Terminal, method performed by terminal, and method performed by serving cell

Abstract

A terminal, a method performed by the terminal, and a method performed by a serving cell are provided. The terminal receives a measurement configuration from a serving cell, measures at least one received signal based on the measurement configuration, and determines whether an event corresponding to the measurement configuration has occurred. Based on determining that the event has occurred, the terminal generates a potential vector by encoding a measurement result set including a plurality of measurement results, and in response to the occurrence of the event, transmits a measurement report generated based on the potential vector to the serving cell.

Description

Technical Field

[0001] The embodiments relate to a terminal, a method performed by a terminal, and a method performed by a serving cell. Background Art

[0002] The communication system can support switching between cells without disconnecting the data flow in the connected state to ensure the mobility of the terminal. During the reception of the service, the terminal can periodically measure the signal received from the currently connected serving cell (or source cell), and can report the measurement results to the serving cell when a handover-related event specified by the serving cell occurs.

[0003] However, when the terminal sends a measurement report for handover to the serving cell, the amount of data of the measurement results included in the measurement report may increase the signaling overhead between the terminal and the serving cell. In addition, the reliability of handover-related events occurring in a specific communication environment may be low, resulting in unnecessary handovers, thereby wasting wireless resources. Summary of the invention

[0004] The embodiment provides a terminal for performing a handover procedure, a method performed by the terminal, and a method performed by a serving cell, in which resources used for measurement reporting for handover are efficiently used, so that unnecessary handovers are reduced.

[0005] A method performed by a terminal is provided, the method comprising: receiving a measurement configuration from a serving cell; measuring at least one received signal based on the measurement configuration; determining whether an event corresponding to the measurement configuration has occurred; based on determining that the event has occurred, generating a potential vector by encoding a measurement result set including a plurality of measurement results; and in response to the occurrence of the event, sending a measurement report generated based on the potential vector to the serving cell.

[0006] Also provided is a method performed by a serving cell, the method comprising: sending a measurement configuration to a terminal; receiving a measurement report corresponding to the measurement configuration and based on an autoencoder from the terminal; obtaining a measurement result set including multiple measurement results by decoding a latent vector included in the measurement report; and controlling a next switching step based on the measurement result set.

[0007] In addition, a terminal is provided, comprising: a memory configured to store a neural network model trained based on an autoencoder; and a processor configured to generate a latent vector from a measurement result set including multiple measurement results for switching by using the neural network model, and determine whether to send a measurement report to a serving cell based on the latent vector. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Embodiments will be more clearly understood from the following detailed description taken in conjunction with the accompanying drawings, in which:

[0009] Figure 1 is a diagram showing a communication system according to an embodiment;

[0010] Figure 2A is a block diagram schematically showing a terminal according to an embodiment, and Figure 2B is a diagram showing an autoencoder used as a base component according to an embodiment;

[0011] Figure 3 is a flow chart showing a handover process in a communication system according to an embodiment;

[0012] Figure 4A is a diagram schematically showing an implementation of a terminal and a serving cell according to an embodiment, and Figure 4B is a diagram schematically showing an implementation of a terminal according to an embodiment;

[0013] Figure 5A is a flowchart showing a handover process in a communication system according to an embodiment, and Figure 5B It shows Figure 5A A flowchart of a detailed embodiment of operation S141;

[0014] Figure 6 is a diagram schematically showing an implementation of a terminal according to an embodiment;

[0015] Figure 7 is a flow chart showing a handover process in a communication system according to an embodiment;

[0016] Figure 8 is a diagram schematically showing an implementation manner of a terminal and a serving cell according to an embodiment;

[0017] Fig.9A is a flowchart showing a handover process in a communication system according to an embodiment, and Fig. 9B It shows Fig.9A A flowchart of a detailed embodiment of operation S260;

[0018] Fig.10 is a flowchart illustrating a method performed by a terminal according to an embodiment;

[0019] Fig.11 is a flow chart showing a handover process in a communication system according to an embodiment;

[0020] Fig.12 is a flowchart showing a switching operation in a communication system according to an embodiment;

[0021] Fig.13A and Fig. 13B is a flow chart showing a method for generating a neural network model based on an autoencoder in a communication system according to an embodiment;

[0022] Fig.14 is a block diagram showing an electronic device according to an embodiment; and

[0023] Fig.15 is a diagram illustrating a communication device for performing a handover operation according to an embodiment. DETAILED DESCRIPTION

[0024] Hereinafter, embodiments will be described with reference to the accompanying drawings.

[0025] Figure 1 is a diagram showing a communication system according to an embodiment. The embodiment may be applicable to both a homogeneous network including cells of the same type and a heterogeneous network including cells of different types. As a non-limiting example, the communication system may be a fifth generation (5G) system, a long term evolution (LTE) system, a code division multiple access (CDMA) system, a global system for mobile communications (GSM) system, a wireless local area network (WLAN) system, or any other wireless communication system. In the following, it is assumed that the communication system is a 5G system, an LTE system, or a system capable of supporting communications based on 5G and LTE. However, it should be understood that the embodiment is not limited thereto and may also be applicable to newly proposed next generation communication technologies.

[0026] refer to Figure 1 , the communication system may include a plurality of macro cells MC1 to MC4, a plurality of small cells SC1 to SC7 and a terminal TM. The terminal TM may be a wireless communication device, may be mobile, and may send and receive data and / or control information by communicating with the macro cells MC1 to MC4 and the small cells SC1 to SC7. The terminal TM may be referred to as, for example, a user equipment (UE), a mobile station (MS), a mobile terminal (MT), a user terminal, a subscriber station (SS), a wireless device, a handheld device, a communication device, a device, etc.

[0027] Each of the macro cells MC1 to MC4 and the small cells SC1 to SC7 may generally refer to a fixed station that communicates with the terminal TM and / or another cell, and may exchange data and control information by communicating with the terminal TM and / or another cell. Compared with the small cells SC1 to SC7, the macro cells MC1 to MC4 may correspond to a larger coverage area. The coverage area refers to a range in which a certain cell can provide a network to a user (or terminal).

