Method, communication device and infrastructure equipment

Abstract

A method of transmitting data via a wireless communication network by a communication device is provided. The method comprises: receiving, at a transmission buffer, data for transmission via a wireless access interface of a wireless communication network; determining an initial buffer size required for transmitting the data based on the data volume in the transmission buffer; reporting an initial buffer status to the infrastructure equipment based on an absolute value of the initial buffer size; determining an updated buffer size required for data transmission based on a change in the amount of data in the transmission buffer; and reporting the updated buffer status to the infrastructure equipment based on a difference between the updated buffer size and the initial buffer size.

Description

Technical Field

[0001] The present disclosure relates to a communication apparatus, infrastructure equipment, and method for transmitting and receiving data via a wireless communication network and for reporting buffer status. This application claims the Paris Convention priority of European Patent Application No. 23156160.6, filed on February 10, 2023, the entire contents of which are incorporated herein by reference. Background Art

[0002] The "background" description provided herein is for the purpose of generally presenting the context of the present disclosure. To the extent described in this background section, the work of the presently designated inventors and aspects of the description that may not qualify as prior art at the time of filing are neither explicitly nor implicitly admitted as prior art with respect to the present invention.

[0003] The wireless communication systems defined by 3GPP are capable of supporting more complex services than the simple voice and messaging services provided by previous generations of mobile telecommunication systems. For example, using the improved radio interface and enhanced data rates provided by UMTS and Long Term Evolution (LTE) systems, users are able to enjoy high data rate applications such as mobile video streaming and mobile video conferencing, which were previously only available via fixed-line data connections. Consequently, the demand for deploying such networks is strong, and the coverage of these networks, i.e., the geographic locations where access to the network is possible, is expected to increase even more rapidly.

[0004] It is expected that future wireless communication networks will routinely and efficiently support communications with a wider range of devices associated with a wider range of data traffic profiles and types than current systems are optimized to support. For example, it is expected that future wireless communication networks will efficiently support communications with devices including reduced complexity devices, machine type communication (MTC) devices, high-resolution video displays, virtual reality headsets, etc. Some of these different types of devices may be deployed in very large numbers, such as low-complexity devices used to support the "Internet of Things", and may generally be associated with the transmission of relatively small amounts of data with relatively high delay tolerance.

[0005] In view of this, it is expected that there is a desire for future wireless communication networks (e.g., wireless communication networks that may be referred to as 5G or New Radio (NR) systems / New Radio Access Technology (RAT) systems [1]) and future iterations / releases of existing systems to effectively support buffer status reporting for scheduling uplink data transmission in the wireless communication networks. Summary of the Invention

[0006] The present disclosure may help solve or alleviate at least some of the above-mentioned problems.

[0007] Embodiments of the present technology may provide a method for transmitting data via a wireless communication network by a communication device. The method includes: receiving data for transmission via a wireless access interface of the wireless communication network at a transmission buffer; determining an initial buffer size required for transmitting the data based on an amount of data in the transmission buffer; reporting an initial buffer status to an infrastructure device based on an absolute value of the initial buffer size; determining an updated buffer size required for data transmission based on a change in the amount of data in the transmission buffer; and reporting an updated buffer status to the infrastructure device based on a difference between the updated buffer size and the initial buffer size.

[0008] An exemplary embodiment may also provide a method for receiving data via a wireless communication network through an infrastructure device, the method comprising: receiving an initial buffer status from a communication device; determining an initial buffer size for receiving data based on the initial buffer status; receiving an updated buffer status from the communication device; determining a difference between the updated buffer size and the initial buffer size based on the updated buffer status; and determining an updated buffer size for data reception by adding the difference to the initial buffer size.

[0009] 5G Extended Reality (XR) services are a combination of 5G network technologies and Extended Reality (XR) technologies such as Virtual Reality (VR), Augmented Reality (AR), and Mixed Reality (MR). Buffer status reporting is conventionally performed based on a Buffer Status (BS) table that contains a mapping of the buffer size level calculated by the data volume calculation process according to TS 38.322 [2] and TS 38.323 [3] to the buffer size field of the Buffer Status Report (BSR). Given the need for better scheduling of UEs using XR services, enhanced buffer status reporting is required to provide increased accuracy for compression and prediction of buffer size.

[0010] Embodiments of the present technology may provide enhanced buffer status reporting with reduced quantization error and provide increased accuracy for compression and prediction of buffer sizes. Thus, capacity gains can be enhanced and better scheduling of UEs can be achieved.

[0011] Various aspects and features of the present disclosure are defined in the following claims.

[0012] It should be understood that the foregoing general description and the following detailed description are exemplary rather than restrictive of the present technology. The described embodiments and other advantages will be best understood by reference to the following detailed description taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] A more complete appreciation of the present disclosure and many of its attendant advantages will be readily obtained as the present disclosure becomes better understood by reference to the following detailed description when considered in conjunction with the accompanying drawings, wherein like reference numerals designate identical or corresponding parts throughout the several views, and:

[0014] Figure 1 schematically illustrates some aspects of an LTE-type wireless communication system that may be configured to operate in accordance with certain embodiments of the present disclosure;

[0015] Figure 2 schematically illustrates some aspects of a new radio access technology (RAT) wireless communication system that may be configured to operate in accordance with certain embodiments of the present disclosure;

[0016] Figure 3 Schematic block diagram illustrating entities within a communication device and infrastructure equipment that may be configured to operate according to exemplary embodiments of the present technology;

[0017] Figure 4 A flowchart illustrating a process performed by a communication device according to an exemplary embodiment of the present technology;

[0018] Figure 5 A flowchart illustrating a process performed by an infrastructure device according to an exemplary embodiment of the present technology;

[0019] Figure 6 is a schematic block diagram illustrating modeling entities within a communication device adapted according to exemplary embodiments of the present technology;

[0020] Figure 7 is a schematic block diagram illustrating modeling entities within infrastructure equipment suitable for operating in accordance with exemplary embodiments of the present technology;

[0021] Figure 8 is a schematic block diagram illustrating modeling entities within a communication device and infrastructure equipment adapted according to exemplary embodiments of the present technology; and

[0022] Figure 9 A flow chart illustrating a process performed by modeling entities within a communication device and infrastructure equipment adapted according to exemplary embodiments of the present technology. DETAILED DESCRIPTION

[0023] Advanced long-term evolution radio access technology (4G)

[0024] Figure 1 A schematic diagram is provided illustrating some basic functionality of a mobile telecommunications network / system 100 that typically operates according to LTE principles, but which may also support other radio access technologies, and which may be suitable for implementing embodiments of the present disclosure as described herein. Figure 1Certain aspects of the various elements of the telecommunications network and their corresponding modes of operation are well known and defined in relevant standards managed by the 3GPP (RTM) body and are also described in a number of books on the subject (e.g., Holma H. ​​and Toskala A. [4]). It will be appreciated that operational aspects of the telecommunications network discussed herein that are not specifically described (e.g., relating to specific communication protocols and physical channels used for communication between the different elements) may be implemented in accordance with any known techniques, e.g., in accordance with the relevant standards and known recommendations for modifications and additions to the relevant standards.

[0025] Network 100 includes multiple base stations 101 connected to a core network portion 102. Each base station provides coverage 103 (e.g., a cell) within which data can be transmitted to and from communication devices 104. Data is transmitted from a base station 101 to a communication device 104 within its respective coverage area 103 via a radio downlink. Data is transmitted from a communication device 104 to a base station 101 via a radio uplink. Core network portion 102 routes data to and from communication devices 104 via the respective base stations 101 and provides functions such as authentication, mobility management, and billing. A communication device may also be referred to as a mobile station, user equipment (UE), user terminal, mobile radio, terminal device, etc. A base station, as an example of network infrastructure equipment / network access node, may also be referred to as a transceiver station / nodeB / e-nodeB, g-nodeB (gNB), etc. In this regard, different terms are often associated with different generations of wireless telecommunication systems for elements providing broadly comparable functionality. However, as explained below, exemplary embodiments of the present disclosure may be implemented equally across different generations of wireless telecommunication systems (such as 5G or New Radio), and for simplicity, certain terminology may be used regardless of the underlying network architecture. That is, the use of particular terminology associated with certain exemplary implementations is not intended to indicate that these implementations are limited to the particular generation of networks to which that particular terminology may be most associated.

[0026] New wireless access technology (5G)

[0027] Figure 2 is a schematic diagram illustrating a network architecture of a new RAT wireless communication network / system 200 based on previously proposed methods, which may also be adapted to provide functionality according to embodiments of the present disclosure described herein. Figure 2The new RAT network 200 shown in the figure comprises a first communication cell 201 and a second communication cell 202. Each communication cell 201, 202 comprises a control node (centralized unit) 221, 222 that communicates with a core network component 210 via respective wired or wireless links 251, 252. The respective control nodes 221, 222 also each communicate with a plurality of distributed units (radio access nodes / remote transmission and reception points (TRP)) 211, 212 in their respective cells. Again, these communications may be via respective wired or wireless links. The distributed units 211, 212 are responsible for providing a wireless access interface for communication devices connected to the network. Each distributed unit 211, 212 has a coverage area (radio access coverage area) 241, 242, wherein the sum of the coverage areas of the distributed units under the control of the control node together defines the coverage of the respective communication cells 201, 202. Each distributed unit 211 , 212 includes a transceiver circuit for transmitting and receiving wireless signals and a processor circuit configured to control the respective distributed unit 211 , 212 .

