Transmitting equipment, receiving equipment, transmitting method and receiving method

Abstract

To provide a transmission method and a reception method of a learning reference signal sequence, and a corresponding transmission device and a reception device.SOLUTION: A transmission device includes a processing unit for determining a learning reference signal sequence including a signal related to constellation points and non-construction points in a constellation, and a transmission unit for transmitting information about the learning reference signal sequence.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to the field of wireless communication, and more particularly to a method for transmitting and receiving a training reference signal sequence and corresponding transmitting and receiving devices. [Background technology]

[0002] Deep neural network receivers (DNN-receivers) are an important research direction for B5G / 6G. Traditional deep neural network receivers or receiving devices use offline training, so when the transmission environment between offline training and online deployment is the same, the traditional deep neural network receiver can achieve excellent performance. However, when the transmission environment of online deployment changes, the performance of the traditional deep neural network receiver will be impaired. Summary of the Invention [Problem to be solved by the invention]

[0003] To improve the adaptability of DNNs to various environments, different training environments are introduced during offline training, thereby improving their adaptability to the environment during online deployment. Specifically, more datasets can be introduced during offline training, but this requires generating a large amount of data during training, which results in increased training complexity. Furthermore, multi-task and transfer learning can be introduced for different training environments, but when the network is transferred to multiple environments, its performance in a single environment is sacrificed.

[0004] Meanwhile, it has been proposed to perform online training on a model using samples acquired online, thereby improving its adaptability to various environments. Online training on a model using samples acquired online may include incremental learning-based online training and meta-learning-based online training. Specifically, incremental learning-based online training refers to updating network weights by adding samples acquired online as incremental samples to a training set. However, because the online samples are incremental, a large number of online samples are required to adapt to new environments. Meta-learning-based online training refers to designing a network model or an optimization algorithm to enable the network model to be more adaptable to new tasks, and updating the network using online samples during online deployment. Compared to other training methods, meta-learning-based online training takes into account the efficiency of online training at the initial stage of model design or optimization algorithm design, resulting in a higher adaptability to new environments, a smaller number of online samples required, and better performance after online learning.

[0005] In meta-learning-based online training, the receiver learns the dynamic environment using online training samples sent by the sender, where the online training samples are also called learning reference signal (LRS) sequences. Currently, there are no LRS sequence design guidelines or LRS sequence design methods that follow these guidelines. Therefore, current approaches employ heuristic design for LRS sequences. In other words, LRS sequences are constructed by randomly selecting constellation points by the sender. LRS sequences constructed by randomly selecting constellation points cannot optimize performance after online updates. [Means for solving the problem]

[0006] According to one aspect of the present disclosure, there is provided a transmitting device, the transmitting device including: a processing unit for determining a training reference signal sequence including signals related to constellation points and non-constellation points in a constellation; and a transmitting unit for transmitting information related to the training reference signal sequence.

[0007] According to another aspect of the present disclosure, there is provided a receiving device, the receiving device including: a receiving unit for receiving information related to a training reference signal sequence including signals related to constellation points and non-constellation points in a constellation; and a processing unit for updating network parameters of the receiving device based on the received information related to the training reference signal sequence to perform data detection.

[0008] According to another aspect of the present disclosure, there is provided a transmission method, the transmission method comprising: determining a training reference signal sequence including signals related to constellation points and non-constellation points in a constellation; and transmitting information related to the training reference signal sequence.

[0009] According to another aspect of the present disclosure, there is provided a receiving method, the receiving method comprising the steps of receiving information on a training reference signal sequence including signals related to constellation points and non-constellation points in a constellation, and updating network parameters of the receiving device based on the received information on the training reference signal sequence to perform data detection. [Brief explanation of the drawings]

[0010] The above and other objects, features, and advantages of the present disclosure will become more apparent from the detailed description of the embodiments of the present disclosure with reference to the accompanying drawings. The accompanying drawings are used to provide a further understanding of the embodiments of the present disclosure, constitute a part of the specification, and are used to explain the present disclosure together with the embodiments of the present disclosure, but are not intended to limit the present disclosure. In the drawings, the same reference numerals generally represent the same elements or steps. [Figure 1] FIG. 2 is an exemplary block diagram of a transmitting device according to one embodiment of the present disclosure. [Figure 2] FIG. 2 is a schematic diagram of an LRS sequence according to one example of the present disclosure. [Figure 3A] FIG. 10 is a schematic diagram of one candidate reference signal sequence table. [Figure 3B] FIG. 10 is a schematic diagram of another candidate reference signal sequence table. [Figure 4] FIG. 2 is an exemplary block diagram of a receiving device according to one embodiment of the present disclosure. [Figure 5] 1 is a flowchart of a transmission method according to one embodiment of the present disclosure. [Figure 6] 4 is a flowchart of a receiving method performed by a receiving device according to one embodiment of the present disclosure. [Figure 7] FIG. 1 is a schematic diagram of a hardware structure of a device according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0011] To further clarify the objectives, technical solutions, and advantages of the present disclosure, exemplary embodiments according to the present disclosure will be described in detail below with reference to the accompanying drawings. In the drawings, the same reference numerals always refer to the same elements. It should be understood that the embodiments described herein are for illustrative purposes only and should not be construed as limiting the scope of the present disclosure. It should be noted that the terminal described herein may include various types of terminals, such as a user equipment (UE), a mobile terminal (also referred to as a mobile station), or a fixed terminal. However, for convenience, the terms terminal and UE may be used interchangeably hereinafter. It should be noted that the terms receiver and receiving equipment may be used interchangeably hereinafter.