[0028] Each of the macro cells MC1 to MC4 and the small cells SC1 to SC7 may be referred to as, for example, a base station, a Node B, an evolved Node B (eNB), a sector, a site, a base transceiver system (BTS), an access point (AP), a relay node, a remote radio head (RRH), a radio unit (RU), etc. Although Figure 1 The communication system is shown to include only macro cells MC1 to MC4 and small cells SC1 to SC7 corresponding to different coverage areas, but this is only an example, and therefore, the embodiment is not limited thereto. The communication system may also include super-large cells, micro cells, pico cells, femto cells, etc. corresponding to coverage areas of various sizes, respectively.

[0029] Due to the mobility of the terminal TM, handover from macro cell to macro cell, from macro cell to small cell, from small cell to macro cell or from small cell to small cell may occur. For handover, a handover procedure specified in the 3GPP standard may be performed between the terminal, the serving cell and the target cell.

[0030] The terminal TM may receive a radio resource control (RRC) reconfiguration message including a measurement configuration from a serving cell to which the current RRC connection is provided for network services. The measurement configuration may be a prior arrangement for the terminal TM to perform an operation of measuring a signal received from a serving cell and sending a measurement report to the serving cell, and may include a report configuration and information about a measurement object to be measured by the terminal TM. For example, the measurement object information may include a frequency band, candidate target cell information, a blacklist, etc., and the report configuration may include information indicating a specific event (or condition) under which the terminal TM will send a measurement report.

[0031] There may be multiple events for triggering measurement reports during the switching process. For example, multiple events may include "A1" to "A6" events, "B1" events and "B2" events, "I1" events, and "C1" events and "C2" events. In this article, events may also be referred to as switching-related events. In order to briefly describe each event, the "A1" event occurs when the signal measurement value of the serving cell becomes better than a threshold, the "A2" event occurs when the signal measurement value of the serving cell becomes worse than a threshold, the "A3" event occurs when the signal measurement value of the neighboring cell becomes better than the signal measurement value of the special cell (or SpCell) by an offset, the "A4" event occurs when the signal measurement value of the neighboring cell becomes better than the threshold, the "A5" event occurs when the signal measurement value of the special cell becomes worse than the first threshold and the signal measurement value of the neighboring cell becomes better than the second threshold, and the "A6" event occurs when the signal measurement value of the neighboring cell becomes better than the signal measurement value of the secondary cell (or SCell) by an offset. The "B1" event occurs when the signal measurement value of the neighboring cell between radio access technologies (RATs) becomes better than a threshold, and the "B2" event occurs when the signal measurement value of the primary cell (or PCell) becomes worse than a first threshold and the signal measurement value of the neighboring cell between RATs becomes better than a second threshold. The "I1" event occurs when the measurement value of the interference signal becomes higher than a threshold. The "C1" event occurs when the channel busy rate of the new radio (NR) side link is higher than a threshold, and the "C2" event occurs when the channel busy rate of the NR side link is lower than a threshold. The reporting configuration may refer to any one of the above-mentioned multiple events. However, the above-mentioned multiple events are merely examples, and the embodiments may be applicable to other events promised or defined in future standards.

[0032] The embodiment may be applicable to all events in the plurality of events. In addition, the embodiment may be selectively applicable to only some events in the plurality of events. The detailed embodiment of this aspect is described below with reference to FIGS. 2 to 10 .

[0033] In some embodiments, the serving cell may provide the terminal TM with a measurement ID having a value according to a combination of the measurement object information and the report configuration as a measurement configuration. In this case, the terminal TM may identify the frequency band, candidate target cell information, and blacklist related to at least one signal to be measured based on the value of the measurement ID, and may also identify a specific event that requires (or triggers) a measurement report to be sent.

[0034] The terminal TM may measure at least one received signal based on the measurement configuration received from the serving cell. In detail, the terminal TM may measure at least one signal corresponding to the measurement object information included in the measurement configuration. The terminal TM may perform a measurement operation and may periodically or aperiodically determine whether an event corresponding to or matching the measurement configuration has occurred. In detail, the terminal TM may determine whether a specific event indicated in a report configuration included in the measurement configuration has occurred.

[0035] In an embodiment, when a specific event has occurred, the terminal TM may generate a latent vector by encoding the measurement result. For example, the terminal TM may generate a latent vector by using a first neural network model corresponding to an encoder trained based on an autoencoder. Figure 2B A detailed embodiment of the autoencoder is described. In this article, the latent vector may be data representing at least one characteristic of the measurement result, and may be compressed to be smaller than the data size of the measurement result.

[0036] In an embodiment, in response to the occurrence of a specific event, the terminal TM may send a measurement report generated based on a latent vector to a serving cell. For example, when a second neural network model corresponding to a decoder trained based on an autoencoder is stored in a serving cell, the terminal TM may send a measurement report including a latent vector to the serving cell. As another example, when the second neural network model is not stored in the serving cell, the terminal TM may send a measurement report including at least one measurement result instead of a latent vector to the serving cell. In this document, the function of the terminal TM sending a measurement report including a latent vector to a serving cell may be referred to as a measurement report function based on an autoencoder.

[0037] In some embodiments, when a specific event has occurred, the terminal TM may not immediately send a measurement report to the serving cell, but may actively determine whether the occurrence of the specific event is reliable by using a potential vector. That is, during the measurement operation for switching in a specific communication environment, incorrect measurement results may be temporarily generated, and incorrect switching to the target cell may be performed due to these incorrect measurement results, and therefore, a so-called ping-pong situation may occur, that is, it is necessary to perform switching back to the previous serving cell. Therefore, according to the embodiment, the terminal TM can generate an event occurrence reliability based on a potential vector, the event occurrence reliability can be compared with a preset reference value, and it is determined whether to send a measurement report to the serving cell based on the comparison result. The event occurrence reliability may be referred to as an event occurrence reliability value in this article. In this article, the function of the terminal TM to actively determine whether to send a measurement report even when a specific event has occurred may be referred to as a measurement report determination function. In some embodiments, the serving cell rather than the terminal TM can determine whether the occurrence of a specific event is reliable based on a potential vector.