[0028] In terms of a wide range of top features, Figure 2 The core network component 210 of the new RAT communication network shown in FIG can be broadly viewed as being similar to the core network component 210 of the new RAT communication network shown in FIG. Figure 1 The core network 102 shown in FIG. 1 corresponds to the core network 102 shown in FIG. 1 , and the corresponding control nodes 221, 222 and their associated distributed units / TRPs 211, 212 can be broadly viewed as providing Figure 1 The term network infrastructure equipment / access node may be used to encompass these elements of a wireless communication system as well as more conventional base station-type elements. Depending on the application being implemented, the responsibility for scheduling transmissions over the radio interface between the various distributed units and the communication devices may lie with the control node / centralized unit and / or the distributed units / TRP.

[0029] Figure 2 It is shown that a communication device or UE 260 is within the coverage of a first communication cell 201. Therefore, the communication device 260 can exchange signaling with the first control node 221 in the first communication cell via one of the distributed units 211 associated with the first communication cell 201. In some cases, communications of a given communication device are routed through only one of the distributed units, but it should be understood that in some other implementations, such as in soft handover scenarios and other scenarios, communications associated with a given communication device may be routed through more than one distributed unit.

[0030] exist Figure 2In the example, two communication units 201, 202 and one communication device 260 are shown for simplicity, but it should of course be recognized that in practice the system may include a large number of communication units (all supported by corresponding control nodes and multiple distributed units) serving a large number of communication devices.

[0031] It should also be understood that Figure 2 It only represents one example of an architecture for a proposed new RAT communication system, wherein the method according to the principles described herein may be employed, and the functions disclosed herein may also be applied to wireless communication systems with different architectures.

[0032] Thus, example embodiments of the present disclosure as discussed herein may be implemented according to various different architectures, such as Figure 1 and Figure 2 ) are implemented in a wireless telecommunications system / network. Therefore, it should be recognized that the specific wireless communication architecture in any given implementation is not of primary significance to the principles described herein. In this regard, certain embodiments of the present disclosure may be generally described in the context of communications between network infrastructure equipment / access nodes and communication devices, where the specific characteristics of the network infrastructure equipment / access nodes and communication devices will depend on the network infrastructure equipment being implemented. For example, in some cases, the network infrastructure equipment / access nodes may include base stations adapted to provide functionality according to the principles described herein, such as Figure 1 The LTE-type base station 101 shown in FIG, and in other examples, the network infrastructure equipment / access node may include a device adapted to provide functionality according to the principles described herein. Figure 2 Control units / control nodes 221, 222 and / or TRPs 211, 212 of the kind shown in .

[0033] In a 5G network, the CU 221 in combination with one or more DUs 213, 216 and one or more TRPs 211, 212 may form a base station or gNB 301 of the radio network portion of a 5G radio access network (RAN). Figure 3 In FIG, the gNB 301 formed by one or more TRPs 211, 212, one or more DUs 213, 216, and CU 221 can be represented in a simplified form as including a transmitter circuit 302, a receiver circuit 303, an antenna 304, and a controller circuit or control processor 305, which is operable to control the transmitter 302 and the wireless receiver 303 to transmit radio signals to and receive radio signals from one or more UEs 311 within a cell 320. The transmitter circuit 302 and the receiver circuit 303 can be implemented together to form a wireless transceiver 306. Figure 3As shown, exemplary UE 301 is shown to include corresponding receiver circuitry 313, transmitter circuitry 312, an antenna, and controller circuitry 315. Transmitter circuitry 312 and receiver circuitry 313 may be implemented together to form a wireless transceiver 316. Controller circuitry 315 is configured to control transmitter circuitry 312 to transmit signals representing uplink data to a wireless communication network via a radio access interface formed by gNB 301, as indicated by arrow 330. Controller circuitry or control processor 315 is also configured to control receiver circuitry 313 to receive downlink data, as signals transmitted by transmitter 302, indicated by arrow 331, and received by receiver 313 in accordance with conventional operation.

[0034] The transmitter circuits 302, 312 and the receiver circuits 303, 313 (as well as other transmitters, receivers, and transceivers described in accordance with the examples and embodiments of the present disclosure) may include radio frequency filters and amplifiers as well as signal processing components and devices to transmit and receive radio signals, for example, in accordance with the 5G / NR standard. The controller circuits 305, 315 (as well as other controllers described in accordance with the examples and embodiments of the present disclosure) may be, for example, a microprocessor, a CPU, or a dedicated chipset configured to execute instructions stored on a computer-readable medium, such as a non-volatile memory. The processing steps described herein may be performed, for example, by a microprocessor in combination with a random access memory, operating in accordance with instructions stored on a computer-readable medium. For ease of representation, the transmitter, receiver, and controller are shown in FIG. Figure 3 Schematically shown as separate elements in the figures. However, it will be appreciated that the functionality of these elements may be provided in a variety of different ways, for example, using one or more appropriately programmed programmable computers, or one or more appropriately configured application specific integrated circuits / circuits / chips / chip sets. As will be appreciated, infrastructure equipment / TRPs / base stations and UEs / communication devices will typically include various other elements associated with their operational functionality.

[0035] Buffer Status Report

[0036] like Figure 3As shown, when UE 311 transmits its uplink data in its transmit buffer 317 using a PUSCH scheduled by an uplink grant from gNB 301, it may be configured to transmit a buffer status report (BSR) in the scheduled PUSCH to indicate the size of the data in transmit buffer 317 to gNB 301 so that gNB 301 can schedule additional PUSCHs for UE 311. gNB 301 may be configured to determine the size of receive buffer 307 based on the BSR received from UE 311. In some embodiments, transmit buffer 317 in UE 311 may be an RLC transmit buffer for temporarily storing RLC layer uplink data to be transmitted to gNB 301.

[0037] TR 38.835 [5] discloses an enhanced BS reporting scheme for sending BSR. For example, the enhanced BS reporting scheme may support traditional dynamic scheduling with traditional BSR. In another example, the enhanced BS reporting scheme may support accurate buffer sizes and a new buffer status table (BS table) with finer granularity. In another example, the enhanced BS reporting scheme may provide an XR-specific BS reporting mechanism to minimize scheduling delays. TR 38.835 also describes the improved capacity performance achieved by the enhanced BS reporting scheme compared to traditional BSR. It is concluded that BSR enhancements may include at least a new BS table and delayed reporting of buffered data in the uplink.

[0038] It should be understood that the new BS report can support:

[0039] · Informs precise data rate and reduces quantization errors

[0040] The latency or lifetime of the reported data

[0041] Increased frequency of reporting BSRs

[0042] While the UE sends a BSR based on the existing known framework as explained in the MAC specification TS 38.321 [6], the XR specific changes to the BSR will require reporting of buffer sizes with finer granularity, with a higher number of bits, a higher reporting frequency and with new information such as the time to live. The time to live represents the time that an application consuming a communication service can continue without an expected message according to TS 22.261 [7]. In traditional reporting, BSR reporting can be periodic or event triggered, whereas in XR services the new BSR reporting can occur more frequently, for example, it can occur every scheduling period. Embodiments of the present technology are desirable for an increased frequency of BSR reporting as it allows signaling of both absolute and incremental values ​​(delta values) and provides the benefits of reducing quantization errors and increasing the accuracy of compression and prediction of buffer sizes.

[0043] Figure 4 A flowchart illustrating a process performed by a communication device (UE) 311 according to an embodiment of the present technology is shown.

[0044] The process begins at step S402, where UE 311 identifies data to be transmitted. For example, the amount of data in a transmission buffer is obtained. Conventionally, the transmission buffer forms part of the radio link control (RLC) layer, where the RLC layer uses the amount of data in the transmission buffer to identify uplink resources that need to be allocated to UE 311 to transmit uplink data.

[0045] The process continues to step S404, where the buffer size to be reported in the BSR is determined based on the amount of data to be transmitted in the RLC transmit buffer. For example, the data amount may be the buffer size. In some embodiments, UE 311 may report the actual value of the buffer size to gNB 301. In some embodiments, UE 311 may report a predicted value of the buffer size to gNB 301, in which case gNB 301 may use the predicted value to reserve resources for the near future.

[0046] In step S406, a determination is made as to whether UE 311 is transmitting a BSR to gNB 301 for the first time using the PUSCH scheduled by gNB 301. A BSR is transmitted to gNB 301 using a MAC Control Element (MAC-CE), and a BS report can be triggered, for example, by the arrival of new data on a logical channel with a higher priority than previously stored in the buffer. In this case, UE 311 will transmit a regular BSR. If the number of padding bits during normal PUSCH transmission leaves sufficient space for transmitting a BSR, UE 311 will transmit a padding BSR. BSR transmission can also be performed at regular intervals during uplink data transmission, and UE 311 will transmit a periodic BSR. If this is the first time UE 311 is transmitting a BSR using the scheduled PUSCH, control proceeds to step S408; otherwise, control transfers to step S410.

[0047] In step S408, UE 311 formats a first BSR based on the absolute value of the buffer size determined in step S404.

[0048] At step S410, UE 311 sends a delta value representing the buffer size compared to the absolute value for subsequent BSRs. For example, if UE 311 reports a buffer size of 10MB in the first BSR, subsequent BSRs will report the change in buffer status relative to 10MB. For example, if an 8-bit buffer size field is used and the buffer status now becomes 8MB, then 8 bits in the BSR MAC-CE are used to represent a finer delta value, i.e., a -2MB change relative to 10MB.