[0012] The meta-learning-based online training can include metric-based meta-learning, model-based meta-learning, and optimization-based meta-learning. In the embodiments of the present disclosure, optimization-based meta-learning can be used. Optimization-based meta-learning improves the neural network steepest descent algorithm, allowing the online steepest descent algorithm to converge quickly in the case of small samples. The receiving device in the embodiments of the present application is a deep neural network-based receiving device (DNN-receiver). More specifically, the receiving device in the embodiments of the present application may be an optimization-based meta-learning receiving device.

[0013] A conventional optimization-based meta-learning method is "Model-Agnostic Meta-Learning (MAML)." The objective of offline training in MAML is to find a neural network parameter set that minimizes the distance from the neural network parameter set to the optimal parameter manifold for tasks 1 and 2. For example, task 1 may be communication in an indoor environment, and task 2 may be communication in an outdoor environment. The objective of offline training of a deep neural network-based receiving device according to an embodiment of the present application is to find a neural network parameter set that minimizes the distance from the neural network parameter set to the optimal parameter manifold for communication in an indoor environment and communication in an outdoor environment. This allows the neural network parameter set to be quickly updated to a parameter set suitable for a specific task during online deployment.

[0014] Hereinafter, a transmitting device 100 according to an embodiment of the present disclosure will be described with reference to Fig. 1. As shown in Fig. 1, the transmitting device 100 according to an embodiment of the present disclosure may include a processing unit 110 and a transmitting unit 120. The transmitting device 100 may further include other components in addition to the processing unit and the transmitting unit, but these components are not relevant to the content of the embodiment of the present disclosure, and therefore, illustrations and descriptions thereof will be omitted here.

[0015] As shown in FIG. 1, the processing unit 110 may determine a training reference signal (LRS) sequence including signals related to constellation points and non-constellation points in a constellation. 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. In the example shown in FIG. 2, the length of the LRS sequence is 2. As shown in FIG. 2, the first LRS symbol 210 in the LRS sequence is a constellation point, while the 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, symbols designated as constellation points are easy samples (i.e., sharp samples) and are important for early network training, while symbols designated as non-constellation points are hard samples (i.e., fuzzy samples) and are important for later network training.

[0016] According to another example of the present disclosure, one or more candidate reference signal sequences may be generated based on the constellation and a maximized loss function, where the loss function is the difference between a loss calculated based on training data received using a parameter set of the neural network before the training reference signal sequence is updated (hereinafter referred to as the "loss before update") and a loss calculated based on training data received using a parameter set of the neural network after the training reference signal sequence is updated (hereinafter referred to as the "loss after update"). Processing unit 110 may determine the learning reference signal sequence from the generated one or more candidate reference signal sequences.

[0017] Specifically, one or more candidate reference signal sequences may be generated in advance by a first training device and a second training device. For example, the first training device and the second training device may be specific modules included in a base station. Alternatively, the first training device and the second training device may be dedicated devices located at the base station. To acquire a real channel between the first training device and the second training device, the first training device may be located at the base station, and the second device may be located at a location where a user equipment can appear in a cell to which the base station belongs.

[0018] During offline training, the first training device may generate a training reference signal sequence based on the constellation and transmit it 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 obtain a loss before updating by detecting training data using an initial neural network parameter set and calculating a loss based on the detected training data. The second training device may further obtain an updated loss by updating the neural network parameter set based on the received training reference signal sequence, detecting training data using the updated neural network parameter set, and calculating a loss based on the detected training data. Next, the second training device may calculate a loss function, i.e., calculate the difference between the loss before updating and the loss after updating, and return the result to the first training device. The first training device may update the training reference signal sequence based on the constellation and the loss function. The training reference signal sequence update process is repeated until a predetermined condition is satisfied, and the obtained sequence is designated as a candidate reference signal sequence.

[0019] In the embodiments of the present disclosure, the loss function may be any function. For example, the loss may be the cross entropy of data, a bit error rate (BER), etc. However, in the embodiments of the present disclosure, the loss function is not limited to these and may be adjusted based on actual needs or performance indicators.

[0020] 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 generation network may have a single hidden layer and a linear activation function. The processing unit 110 may generate multiple candidate reference signal sequences with similar performance for the same constellation or codebook by randomly initializing the reference signal sequence generation network. For example, the reference signal sequence generation network may generate multiple candidate reference signal sequences with similar performance for the same constellation or codebook using different initialization parameters. Note that the processing unit 110 may further generate multiple candidate reference signal sequences corresponding to multiple constellations or codebooks, respectively, using the reference signal sequence generation network. For example, when offline training is performed using the first training device and the second training device described above, the reference signal sequence generation network may be configured in the first training device. The first training device may generate candidate reference signal sequences using the reference signal sequence generation network. The second training device may calculate a loss function based on the candidate reference signal sequences generated by the first training device using the reference signal sequence generation network. Specifically, the second training device may calculate the loss function based on the original candidate reference signal sequences (i.e., the candidate reference signal sequences not transmitted through the channel) generated by the first training device and the candidate reference signal sequences transmitted through the channel. For example, in the case of online deployment, the reference signal sequence generation network may be configured in the transmitting device.