[0038] In an embodiment, the terminal TM may support at least one of a measurement report function based on an automatic encoder and a measurement report determination function, and in order to notify the serving cell of this, the terminal TM may provide capability information to the serving cell, the capability information indicating at least one handover-related function that the terminal TM may support. In some embodiments, the terminal TM may send the capability information to the serving cell through an RRC signaling step or a certain step with the serving cell. In an embodiment, the serving cell may process the measurement report received from the terminal TM based on the capability information of the terminal TM.

[0039] For example, the serving cell can determine that the terminal TM can support the measurement report function based on the autoencoder through the capability information of the terminal TM, and can recognize that the measurement report to be received from the terminal TM later includes a latent vector. Thereafter, the serving cell can decode the latent vector of the measurement report received from the terminal TM by using the second neural network model.

[0040] For example, the serving cell can determine that the terminal TM can support the measurement report function through the capability information of the terminal TM, and can recognize that the measurement report to be received from the terminal TM later has high reliability. Thereafter, the serving cell can control the next handover step, such as a handover operation, by making full use of the measurement report received from the terminal TM without separately analyzing the measurement report.

[0041] The terminal TM according to the embodiment can encode the measurement results required for handover of the serving cell into a latent vector by using a neural network model trained based on an autoencoder, and send the latent vector to the serving cell, thereby significantly reducing the signaling overhead for handover.

[0042] In addition, the terminal TM according to the embodiment can determine whether the event that has occurred is reliable based on the potential vector, and can actively determine whether to send a measurement report based on the determination result, thereby reducing unnecessary handovers. Therefore, the wireless resources of the communication system can be used efficiently.

[0043] Figure 2A is a block diagram schematically showing a terminal 100 according to an embodiment, and Figure 2B is a diagram showing an autoencoder used as a basic component according to an embodiment. Figure 2B Embodiments of the autoencoder are shown for example purposes only, and the autoencoder is not limited thereto.

[0044] refer to Figure 2A, the terminal 100 may include multiple antennas AT, a radio frequency (RF) integrated circuit (IC) 110, a baseband IC 120, a processor 130, and a memory 140. In an exemplary embodiment, the processor 130 is a single processor, while in another embodiment, the processor 130 includes at least one processor. Figure 2A The illustrated implementation of the terminal 100 is merely an example, and thus, the embodiments are not limited thereto, and the terminal 100 may include more or fewer components.

[0045] RFIC 110 can perform functions such as frequency band conversion and amplification of signals so as to send and receive signals via wireless channels by using antenna AT. In detail, RFIC 110 can up-convert the baseband signal provided from baseband IC 120 into an RF band signal and send the RF band signal through antenna AT, and can down-convert the RF band signal received by antenna AT into a baseband signal. For example, RFIC 110 may include a transmit filter, a receive filter, an amplifier, a mixer, an oscillator, a digital-to-analog converter (DAC), an analog-to-digital converter (ADC), etc. In addition, RFIC 110 may also include a plurality of RF chains (not shown), and beamforming may be performed by using antenna AT. For beamforming, RFIC 110 may adjust the phase and amplitude of each signal sent and received by antenna AT. In addition, RFIC 110 may perform multiple-input multiple-output (MIMO), and may receive multiple layers of signals when performing MIMO operations.

[0046] The baseband IC 120 may perform conversion operations between baseband signals and bit streams according to the physical layer standard of the system. For example, the baseband IC 120 may generate complex symbols by encoding and modulating the transmission bit stream during data transmission. In addition, the baseband IC 120 may recover the received bit stream by demodulating and decoding the baseband signal provided from the RFIC 110 during data reception.

[0047] As described above, RFIC 110 and baseband IC 120 can send and receive signals. Each of RFIC 110 and baseband IC 120 can be referred to as a transmitter, a receiver, a transceiver, or a communicator. In addition, at least one of RFIC 110 and baseband IC 120 can include multiple communication modules to support a variety of different wireless access technologies. In addition, at least one of RFIC 110 and baseband IC 120 can include different communication modules for processing signals of different frequency bands. For example, different wireless access technologies can include NR technology, LTE technology, etc. In addition, different frequency bands can include ultra-high frequency bands, millimeter wave bands, etc. Terminal 100 can communicate with a cell by using RFIC 110 and baseband IC 120.

[0048] The processor 130 may control the overall operation of the terminal 100. In an embodiment, the processor 130 may include a switching circuit 132 that may perform at least one of encoding a measurement result for switching and determining whether to send a measurement report to a serving cell.

[0049] The memory 140 may store data such as basic programs and applications for the operation of the terminal 100. In an embodiment, the memory 140 may store an autoencoder-based neural network model 142. The neural network model 142 may include a first neural network model corresponding to an encoder trained based on the autoencoder and a second neural network model corresponding to a decoder trained based on the autoencoder.

[0050] Further references Figure 2B The autoencoder is described in detail. The autoencoder may include an encoder and a decoder. The encoder may be a cognitive network and may convert the input data x 1 、x 2 and x 3 is converted into a latent vector (or internal representation), and the decoder can be a generative network and can convert the latent vector into output data y 1 ,y 2 and 3 For example, in an autoencoder, the number of neurons in the input layer and the output layer can be the same, and the number of neurons in the hidden layer can be less than the number of neurons in the input layer or the output layer. An autoencoder can perform unsupervised learning to generate an output signal indicating input data x 1 、x 2 and x 3 The latent vector of the important features of the neural network model is generated. The neural network model can be called an autoencoder-based neural network model.

[0051] An autoencoder suitable for use in an embodiment may correspond to any of an “undercomplete” autoencoder, a “stacked” autoencoder, a “denoising” autoencoder, a “sparse” autoencoder, a “variational” autoencoder, and the like.