[0049] In some embodiments, each logical channel transmits a buffer size change so that if the buffer contains high-priority data, this high-priority data is indicated, allowing for rapid clearing of high-priority buffered data. In this case, the reference for the absolute value can be the last absolute value of the reported buffer size or a new reference value such as "0." Because the same number of bits are used to represent the incremental value, quantization errors in the reported buffer size are reduced.

[0050] In step S412, UE 311 transmits data and a BSR as a MAC-CE to gNB 301. The BSR may contain the absolute value or incremental value of the buffer in a known format.

[0051] In some embodiments, XR requires frequent buffer status reporting, and in some cases, buffer status reporting is performed in every scheduling cycle, and the buffer does not disappear from one instance of the schedule.

[0052] Figure 5 A flowchart of a process performed by an infrastructure device (gNB) 301 according to an embodiment of the present technology is shown.

[0053] The process starts at step S502, where gNB 301 receives data and BSR from UE 311.

[0054] In step S504, it is determined whether gNB 301 has received a BSR in the scheduled PUSCH for the first time. If so, control continues to step S506, otherwise control passes to step S508.

[0055] As step S506, gNB 301 regards the BS field value in the BSR as the absolute value of the transmission buffer size in UE 311 and determines its reception buffer size accordingly.

[0056] In step S508, the gNB 301 considers the BS field value in the BSR as the incremental value of the UE's transmit buffer size and calculates the corresponding absolute value by adding the incremental value to the reference buffer size obtained from the first BSR received in the scheduled PUSCH. The gNB 301 then determines its receive buffer size based on the calculated absolute value.

[0057] index BS value …… …… 46 ≤181 47 ≤193 …… …… 243 ≤47087187 244 ≤46182206 …… ……

[0058] Table 1

[0059] Table 1 shows an exemplary mapping of the buffer size level (BS value) to the buffer size field (8-bit index). As shown, the step size increases significantly with the BS value. For example, when the buffer size is between 181 bytes and 193 bytes, the step size is 12 bytes. When the buffer size reaches the range of 47,087,187 bytes to 46,182,206 bytes, the step size increases to 2,815,019 bytes.

[0060] To represent the buffer size with higher accuracy and smaller step sizes, the number of bits required would have to be increased from the current 5 or 8 bits in TR 38.321 to, for example, 12, 16, or more bits. However, the BSR overhead would also increase due to the additional bits. If the BSR were sent at the beginning of each scheduling period, the cumulative overhead would be very large. By sending incremental values ​​of the buffer size rather than absolute values ​​in the BSR, as described in embodiments of the present technology, the granularity of the reported buffer size can be significantly improved, and the buffer size can be communicated from UE 311 to gNB 301 with greater accuracy.

[0061] In some embodiments, AI / ML is used for compression and decompression of the BSR to represent higher accuracy in the reporting by still using 8-bit traditional reporting, as will be further described below.

[0062] AI / Machine Learning Enhances BSR

[0063] Figure 6 is a schematic block diagram illustrating modeling entities within a communication device (UE) 611 adapted according to exemplary embodiments of the present technology.

[0064] exist Figure 66, an example UE 611 is shown as including corresponding receiver circuitry 313, transmitter circuitry 612, antenna 614, controller circuitry 615, and artificial intelligence (AI) / machine learning (ML) model 618. Transmitter circuitry 612 and receiver circuitry 613 may be implemented together to form a wireless transceiver 616. Controller circuitry 615 is configured to control transmitter circuitry 612 to transmit signals representing uplink data to a wireless communication network via a radio access interface formed by gNB 601, as indicated by arrow 630. Controller circuitry 615 is also configured to control receiver circuitry 313 to receive downlink data, as signals transmitted by transmitter 602, as indicated by arrow 631, and received by receiver 613 in accordance with conventional operation. When UE 611 transmits its uplink data in transmit buffer 617 using a PUSCH scheduled by uplink grant signaling from gNB 601, it can send a BSR in the scheduled PUSCH to indicate the size of transmit buffer 617 to gNB 601 so that gNB 601 can schedule additional PUSCHs for UE 611. In some embodiments, controller circuitry 615 determines the buffer size based on a model 618 derived using machine learning techniques described below. For any allowable combination of input values, model 618 determines the buffer size for the data arriving in transmit buffer 617. In some embodiments, the value of the buffer size (BS) field, which defines a range of buffer sizes, can be determined directly. In this case, model 618 can apply a classification process to classify the input value as corresponding to exactly one of the predetermined BS field values.

[0065] In some embodiments of the present technology, the UE 611 may receive a representation of the model 618 stored in a memory (not shown) of the UE 611. Preferably, the memory is a non-volatile memory.

[0066] In some embodiments, the AI / ML BSR prediction or compression algorithm considers the following factors on the transmitter side (based on UE's information):

[0067] RSRP / RSRQ value

[0068] Channel state information (CQI and calculated SRS)

[0069] Power Headroom (PHR)

[0070] Data lifetime

[0071] Frequency of BSR reporting

[0072] Application layer status such as L4S

[0073] The size of the current resource allocation (#RB, MCS, TBS, etc.)

[0074] Based on one or more of these factors, UE 611 determines the buffer size and compresses the information by encoding it into a buffer size value based on the buffer status table. The buffer size value represents the absolute value of the buffer size when the BSR is first transmitted. For subsequent BSRs, the buffer size value represents the incremental value of the updated buffer size compared to the absolute value of the buffer size reported in the first BSR.

[0075] In some embodiments, UE 611 may use the above UE-based information as well as other factors provided by gNB 601 (gNB-based information), such as:

[0076] Cell load

[0077] Uplink interference

[0078] Application layer status such as L4S

[0079] According to an embodiment of the present technology, the UE-side AI / ML model 618 operates on the UE 611 based on the factors described above, and the output of the AI / ML model 618 can be used to predict the near-term BSR and adjust the BSR report by taking the predicted value into account.

[0080] In some embodiments, the gNB 601 can provide assistance information to the UE 611 for the UE-side model 618 to operate based on gNB-side parameters such as gNB load, congestion, and interference. The gNB 601 is transparent and will process both predicted and non-predicted values ​​in the BSR in the same manner.

[0081] In some embodiments, the UE 611 can indicate whether the buffer size value is a predicted or actual buffer size. If the BSR is based on a predicted value, the gNB 601 uses this information to reserve resources for the near future. The gNB 601 can also perform a plausibility check on the confidence level of the predicted value.

[0082] Figure 7 is a schematic block diagram illustrating modeled entities within an infrastructure equipment (gNB) 701 suitable for operating according to exemplary embodiments of the present technology.

[0083] exist Figure 7, gNB 701 is shown as including transmitter circuitry 702, receiver circuitry 703, antenna 704, artificial intelligence (AI) / machine learning (ML) model 708, and controller circuitry or control processor 305, which is operable to control transmitter 302 and wireless receiver 303 to transmit and receive radio signals to one or more UEs 711. Transmitter circuitry 702 and receiver circuitry 703 may be implemented together to form a wireless transceiver 706.

[0084] When gNB 701 receives a BSR from UE 711 on a scheduled PUSCH, it obtains information regarding the size of transmit buffer 717 in UE 711. This allows gNB 701 to allocate a receive buffer for receiving data and schedule additional PUSCHs for UE 711. In some embodiments, controller circuitry 705 determines the buffer size based on a model 708 derived using machine learning techniques described below. For any allowable combination of input values, model 708 determines the buffer size for the data to be received. In some embodiments, the value of the buffer size (BS) field may be predicted to define a range of buffer sizes. In this case, model 708 may apply a classification process to classify the input value as corresponding to exactly one of the predetermined BS field values.

[0085] In some embodiments of the present technology, the gNB 701 may receive a representation of the model 708 stored in a memory (not shown) of the gNB 701. Preferably, the memory is a non-volatile memory.

[0086] In some embodiments, the AI / ML BSR prediction or decompression algorithm may consider the following factors on the UE 711 (based on the UE's information):

[0087] RSRP / RSRQ value

[0088] Channel state information (CQI and calculated SRS)

[0089] Power Headroom (PHR)

[0090] Data lifetime

[0091] Frequency of BSR reporting

[0092] Application layer status such as L4S

[0093] The size of the current resource allocation (#RB, MCS, TBS, etc.)

[0094] Based on one or more of these factors, the gNB 701 decompresses the buffer size information by decoding the received buffer size value based on the buffer status table. The buffer size value represents the absolute value of the buffer size when the BSR is first received. For subsequent BSRs, the buffer size value represents the incremental value of the updated buffer size compared to the absolute value of the buffer size reported in the initial BSR.

[0095] In some embodiments, the decompressor in the gNB 701 may use the above-mentioned UE-based information as well as other factors about the gNB 701 (gNB-based information), such as:

[0096] Cell load

[0097] Uplink interference

[0098] Application layer status such as L4S

[0099] According to embodiments of the present technology, gNB 701 uses the above information as input and then generates a predicted BSR for the near future based on gNB-side model 708. In some embodiments, if gNB 701 is confident in the predicted value, gNB 701 may instruct UE 711 to skip BSR reporting for a certain period of time. This information or command may be sent to UE 711 via MAC-CE or PHY signaling. Upon receiving the skip command, UE 711 will skip sending BSRs for a predetermined duration.

[0100] For incoming data, if a PDU from a PDU set is dropped, the associated data from the logical channel should also be dropped. Any such data drop will have an impact on the BSR prediction on the gNB side. Therefore, in some embodiments, when a packet from a PDU set is dropped, a BSR is triggered at the next available instance of a BSR report.

[0101] Figure 8 is a schematic block diagram illustrating modeled entities in a communication device (UE) 811 and an infrastructure equipment (gNB) 801 adapted according to exemplary embodiments of the present technology.