[0021] According to another example of the present disclosure, the transmitting device and the receiving device may include a storage unit for storing one or more generated candidate reference signal sequences. For example, the generated one or more candidate reference signal sequences and their corresponding sequence indexes may be stored in a candidate reference signal sequence table. FIGS. 3A and 3B show schematic diagrams of candidate reference signal sequence tables. In the example shown in FIG. 3A, 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 table includes multiple subtables, such as candidate reference signal sequence subtable 321 and candidate reference signal sequence subtable 322, each of which corresponds to one or more codebooks / constellations and includes one or more candidate reference signal sequences corresponding to the codebook / constellation. The example shown in FIG. 3B shows candidate reference signal sequence subtables for one codebook / constellation structure, which improves instruction flexibility and reduces signaling overhead.

[0022] Returning to Fig. 1, the transmitting unit 120 can transmit information about the training reference signal sequence determined by the processing unit 110. The receiving device can update a parameter set of the device's neural network based on the information about the training reference signal sequence transmitted by the transmitting unit 120 to perform data detection.

[0023] According to one example of the present disclosure, the information about the training reference signal sequence includes a sequence index of the determined training reference signal sequence and at least one of the determined training reference signal sequence. 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 training reference signal sequence determined by the processing unit 110. The transmitting unit 120 may transmit the training reference signal sequence corresponding to the sequence index.

[0024] According to another example of the present disclosure, the training reference signal sequence may be preset. In this case, there is no need to transmit a sequence index, and the information on the training reference signal sequence may include the training reference signal sequence determined by the processing unit 110. For example, in the case of uplink transmission, i.e., when the transmitting device is a user equipment and the receiving device is a base station, the base station may preset a training reference signal sequence for the user equipment to use, so that the user equipment may transmit the training reference signal sequence according to the preset setting. According to another example of the present disclosure, the transmitting unit 120 may transmit the training reference signal sequence together with data. For example, when the transmitting unit 120 intends to transmit data, it may first transmit the training reference signal sequence and then transmit the data. Alternatively, the transmitting unit 120 may transmit information on the training reference signal sequence based on a trigger signal. For example, the receiving device or the transmitting device may trigger transmission of the training reference signal sequence when it detects that the channel environment has changed or is lower than expected. For example, the receiving device calculates a loss based on the received data, and if the loss satisfies a predetermined condition, triggers the transmitting device to transmit a training reference signal sequence. Specifically, if the loss is a bit error rate (BER), the receiving device triggers the transmission of a training reference signal sequence if the loss is equal to or greater than a predetermined threshold. If the loss is a cross entropy, the receiving device triggers the transmission of a training reference signal sequence if the loss is equal to or less than a predetermined threshold.

[0025] In an example according to the invention, the training reference signal sequence transmitted by the transmitting unit may be used to replace a reference signal in a legacy system, and the receiving device may perform channel estimation based on the received training reference signal sequence. For example, the training reference signal sequence may replace a cell reference signal (CRS), a demodulation reference signal (DMRS), a channel state reference signal (CSI-RS), etc. in the legacy system. Also, for example, the training reference signal sequence may replace a reference channel, such as SSB or SRS, or a broadcast channel, etc. in the legacy system.

[0026] Hereinafter, a receiving device 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, the receiving device 400 according to one embodiment of the present disclosure may include a receiving unit 410 and a processing unit 420. The receiving device 400 may further include other components in addition to the processing unit and the receiving unit, but these components are not relevant to the content of the embodiment of the present disclosure, and therefore will not be shown or described here.

[0027] As shown in Fig. 4, the receiving unit 410 may receive information about a training reference signal sequence, which includes signals related to constellation points and non-constellation points in a constellation. The training reference signal sequence has been described in detail above with reference to Fig. 2, and therefore will not be described again here.

[0028] The processing unit 420 updates a parameter set of a neural network of the receiving device 400 based on information about the training reference signal sequence to perform data detection. According to one example of the present disclosure, the processing unit 420 obtains a training reference signal sequence that is not transmitted through a channel based on information about the training reference signal sequence, and updates a parameter set of the neural network of the receiving device based on the training reference signal sequence that is transmitted through the channel and the training reference signal sequence that is not transmitted through the channel.

[0029] Specifically, as described above, the information on the training reference signal sequence may include at least one of a sequence index of the training reference signal sequence and the training reference signal sequence. The processing unit 420 may acquire a training reference signal sequence that is not transmitted through the channel from one or more pre-stored candidate reference signal sequences based on the sequence index. Alternatively, if the training reference signal sequence is preset, the pre-set training reference signal sequence may be acquired and used as the training reference signal sequence that is not transmitted through the channel. The processing unit 420 may use the training reference signal sequence transmitted from the transmitting device and received by the receiving unit 410 as the training reference signal sequence transmitted through the channel. The processing unit 420 may update a parameter set of a neural network of the receiving device based on the training reference signal sequence transmitted through the channel and the training reference signal sequence that is not transmitted through the channel. According to an example of the present disclosure, after performing data detection using the updated parameter set of the neural network, the processing unit 420 may determine whether a loss calculated based on the detected data satisfies a predetermined condition. The receiving device may further include a transmitting unit. If the processing unit 420 determines that the loss satisfies a predetermined condition, the transmitting unit can transmit a trigger signal for a training reference signal sequence to the transmitting device. For example, if the loss is a bit error rate (BER), the processing unit 420 can determine that the loss satisfies the predetermined condition if the loss is equal to or greater than a predetermined threshold, and the transmitting unit can transmit a trigger signal for a training reference signal sequence to the transmitting device. For example, if the loss is a cross entropy, the processing unit 420 can determine that the loss satisfies the predetermined condition if the loss is equal to or less than a predetermined threshold, and the transmitting unit can transmit a trigger signal for a training reference signal sequence to the transmitting device.