[0052] Return to reference Figure 2A In an embodiment, the switching circuit 132 may generate a potential vector by encoding the measurement results for the switching based on the neural network model stored in the memory 140. The switching circuit 132 may generate a measurement report including the potential vector, and may send the measurement report to the serving cell through the antenna AT, the RFIC 110, and the baseband IC 120.

[0053] In an embodiment, the switching circuit 132 may determine whether to send a measurement report to the serving cell based on a potential vector. When a specific event indicated in the received measurement configuration has occurred, the switching circuit 132 may determine whether the occurrence of the event is reliable by using the potential vector. In detail, the switching circuit 132 may generate an event occurrence confidence based on the potential vector, and when the event occurrence confidence is greater than a reference value, it may be determined to send a measurement report to the serving cell. In addition, when the event occurrence confidence is less than or equal to the reference value, the switching circuit 132 may determine not to send a measurement report to the serving cell, may perform a measurement operation, and may continuously monitor whether a specific event has occurred.

[0054] In an embodiment, the switching circuit 132 may manage the switching related functions supported by the terminal 100. For example, the switching circuit 132 may manage the functions that the terminal 100 may support among the measurement report function and the measurement report determination function based on the automatic encoder, and may generate capability information indicating the functions that the terminal 100 may support. The switching circuit 132 may send the capability information to the serving cell through the antenna AT, the RFIC 110, and the baseband IC 120.

[0055] In an embodiment, the switching circuit 132 may provide the serving cell with the second neural network model required for decoding the latent vector. In some embodiments, the serving cell may generate the second neural network model independently from the terminal 100.

[0056] Figure 3 is a flow chart showing a handover process in a communication system according to an embodiment. The description made below may be based on specifications defined in 3GPP standards (eg, 3GPP TS 38.331).

[0057] refer to Figure 3 , in operation S100 , the terminal 100 in the RRC connection state with the serving cell 10 may receive a network service from the serving cell 10 by operating in a connected mode.

[0058] In operation S110, the serving cell 10 may send a measurement configuration to the terminal 100. The serving cell 10 may determine the measurement object information and the report configuration associated with the specific measurement ID (or measID) in the measurement configuration, and may send an RRC reconfiguration message including the measurement configuration to the terminal 100. The report configuration may indicate a specific event that triggers the terminal 100 to perform a measurement report, and may indicate which object the terminal 100 measures as the electric field strength. For example, the report configuration may indicate which of the reference signal received power (RSRP), the reference signal received quality (RSRQ), the signal to interference plus noise ratio (SINR), the received signal code power (RSCP), and the received energy from pilot signal to noise density (EcN0) will be measured as the electric field strength. In this article, an embodiment in which the RSRP of the terminal 100 is measured is mainly described, but it should be understood that this is an example and the embodiment is not limited thereto.

[0059] In operation S120, the terminal 100 may measure at least one signal based on the measurement configuration received in operation S100. The terminal 100 may generate multiple measurement results by measuring at least one signal indicated in the measurement configuration. The multiple measurement results may include measurement results of the electric field strength (such as RSRP) indicated in the measurement configuration, and at least one measurement result predetermined to generate a potential vector. In this document, the multiple measurement results may be referred to as a measurement result set.

[0060] In operation S130, the terminal 100 may determine whether an event corresponding to or matching the measurement configuration received in operation S100 has occurred. For example, the terminal 100 may determine whether an event occurrence condition is satisfied based on a measurement result corresponding to an electric field strength (e.g., RSRP) among a plurality of measurement results generated in operation S120.

[0061] When the answer is "NO" in operation S130, the terminal 100 may re-perform operation S120.

[0062] When the answer is "yes" in operation S130, in operation S140, the terminal 100 may generate a latent vector by encoding the measurement result generated in operation S120. For example, the terminal 100 may generate a latent vector from the measurement result by using a first neural network model corresponding to an encoder trained based on an autoencoder.

[0063] In operation S150, the terminal 100 may transmit a measurement report including the potential vector generated in operation S140 to the serving cell 10. The terminal 100 may know in advance through certain signaling (eg, RRC signaling) that the serving cell 10 may decode the potential vector.

[0064] In operation S160, the serving cell 10 may obtain a plurality of measurement results by decoding the latent vector. For example, the serving cell 10 may obtain the measurement results from the latent vector by using a second neural network model corresponding to a decoder trained based on an autoencoder. In an embodiment, the serving cell 10 may receive the second neural network model from the terminal 100. In some embodiments, the serving cell 10 may directly generate the second neural network model by learning based on the autoencoder.

[0065] In operation S170 , the serving cell 10 may control a next handover procedure based on the measurement result obtained in operation S160 .

[0066] Figure 4A is a diagram schematically showing an implementation of a terminal 100 and a serving cell 10 according to an embodiment, and Figure 4B is a diagram schematically showing an implementation of the terminal 100 according to an embodiment. Figure 4B Assume an embodiment in which the serving cell 10 is unable to decode the latent vector.

[0067] refer to Figure 4A , the terminal 100 may include an encoder 132_1, and the serving cell 10 may include a decoder 13_1. The encoder 132_1 may correspond to the first neural network model, and the decoder 13_1 may correspond to the second neural network model.

[0068] In an embodiment, the terminal 100 may input a plurality of measurement results MR_1 to MR_n generated by a measurement operation for handover as a measurement result set MR_G to the encoder 132_1. The encoder 132_1 may generate a potential vector Z by encoding the measurement result set MR_G. For example, the plurality of measurement results MR_1 to MR_n may include measurement results of a signal corresponding to at least one of the serving cell 10 and the candidate target cell for handover. The signal may be identified by a measurement configuration received from the serving cell 10.

[0069] In an embodiment, terminal 100 may send latent vector Z to serving cell 10, and serving cell 10 may input latent vector Z to decoder 13_1. Decoder 13_1 may decode latent vector Z to obtain a measurement result set MR_G' including a plurality of measurement results MR_1' to MR_n'.

[0070] Further references Figure 4B , the terminal 100 may include an encoder 132_1, a decoder 132_2, and a reporting circuit 132_3. The encoder 132_1 may correspond to the first neural network model, and the decoder 132_2 may correspond to the second neural network model.