[0102] exist Figure 88, UE 811 is shown as including corresponding receiver circuitry 813, transmitter circuitry 812, antenna 814, controller circuitry 815, and artificial intelligence (AI) model 818. Transmitter circuitry 812 and receiver circuitry 813 may be implemented together to form a wireless transceiver 816. Controller circuitry 815 is configured to control transmitter circuitry 812 to transmit signals representing uplink data to a wireless communication network via a radio access interface formed by gNB 801, as indicated by arrow 830. Controller circuitry 815 is also configured to control receiver circuitry 813 to receive downlink data, as signals transmitted by transmitter 802, as indicated by arrow 831, and received by receiver 813 in accordance with conventional operation.

[0103] The gNB 801 is shown as including transmitter circuitry 802, receiver circuitry 803, antenna 804, an AI module 808, and a controller circuit or control processor 805 operable to control the transmitter 802 and wireless receiver 803 to transmit and receive radio signals to one or more UEs 811. The transmitter circuitry 802 and receiver circuitry 803 may be implemented together to form a wireless transceiver 806.

[0104] When UE 811 transmits its uplink data in transmit buffer 817 using a PUSCH scheduled by an uplink grant from gNB 801, it can be configured to send a BSR in the scheduled PUSCH to indicate the size of transmit buffer 817 to gNB 801, allowing gNB 801 to schedule additional PUSCHs for UE 811. In some embodiments, controller circuitry 815 determines the buffer size based on a model 818 derived using machine learning techniques described below. For any allowable combination of input values, model 818 determines the buffer size for the data to be transmitted based on the data arriving in transmit buffer 817. In some embodiments, the value of the buffer size (BS) field, which defines a range of buffer sizes, can be determined directly. In this case, model 818 can apply a classification process to classify the input value as corresponding to exactly one of the predetermined BS field values.

[0105] In some embodiments of the present technology, the UE 811 may receive a representation of the model 818 stored in a memory (not shown) of the UE 811. Preferably, the memory is a non-volatile memory.

[0106] When gNB 801 receives a BSR from UE 811 on a scheduled PUSCH, it can obtain information associated with the size of transmit buffer 817 in UE 811. This allows gNB 801 to allocate a receive buffer for receiving data and schedule additional PUSCHs for UE 811. In some embodiments, controller circuitry 805 determines the buffer size based on a model 808 derived using machine learning techniques described below. For any allowable combination of input values, model 808 determines the buffer size for the data to be received. In some embodiments, the value of the buffer size (BS) field can be predicted to define a range of buffer sizes. In this case, the model can apply a classification process to classify the input value as corresponding to exactly one of the predetermined BS field values.

[0107] In some embodiments of the present technology, the gNB 801 may receive a representation of the model 808 stored in a memory (not shown) of the gNB 801. Preferably, the memory is a non-volatile memory.

[0108] In some embodiments, the AI / ML BSR prediction, compression or decompression algorithm may consider the following factors on the UE side (based on UE information):

[0109] RSRP / RSRQ value

[0110] Channel state information (CQI and calculated SRS)

[0111] Power Headroom (PHR)

[0112] Data lifetime

[0113] Frequency of BSR reporting

[0114] Application layer status such as L4S

[0115] The size of the current resource allocation (#RB, MCS, TBS, etc.)

[0116] Based on one or more of these factors, the UE 811 determines the buffer size and compresses the information by encoding the information into a buffer size value based on a buffer status table, and the gNB 801 decompresses the buffer size information by decoding the received buffer size value based on the same buffer status table.

[0117] The buffer size value represents the absolute value of the buffer size when the BSR is first sent. For subsequent BSRs, the buffer size value represents the incremental value of the updated buffer size compared to the absolute value of the buffer size reported in the first BSR.

[0118] In some embodiments, UE 811 and / or gNB 801 may use the above-mentioned UE-based information as well as other factors from the gNB side (gNB-based information), such as:

[0119] Cell load

[0120] Uplink interference

[0121] Application layer status such as L4S

[0122] According to embodiments of the present technology, the above information is used to establish key performance indicators (KPIs) or loss functions for the compressor and decompressor and train the AI ​​model. In some embodiments, once the gNB 801 is confident in the predicted BSR value, the UE 811 may be asked to skip several iterations of the BSR.

[0123] In some embodiments, UE 811 can report actual or predicted values ​​for the buffer size. If UE 811 reports a predicted value, gNB 801 should be aware that it is a prediction and how UE 811 arrived at that predicted value. In other words, the decompressor in gNB 801 needs to know the rules used by the compressor in UE 811.

[0124] In some embodiments, this may be achieved by the UE 811 indicating a confidence level for the predicted value as a percentage and based on historical values ​​of the BSR. If the actual value deviates from the predicted value, the UE 811 sends a new BSR.

[0125] For incoming data, if a PDU from a PDU set is dropped, the associated data from the logical channel should also be dropped. Any such data drop will have an impact on the BSR prediction on the gNB side. Therefore, in some embodiments, when a packet from a PDU set is dropped, a BSR is triggered at the next available instance of a BSR report.

[0126] Training AI / ML models

[0127] In some embodiments, the controllers 805 and 815 of the gNB 801 and the UE 811, respectively, include machine learning-based models 808 and 818. The machine learning can be performed separately (e.g., offline).

[0128] The resulting representation of the model may be stored in non-volatile memory on the UE 811 and / or gNB 801. In some embodiments, the representation of the model is transmitted to the communication device (UE) 811 (and in some embodiments, the infrastructure equipment (gNB) 801).

[0129] In some embodiments, the training of the machine learning model may be aimed at minimizing a loss function calculated based on the input parameter values ​​and the selected BSR table. That is, the model may be iterated over multiple different values ​​of the input parameters, and for each set of input parameter values, the loss function is evaluated against a different BSR table.

[0130] In some embodiments, the loss function may be associated with the performance gap between the predicted buffer size and the actual buffer size. For example, the loss function may be defined as E=f[PBS, ABS], where PBS represents the predicted buffer size, ABS represents the actual buffer size used by the data, and f[...] represents the loss function definition. For example, ABS may represent the buffer size occupied by successfully transmitted or received data bits. In some embodiments, the function f[...] corresponds to a mean squared error function E[..,], such that E[PBS, ABS] is defined as the average of the squared differences between PBS and ABS. In some embodiments, the loss function for compression, decompression, and prediction of BSR may be implemented based on squared generalized cosine similarity (SGCS). In some other embodiments, the loss function may be any other suitable function.

[0131] In some embodiments, the model includes multiple weights associated with units and can be trained according to the principles of known backpropagation methods. For example, initially, an output (loss function) is determined based on a set of input values ​​(forward propagation) and a test data set. Then, the partial derivative (gradient) of the loss function with respect to the weight W from the output layer unit to the input layer unit (backpropagation) is calculated. Finally, the model updates the weight W based on the gradient of the backpropagation.

[0132] In some embodiments, training generates a model for estimating buffer sizes for any input combination of BS tables and input parameter values. For example, BSTBL represents an index to a particular BS table, and I1...I N Represents the input parameter value, the model can determine the function f to estimate the expected loss E=f(BSTBL,I1...I N ). Thus, in operation, the controllers 305, 304 may evaluate the expected losses E of a plurality of different BS tables in conjunction with given input parameter values ​​and select the combination that gives the lowest loss.

[0133] In some embodiments, the model can provide classification. For example, the model can execute a function whose output is a vector, where each element of the vector represents a different BS field value of the BS table, such that for a given combination of input values, only the one element of the vector corresponding to the most significant BS field value is equal to 1, while the other elements have a value of zero. Thus, training can determine the internal weights of nodes within a conventional classification neural network.

[0134] In some cases, the performance gap between the predicted and actual buffer sizes can become larger, and the AI ​​algorithm may face difficulties in predicting the buffer size, for example, when a large error in the loss function occurs. Therefore, a fallback operation is required. In some embodiments, BSR prediction can be stopped or paused for a period of time, and the UE can send actual BSRs more frequently. In some embodiments, an effective solution is provided by actively controlling the buffer size. For example, if the prediction error becomes large, the Active Queue Management (AQM) function can be enabled in the UE buffer or the gNB scheduler buffer. In this case, if the queue length becomes larger and the risk of buffer overflow is high, some packets in the queue may be intentionally dropped. Thus, buffer overflow and / or congestion can be avoided.

[0135] Figure 9 A flow chart illustrating a process performed by a modeling entity in a communication device (UE) 811 and / or infrastructure equipment (gNB) 801 adapted according to embodiments of the present technology is shown.

[0136] Figure 9 The process begins at step S902 by determining the values ​​of one or more input parameters. These may be determined in a deterministic manner (e.g., by selecting the next value in the sequence from a predetermined range of values ​​for each respective input parameter) or may be randomly selected. The method of selection may be different for different parameters: for example, the parameters may be randomly selected or may be incrementally increased. According to embodiments of the present technology, input parameters may be employed on the UE side (based on information about the UE), for example:

[0137] RSRP / RSRQ value

[0138] Channel state information (CQI and calculated SRS)

[0139] Power Headroom (PHR)

[0140] Data lifetime

[0141] Frequency of BSR reporting

[0142] Application layer status such as L4S

[0143] The size of the current resource allocation (#RB, MCS, TBS, etc.)