[0030] In the transmitting device and the receiving device according to the embodiment of the present disclosure, a training reference signal sequence including signals related to constellation points and non-constellation points in a constellation is generated and transmitted to a receiver, so that the parameters of a neural network can be better trained in different scenes and channel environments. In addition, a candidate reference signal sequence can be generated based on a difference between a loss calculated based on training data received using a parameter set of the neural network before the training reference signal sequence is updated and a loss calculated based on training data received using a parameter set of the neural network after the training reference signal sequence is updated, and the training reference signal sequence can be further obtained to maximize the accuracy of data detection after the receiving device updates the network parameters online.

[0031] In an embodiment according to the present disclosure, the transmitting device may be a base station, and the receiving device may be a user equipment (UE), 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 generation network may be configured in the base station. Specifically, the base station may configure multiple types of networks with different structures, parameters, and coefficients to accommodate different scenes and channel environments. The base station determines which network to use and determines a training reference signal sequence based on system configurations such as bands and bandwidths and / or signals received by the UE (e.g., RACH signals). The base station may configure a training reference signal sequence used by the UE via signaling so that the UE can update the parameters of its neural network online. Correspondingly, the UE inputs the received training reference signal sequence into the original training network to update the parameters of the DNN receiver. Then, the UE inputs the received downlink data into the updated DNN receiver for data detection.

[0032] For example, in the case of uplink transmission, the transmitting device is a user equipment (UE) and the receiving device is a base station. An LRS generating network may be configured in the UE. Specifically, the UE may configure multiple types of LRS generating networks with different structures, parameters, and coefficients to accommodate different scene and channel environments. Some or all of the network architectures, parameters, and network unit coefficients of the LRS generating networks may be configured by the base station via signaling. Correspondingly, a DNN receiver may be configured in the base station device, which updates DNN receiver parameters based on the received training reference signal sequence to perform data detection.

[0033] It should be noted that in the embodiments of the present disclosure, the training reference signal sequence may be intended for a wireless transmission environment of a single user equipment, or may be intended for a wireless transmission environment of multiple users.

[0034] It should be noted that in the embodiments of the present disclosure, the training reference signal sequence may be cell-specific, i.e., user equipments in the same cell use the same training reference signal sequence. Alternatively, the training reference signal sequence may be user equipment-specific, i.e., different users use different training reference signal sequences according to changes in their own environments.

[0035] In addition, in a CU (Centralized Unit)-DU (Distributed Unit) network architecture, the receiver DNN according to the above-described embodiments of the present disclosure may be located in the CU or the DU. When performing online adjustment to DNN network parameters based on a training reference sequence, different situations may be considered.

[0036] Specifically, if the receiver DNN is located in the CU, the DU may transmit information about the received training reference sequence signal to the CU, and the CU may complete online updating of the DNN parameters. On the other hand, if the receiver DNN is located in the DU, the DU may adjust the DNN network parameters based on the information about the received training reference signal, transmit the adjusted DNN parameters to the CU, and the CU may distribute the parameters to other DUs that need to be updated.

[0037] A transmitting method according to an embodiment of the present disclosure will now be described with reference to Fig. 5. Fig. 5 is a flowchart of a transmitting method 500 according to one embodiment of the present disclosure. The steps of the transmitting method 500 correspond to the operations of the transmitting device 100 described above with reference to the figures, and therefore, detailed descriptions of the same will be omitted here for convenience.

[0038] As shown in Figure 5, in step S501, a training reference signal sequence is determined, which includes signals related to constellation points and non-constellation points in the constellation. Different symbols have different importance when training a deep neural network (DNN). For example, symbols designated as constellation points are easy samples (i.e., sharp samples) and are important for the early stage of network training, while symbols designated as non-constellation points are hard samples (i.e., fuzzy samples) and are important for the later stage of network training.

[0039] According to another example of the present disclosure, one or more candidate reference signal sequences can be generated based on a constellation and a maximized loss function, where the loss function is the difference between a loss calculated based on training data received using a parameter set of a neural network before the training reference signal sequence is updated (hereinafter referred to as the "loss before update") and a loss calculated based on training data received using a parameter set of the neural network after the training reference signal sequence is updated (hereinafter referred to as the "loss after update"). In step S501, the learning reference signal sequence can be determined from the generated one or more candidate reference signal sequences.

[0040] Specifically, the method illustrated in FIG. 5 may further include a step of pre-generating one or more candidate reference signal sequences by a first training device and a second training device. During offline training, the first training device may generate a training reference signal sequence based on the constellation and transmit it 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 obtain a loss before updating by detecting training data using an initial neural network parameter set and calculating a loss based on the detected training data. The second training device may further obtain an updated loss by updating the neural network parameter set based on the received training reference signal sequence, detecting training data using the updated neural network parameter set, and calculating a loss based on the detected training data. Next, the second training device may calculate a loss function, i.e., the difference between the loss before updating and the loss after updating, and return the calculated loss to the first training device. The first training device may update the training reference signal sequence based on the loss function. The above training reference signal sequence updating process is repeated until a predetermined condition is met, and the resulting sequence is taken as a candidate reference signal sequence.