[0071] In an embodiment, the terminal 100 may input a plurality of measurement results MR_1 to MR_n generated through a measurement operation for handover to the encoder 132_1 as a measurement result set MR_G. The encoder 132_1 may generate a latent vector Z by encoding the measurement result set MR_G.

[0072] In an embodiment, the terminal 100 may input the latent vector Z to the decoder 132_2. The decoder 132_2 may output a plurality of measurement results MR_1 to MR_n by decoding the latent vector Z.

[0073] In an embodiment, the reporting circuit 132_3 may receive the plurality of measurement results MR_1 to MR_n, and may generate a measurement report including at least one of the plurality of measurement results MR_1 to MR_n.

[0074] The terminal 100 may use the generated potential vector Z to determine whether to perform measurement reporting or any operation for efficiently performing handover.

[0075] Figure 5A is a flowchart showing a handover process in a communication system according to an embodiment, and Figure 5B It shows Figure 5A Flow chart of a detailed embodiment of operation S141.

[0076] refer to Figure 5A In operation S100 , the terminal 100 may operate in a connection mode with the serving cell 10 .

[0077] In operation S110 , the serving cell 10 may transmit a measurement configuration to the terminal 100 .

[0078] In operation S120, the terminal 100 may measure at least one signal based on the measurement configuration received in operation S100.

[0079] In operation S130 , the terminal 100 may determine whether an event matching the measurement configuration received in operation S100 has occurred.

[0080] When the answer is "NO" in operation S130, the terminal 100 may re-perform operation S120.

[0081] When the answer is 'yes' in operation S130, in operation S140, the terminal 100 may generate a latent vector by encoding the measurement result generated in operation S120.

[0082] In operation S141, the terminal 100 may determine whether to perform a measurement report based on the potential vector generated in operation S140. For example, the terminal 100 may generate an event occurrence confidence from the potential vector, may compare the event occurrence confidence with a reference value, and may determine whether to perform a measurement report based on the comparison result. That is, only when the event occurrence confidence generated from the potential vector is greater than the reference value, the terminal 100 may send a measurement report including the potential vector to the serving cell.

[0083] When the answer is "NO" in operation S141, the terminal 100 may re-perform operation S120.

[0084] When the answer in operation S141 is "yes", in operation S150, the terminal 100 may send a measurement report including the potential vector to the serving cell 10. In some embodiments, when the serving cell 10 cannot decode the potential vector, as in the above Figure 4B As in the embodiment of FIG. 1 , the terminal 100 may send a measurement report generated by decoding the latent vector to the serving cell 10.

[0085] In operation S160, the serving cell 10 may obtain a plurality of measurement results by decoding the latent vectors.

[0086] In operation S170, the serving cell 10 may control the next handover step based on the measurement result obtained in operation S160. For example, the serving cell 10 may fully use the obtained measurement result by trusting the occurrence of an event that has triggered the measurement report received from the terminal 100. The serving cell 10 may know in advance that the terminal 100 supports the measurement report determination function.

[0087] Further references Figure 5B , Figure 5A Operation S141 may include operations S141_1 to S141_4.

[0088] In operation S141_1, the terminal 100 may generate event occurrence confidence based on the latent vector.

[0089] In operation S141_2, the terminal 100 may compare the event occurrence reliability with a reference value. For example, the terminal 100 may determine whether the event occurrence reliability is greater than the reference value.

[0090] When the answer is “yes” in operation S141_2 , the terminal 100 may determine to send a measurement report including a potential vector to the serving cell 10 in operation S141_3 .

[0091] When the answer is “No” in operation S141_2 , the terminal 100 may determine not to transmit a measurement report including a potential vector to the serving cell 10 in operation S141_4 .

[0092] Figure 6 is a diagram schematically showing an implementation of the terminal 100 according to an embodiment. Figure 6 An embodiment assuming that an "A3" event occurs is shown. However, this is merely an example, and thus, the embodiment is not limited thereto, and the type of measurement result generated by the measurement operation may vary depending on the type of event.

[0093] refer to Figure 6 , the terminal 100 may include an encoder 132_1 and a report determination circuit 132_4. The encoder 132_1 may correspond to the above-mentioned first neural network model.

[0094] In an embodiment, the terminal 100 may generate a first measurement result MR_RSRP, a second measurement result MR_TA, and a third measurement result MR_PMI for the serving cell and the candidate target cell through a measurement operation for switching. For example, the first measurement result MR_RSRP may include the RSRP of the signal of the serving cell and the RSRP of the signal of the candidate target cell. The second measurement result MR_TA may include timing advance information between the terminal 100 and the serving cell and timing advance information between the terminal 100 and the candidate target cell. In addition, the third measurement result MR_PMI may include precoding matrix indication (PMI) information of the serving cell and PMI information of the candidate target cell. In some embodiments, the second measurement result MR_TA may be replaced with any one of the results measured by various methods to indicate the distance from each of the serving cell and the candidate target cell to the terminal 100, and the third measurement result MR_PMI may be replaced with any one of the results measured by various methods to indicate the direction toward the serving cell and the direction toward the candidate target cell relative to the terminal 100.

[0095] In an embodiment, the terminal 100 may input the first to third measurement results MR_RSRP, MR_TA, and MR_PMI to the encoder 132_1 as a measurement result set MR_G. The encoder 132_1 may generate a latent vector Z by encoding the measurement result set MR_G.

[0096] In an embodiment, the report determination circuit 132_4 may determine whether to send a measurement report to the serving cell based on the potential vector Z. For example, the report determination circuit 132_4 may generate an event occurrence confidence by using the potential vector Z, and when the event occurrence confidence is greater than a reference value, it may be determined to send a measurement report to the serving cell.

[0097] Figure 7 is a flowchart illustrating a handover process in a communication system according to an embodiment. Figure 7 The embodiment corresponds to an embodiment in which the serving cell 10 instead of the terminal 100 generates the event occurrence confidence based on the potential vector.

[0098] refer to Figure 7 In operation S100 , the terminal 100 may operate in a connection mode with the serving cell 10 .

