control communication performed by the communication device 740 and to control reading and/or writing of data in the memory 720 and the storage 730.

The processor 710 causes, for example, an operating system to operate to control the entire computer. The processor 710 may be configured by a Central Processing Unit (CPU) including an interface with a peripheral device, a control device, an arithmetic device, a register, and the like. For example, the above-described determining unit, adjusting unit, and the like may be implemented by the processor 710.

Further, the processor 710 reads out the program (program code), the software module, the data, and the like from the storage 730 and/or the communication device 740 to the memory 720, and executes various processes according to them. As the program, a program that causes a computer to execute at least a part of the operations described in the above embodiments may be used. For example, the processing units of the receiving device and the transmitting device may be implemented by a control program stored in the memory 720 and operated by the processor 710, and may be implemented similarly for other functional blocks.

The Memory 720 is a computer-readable recording medium, and may be configured by at least one of a Read Only Memory (ROM), a Programmable Read Only Memory (EPROM), an Electrically Programmable Read Only Memory (EEPROM), a Random Access Memory (RAM), and other suitable storage media. Memory 720 may also be referred to as registers, cache, main storage (primary storage), etc. The memory 720 may store executable programs (program codes), software modules, and the like for implementing the methods according to an embodiment of the present disclosure.

The memory 730 is a computer-readable recording medium, and may be configured by at least one of a flexible disk (floppy disk), a floppy (registered trademark) disk (floppy disk), a magneto-optical disk (for example, a compact Disc read only memory (CD-rom), etc.), a digital versatile Disc, a Blu-ray (registered trademark) optical disk), a removable disk, a hard disk drive, a smart card, a flash memory device (for example, a card, a stick, a key driver), a magnetic stripe, a database, a server, and other suitable storage media. Memory 730 may also be referred to as a secondary storage device.

The communication device 740 is hardware (transmission/reception device) for performing communication between computers via a wired and/or wireless network, and is also referred to as a network device, a network controller, a network card, a communication module, or the like. Communication device 740 may include high Frequency switches, duplexers, filters, Frequency synthesizers, and the like, for example, to implement Frequency Division Duplexing (FDD) and/or Time Division Duplexing (TDD). For example, the transmitting unit, the receiving unit, and the like described above can be realized by the communication device 740.

The input device 750 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that accepts input from the outside. The output device 760 is an output device (for example, a display, a speaker, a Light Emitting Diode (LED) lamp, or the like) that outputs to the outside. The input device 750 and the output device 760 may be integrated (e.g., a touch panel).

The processor 710, the memory 720, and other devices are connected by a bus 770 for communicating information. The bus 770 may be constituted by a single bus or different buses between devices.

Further, the transmitting apparatus and the receiving apparatus may include hardware such as a microprocessor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Programmable Logic Device (PLD), a Field Programmable Gate Array (FPGA), and part or all of the functional blocks may be implemented by the hardware. For example, the processor 710 may be installed through at least one of these hardware.

(modification example)

In addition, terms described in the present specification and/or terms necessary for understanding the present specification may be interchanged with terms having the same or similar meanings. For example, the channels and/or symbols may also be signals (signaling). Furthermore, the signal may also be a message. The reference signal may be simply referred to as rs (reference signal), and may be referred to as Pilot (Pilot), Pilot signal, or the like according to the applicable standard. Further, a Component Carrier (CC) may also be referred to as a cell, a frequency Carrier, a Carrier frequency, and the like.

Note that information, parameters, and the like described in this specification may be expressed as absolute values, relative values to predetermined values, or other corresponding information. For example, the radio resource may be indicated by a prescribed index. Further, the formulas and the like using these parameters may also be different from those explicitly disclosed in the present specification.

The names used for parameters and the like in the present specification are not limitative in any way. For example, the various channels (Physical uplink control Channel (PUCCH), Physical Downlink Control Channel (PDCCH), etc.) and information elements may be identified by any suitable names, and thus the various names assigned to the various channels and information elements are not limited in any way.

Information, signals, and the like described in this specification can be represented using any of a variety of different technologies. For example, data, commands, instructions, information, signals, bits, symbols, chips, and the like that may be referenced throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or photons, or any combination thereof.

Further, information, signals, and the like may be output from an upper layer to a lower layer, and/or from a lower layer to an upper layer. Information, signals, etc. may be input or output via a plurality of network nodes.

The input or output information, signals, and the like may be stored in a specific place (for example, a memory) or may be managed by a management table. The information, signals, etc. that are input or output may be overwritten, updated or supplemented. The output information, signals, etc. may be deleted. The input information, signals, etc. may be sent to other devices.

The information notification is not limited to the embodiments and modes described in the present specification, and may be performed by other methods. For example, the notification of the Information may be implemented by physical layer signaling (e.g., Downlink Control Information (DCI), Uplink Control Information (UCI)), upper layer signaling (e.g., Radio Resource Control (RRC) signaling, broadcast Information (Master Information Block, System Information Block (SIB), etc.), Medium Access Control (MAC) signaling), other signals, or a combination thereof.