[0041] 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 generation network may have a single hidden layer and a linear activation function. The reference signal sequence generation network may generate multiple candidate reference signal sequences with similar performance for the same constellation or codebook by randomly initializing the network. For example, the reference signal sequence generation network may generate multiple candidate reference signal sequences with similar performance for the same constellation or codebook using different initialization parameters. Note that the reference signal sequence generation network may generate multiple candidate reference signal sequences corresponding to multiple constellations or codebooks, respectively. For example, when offline training is performed using the first training device and the second training device described above, the reference signal sequence generation network may be configured in the first device. The first training device may generate candidate reference signal sequences using the reference signal sequence generation network. The second training device may calculate a loss function based on the candidate reference signal sequences generated by the first training device using the reference signal sequence generation network. Specifically, the second training device may calculate the loss function based on the original candidate reference signal sequence (i.e., the candidate reference signal sequence not transmitted through the channel) generated by the first training device and the candidate reference signal sequence transmitted through the channel. For example, in online deployment, the reference signal sequence generation network may be configured in the transmitting device.

[0042] 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 and a sequence index corresponding to each sequence may be stored in a candidate reference signal sequence table.

[0043] In step S502, information about the training reference signal sequence is transmitted. According to one example of the present disclosure, the information about the training reference signal sequence includes a sequence index of the determined training reference signal sequence and at least one of the determined training reference signal sequences. For example, as described above, there may be a plurality of candidate reference signal sequences. In this case, in step S502, a sequence index may be transmitted to indicate the training reference signal sequence determined in step S501. In step S502, a training reference signal sequence corresponding to the sequence index may be transmitted.

[0044] According to another example of the present disclosure, the training reference signal sequence may be preset. In this case, there is no need to transmit a sequence index. The training reference signal sequence information may include a determined training reference signal sequence. For example, in the case of uplink transmission, i.e., when the transmitting device is a user equipment and the receiving device is a base station, the base station may preset the training reference signal sequence to be used for the user equipment, so that the user equipment may transmit the training reference signal sequence according to the preset setting.

[0045] According to another example of the present disclosure, in step S502, a training reference signal sequence may be transmitted together with the data. For example, in step S502, when transmitting data, the training reference signal sequence may be transmitted first, and then the data may be transmitted. Alternatively, in step S502, information about the training reference signal sequence may be transmitted based on a trigger signal. For example, the receiving device or the transmitting device may trigger transmission of the training reference signal sequence when it detects that the channel environment has changed or is lower than expected. For example, the receiving device may calculate a loss based on the received data, and if the loss meets a predetermined condition, trigger the transmitting device to transmit the training reference signal sequence.

[0046] In an example according to the invention, the training reference signal sequence transmitted by the transmitting unit may be used to replace a reference signal in a legacy system, and the receiving device may perform channel estimation based on the received training reference signal sequence. For example, the training reference signal sequence may replace a cell reference signal (CRS), a demodulation reference signal (DMRS), a channel state reference signal (CSI-RS), etc. in the legacy system. Also, for example, the training reference signal sequence may replace a reference channel, such as SSB or SRS, or a broadcast channel, etc. in the legacy system.

[0047] Hereinafter, a receiving method 600 according to an embodiment of the present disclosure will be described with reference to Fig. 6. Fig. 6 is a flowchart of the receiving method 600 performed by a receiving device according to an embodiment of the present disclosure, which may be a deep neural network-based receiver (DNN-receiver). The steps of the receiving method 600 correspond to the operations of the receiving device 400 described above with reference to Fig. 4, and therefore, detailed descriptions of the same will be omitted here for convenience.

[0048] 6, in step S601, information about a training reference signal sequence including signals related to constellation points and non-constellation points in a constellation is received. Then, in step S602, a parameter set of a neural network of a receiving device is updated based on the information about the received training reference signal sequence to perform data detection. According to one example of the present disclosure, in step S602, a training reference signal sequence that is not transmitted through a channel may be obtained based on the information about the training reference signal sequence, and a parameter set of the neural network of the receiving device may be updated based on the training reference signal sequence transmitted through the channel and the training reference signal sequence that is not transmitted through the channel.

[0049] Specifically, as described above, the information on the training reference signal sequence may include at least one of a sequence index of the training reference signal sequence and the training reference signal sequence. In step S602, a training reference signal sequence that is not transmitted via a channel may be acquired from one or more pre-stored candidate reference signal sequences based on the sequence index. Alternatively, if a training reference signal sequence is preset, the preset training reference signal sequence may be acquired as the training reference signal sequence that is not transmitted via a channel. Note that in step S602, the training reference signal sequence transmitted from the transmitting device and received in step S601 may be set as the training reference signal sequence transmitted via a channel. In step S602, a parameter set of a neural network of the receiving device may be updated based on the training reference signal sequence that is transmitted via a channel and the training reference signal sequence that is not transmitted via a channel.

[0050] According to an example of the present disclosure, the receiving method 600 may further include, after performing data detection using the updated neural network parameter set, determining whether a loss calculated based on the detected data satisfies a predetermined condition. If it is determined that the loss satisfies the predetermined condition, transmitting a trigger signal for a training reference signal sequence to the transmitting device. For example, if the loss is a bit error rate (BER), the processing unit 420 may determine that the loss satisfies the predetermined condition if the loss is equal to or greater than a predetermined threshold, and the transmitting unit may transmit a trigger signal for a training reference signal sequence to the transmitting device. For example, if the loss is a cross entropy, the processing unit 420 may determine that the loss satisfies the predetermined condition if the loss is equal to or less than a predetermined threshold, and the transmitting unit may transmit a trigger signal for a training reference signal sequence to the transmitting device.