[0144] In some embodiments, the input parameters may be further obtained on the gNB side (based on gNB information), for example:

[0145] Cell load

[0146] Uplink interference

[0147] Application layer status such as L4S

[0148] At step S904, a buffer status table (BS table) is selected. This can be selected randomly, based on the current version of the model, or in a deterministic manner (e.g., sequentially from a set of predetermined formats).

[0149] At step S906, a loss function corresponding to the BS table selected at step S904 and the input parameter values ​​selected at step S902 is determined. Any suitable loss function may be used.

[0150] The loss function may be determined through simulation or by obtaining data corresponding to actual data transmission.

[0151] Based on the loss function determined in step S906, the model is updated. The update can be automatic according to known machine learning techniques. For example, if the current (unupdated) model indicates that a particular format should be selected for the input parameter values ​​selected in step S902, and it is determined that the loss function determined in step S906 is lower for the BS table selected in step S904 than for the BS table currently recommended by the model, the model can be updated so that, for the input parameter values ​​selected in step S902, the BS table selected in step S904 is recommended by the updated model.

[0152] In step S910, it is determined whether more BS tables are considered for the same input parameter value. If so, control returns to step S904, otherwise control continues to step S912.

[0153] In step S912, it is determined whether additional input parameter values ​​are to be considered. If so, control returns to step S902, otherwise control passes to step S914.

[0154] In step S914 , a representation of the updated model is stored, for example on a computer-readable medium.

[0155] In step S916, the representation of the model is transmitted to the communication device and / or infrastructure equipment. The transmission in step S916 may be via a wireless access interface (such as via Figure 8 ) or via a wired interface (such as during the manufacturing process).

[0156] The representation of the model transmitted at step S916 may be a reduced representation of the model stored at step S914. For example, the model stored at step S916 may include an indication of the value of the loss function determined at step S906, while the reduced model representation transmitted at step S916 may only provide a means for determining the buffer size based on the input parameter values.

[0157] According to embodiments of this technology, a loss function is generated based on input parameters (UE-based information and / or gNB-based information) to adapt to the performance gap between the transmitter and receiver. The AI ​​model can be updated and predictions can be frequently adjusted based on the loss function.

[0158] Model ID processing

[0159] According to an embodiment of the present technology, only one AI model is configured and used during the lifetime of a connection / service. In some embodiments, different models are supported for a connection, for example, the network may want to increase the accuracy of the BSR and therefore switch from the legacy BSR to the new BSR. In this case, the model ID is configured using RRC signaling, and any switch between these models can occur via MAC or PHY signaling. Therefore, the AI / ML model can be represented by a unique model ID. Separate model IDs can be assigned to the legacy BSR with AI / ML enhancements, the new BSR table with AI / ML enhancements, the one-sided model, and the two-sided model, respectively. In this way, model switching can be communicated by sending the model ID instead of the entire configuration of the configuration / sharing model.

[0160] In some embodiments, the new BSR table implicitly configures the new model ID. In some embodiments, the BSR switch command means that the new model ID is currently being used.

[0161] UE and gNB capability handling during UE mobility

[0162] According to an embodiment of the present technology, the UE may switch from the old BSR scheme to the new BSR scheme during a handover procedure from a source gNB to a target gNB in ​​the following cases:

[0163] If the UE supports the XR-based BSR scheme and AI / ML enhancement for BSR in the source gNB, but the target gNB does not support the XR-based BSR scheme or AI / ML enhancement for BSR.

[0164] If the UE supports one AI / ML model (or AI / ML model ID) in the source gNB, but the target gNB is configured with a different AI / ML model. This can also happen within the same gNB, as different distributed units (DUs) can support different capabilities but still be connected to the same centralized unit (CU).

[0165] Switching between the old and new BSR schemes can be handled using the same signaling used for handover between gNBs of different capabilities, for example, using incremental signaling or setup / release signaling. Switching between AI / ML models (or AI / ML model IDs) can be handled in the same manner.

[0166] If a single-sided model (UE-side model or gNB-side model) is used in the source gNB, there is no problem even if the target gNB does not support it. This is because the single-sided model has no interoperability issues.

[0167] However, if the source gNB uses a two-sided model and the target gNB does not use the same AI / ML model, there are two options according to some embodiments:

[0168] Source gNB releases AI / ML model configuration before handover: This option requires the source gNB to be aware of the target cell capabilities. It will require additional signaling.

[0169] UE releases AI / ML model configuration upon receiving Handover Command: If the target gNB supports the same configuration as the source gNB, this option will result in unnecessary release / setup. However, this issue can be overcome if the target gNB provides an indication of whether the source configuration needs to be retained. For example, the target gNB may not provide a model or provide a new model in the Handover Command.

[0170] It should be understood that although in some aspects the present disclosure focuses on implementations in LTE and / or 5G networks for the purpose of providing specific examples, the same principles are applicable to other wireless telecommunication systems. Thus, even though the terminology used herein is generally the same as or similar to that of the LTE and 5G standards, the teachings are not limited to current versions of LTE and 5G and are equally applicable to any suitable arrangements that are not based on LTE or 5G and / or conform to any other future versions of LTE, 5G, or other standards.

[0171] It should be noted that the various exemplary methods discussed herein may rely on predetermined / predefined information known to both the base station and the communication device. It should be understood that, in general, such predetermined / predetermined information may be established, for example, by definition in the operating standard of the wireless telecommunications system, or in signaling previously exchanged between the base station and the communication device (e.g., in system information signaling), or in association with radio resource control setup signaling, or in information stored in a SIM application. That is, the specific manner in which the relevant predefined information is established and shared between the various elements of the wireless communication system is not of primary significance to the operational principles described herein. It should be further noted that, unless the context requires otherwise, the various exemplary methods discussed herein rely on information exchanged / transmitted between the various elements of the wireless telecommunications system, and it will be understood that such communication may generally be conducted according to conventional techniques, e.g., with respect to the particular signaling protocol and the type of communication channel used. That is, the specific manner in which the relevant information is exchanged between the various elements of the wireless telecommunications system is not of primary significance to the operational principles described herein.

[0172] It will be understood that the principles described herein are applicable not only to certain types of communication devices, but may be more generally applied to any type of communication device; for example, the approach is not limited to URLLC / IIoT devices or other low-latency communication devices, but may be more generally applied to any type of communication device that operates, for example, utilizing a wireless link to a communication network.

[0173] It will be further understood that the principles described herein are applicable not only to LTE-based or 5G / NR-based wireless communication systems, but also to any type of wireless telecommunication system that supports dynamic scheduling of shared communication resources.

[0174] Further particular and preferred aspects of the invention are set out in the accompanying independent and dependent claims.It will be appreciated that features of the dependent claims may be combined with features of the combined independent claims other than those explicitly set out in a claim.

[0175] Therefore, the above discussion discloses and describes only exemplary embodiments of the present invention. As will be appreciated by those skilled in the art, the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, the disclosure of the present invention is intended to be illustrative, but not limiting, of the scope of the present invention and the other claims. The present disclosure (including any readily discernible variations of the teachings herein) defines, in part, the scope of the aforementioned claim terms so that no inventive subject matter is dedicated to the public.

[0176] The corresponding features of the present disclosure are defined by the following numbered paragraphs:

[0177] Paragraph 1. A method for transmitting data via a wireless communication network by a communication device, comprising:

[0178] receiving, at a transmission buffer, data for transmission via a wireless access interface of a wireless communication network,

[0179] Based on the amount of data in the transmit buffer, determine the initial buffer size required to transmit that data,

[0180] Based on the absolute value of the initial buffer size, reporting the initial buffer status to the infrastructure device,

[0181] determining an updated buffer size required for data transmission based on a change in the amount of data in the transmission buffer, and

[0182] An updated buffer status is reported to the infrastructure device based on the difference between the updated buffer size and the initial buffer size.

[0183] Paragraph 2. A method according to paragraph 1, wherein the initial buffer status and the updated buffer status are reported per logical channel, and the priority of the data is indicated in the initial buffer status and the updated buffer status per logical channel, wherein high priority data in the transmission buffer is cleared first.

[0184] Paragraph 3. The method according to paragraph 1, further comprising the following steps:

[0185] In case the infrastructure equipment is satisfied with the accuracy of the buffer status reported by the communication device, the communication device skips reporting of the buffer status for a predetermined time period based on a command from the infrastructure equipment.

[0186] Paragraph 4. The method of paragraph 1, wherein reporting of the buffer status is triggered when a packet from the PDU set is dropped.

[0187] Paragraph 5. A communication device comprising:

[0188] transceiver circuitry configured to transmit data via a wireless communication network; and

[0189] A controller circuit is configured, in conjunction with the transceiver circuit, to:

[0190] receiving, at a transmission buffer, data for transmission via a wireless access interface of a wireless communication network,

[0191] Based on the amount of data in the transmit buffer, determine the initial buffer size required to transmit that data,

[0192] Based on the absolute value of the initial buffer size, reporting the initial buffer status to the infrastructure device,

[0193] determining an updated buffer size required for data transmission based on the change in the amount of data in the transmission buffer, and

[0194] An updated buffer status is reported to the infrastructure device based on the difference between the updated buffer size and the initial buffer size.

[0195] Paragraph 6. A circuit for a communication device, comprising:

[0196] transceiver circuitry configured to transmit data via a wireless communication network; and

[0197] A controller circuit is configured, in conjunction with the transceiver circuit, to:

[0198] receiving, at a transmission buffer, data for transmission via a wireless access interface of a wireless communication network,

[0199] Based on the amount of data in the transmit buffer, determine the initial buffer size required to transmit that data,

[0200] Based on the absolute value of the initial buffer size, reporting the initial buffer status to the infrastructure device,

[0201] determining an updated buffer size required for data transmission based on the change in the amount of data in the transmission buffer, and

[0202] An updated buffer status is reported to the infrastructure device based on the difference between the updated buffer size and the initial buffer size.

