WO2022073162A1

WO2022073162A1 - Compression and segmenting for communicating parameters of a neural network configuration

Info

Publication number: WO2022073162A1
Application number: PCT/CN2020/119864
Authority: WO
Inventors: Ruiming Zheng; Yu Zhang; Hao Xu; Qiaoyu Li
Original assignee: Qualcomm Incorporated
Priority date: 2020-10-08
Filing date: 2020-10-08
Publication date: 2022-04-14

Abstract

A method of wireless communication by a user equipment (UE) includes segmenting a UE neural network configuration message to form an RRC neural network configuration message. The method also includes compressing the RRC neural network configuration message to form a compressed RRC configuration message. The method further includes transmitting the compressed RRC neural network configuration message to a base station.

Description

COMPRESSION AND SEGMENTING FOR COMMUNICATING PARAMETERS OF A NEURAL NETWORK CONFIGURATION

FIELD OF THE DISCLOSURE

Aspects of the present disclosure generally relate to wireless communications, and more particularly to a fifth generation (5G) new radio (NR) compression and segmenting of signals for communicating parameters of a neural network configuration.

BACKGROUND

Wireless communications systems are widely deployed to provide various telecommunications services such as telephony, video, data, messaging, and broadcasts. Typical wireless communications systems may employ multiple-access technologies capable of supporting communications with multiple users by sharing available system resources (e.g., bandwidth, transmit power, and/or the like) . Examples of such multiple-access technologies include code division multiple access (CDMA) systems, time division multiple access (TDMA) systems, frequency-division multiple access (FDMA) systems, orthogonal frequency-division multiple access (OFDMA) systems, single-carrier frequency-division multiple access (SC-FDMA) systems, time division synchronous code division multiple access (TD-SCDMA) systems, and long term evolution (LTE) . LTE/LTE-Advanced is a set of enhancements to the universal mobile telecommunications system (UMTS) mobile standard promulgated by the Third Generation Partnership Project (3GPP) .

A wireless communications network may include a number of base stations (BSs) that can support communications for a number of user equipment (UEs) . A user equipment (UE) may communicate with a base station (BS) via the downlink and uplink. The downlink (or forward link) refers to the communications link from the BS to the UE, and the uplink (or reverse link) refers to the communications link from the UE to the BS. As will be described in more detail, a BS may be referred to as a Node B, a gNB, an access point (AP) , a radio head, a transmit and receive point (TRP) , a new radio (NR) BS, a 5G Node B, and/or the like.

The above multiple access technologies have been adopted in various telecommunications standards to provide a common protocol that enables different user equipment to communicate on a municipal, national, regional, and even global level. New Radio (NR) , which may also be referred to as 5G, is a set of enhancements to the LTE mobile standard promulgated by the Third Generation Partnership Project (3GPP) . NR is designed to better support mobile broadband Internet access by improving spectral efficiency, lowering costs, improving services, making use of new spectrum, and better integrating with other open standards using orthogonal frequency division multiplexing (OFDM) with a cyclic prefix (CP) (CP-OFDM) on the downlink (DL) , using CP-OFDM and/or SC-FDM (e.g., also known as discrete Fourier transform spread OFDM (DFT-s-OFDM) ) on the uplink (UL) , as well as supporting beamforming, multiple-input multiple-output (MIMO) antenna technology, and carrier aggregation.

Artificial neural networks may comprise interconnected groups of artificial neurons (e.g., neuron models) . The artificial neural network may be a computational device or represented as a method to be performed by a computational device. Convolutional neural networks, such as deep convolutional neural networks, are a type of feed-forward artificial neural network. Convolutional neural networks may include layers of neurons that may be configured in a tiled receptive field. It would be desirable to apply neural network processing to wireless communications to achieve greater efficiencies.

SUMMARY

A method of wireless communication by a user equipment (UE) is described. The method includes compressing an RRC neural network configuration message to form a compressed RRC configuration message. The method also includes transmitting the compressed RRC configuration message to a base station.

A method of wireless communication by a base station is described. The method includes updating parameters for configuration of an artificial neural network. The method also includes segmenting configuration messages including the updated parameters for configuration of the artificial neural network to form segmented configuration messages. The method further includes compressing the segmented configuration messages to form compressed, segmented configuration messages. The method also includes transmitting, via radio resource control (RRC) signaling, the compressed, segmented configuration messages to a user equipment (UE) .

An apparatus for wireless communication by a user equipment (UE) is described. The apparatus includes means for segmenting a UE neural network configuration message to form a radio resource control (RRC) neural network configuration message. The apparatus also includes means for compressing the RRC neural network configuration message to form a compressed RRC configuration message. The apparatus further includes means for transmitting the compressed RRC neural network configuration message to a base station.