[0051] In the transmission method and the receiving method according to the embodiment of the present disclosure, a training reference signal sequence including signals related to constellation points and non-constellation points in a constellation is generated and transmitted to a receiver, which can be better suited to training the parameters of a neural network in different scenes and channel environments. Furthermore, a candidate reference signal sequence can be generated based on a difference between a loss calculated based on training data received using a parameter set of the neural network before the training reference signal sequence is updated and a loss calculated based on training data received using a parameter set of the neural network after the training reference signal sequence is updated, and the training reference signal sequence can be further obtained to maximize the accuracy of data detection after the receiving device updates the network parameters online.

[0052] <Hardware structure> The block diagrams used to explain the above embodiments show functional blocks. These functional blocks (structural units) are realized by any combination of hardware and / or software. Furthermore, the means for realizing each functional block is not particularly limited. That is, each functional block may be realized by a single device that is physically and / or logically coupled, or may be realized by two or more physically and / or logically separated devices that are connected directly and / or indirectly (for example, wired and / or wirelessly) to each other.

[0053] For example, a device (e.g., a transmitting device, a receiving device, etc.) according to an embodiment of the present disclosure may function as a computer that executes processing of the wireless communication method according to the present disclosure. Fig. 7 is a schematic diagram of the hardware structure of a device 700 (a base station or a terminal) according to an embodiment of the present disclosure. The device 700 (a base station or a 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, etc.

[0054] In the following description, the term "apparatus" can be replaced with circuit, device, unit, etc. The hardware structure of the user terminal and base station may be configured to include one or more of the apparatuses shown in the figures, or may be configured to exclude some of the apparatuses.

[0055] For example, although only one processor 710 is shown in the figure, multiple processors may be provided. Furthermore, processing may be performed by one processor, or by one or more processors simultaneously, sequentially, or otherwise. Additionally, processor 710 may be implemented on one or more chips.

[0056] Each function of device 70 is realized, for example, by loading specific software (programs) onto hardware such as processor 710 and memory 720, causing processor 710 to perform calculations and control communication via communication device 740 and reading and / or writing of data in memory 720 and storage 730.

[0057] The processor 710, for example, runs an operating system to control the entire computer. The processor 710 may be configured as a central processing unit (CPU) including an interface with peripheral devices, a control device, an arithmetic unit, a register, etc. For example, the above-mentioned decision unit, adjustment unit, etc. may be realized by the processor 710.

[0058] The processor 710 reads programs (program codes), software modules, data, etc. from the storage 730 and / or the communication device 740 into the memory 720 and executes various processes in accordance with the programs. The programs used are those that cause a computer to execute at least some of the operations described in the above embodiments. 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 run by the processor 710, and similar implementations may be used for other functional blocks.

[0059] The memory 720 is a computer-readable recording medium and may be composed of at least one of, for example, 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. The memory 720 may also be referred to as a register, a cache, a main memory, etc. The memory 720 may store an executable program (program code), a software module, etc. for implementing a method according to one embodiment of the present disclosure.

[0060] Storage 730 is a computer-readable recording medium and may be comprised of at least one of, for example, a flexible disk, a floppy disk, a magneto-optical disk (e.g., a compact disc (CD-ROM), a digital universal disc, a Blu-ray disc), a removable disk, a hard drive, a smart card, a flash memory device (e.g., a card, stick, key driver), a magnetic strip, a database, a server, and other suitable storage media. Storage 730 may also be referred to as a secondary storage device.

[0061] The communication device 740 is hardware (transmitting and receiving equipment) for communicating between computers via a wired and / or wireless network, and is also referred to as, for example, a network device, a network controller, a network card, a communication module, etc. The communication device 740 may be configured to include a high-frequency switch, a duplexer, a filter, a frequency synthesizer, etc. to realize, for example, Frequency Division Duplex (FDD) and / or Time Division Duplex (TDD). For example, the above-mentioned transmitting unit, receiving unit, etc. may be realized by the communication device 740.

[0062] The input device 750 is an input device (for example, a keyboard, mouse, microphone, switch, button, sensor, etc.) that receives input from the outside. The output device 760 is an output device (for example, a display, speaker, light emitting diode (LED) lamp, etc.) that outputs to the outside. The input device 750 and the output device 760 may be integrated into one structure (for example, a touch panel).

[0063] Each device, such as the processor 710 and the memory 720, is connected via a bus 770 for communicating information. The bus 770 may be configured as a single bus or may be configured as different buses between the devices.

[0064] The transmitting device and the receiving device may include hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), and a field programmable gate array (FPGA), and some or all of the functional blocks may be realized through the hardware. For example, the processor 710 may be installed by at least one of these pieces of hardware.

[0065] (Variation) Note that terms explained in this specification and / or terms necessary for understanding this specification may be replaced with terms having the same or similar meanings. For example, a channel and / or a symbol may be a signal (signaling). A signal may also be a message. A reference signal may be abbreviated as RS (Reference Signal) and may also be called a pilot, pilot signal, etc. depending on the applicable standard. A component carrier (CC) may also be called a cell, frequency carrier, carrier frequency, etc.