[0203] Paragraph 7. A method of receiving data via a wireless communication network by an infrastructure device, comprising:

[0204] receiving an initial buffer state from a communication device,

[0205] Based on the initial buffer state, determine the initial buffer size for receiving data,

[0206] receiving an updated buffer status from a communication device,

[0207] Based on the updated buffer state, determining the difference between the updated buffer size and the initial buffer size, and

[0208] By adding the difference to the initial buffer size, an updated buffer size for data reception is determined.

[0209] Paragraph 8. A method according to paragraph 7, wherein the initial buffer status and the updated buffer status are received per logical channel, and the priority of the data is indicated in the initial buffer status and the updated buffer status per logical channel, wherein high priority data in the receive buffer is cleared first.

[0210] Paragraph 9. An infrastructure device forming part of a wireless communication network, the infrastructure device comprising:

[0211] a transceiver circuit configured to receive data from a communication device; and

[0212] A controller circuit configured, in conjunction with the transceiver circuit, to:

[0213] receiving an initial buffer state from a communication device,

[0214] Based on the initial buffer state, determine the initial buffer size for receiving data,

[0215] receiving an updated buffer status from a communication device,

[0216] Based on the updated buffer state, determining the difference between the updated buffer size and the initial buffer size, and

[0217] By adding the difference to the initial buffer size, an updated buffer size for data reception is determined.

[0218] Paragraph 10. A circuit for use in infrastructure equipment, the infrastructure equipment forming part of a wireless communication network, the infrastructure equipment comprising:

[0219] a transceiver circuit configured to receive data from a communication device; and

[0220] A controller circuit is configured, in conjunction with the transceiver circuit, to:

[0221] receiving an initial buffer state from a communication device,

[0222] Based on the initial buffer state, determine the initial buffer size for receiving data,

[0223] receiving an updated buffer status from a communication device,

[0224] Based on the updated buffer state, determining the difference between the updated buffer size and the initial buffer size, and

[0225] By adding the difference to the initial buffer size, an updated buffer size for data reception is determined.

[0226] Paragraph 11. A wireless communication system comprising the communication apparatus according to Paragraph 5 and the infrastructure equipment according to Paragraph 9.

[0227] Paragraph 12. A computer program comprising instructions which, when loaded onto a computer, cause the computer to perform the method according to paragraph 1 or paragraph 7.

[0228] Paragraph 13. A non-transitory computer-readable storage medium storing the computer program according to Paragraph 12.

[0229] Paragraph 14. A method of transmitting data via a wireless communication network by a communication device, comprising:

[0230] receiving, at a transmission buffer, data for transmission via a wireless access interface of a wireless communication network,

[0231] Determine the value of one or more input parameters,

[0232] using a first model at the communication device, predicting an initial buffer size required to transmit the data based on the amount of data in the buffer and the value of each of one or more input parameters,

[0233] Based on the absolute value of the initial buffer size, reporting the initial buffer status to the infrastructure device,

[0234] predicting an updated buffer size required for data transmission using the first model based on the updated amount of data in the transmission buffer and the updated value of each of the one or more input parameters, and

[0235] reporting an updated buffer state to the infrastructure device based on the difference between the updated buffer size and the initial buffer size,

[0236] Wherein, the first model is trained using machine learning.

[0237] Paragraph 15. A method according to paragraph 14, wherein the initial buffer status and the updated buffer status are reported per logical channel, and the priority of the data is indicated in the initial buffer status and the updated buffer status per logical channel, wherein high priority data in the transmission buffer is cleared first.

[0238] Paragraph 16. The method of paragraph 14, wherein the input parameters include parameters of the communication device and include one or more of the following:

[0239] RSRP value,

[0240] RSRQ value,

[0241] Channel state information,

[0242] Power headroom,

[0243] The lifetime of the data,

[0244] Report the frequency of buffer status reports,

[0245] Application layer status, and

[0246] The size of the current resource allocation.

[0247] Paragraph 17. The method of paragraph 16, wherein the input parameters further include parameters provided by infrastructure equipment and include one or more of the following:

[0248] Cell load,

[0249] congestion,

[0250] Uplink interference, and

[0251] Application layer status.

[0252] Paragraph 18. The method according to paragraph 14, further comprising the steps of: generating a loss function for adapting to the performance gap between the buffer size predicted by the first model and the actual buffer size, and

[0253] Based on the loss function, the first model is updated.

[0254] Paragraph 19. The method of paragraph 14, further comprising the steps of: generating a loss function for adapting to a performance gap between a buffer size predicted by the first model and a buffer size predicted by a second model, wherein the infrastructure device uses the second model to predict a buffer size required for data reception, and the second model is trained using machine learning, and

[0255] Based on the loss function, the first model is updated.

[0256] Paragraph 20. The method according to paragraph 14, further comprising the steps of:

[0257] In case the infrastructure equipment is satisfied with the accuracy of the buffer status reported by the communication device, the communication device skips reporting of the buffer status for a predetermined time period based on a command from the infrastructure equipment.

[0258] Paragraph 21. The method according to paragraph 14, further comprising the steps of:

[0259] The communication device indicates to the infrastructure equipment whether the reported buffer status is based on a predicted buffer size and how the predicted buffer size is calculated.

[0260] Paragraph 22. The method of paragraph 14, wherein reporting of the buffer status is triggered when a packet from the PDU set is dropped.

[0261] Paragraph 23. The method according to paragraph 14, further comprising the steps of:

[0262] The communication apparatus increases the accuracy of the buffer status reporting by switching from a conventional buffer status reporting scheme to a new buffer status reporting scheme based on a switching command received from an infrastructure device.

[0263] Paragraph 24. The method of paragraph 23, wherein the switch command indicates that a new model is being used.

[0264] Paragraph 25. A method according to any of paragraphs 14 to 24, wherein the communication device is user equipment.

[0265] Paragraph 26. The method according to paragraph 25, further comprising the steps of:

[0266] In case of handover, if the target radio network infrastructure equipment does not have a model or uses a different model than the model at the source radio network infrastructure equipment, the user equipment releases the model configuration upon receiving the handover command.

[0267] Paragraph 27. A communication device comprising:

[0268] transceiver circuitry configured to transmit data via a wireless communication network; and

[0269] A controller circuit configured, in conjunction with the transceiver circuit, to:

[0270] receiving, at a transmission buffer, data for transmission via a wireless access interface of a wireless communication network,

[0271] Determine the value of one or more input parameters,

[0272] using a first model at the communication device, predicting an initial buffer size required to transmit the data based on the amount of data in the buffer and the value of each of one or more input parameters,

[0273] Based on the absolute value of the initial buffer size, reporting the initial buffer status to the infrastructure device,

[0274] predicting an updated buffer size required for data transmission using a first model based on the updated amount of data in the transmission buffer and the updated value of each of the one or more input parameters, and

[0275] reporting an updated buffer state to the infrastructure device based on the difference between the updated buffer size and the initial buffer size,

[0276] Wherein, the first model is trained using machine learning.

[0277] Paragraph 28. A circuit for a communication device, comprising:

[0278] a transceiver circuit configured to transmit data via a wireless communication network, and

[0279] A controller circuit configured, in conjunction with the transceiver circuit, to:

[0280] receiving, at a transmission buffer, data for transmission via a wireless access interface of a wireless communication network,

[0281] Determine the value of one or more input parameters,

[0282] using a first model at the communication device, predicting an initial buffer size required to transmit the data based on the amount of data in the buffer and the value of each of one or more input parameters,

[0283] Based on the absolute value of the initial buffer size, reporting the initial buffer status to the infrastructure device,

[0284] predicting an updated buffer size required for data transmission using a first model based on the updated amount of data in the transmission buffer and the updated value of each of the one or more input parameters, and

[0285] reporting an updated buffer state to the infrastructure device based on the difference between the updated buffer size and the initial buffer size,

[0286] Wherein, the first model is trained using machine learning.

[0287] 29. A method of receiving data via a wireless communication network by an infrastructure device, comprising:

[0288] receiving an initial buffer state from a communication device,

[0289] Determine the value of one or more input parameters,

[0290] using a first model at the infrastructure device, predicting an initial buffer size required to receive the data based on the initial buffer state and the values ​​of each of the one or more input parameters,

[0291] receiving an updated buffer status from a communication device,

[0292] Based on the updated buffer state, determining the difference between the updated buffer size and the initial buffer size,

[0293] Determine the updated buffer size by adding this difference to the initial buffer size,

[0294] predicting a receive buffer size for data reception based on the updated buffer size and the updated value of each of the one or more input parameters, and

[0295] Wherein, the first model is trained using machine learning.

[0296] Paragraph 30. A method according to paragraph 29, wherein the initial buffer status and the updated buffer status are reported per logical channel, and the priority of the data is indicated in the initial buffer status and the updated buffer status per logical channel, wherein high priority data in the receive buffer is cleared first.

[0297] Paragraph 31. The method of paragraph 29, wherein the input parameters include parameters of the communication device and include one or more of the following:

[0298] RSRP value,

[0299] RSRQ value,

[0300] Channel state information,

[0301] Power headroom,

[0302] The lifetime of the data,

[0303] Report the frequency of buffer status reports,

[0304] Application layer status, and

[0305] The size of the current resource allocation.