[0099] In operation S110 , the serving cell 10 may transmit a measurement configuration to the terminal 100 .

[0100] In operation S120 , the terminal 100 may measure at least one signal based on the received measurement configuration in operation S100 .

[0101] In operation S130 , the terminal 100 may determine whether an event matching the received measurement configuration in operation S100 has occurred.

[0102] When the answer is "NO" in operation S130, the terminal 100 may re-perform operation S120.

[0103] When the answer is 'yes' in operation S130, in operation S140, the terminal 100 may generate a latent vector by encoding the measurement result generated in operation S120.

[0104] In operation S150 , the terminal 100 may transmit a measurement report including a potential vector to the serving cell 10 .

[0105] In operation S180 , the serving cell 10 may generate event occurrence confidence based on the potential vector.

[0106] In operation S190, the serving cell 10 may control the next handover step based on the event occurrence confidence. For example, when the event occurrence confidence is less than or equal to the reference value, the serving cell 10 may determine that the event that has occurred is unreliable, and may resend the measurement configuration to the terminal 100 in operation S110. When the event occurrence confidence is greater than the reference value, the serving cell 10 may control the execution of the next handover step.

[0107] Figure 8is a diagram schematically showing an implementation of the terminal 100 and the serving cell 10 according to an embodiment. Figure 8 Shows Figure 7 The implementation method of the serving cell 10 is as follows.

[0108] refer to Figure 8 , the terminal 100 may include an encoder 132_1, and the serving cell 10 may include a decoder 13_1 and a switching control circuit 13_2. The encoder 132_1 may correspond to the above-mentioned first neural network model, and the decoder 13_1 may correspond to the above-mentioned second neural network model.

[0109] In an embodiment, the terminal 100 may input a plurality of measurement results MR_1 to MR_n generated through a measurement operation for handover to the encoder 132_1 as a measurement result set MR_G. The encoder 132_1 may generate a latent vector Z by encoding the measurement result set MR_G.

[0110] In an embodiment, the terminal 100 may send a potential vector Z to the serving cell 10, and the serving cell 10 may input the potential vector Z to the switching control circuit 13_2. The switching control circuit 13_2 may generate an event occurrence confidence based on the potential vector Z, and may control the next switching step based on the event occurrence confidence. For example, when the event occurrence confidence is greater than a reference value, the switching control circuit 13_2 may provide the potential vector Z to the decoder 13_1. When the event occurrence confidence is less than or equal to the reference value, the switching control circuit 13_2 may not provide the potential vector Z to the decoder 13_1, and may resend the previously sent measurement configuration to the terminal 100.

[0111] In an embodiment, the decoder 13_1 may obtain a measurement result set MR_G' including a plurality of measurement results MR_1' to MR_n' by decoding the latent vector Z provided from the switching control circuit 13_2. The switching control circuit 13_2 may control the next switching step by using the measurement result set MR_G'.

[0112] Fig.9A is a flowchart showing a handover process in a communication system according to an embodiment, and Fig. 9B It shows Fig.9A Flow chart of a detailed embodiment of operation S260. Fig.9A and Fig. 9B Assume an embodiment in which measurement reporting is performed according to the occurrence of an "A3" event.

[0113] refer to Fig.9A In operation S200, the terminal 100 may measure a signal of the serving cell 10 and a signal of the candidate target cell based on the measurement configuration.

[0114] In operation S210, the terminal 100 may determine that an 'A3' event has occurred based on at least one measurement result generated in operation S200.

[0115] In operation S220, the terminal 100 may generate a latent vector by encoding the measurement result generated in operation S200. For example, the terminal 100 may input the measurement result into a first neural network model corresponding to an encoder trained based on an autoencoder, and may obtain an output from the first neural network model. The output may be a latent vector.

[0116] In operation S230, the terminal 100 may determine whether the "A3" event occurrence reliability is greater than a reference value. For example, the terminal 100 may generate the "A3" event occurrence reliability based on the latent vector generated in operation S220 and compare the "A3" event occurrence reliability with the reference value.

[0117] In operation S240 , the terminal 100 may transmit a measurement report including the potential vector generated in operation S220 to the serving cell 10 .

[0118] In operation S250, the serving cell 10 may obtain a measurement result by decoding the latent vector. For example, the serving cell 10 may input the latent vector into a second neural network model corresponding to a decoder trained based on an autoencoder, and may obtain an output from the second neural network model. The output may be a measurement result.

[0119] In operation S260, the serving cell 10 may determine to switch to the target cell based on the measurement result obtained in operation S250. For example, the serving cell 10 may recognize that the terminal 100 supports the measurement report determination function and may trust the measurement report received from the terminal 100. Therefore, in response to receiving the measurement report, the serving cell 10 may determine that the handover is performed due to the occurrence of the "A3" event and may determine the target cell by referring to the measurement result.

[0120] Further references Fig. 9B , operation S260 may include operations S261 to S264.

[0121] In operation S261 , the serving cell 10 may refer to the capability information of the terminal 100 .

[0122] In operation S262, the serving cell 10 may determine whether the terminal 100 supports a measurement report determination function based on the capability information.

[0123] When the answer is "yes" in operation S262, the serving cell 10 may skip a separate analysis operation on the measurement result in operation S263. The serving cell 10 may determine handover according to the occurrence of the "A3" event based only on the reception of the measurement report.

[0124] When the answer is "No" in operation S262, the serving cell 10 may perform an analysis operation on the measurement result in operation S264. The serving cell 10 may determine handover according to the occurrence of the "A3" event based on the analysis result.

[0125] Fig.10 is a flowchart illustrating a method performed by a terminal according to an embodiment.

[0126] refer to Fig.10 , in operation S300, the terminal may receive a measurement configuration from a serving cell.

[0127] In operation S310, the terminal may determine an event matching a measurement configuration.