[0066] Furthermore, the information, parameters, etc. described in this specification may be expressed as absolute values, relative values ​​from a predetermined value, or other corresponding information. For example, a radio resource may be indicated by a predetermined index. The formulas using these parameters, etc. may differ from those explicitly disclosed in this specification.

[0067] The names used herein for parameters and the like are not intended to be limiting in any way. For example, the various channels (e.g., Physical Uplink Control Channel (PUCCH), Physical Downlink Control Channel (PDCCH), etc.) and information units may be identified by any suitable names, and therefore the various names assigned to these various channels and information units are not intended to be limiting in any way.

[0068] The information, signals, etc. described herein may be represented using any of a variety of different technologies. For example, data, instructions, commands, information, signals, bits, symbols, chips, etc. that may be referred to throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or magnetic particles, optical fields or photons, or any combination thereof.

[0069] Furthermore, information, signals, etc. may be output from a higher layer to a lower layer and / or from a lower layer to a higher layer. Information, signals, etc. may be input / output via multiple network nodes.

[0070] Input and output information, signals, etc. may be stored in a specific location (for example, memory) or may be managed using a management table. Input and output information, signals, etc. may be overwritten, updated, or added. Output information, signals, etc. may be deleted. Input information, signals, etc. may be transmitted to another device.

[0071] The notification of information is not limited to the aspects / embodiments described in this specification and may be performed by other methods. For example, the notification of information may be performed by physical layer signaling (e.g., downlink control information (DCI), uplink control information (UCI)), higher layer signaling (e.g., radio resource control (RRC) signaling, broadcast information (master information block (MIB), system information block (SIB), etc.), medium access control (MAC) signaling), other signals, or a combination thereof.

[0072] The physical layer signaling may be called L1 / L2 (first layer / second layer) control information (L1 / L2 control signal), L1 control information (L1 control signal), etc. The RRC signaling may be called an RRC message, such as an RRC connection setup message or an RRC connection reconfiguration message. The MAC signaling may be transmitted using a MAC control unit (MAC CE (Control Element)), for example.

[0073] Furthermore, notification of specified information (e.g., notification that "it is X") is not limited to explicit notification, but may be made implicitly (e.g., by not notifying the specified information or by notifying other information).

[0074] The determination may be made by a value represented by one bit (0 or 1), by a Boolean value represented by true or false, or by a comparison of numerical values ​​(e.g., comparison with a predetermined value).

[0075] Software shall be construed broadly to mean commands, command sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, threads of execution, steps, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.

[0076] It should be noted that software, commands, information, etc. may be transmitted or received via a transmission medium. For example, if the software is transmitted from a website, server, or other remote source using wired technologies (such as coaxial cable, optical cable, twisted pair, Digital Subscriber Line (DSL)), and / or wireless technologies (such as infrared, microwave, etc.), these wired and / or wireless technologies are included within the definition of transmission media.

[0077] As used herein, terms such as "system" and "network" may be used interchangeably.

[0078] In this specification, the terms "base station (BS)," "radio base station," "eNB," "gNB," "cell," "sector," "cell group," "carrier," and "component carrier" may be used interchangeably. A base station may also be called a fixed station, NodeB, eNodeB (eNB), access point, transmission point, reception point, femtocell, small cell, etc.

[0079] A base station can accommodate one or more (e.g., three) cells (also called sectors). When a base station accommodates multiple cells, the overall coverage area of ​​the base station can be divided into multiple smaller areas, and each smaller area can also be provided with communication service by a base station subsystem (e.g., a small indoor base station (RF remote head (RRH, remote radio head)). The term "cell" or "sector" refers to a part or the entire coverage area of ​​a base station and / or a base station subsystem that provides communication service in this coverage.

[0080] As used herein, the terms "Mobile Station (MS)," "user terminal," "User Equipment (UE)," and "terminal" may be used interchangeably. A mobile station may also be referred to by those skilled in the art as a subscriber station, mobile unit, subscriber unit, wireless unit, remote unit, mobile device, wireless device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless terminal, remote terminal, handset, user agent, mobile client, client, or some other suitable terminology.

[0081] Furthermore, a radio base station in this specification may be replaced with a user terminal. For example, the aspects / present embodiments of the present disclosure may be applied to a configuration in which communication between a radio base station and a user terminal is replaced with communication between multiple user terminals (D2D, Device-to-Device). In this case, the functions possessed by the first communication device or the second communication device in the above-mentioned device 800 may be considered to be functions possessed by the user terminal. Furthermore, terms such as "uplink" and "downlink" may be replaced with "side." For example, an uplink channel may be replaced with a side channel.

[0082] Similarly, the user terminal in this specification may be replaced with a radio base station, in which case the functions of the user terminal described above may be functions of the first communication device or the second communication device.

[0083] In this specification, an operation that is described as being performed by a base station may be performed by its upper node in some cases. It is apparent that in a network including one or more 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 (such as, but not limited to, a mobility management entity (MME) and a serving gateway (S-GW)), or a combination thereof.

[0084] Each aspect / embodiment described herein may be used alone or in combination, and may be switched during execution. The processing steps, sequences, flowcharts, etc. of each method / embodiment described herein may be rearranged as long as there is no contradiction. For example, with respect to the methods described herein, various step units are given in an exemplary order, but are not limited to the given specific order.