A user equipment (UE) includes a processor and a memory coupled with the processor. The UE also includes instructions stored in the memory. When the instructions are executed by the processor the UE is operable to segment a UE neural network configuration message to form a radio resource control (RRC) neural network configuration message. The UE is also operable to compress the RRC neural network configuration message to form a compressed RRC configuration message. The UE is further operable to transmit the compressed RRC neural network configuration message to a base station.

A non-transitory computer-readable medium includes program code recorded thereon. The program code is executed by a processor. The non-transitory computer-readable medium includes program code to segment a UE neural network configuration message to form a radio resource control (RRC) neural network configuration message. The non-transitory computer-readable medium also includes program code to compress the RRC neural network configuration message to form a compressed RRC configuration message. The non-transitory computer-readable medium further includes program code to transmit the compressed RRC neural network configuration message to a base station.

A apparatus for wireless communication by a user equipment (UE) is described. The apparatus includes means for compressing an RRC neural network configuration message to form a compressed RRC configuration message. The apparatus also includes means for transmitting the compressed RRC configuration message to a base station.

A user equipment (UE) includes a processor and a memory coupled with the processor. The UE also includes instructions stored in the memory. When the instructions are executed by the processor the UE is operable to compress an RRC neural network configuration message to form a compressed RRC configuration message. The processor the UE is also operable to transmit the compressed RRC configuration message to a base station.

A non-transitory computer-readable medium includes program code recorded thereon. The program code is executed by a processor. The non-transitory computer-readable medium includes program code to compress an RRC neural network configuration message to form a compressed RRC configuration message. The non-transitory computer-readable medium also includes program code to transmit the compressed RRC configuration message to a base station.

An apparatus for wireless communication by a base station is described. The apparatus includes means for updating parameters for configuration of an artificial neural network. The apparatus also includes means for segmenting configuration messages including the updated parameters for configuration of the artificial neural network to form segmented configuration messages. The apparatus further includes means for compressing the segmented configuration messages to form compressed, segmented configuration messages. The apparatus also includes means for transmitting, via radio resource control (RRC) signaling, the compressed, segmented configuration messages to a user equipment (UE) .

A base station includes a processor and a memory coupled with the processor. The base station also includes instructions stored in the memory. When the instructions are executed by the processor, the base station is operable to update parameters for configuration of an artificial neural network. The base station is also operable to transmit to segment configuration messages including the updated parameters for configuration of the artificial neural network to form segmented configuration messages. The base station is further operable to compress the segmented configuration messages to form compressed, segmented configuration messages. The base station is also operable to transmit, via radio resource control (RRC) signaling, the compressed, segmented configuration messages to a user equipment (UE) .

A non-transitory computer-readable medium includes program code recorded thereon. The program code is executed by a processor. The non-transitory computer-readable medium includes program code to update parameters for configuration of an artificial neural network configuration. The non-transitory computer-readable medium also includes program code to segment configuration messages including the updated parameters for configuration of the artificial neural network to form segmented configuration messages. The non-transitory computer-readable medium further includes program code to compress the segmented configuration messages to form compressed, segmented configuration messages. The non-transitory computer-readable medium also includes program code to transmit, via radio resource control (RRC) signaling, the compressed, segmented configuration messages to a user equipment (UE) .

Aspects generally include a method, apparatus, system, computer program product, non-transitory computer-readable medium, user equipment, base station, wireless communication device, and processing system as substantially described with reference to and as illustrated by the accompanying drawings and specification.

The foregoing has outlined rather broadly the features and technical advantages of examples according to the disclosure in order that the detailed description that follows may be better understood. Additional features and advantages will be described. The conception and specific examples disclosed may be readily utilized as a basis for modifying or designing other structures for carrying out the same purposes of the present disclosure. Such equivalent constructions do not depart from the scope of the appended claims. Characteristics of the concepts disclosed, both their organization and method of operation, together with associated advantages will be better understood from the following description when considered in connection with the accompanying figures. Each of the figures is provided for the purposes of illustration and description, and not as a definition of the limits of the claims.

BRIEF DESCRIPTION OF THE DRAWINGS

So that features of the present disclosure can be understood in detail, a particular description may be had by reference to aspects, some of which are illustrated in the appended drawings. It is to be noted, however, that the appended drawings illustrate only certain aspects of this disclosure and are therefore not to be considered limiting of its scope, for the description may admit to other equally effective aspects. The same reference numbers in different drawings may identify the same or similar elements.

FIGURE 1 is a block diagram conceptually illustrating an example of a wireless communications network, in accordance with various aspects of the present disclosure.

FIGURE 2 is a block diagram conceptually illustrating an example of a base station in communication with a user equipment (UE) in a wireless communications network, in accordance with various aspects of the present disclosure.