In addition, physical layer signaling may also be referred to as L1/L2 (layer 1/layer 2) control information (L1/L2 control signals), L1 control information (L1 control signals), and the like. The RRC signaling may also be referred to as an RRC message, and may be, for example, an RRC Connection Setup (RRC Connection Setup) message, an RRC Connection Reconfiguration (RRC Connection Reconfiguration) message, or the like. The MAC signaling may be notified by a MAC Control Element (MAC CE (Control Element)), for example.

Note that the notification of the predetermined information (for example, the notification of "X") is not limited to be explicitly performed, and may be implicitly performed (for example, by not performing the notification of the predetermined information or by performing the notification of other information).

The determination may be performed by a value (0 or 1) represented by 1 bit, may be performed by a true-false value (boolean value) represented by true (true) or false (false), or may be performed by comparison of numerical values (for example, comparison with a predetermined value).

Software, whether referred to as software, firmware, middleware, microcode, hardware description language, or by other names, is to be broadly construed to refer to commands, command sets, code segments, program code, programs, subroutines, software modules, applications, software packages, routines, subroutines, objects, executables, threads of execution, steps, functions, and the like.

Further, software, commands, information, and the like may be transmitted or received via a transmission medium. For example, when the software is transmitted from a website, server, or other remote source using a wired technology (e.g., coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL, microwave, etc.) and/or a wireless technology (e.g., infrared, microwave, etc.), the wired technology and/or wireless technology are included in the definition of transmission medium.

The terms "system" and "network" as used in this specification may be used interchangeably.

In the present specification, terms such as "Base Station (BS)", "radio Base Station", "eNB", "gNB", "cell", "sector", "cell group", "carrier", and "component carrier" are used interchangeably. A base station may also be referred to by terms such as a fixed station (fixed station), NodeB, eNodeB (eNB), access point (access point), transmission point, reception point, femto cell, and small cell.

A base station may accommodate one or more (e.g., three) cells (also referred to as sectors). When a base station accommodates multiple cells, the entire coverage area of the base station may be divided into multiple smaller areas, and each smaller area may also provide communication services through a base station subsystem (e.g., an indoor small Radio Head (RRH)). The term "cell" or "sector" refers to a portion or the entirety of the coverage area of a base station and/or base station subsystem that is in communication service within the coverage area.

In this specification, terms such as "Mobile Station (MS)", "User terminal (User terminal)", "User Equipment (UE)", and "terminal" may be used interchangeably. A mobile station is also sometimes referred to by those skilled in the art as a subscriber station, mobile unit, subscriber unit, wireless unit, remote unit, mobile device, wireless communications device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless terminal, remote terminal, handset, user agent, mobile client, or by some other appropriate terminology.

In addition, the radio base station in this specification may be replaced with a user terminal. For example, the aspects/embodiments of the present disclosure may be applied to a configuration in which communication between a wireless base station and a user terminal is replaced with communication between a plurality of user terminals (D2D, Device-to-Device). In this case, the functions of the first communication device or the second communication device in the above-described device 800 may be regarded as the functions of the user terminal. Also, words such as "upstream" and "downstream" may be replaced with "side". For example, the uplink channel may be replaced with a side channel.

Also, the user terminal in this specification may be replaced with a radio base station. In this case, the functions of the user terminal may be regarded as functions of the first communication device or the second communication device.

In this specification, it is assumed that a specific operation performed by a base station is sometimes performed by its upper node (upper node) in some cases. It is obvious that in a network including one or more network nodes (network nodes) having a base station, various operations performed for communication with a terminal may be performed by the base station, one or more network nodes other than the base station (for example, a Mobility Management Entity (MME), a Serving-Gateway (S-GW), or the like may be considered, but not limited thereto), or a combination thereof.

The embodiments and modes described in this specification may be used alone or in combination, or may be switched during execution. Note that, as long as there is no contradiction between the processing steps, sequences, flowcharts, and the like of the embodiments and the embodiments described in the present specification, the order may be changed. For example, with respect to the methods described in this specification, various elements of steps are presented in an exemplary order and are not limited to the particular order presented.

The aspects/embodiments described in this specification can be applied to a mobile communication system using Long Term Evolution (LTE), Long Term Evolution Advanced (LTE-a), Long Term Evolution-Beyond (LTE-B), LTE-Beyond (SUPER 3G), international mobile telecommunications Advanced (IMT-Advanced), 4th generation mobile telecommunications system (4G, 4th generation mobile telecommunications system), 5th generation mobile telecommunications system (5G, 5th generation mobile telecommunications system), Future Radio Access (FRA, Future Radio Access), New Radio Access Technology (New-RAT, Radio Access Technology), New Radio (NR, New Radio), New Radio Access (NX, New Access), New generation Radio Access (FX, function, global Radio registration system (GSM), global System for Mobile communications), code division multiple access 3000(CDMA3000), Ultra Mobile Broadband (UMB), IEEE 920.11(Wi-Fi (registered trademark)), IEEE 920.16(WiMAX (registered trademark)), IEEE 920.20, Ultra WideBand (UWB, Ultra-WideBand), Bluetooth (registered trademark)), other appropriate wireless communication methods, and/or a next generation System extended based thereon.