[0085] The aspects / present embodiments described in this specification are based on Long Term Evolution (LTE), Advanced Long Term Evolution (LTE-A, LTE-Advanced), Super Long Term Evolution (LTE-B, LTE-Beyond), Super 3G, International Mobile Telecommunications System (IMT-Advanced), 4th generation mobile communication system (4G), 5th generation mobile communication system (5G), Future Radio Access (FRA), New Radio Access Technology (New-RAT, Radio Access Technology), New Radio (NR), New radio access (NX), Next Generation Radio Access (FX), Global System for Mobile communications (GSM (registered trademark)), Code Division Multiple Access 3000 (CDMA3000), Ultra Mobile Broadband (UMB), and the like. The present invention may be applied to systems that use IEEE 920.11 (Wi-Fi (registered trademark)), IEEE 920.16 (WiMAX (registered trademark)), IEEE 920.20, Ultra-Wide Band (UWB), Bluetooth (registered trademark), or other suitable wireless communication methods, and / or next-generation systems that are based on and extend these.

[0086] As used herein, "based on" does not mean "based only on," unless expressly stated otherwise. That is, the phrase "based on" means both "based only on" and "based at least on."

[0087] As used herein, any reference to units using names such as "first," "second," etc., does not generally limit the quantity or order of those units. These names may be used herein as a convenient way to distinguish between two or more units. Thus, a reference to a first unit and a second unit does not imply that only two units may be employed therein, or that the first unit must precede the second unit.

[0088] As used herein, the term "determining" may encompass a wide variety of actions. For example, "determining" may be considered to be calculating, computing, processing, deriving, investigating, looking up (e.g., looking up in a table, database, or another data structure), ascertaining, etc. Also, "determining" may be considered to be receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, accessing (e.g., accessing data in memory), etc. Also, "determining" may be considered to be resolving, selecting, choosing, establishing, comparing, etc. In other words, a "judgment (decision)" may be considered to be a "judgment (decision)" of some action.

[0089] As used herein, the terms "connected," "coupled," or any variation thereof, refer to any connection or coupling, direct or indirect, between two or more units, and may include the presence of one or more intermediate units between two units that are "connected" or "coupled" to each other. The coupling or connection between units may be physical, logical, or a combination thereof. For example, "connected" may be substituted with "access." As used herein, when two units are connected, they may be considered to be "connected" or "coupled" to each other using one or more wires, cables, and / or printed electrical connections, as well as using electromagnetic energy having wavelengths in the radio frequency range, microwave range, and / or optical (both visible and invisible) range, as some non-limiting and non-exhaustive examples.

[0090] When the terms "including," "comprising," and variations thereof are used in this specification or the claims, these terms are intended to be inclusive, similar to the term "comprising." Furthermore, when the term "or" is used in the specification or the claims, it is not intended to be an exclusive or.

[0091] Although the present disclosure has been described in detail above, it is clear to those skilled in the art that the present disclosure is not limited to the embodiments described in this specification. The present disclosure can be implemented in modified and altered forms without departing from the spirit and scope of the present disclosure, which are determined by the description of the claims. Therefore, the description in this specification is intended to be illustrative and explanatory and does not have any limiting meaning on the present disclosure.

Claims

1. A transmitting device, a processing unit for determining a training reference signal sequence comprising signals relating to constellation points and non-constellation points in the constellation; a transmitting unit for transmitting information about the training reference signal sequence; Transmitting equipment including.

2. generating one or more candidate reference signal sequences based on the constellation and the loss function; the loss function is a difference between a loss calculated based on training data received using a parameter set of the neural network before the training reference signal sequence is updated and a loss calculated based on training data received using a parameter set of the neural network after the training reference signal sequence is updated; The transmitting device of claim 1 , wherein the processing unit determines the training reference signal sequence from the one or more candidate reference signal sequences.

3. The transmitting device according to claim 1 or 2, wherein the information about the training reference signal sequence includes at least one of a sequence index of the determined training reference signal sequence and the determined training reference signal sequence.

4. The transmitting device of claim 2 , wherein the one or more candidate reference signal sequences are generated using a reference signal sequence generating network.

5. the transmitting unit transmits the training reference signal sequence together with data; or The transmitting device according to claim 1 or 2, wherein the transmitting unit transmits information about the training reference signal sequence based on a trigger signal.

6. a receiving unit for receiving information about a training reference signal sequence including signals about constellation points and non-constellation points in the constellation; a processing unit for updating a parameter set of a neural network of the receiving device based on information about the received training reference signal sequence to perform data detection; Including receiving equipment.

7. 7. The receiving device of claim 6, wherein the processing unit obtains a training reference signal sequence transmitted through a channel and a training reference signal sequence not transmitted through a channel based on information about the training reference signal sequence, and updates a parameter set of a neural network of the receiving device based on the training reference signal sequence transmitted through the channel and the training reference signal sequence not transmitted through the channel.

8. The processing unit further determines whether the loss calculated based on the detected data satisfies a predetermined condition; The receiving device is 8. The receiving device according to claim 6 or 7, further comprising a sending unit for sending a trigger signal for a training reference signal sequence to a sending device if the loss satisfies a predetermined condition.

9. determining a training reference signal sequence comprising signals relating to constellation points and non-constellation points in the constellation; transmitting information about the training reference signal sequence; A transmission method including:

10. receiving information about a training reference signal sequence including signals relating to constellation points and non-constellation points in a constellation; updating a parameter set of a neural network of the receiving device based on information about the received training reference signal sequence to perform data detection; receiving method, including

Images