[0306] Paragraph 32. The method of paragraph 31, wherein the input parameters include parameters of infrastructure equipment and include one or more of the following:

[0307] Cell load,

[0308] congestion,

[0309] Uplink interference, and

[0310] Application layer status.

[0311] Paragraph 33. The method of Paragraph 29 further comprises the steps of: generating a loss function for adapting to the performance gap between the buffer size predicted by the first model and the actual buffer size, and

[0312] Based on the loss function, the first model is updated.

[0313] Paragraph 34. The method of paragraph 29 further comprises the steps of: generating a loss function for adapting to a performance gap between a buffer size predicted by the first model and a buffer size predicted by a second model, wherein the communication device uses the second model to predict a buffer size required for data transmission, and the second model is trained using machine learning, and

[0314] Based on the loss function, the first model is updated.

[0315] Paragraph 35. The method according to paragraph 29, further comprising the steps of:

[0316] In case the infrastructure equipment is satisfied with the accuracy of the buffer size predicted by the infrastructure equipment, the infrastructure equipment instructs the communication device to skip reporting of the buffer status for a predetermined time period.

[0317] Paragraph 36. The method according to paragraph 29, further comprising the steps of:

[0318] The infrastructure equipment then receives from the communication device an indication of whether the reported buffer status is based on a predicted buffer size and the manner in which the predicted buffer size was calculated.

[0319] Paragraph 37. The method of paragraph 29, wherein reporting of the buffer status is triggered when a packet from the PDU set is dropped.

[0320] Paragraph 38. The method according to paragraph 29, further comprising the steps of:

[0321] The infrastructure equipment improves the accuracy of the buffer status reporting by sending a switching command to the communication device to instruct the communication device to switch from the traditional buffer status reporting scheme to the new buffer status reporting scheme.

[0322] Paragraph 39. The method of paragraph 29, wherein the switch command indicates that a new model is being used.

[0323] Paragraph 40. The method according to paragraph 29, further comprising the steps of:

[0324] In case of a handover, if the target radio network infrastructure device does not have a model or uses a different model than the model at the source radio network infrastructure device, the source radio network infrastructure device releases the model configuration before the handover.

[0325] Paragraph 41. The method of paragraph 29, further comprising the steps of:

[0326] In case of a handover, if the target radio network infrastructure device does not have a model or uses a new model that is different from the model at the source radio network infrastructure device, the target radio network infrastructure device either provides no model or provides a new model in the handover command.

[0327] Paragraph 42. An infrastructure device forming part of a wireless communication network, the infrastructure device comprising:

[0328] a transceiver circuit configured to receive data from a communication device; and

[0329] A controller circuit configured, in conjunction with the transceiver circuit, to:

[0330] receiving an initial buffer state from a communication device,

[0331] Determine the value of one or more input parameters,

[0332] using a first model at the infrastructure device, predicting an initial buffer size required to receive the data based on the initial buffer state and the values ​​of each of the one or more input parameters,

[0333] receiving an updated buffer status from a communication device,

[0334] Based on the updated buffer state, determining the difference between the updated buffer size and the initial buffer size,

[0335] Determine the updated buffer size by adding the difference to the initial buffer size, and

[0336] predicting a receive buffer size for data reception based on the updated buffer size and the updated value of each of the one or more input parameters,

[0337] Wherein, the first model is trained using machine learning.

[0338] Paragraph 43. A circuit for use in infrastructure equipment, the infrastructure equipment forming part of a wireless communication network, the infrastructure equipment comprising:

[0339] a transceiver circuit configured to receive data from a communication device; and

[0340] A controller circuit configured, in conjunction with the transceiver circuit, to:

[0341] receiving an initial buffer state from a communication device,

[0342] Determine the value of one or more input parameters,

[0343] using a first model at the infrastructure device, predicting an initial buffer size required to receive the data based on the initial buffer state and the values ​​of each of the one or more input parameters,

[0344] receiving an updated buffer status from a communication device,

[0345] Based on the updated buffer state, determining the difference between the updated buffer size and the initial buffer size,

[0346] By adding this difference to the initial buffer size, an updated buffer size is determined, and

[0347] predicting a receive buffer size for data reception based on the updated buffer size and the updated value of each of the one or more input parameters,

[0348] Wherein, the first model is trained using machine learning.

[0349] Paragraph 44. A wireless communication system comprising the communication apparatus of paragraph 27 and the infrastructure equipment of paragraph 42.

[0350] Paragraph 45. A computer program comprising instructions which, when loaded onto a computer, cause the computer to perform the method according to paragraph 14 or paragraph 29.

[0351] Paragraph 46. A non-transitory computer-readable storage medium storing the computer program according to Paragraph 45.

[0352] References

[0353] [1] 3GPP TS 38.300v.15.2.0 "NR; NR and NG-RAN Overall Description; Stage2 (Release 15)", June 2018

[0354] [2]TS38.322, "Radio Link Control (RLC) protocol specification", Release17

[0355] [3]TS38.323, "Packet Data Convergence Protocol (PDCP) specification", Release 17

[0356] [4]Holma H.and Toskala A, "LTE for UMTS OFDMA and SC-FDMA based radioaccess", John Wiley and Sons, 2009

[0357] [5]TR38.835, “Study on XR enhancements for NR”, Release 18

[0358] [6]TS38.321, "Medium Access Control (MAC) protocol specification", Release 17

[0359] [7]TS22.261 "Service requirements for the 5G system" (Release 17)

Claims

1. A method for transmitting data via a wireless communication network by a communication device, the method comprising: receiving, at a transmission buffer, data for transmission via a wireless access interface of the wireless communication network, determining an initial buffer size required to transmit the data based on the amount of data in the transmission buffer; reporting an initial buffer state to an infrastructure device based on the absolute value of the initial buffer size, determining an updated buffer size required for data transmission based on the change in the amount of data in the transmission buffer, and An updated buffer status is reported to the infrastructure device based on a difference between the updated buffer size and the initial buffer size.

2. The method according to claim 1, wherein The initial buffer status and the updated buffer status are reported per logical channel, and the priority of data is indicated in the initial buffer status and the updated buffer status per logical channel, wherein high priority data in the transmission buffer is cleared first.

3. The method according to claim 1, further comprising the steps of: In a case where the infrastructure equipment is satisfied with the accuracy of the buffer status reported by the communication device, the communication device skips reporting of the buffer status for a predetermined time period based on a command from the infrastructure equipment.

4. The method according to claim 1, wherein When a packet from a PDU set is dropped, reporting of the buffer status is triggered.

5. A communication device comprising: a transceiver circuit configured to transmit data via a wireless communication network, and A controller circuit configured, in conjunction with the transceiver circuit, to: receiving, at a transmission buffer, data for transmission via a wireless access interface of the wireless communication network, determining an initial buffer size required to transmit the data based on the amount of data in the transmission buffer; reporting an initial buffer state to an infrastructure device based on the absolute value of the initial buffer size, determining an updated buffer size required for data transmission based on the change in the amount of data in the transmission buffer, and An updated buffer status is reported to the infrastructure device based on a difference between the updated buffer size and the initial buffer size.

6. A circuit for a communication device, comprising: a transceiver circuit configured to transmit data via a wireless communication network, and A controller circuit configured, in conjunction with the transceiver circuit, to: receiving, at a transmission buffer, data for transmission via a wireless access interface of the wireless communication network, determining an initial buffer size required to transmit the data based on the amount of data in the transmission buffer; reporting an initial buffer state to an infrastructure device based on the absolute value of the initial buffer size, determining an updated buffer size required for data transmission based on the change in the amount of data in the transmission buffer, and An updated buffer status is reported to the infrastructure device based on a difference between the updated buffer size and the initial buffer size.

7. A method of receiving data via a wireless communication network by an infrastructure device, the method comprising: receiving an initial buffer state from a communication device, determining an initial buffer size for receiving the data based on the initial buffer state, receiving an updated buffer status from the communication device, determining a difference between an updated buffer size and the initial buffer size based on the updated buffer status, and The updated buffer size for data reception is determined by adding the difference to the initial buffer size.

8. The method according to claim 7, wherein: The initial buffer status and the updated buffer status are received per logical channel, and the priority of data is indicated in the initial buffer status and the updated buffer status per logical channel, wherein high priority data in the receive buffer is cleared first.

9. An infrastructure device, the infrastructure device forming part of a wireless communication network, the infrastructure device comprising: a transceiver circuit configured to receive data from a communication device; as well as A controller circuit configured, in conjunction with the transceiver circuit, to: receiving an initial buffer state from a communication device, determining an initial buffer size for receiving the data based on the initial buffer state, receiving an updated buffer status from the communication device, Based on the updated buffer state, determining a difference between an updated buffer size and the initial buffer size, and The updated buffer size for data reception is determined by adding the difference to the initial buffer size.

10. A circuit for use in infrastructure equipment, the infrastructure equipment forming part of a wireless communication network, the infrastructure equipment comprising: a transceiver circuit configured to receive data from a communication device; as well as A controller circuit configured, in conjunction with the transceiver circuit, to: receiving an initial buffer state from a communication device, determining an initial buffer size for receiving the data based on the initial buffer state, receiving an updated buffer status from the communication device, Based on the updated buffer state, determining a difference between an updated buffer size and the initial buffer size, and The updated buffer size for data reception is determined by adding the difference to the initial buffer size.

11. A wireless communication system comprising the communication apparatus according to claim 5 and the infrastructure equipment according to claim 9.

12. A computer program comprising instructions which, when loaded onto a computer, cause the computer to perform the method according to claim 1 or claim 7.