The term "according to" used in the present specification does not mean "according only" unless explicitly stated in other paragraphs. In other words, the statement "according to" means both "according to only" and "according to at least".

Any reference to elements using the designations "first", "second", etc. used in this specification is not intended to be a comprehensive limitation on the number or order of such elements. These names may be used in this specification as a convenient way to distinguish between two or more elements. Thus, references to a first unit and a second unit do not imply that only two units may be employed or that the first unit must precede the second unit in several ways.

The term "determining" used in the present specification may include various operations. For example, regarding "determination (determination)", calculation (computing), estimation (computing), processing (processing), derivation (deriving), investigation (analyzing), search (looking up) (for example, a search in a table, a database, or another data structure), confirmation (ascertaining), and the like may be regarded as "determination (determination)". In addition, regarding "determination (determination)", reception (e.g., reception information), transmission (e.g., transmission information), input (input), output (output), access (access) (e.g., access to data in a memory), and the like may be regarded as "determination (determination)". Further, regarding "judgment (determination)", it is also possible to regard solution (solving), selection (selecting), selection (breathing), establishment (evaluating), comparison (comparing), and the like as performing "judgment (determination)". That is, with respect to "determining (confirming)", several actions may be considered as performing "determining (confirming)".

The terms "connected", "coupled" or any variation thereof as used in this specification refer to any connection or coupling, either direct or indirect, between two or more elements, and may include the following: between two units "connected" or "coupled" to each other, there are one or more intermediate units. The combination or connection between the elements may be physical, logical, or a combination of both. For example, "connected" may also be replaced with "accessed". As used in this specification, two units may be considered to be "connected" or "joined" to each other by the use of one or more wires, cables, and/or printed electrical connections, and by the use of electromagnetic energy or the like having wavelengths in the radio frequency region, the microwave region, and/or the optical (both visible and invisible) region, as a few non-limiting and non-exhaustive examples.

When the terms "including", "including" and "comprising" and variations thereof are used in the present specification or claims, these terms are open-ended as in the term "including". Further, the term "or" as used in the specification or claims is not exclusive or.

While the present disclosure has been described in detail above, it will be apparent to those skilled in the art that the present disclosure is not limited to the embodiments described in the present specification. The present disclosure can be implemented as modifications and variations without departing from the spirit and scope of the present disclosure defined by the claims. Accordingly, the description of the present specification is for the purpose of illustration and is not intended to be in any way limiting of the present disclosure.

Claims (10)

1. A transmitting device, comprising:

a processing unit for determining a learning reference signal sequence, wherein the learning reference signal sequence comprises signals relating to constellation points and non-constellation points in a constellation diagram; and

a transmitting unit for transmitting information on the learning reference signal sequence.

2. The transmission apparatus of claim 1, wherein

Generating one or more candidate reference signal sequences based on the constellation map and a maximized loss function, wherein the loss function is a difference between a loss calculated based on training data received using the set of neural network parameters before the training reference signal sequence update and a loss calculated based on training data received using the set of neural network parameters after the training reference signal sequence update,

the processing unit determines the learning reference signal sequence from the one or more candidate reference signal sequences.

3. The transmission apparatus according to claim 1 or 2, wherein

The information on the learning reference signal sequence includes a sequence index of the determined learning reference signal sequence and at least one of the determined learning reference signal sequences.

4. The transmission apparatus of claim 2, wherein

Generating the one or more candidate reference signal sequences using a reference signal sequence generation network.

5. The transmission apparatus according to claim 1 or 2, wherein

The transmitting unit transmits the learning reference signal sequence with data; or

The transmission unit transmits information on the learning reference signal sequence according to a trigger signal.

6. A receiving device, comprising:

a receiving unit configured to receive information on a learning reference signal sequence, wherein the learning reference signal sequence includes signals on constellation points and non-constellation points in a constellation diagram;

and the processing unit is used for updating the neural network parameter set of the receiving equipment according to the received information about the learning reference signal sequence so as to detect data.

7. The receiving device of claim 6, wherein

The processing unit obtains a learning reference signal sequence transmitted through a channel and a learning reference signal sequence not transmitted through the channel according to information on the learning reference signal sequence, and updates the neural network parameter set of the receiving device according to the learning reference signal sequence transmitted through the channel and the learning reference signal sequence not transmitted through the channel.

8. The receiving device of claim 6 or 7, wherein

The processing unit is further configured to determine whether a loss calculated from the detected data satisfies a predetermined condition; and

the receiving apparatus further includes:

a transmitting unit for transmitting a learning reference signal sequence trigger signal to a transmitting device when the loss satisfies a predetermined condition.

9. A transmission method, comprising:

determining a learning reference signal sequence, wherein the learning reference signal sequence comprises signals for constellation points and non-constellation points in a constellation diagram; and

transmitting information on the learning reference signal sequence.

10. A receiving method, comprising:

receiving information on a learning reference signal sequence, wherein the learning reference signal sequence comprises signals on constellation points and non-constellation points in a constellation diagram; and

and updating the neural network parameter set of the receiving device according to the received information about the learning reference signal sequence so as to detect the data.

Images