[0128] In operation S320, the terminal may determine whether the event determined in operation S310 is a target event. Herein, the target event may refer to an event preset to perform an autoencoder-based measurement report among multiple handover-related events. For example, the target event may correspond to an "A3" event among multiple handover-related events. However, this is merely an example, and therefore, the embodiment is not limited thereto, and multiple handover-related events may all correspond to the target event.

[0129] When the answer is "yes" in operation S320, in operation S330, the terminal may perform an autoencoder-based measurement report. Herein, the autoencoder-based measurement report may include at least one of an operation of generating a latent vector by using a neural network model trained based on the autoencoder and an operation of determining whether to perform a measurement report based on the latent vector.

[0130] When the answer is "No" in operation S320, in operation S340, the terminal may perform general measurement reporting.

[0131] Fig.11 is a flowchart illustrating a handover process in a communication system according to an embodiment.

[0132] refer to Fig.11 , in operation S400, the terminal 100 in the RRC connection state with the serving cell 10 may receive a network service from the serving cell 10 by operating in a connected mode.

[0133] In operation S410 , the serving cell 10 may transmit a measurement configuration to the terminal 100 .

[0134] In operation S420, the terminal 100 may perform an auto-encoder-based measurement operation.

[0135] In operation S430 , the terminal 100 may transmit a measurement report including a potential vector to the serving cell 10 .

[0136] In operation S440 , the serving cell 10 may determine to switch to the target cell 20 based on the received measurement report.

[0137] In operation S450 , the serving cell 10 may send a handover request to the target cell 20 .

[0138] In operation S460 , the target cell 20 may control permission (or approval) for handover.

[0139] In operation S470 , the target cell 20 may send a handover response to the serving cell 10 .

[0140] In operation S480, the serving cell 10 may send a handover command to the terminal 100. In response to the handover command, the terminal 100 may suspend existing communication and may release the connection with the serving cell 10.

[0141] In operation S491, the terminal 100 may transmit a random access preamble to the target cell 20 through a random access channel (RACH).

[0142] In operation S492, when the random access preamble is properly received, the target cell 20 may transmit a random access response including a Timing Advance (TA) command, which is information for synchronization, to the terminal 100.

[0143] In operation S493, the terminal 100 may transmit a handover confirmation indicating that the handover has been completed based on the uplink resource allocation information specified in the random access response to the target cell 20, thereby completing the handover procedure.

[0144] Fig.12 is a flowchart showing a switching operation in a communication system according to an embodiment.

[0145] refer to Fig.12 In operation S500, the terminal 800 may send capability information to the serving cell 810, the capability information including first information indicating whether the terminal 800 supports measurement reporting based on an autoencoder. In some embodiments, the capability information may also include second information indicating whether the terminal 800 supports a measurement report determination function using a potential vector.

[0146] In operation S510 , the serving cell 810 may control handover to the terminal 800 based on the capability information received in operation S500 .

[0147] Fig.13A and Fig. 13B is a flowchart illustrating a method for generating a neural network model based on an autoencoder in a communication system according to an embodiment.

[0148] refer to Fig.13A In operation S600, the terminal 900 may generate a neural network model for a measurement report based on an auto encoder. For example, during mass production of the terminal 900, the terminal 900 may learn based on the auto encoder training measurement result to generate a first neural network model corresponding to the encoder and a second neural network model corresponding to the decoder. The first neural network model and the second neural network model may be stored in a certain memory of the terminal 900.

[0149] In operation S610, the terminal 900 may send a second neural network model required to decode a latent vector to the serving cell 910.

[0150] In operation S620, the serving cell 810 may store the second neural network model received in operation S610 in a certain memory.

[0151] In operation S630, the terminal 900 and the serving cell 810 may perform a handover process using a neural network model. For example, the terminal 900 may generate a latent vector from the measurement result using a first neural network model and send the latent vector to the serving cell 810, and the serving cell 810 may obtain the measurement result from the latent vector using a second neural network model.

[0152] Further references Fig. 13B In operation S700, the terminal 900 may generate a first neural network model for autoencoder-based measurement reporting.

[0153] In operation S710, the serving cell 910 may generate a second neural network model for autoencoder-based measurement reporting. Operation S710 may be performed independently of operation S700.

[0154] In operation S720, the terminal 900 and the serving cell 910 may perform a handover procedure using a neural network model including a first neural network model and a second neural network model.

[0155] Fig.14 is a block diagram showing an electronic device 1000 according to an embodiment.

[0156] refer to Fig.14 , the electronic device 1000 may include a memory 1010, a processor unit 1020, an input / output controller 1040, a display 1050, an input device 1060, and a communication processor 1090. Here, a plurality of memories 1010 may be provided. Each component will be described as follows.

[0157] The memory 1010 may include a program storage device 1011 storing a program for controlling the operation of the electronic device 1000 and a data storage device 1012 storing data generated during the operation of the program. The data storage device 1012 may store data required for the operation of the application program 1013 and the switching program 1014. In an embodiment, the data storage device 1012 according to the embodiment may store a neural network model NN trained based on an autoencoder. When the switching program 1014 is executed, the neural network model NN may be used.

[0158] The program storage device 1011 may include an application program 1013 and a switching program 1014. Here, the program included in the program storage device 1011 may be a set of instructions and may be expressed as an instruction set. The application program 1013 may include program codes for running various applications working in the electronic device 1000. That is, the application program 1013 may include codes (or commands) related to various applications driven by the processor 1022. According to an embodiment, the switching program 1014 may include a control code for performing a measurement operation for switching.

[0159] In an embodiment, by executing the handover procedure 1014, the processor 1022 may encode the measurement result to generate a potential vector. In addition, by executing the handover procedure 1014, the processor 1022 may determine whether to perform a measurement report to the serving cell based on the potential vector.

[0160] The electronic device 1000 may include a communication processor 1090, which may perform communication functions for voice communication and data communication. The processor 1022 may receive an RRC reconfiguration message from a source cell through the communication processor 1090, may perform a measurement operation on an object in response to the received RRC reconfiguration message, and may report a measurement result of a specific object that meets a certain condition to the source cell through the communication processor 1090.