13. A non-transitory computer-readable storage medium storing the computer program according to claim 12.

14. A method for transmitting data via a wireless communication network by a communication device, the method comprising: receiving, at a transmission buffer, data for transmission via a wireless access interface of the wireless communication network, Determine the value of one or more input parameters, using a first model at the communication device, predicting an initial buffer size required to transmit the data based on the amount of data in the buffer and the value of each of the one or more input parameters, reporting an initial buffer state to an infrastructure device based on the absolute value of the initial buffer size, predicting an updated buffer size required for data transmission using the first model based on the updated amount of data in the transmission buffer and the updated value of each of the one or more input parameters, and reporting an updated buffer status to the infrastructure device based on a difference between the updated buffer size and the initial buffer size, Wherein, the first model is trained using machine learning.

15. The method according to claim 14, wherein The initial buffer status and the updated buffer status are reported per logical channel, and the priority of data is indicated in the initial buffer status and the updated buffer status per logical channel, wherein high priority data in the transmission buffer is cleared first.

16. The method according to claim 14, wherein The input parameters include parameters of the communication device and include one or more of the following: RSRP value, RSRQ value, Channel state information, Power headroom, The lifetime of the data, Report the frequency of buffer status reports, Application layer status, and The size of the current resource allocation.

17. The method according to claim 16, wherein The input parameters also include parameters provided by the infrastructure equipment, and include one or more of the following: Cell load, congestion, Uplink interference, and Application layer status.

18. The method according to claim 14, further comprising the steps of: generating a loss function for accommodating a performance gap between a buffer size predicted by the first model and an actual buffer size, and Based on the loss function, the first model is updated.

19. The method according to claim 14, further comprising the steps of: generating a loss function for accommodating a performance gap between a buffer size predicted by the first model and a buffer size predicted by a second model, wherein the infrastructure device uses the second model to predict a buffer size required for data reception and the second model is trained using machine learning, and The first model is updated based on the loss function.

20. The method according to claim 14, further comprising the steps of: In a case where the infrastructure equipment is satisfied with the accuracy of the buffer status reported by the communication device, the communication device skips reporting of the buffer status for a predetermined time period based on a command from the infrastructure equipment.

21. The method according to claim 14, further comprising the steps of: The communication device indicates to the infrastructure equipment whether the reported buffer status is based on a predicted buffer size and how the predicted buffer size is calculated.

22. The method according to claim 14, wherein When a packet from a PDU set is dropped, reporting of the buffer status is triggered.

23. The method according to claim 14, further comprising the steps of: The communication apparatus improves the accuracy of buffer status reporting by switching from a conventional buffer status reporting scheme to a new buffer status reporting scheme based on a switching command received from the infrastructure equipment.

24. The method according to claim 23, wherein The switch command indicates that a new model is being used.

25. The method according to any one of claims 14 to 24, wherein The communication device is user equipment.

26. The method according to claim 25, further comprising the steps of: In case of a handover, if the target radio network infrastructure equipment does not have a model or uses a different model than the model at the source radio network infrastructure equipment, the user equipment releases the model configuration upon receiving the handover command.

27. A communication device comprising: a transceiver circuit configured to transmit data via a wireless communication network, and A controller circuit configured, in conjunction with the transceiver circuit, to: receiving, at a transmission buffer, data for transmission via a wireless access interface of the wireless communication network, Determine the value of one or more input parameters, using a first model at the communication device, predicting an initial buffer size required to transmit the data based on the amount of data in the buffer and the value of each of the one or more input parameters, reporting an initial buffer state to an infrastructure device based on the absolute value of the initial buffer size, predicting an updated buffer size required for data transmission using a first model based on the updated amount of data in the transmission buffer and the updated value of each of the one or more input parameters, and reporting an updated buffer status to the infrastructure device based on a difference between the updated buffer size and the initial buffer size, Wherein, the first model is trained using machine learning.

28. A circuit for a communication device, comprising: a transceiver circuit configured to transmit data via a wireless communication network, and A controller circuit configured, in conjunction with the transceiver circuit, to: receiving, at a transmission buffer, data for transmission via a wireless access interface of the wireless communication network, Determine the value of one or more input parameters, using a first model at the communication device, predicting an initial buffer size required to transmit the data based on the amount of data in the buffer and the value of each of the one or more input parameters, reporting an initial buffer state to an infrastructure device based on the absolute value of the initial buffer size, predicting an updated buffer size required for data transmission using a first model based on the updated amount of data in the transmission buffer and the updated value of each of the one or more input parameters, and reporting an updated buffer status to the infrastructure device based on a difference between the updated buffer size and the initial buffer size, Wherein, the first model is trained using machine learning.

29. A method of receiving data via a wireless communication network by an infrastructure device, the method comprising: receiving an initial buffer state from a communication device, Determine the value of one or more input parameters, using a first model at the infrastructure device, predicting an initial buffer size required to receive the data based on the initial buffer state and the values ​​of each of the one or more input parameters, receiving an updated buffer status from the communication device, determining a difference between an updated buffer size and the initial buffer size based on the updated buffer state, determining the updated buffer size by adding the difference to the initial buffer size, and predicting a receive buffer size for data reception based on the updated buffer size and the updated value of each of the one or more input parameters, Wherein, the first model is trained using machine learning.

30. The method according to claim 29, wherein The initial buffer status and the updated buffer status are reported per logical channel, and the priority of data is indicated in the initial buffer status and the updated buffer status per logical channel, wherein high priority data in the receive buffer is cleared first.

31. The method according to claim 29, wherein The input parameters include parameters of the communication device and include one or more of the following: RSRP value, RSRQ value, Channel state information, Power headroom, The lifetime of the data, Report the frequency of buffer status reports, Application layer status, and The size of the current resource allocation.

32. The method according to claim 31, wherein The input parameters include parameters of the infrastructure equipment and include one or more of the following: Cell load, congestion, Uplink interference, and Application layer status.

33. The method according to claim 29, further comprising the steps of: generating a loss function for accommodating a performance gap between a buffer size predicted by the first model and an actual buffer size, and Based on the loss function, the first model is updated.

34. The method according to claim 29, further comprising the steps of: generating a loss function for accommodating a performance gap between a buffer size predicted by the first model and a buffer size predicted by a second model, wherein the communication device uses the second model to predict a buffer size required for data transmission and the second model is trained using machine learning, and The first model is updated based on the loss function.

35. The method according to claim 29, further comprising the steps of: In case the infrastructure equipment is satisfied with the accuracy of the buffer size predicted by the infrastructure equipment, the infrastructure equipment instructs the communication device to skip reporting of the buffer status for a predetermined time period.

36. The method according to claim 29, further comprising the steps of: The infrastructure equipment receives from the communication device an indication of whether the reported buffer status is based on a predicted buffer size and a manner in which the predicted buffer size is calculated.

37. The method of claim 29, wherein: When a packet from a PDU set is dropped, reporting of the buffer status is triggered.

38. The method according to claim 29, further comprising the steps of: The infrastructure equipment improves the accuracy of the buffer status reporting by sending a switching command to the communication device to instruct the communication device to switch from the traditional buffer status reporting scheme to the new buffer status reporting scheme.

39. The method according to claim 29, wherein The Toggle command indicates that the new model is in use.

40. The method according to claim 29, further comprising the steps of: In case of a handover, if the target radio network infrastructure device does not have a model or uses a different model than the model at the source radio network infrastructure device, the source radio network infrastructure device releases the model configuration before the handover.

41. The method according to claim 29, further comprising the steps of: In case of a handover, if the target radio network infrastructure device does not have a model or uses a new model that is different from the model at the source radio network infrastructure device, the target radio network infrastructure device either provides no model or provides a new model in the handover command.

42. Infrastructure equipment, said infrastructure equipment forming part of a wireless communication network, said infrastructure equipment comprising: a transceiver circuit configured to receive data from a communication device; as well as A controller circuit configured, in conjunction with the transceiver circuit, to: receiving an initial buffer state from a communication device, Determine the value of one or more input parameters, using a first model at the infrastructure device, predicting an initial buffer size required to receive the data based on the initial buffer state and the values ​​of each of the one or more input parameters, receiving an updated buffer status from the communication device, determining a difference between an updated buffer size and the initial buffer size based on the updated buffer state, determining the updated buffer size by adding the difference to the initial buffer size, and predicting a receive buffer size for data reception based on the updated buffer size and the updated value of each of the one or more input parameters, Wherein, the first model is trained using machine learning.

43. A circuit for use in infrastructure equipment, the infrastructure equipment forming part of a wireless communication network, the infrastructure equipment comprising: a transceiver circuit configured to receive data from a communication device; as well as A controller circuit configured, in conjunction with the transceiver circuit, to: receiving an initial buffer state from a communication device, Determine the value of one or more input parameters, using a first model at the infrastructure device, predicting an initial buffer size required to receive the data based on the initial buffer state and the values ​​of each of the one or more input parameters, receiving an updated buffer status from the communication device, determining a difference between an updated buffer size and the initial buffer size based on the updated buffer state, determining the updated buffer size by adding the difference to the initial buffer size, and predicting a receive buffer size for data reception based on the updated buffer size and the updated value of each of the one or more input parameters, Wherein, the first model is trained using machine learning.

44. A wireless communication system comprising the communication apparatus according to claim 27 and the infrastructure equipment according to claim 42.

45. A computer program comprising instructions which, when loaded onto a computer, cause the computer to perform the method according to claim 14 or claim 29.

46. ​​A non-transitory computer-readable storage medium storing the computer program according to claim 45.

Images