[0161] The peripheral device interface 1023 may control the connection between the input / output controller 1040, the communication processor 1090, the processor 1022, and the memory interface 1021. The processor 1022 uses at least one software program to control multiple cells to provide corresponding services. In this case, the processor 1022 may run at least one program stored in the memory 1010 to provide a service corresponding to the program.

[0162] The input / output controller 1040 may provide an interface between input / output devices such as a display 1050 and an input device 1060 and the peripheral device interface 1023. The display 1050 displays status information, input characters, moving pictures, still pictures, etc. For example, the display 1050 may display information about an application program driven by the processor 1022.

[0163] The input device 1060 may provide input data generated by selection of the electronic device 1000 to the processor unit 1020 through the input / output controller 1040. In this case, the input device 1060 may include a keyboard having at least one hardware button, a touch pad for sensing touch information, etc. For example, the input device 1060 may provide touch information such as a touch, a touch movement, or a touch release sensed through the touch pad to the processor 1022 through the input / output controller 1040.

[0164] Fig.15 is a diagram illustrating a communication device for performing a handover operation according to an embodiment.

[0165] refer to Fig.15 , the home gadget 2100, the home appliance 2120, the entertainment device 2140, and the AP 2200 may perform a switching operation according to an embodiment. A plurality of APs 2200 may be located at various points, respectively, and based on the measurement operation according to an embodiment, the home gadget 2100, the home appliance 2120, and the entertainment device 2140 may be switched from any one of the plurality of APs 2200 to another AP 2200.

[0166] In some embodiments, the home gadgets 2100 , the home appliances 2120 , the entertainment devices 2140 , and the AP 2200 may constitute an Internet of Things (IoT) network system. Fig.15 The communication devices shown are examples only, and it should be understood that other communication devices may be used in Fig.15 Other communication devices not shown in the figure include the wireless communication device according to the embodiment.

[0167] Various changes in form and details may be made therein without departing from the spirit and scope of the appended claims.

Claims

1. A method performed by a terminal, the method comprising: receiving a measurement configuration from a serving cell; measuring at least one received signal based on the measurement configuration; determining whether an event corresponding to the measurement configuration has occurred; generating a latent vector by encoding a measurement set comprising a plurality of measurements based on determining that the event has occurred; as well as In response to the occurrence of the event, a measurement report generated based on the potential vector is sent to the serving cell.

2. The method according to claim 1, wherein: Generating the latent vector includes: inputting the set of measurements into a first neural network model corresponding to an encoder trained based on an autoencoder; and An output of the first neural network model is obtained as the latent vector.

3. The method according to claim 2, wherein: Based on a second neural network model corresponding to a decoder trained based on the autoencoder being stored in the serving cell, the measurement report includes the latent vector.

4. The method according to claim 3, further comprising: generating the second neural network model; as well as Sending the second neural network model to the serving cell. 5 . The method according to claim 3 , further comprising sending capability information to the serving cell, the capability information comprising first information indicating support for a measurement reporting function based on the auto-encoder.

6. The method according to claim 1, wherein: Sending the measurement report to the serving cell includes: generating an event occurrence confidence value based on the potential vector; comparing the event occurrence confidence value with a reference value; and Based on a result of the comparison, it is determined whether to send the measurement report to the serving cell.

7. The method according to claim 6, wherein: The plurality of measurement results include received signal received power, timing advance, and precoding matrix indication corresponding to the serving cell and candidate target cells for handover. 8 . The method according to claim 6 , further comprising sending capability information to the serving cell, the capability information comprising second information indicating support for a measurement report determination function.

9. The method according to claim 6, wherein: Determining whether to send the measurement report to the serving cell includes: determining to send the measurement report to the serving cell based on the event occurrence confidence value being greater than the reference value.

10. The method according to claim 9, wherein: Sending the measurement report to the serving cell further includes: Based on the fact that a second neural network model corresponding to a decoder trained based on an autoencoder is not stored in the serving cell, inputting the latent vector to the second neural network model; and The measurement report including the output of the second neural network model is sent to the serving cell.

11. A method performed by a serving cell, the method comprising: Sending measurement configuration to the terminal; receiving, from the terminal, a measurement report corresponding to the measurement configuration and based on an autoencoder; Obtaining a measurement result set including a plurality of measurement results by decoding a latent vector included in the measurement report; as well as The switching operation is controlled based on the set of measurement results.

12. The method according to claim 11, wherein: The latent vector is generated based on a first neural network model corresponding to an encoder trained based on the autoencoder, and Obtaining the measurement result set includes: inputting the latent vector into a second neural network model corresponding to a decoder trained based on the autoencoder; and The output of the second neural network model is obtained as the measurement result set.

13. The method according to claim 12, wherein: The first neural network model is generated in the terminal, and The second neural network model is generated in the serving cell. 14 . The method according to claim 11 , further comprising receiving capability information from the terminal, the capability information comprising first information indicating support for a measurement reporting function based on the auto-encoder.

15. The method according to claim 14, wherein: The capability information further includes second information indicating support for a measurement report determination function, and Controlling the switching operation is further performed based on the second information.

16. The method according to claim 15, wherein: The measurement configuration includes configuration related to the "A3" event, and The control of the switching operation includes: Determine switching to a target cell based on the second information; and The target cell is determined based on the measurement result set.

17. The method according to claim 11, wherein: The plurality of measurement results include received signal received power, timing advance, and precoding matrix indication corresponding to the serving cell and candidate target cells for handover.

18. A terminal, comprising: A memory configured to store a neural network model trained based on an autoencoder; as well as at least one processor, the at least one processor being configured to: generating a latent vector from a measurement set including a plurality of measurements for a handover by using the neural network model; and A determination is made based on the potential vector whether to send a measurement report to a serving cell.

19. The terminal according to claim 18, wherein: The at least one processor is further configured to: generating an event occurrence confidence of an event related to the measurement report based on the latent vector; as well as Based on whether the event occurrence confidence is greater than a reference value, determining whether to send the measurement report to the serving cell.

20. The terminal according to claim 18, wherein: The measurement report includes the latent vector.

Images