FIGURE 3 illustrates an example implementation of designing a neural network using a system-on-a-chip (SOC) , including a general-purpose processor, in accordance with certain aspects of the present disclosure.

FIGURES 4A, 4B, and 4C are diagrams illustrating a neural network, in accordance with aspects of the present disclosure.

FIGURE 4D is a diagram illustrating an exemplary deep convolutional network (DCN) , in accordance with aspects of the present disclosure.

FIGURE 5 is a block diagram illustrating an exemplary deep convolutional network (DCN) , in accordance with aspects of the present disclosure.

FIGURE 6 is a block diagram illustrating an artificial intelligence (AI) -based end-to-end (E2E) wireless system incorporating a neural network within a transmitter (Tx) and/or a receiver (Rx) , according to aspects of the present disclosure.

FIGURE 7 is a block diagram illustrating an encoded radio resource control (RRC) message.

FIGURE 8 is a call flow diagram illustrating radio resource control (RRC) message segmenting and compression for uplink communication of neural network configuration parameters, according to aspects of the present disclosure.

FIGURE 9 is a call flow diagram illustrating radio resource control (RRC) message segmenting and packet data convergence protocol (PDCP) compression for downlink communication of neural network configuration parameters, according to aspects of the present disclosure.

FIGURE 10 is a block diagram illustrating an encoded radio resource control (RRC) message, according to aspects of the present disclosure.

FIGURE 11 is a flow diagram illustrating a method for compressing and segmenting radio resource control (RRC) messages for uplink communication of neural network configuration parameters, according to aspects of the present disclosure.

FIGURE 12 is a flow diagram illustrating an example process performed, for example, by a user equipment (UE) , in accordance with various aspects of the present disclosure.

FIGURE 13 is a flow diagram illustrating an example process performed, for example, by a base station, in accordance with various aspects of the present disclosure.

DETAILED DESCRIPTION

Various aspects of the disclosure are described more fully below with reference to the accompanying drawings. This disclosure may, however, be embodied in many different forms and should not be construed as limited to any specific structure or function presented throughout this disclosure. Rather, these aspects are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art. Based on the teachings, one skilled in the art should appreciate that the scope of the disclosure is intended to cover any aspect of the disclosure, whether implemented independently of or combined with any other aspect of the disclosure. For example, an apparatus may be implemented or a method may be practiced using any number of the aspects set forth. In addition, the scope of the disclosure is intended to cover such an apparatus or method which is practiced using other structure, functionality, or structure and functionality in addition to or other than the various aspects of the disclosure set forth. It should be understood that any aspect of the disclosure disclosed may be embodied by one or more elements of a claim.

Several aspects of telecommunications systems will now be presented with reference to various apparatuses and techniques. These apparatuses and techniques will be described in the following detailed description and illustrated in the accompanying drawings by various blocks, modules, components, circuits, steps, processes, algorithms, and/or the like (collectively referred to as “elements” ) . These elements may be implemented using hardware, software, or combinations thereof. Whether such elements are implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system.

It should be noted that while aspects may be described using terminology commonly associated with 5G and later wireless technologies, aspects of the present disclosure can be applied in other generation-based communications systems, such as and including 3G and/or 4G technologies.

In an artificial intelligence (AI) -based end-to-end (E2E) wireless system, a neural network may be incorporated within a transmitter (Tx) and a receiver (Rx) of the AI-based E2E wireless system. In one configuration, a transmitter neural network replaces encoding, modulation, and/or precoding components in the transmitter. In addition, a receiver neural network replaces synchronization, channel estimation, detection, demodulation, and/or decoding components in the receiver. In some aspects, the neural network replaces one, some, or all transmitting/receiving modules of the AI-based wireless system. In this aspect of the present disclosure, offline training and online refinement are performed to configure the transmitter neural network (s) and the receiver neural network (s) .

Unfortunately, a parameter size for configuration of the neural networks of an AI-based end-to-end wireless system may be up to tens of megabytes. In current 5G NR systems, a radio resource control (RRC) message is only a few bytes (and not more than a few kilobytes) for an uplink or downlink transmission. Aspects of the present disclosure are directed to segmenting and compressing signals for communicating of neural network configuration parameters of an AI-based wireless system in a 5G NR system.

In aspects of the present disclosure, UE parameters of an AI-based wireless system are segmented and compressed to enabling communication between a UE and a base station (e.g., gNB) . In some aspects, refinement of one or multiple configuration parameters of the neural networks of the AI-based wireless system is performed for improved end-to-end transmission. The neural networks of the AI-based wireless system may be composed of one or any combination of kernels and/or coefficients of the kernels. The combination of the kernels and/or the coefficients of the kernels may be for a certain convolutional layer, a locally-connected layer, or parameters for a certain dense layer.

In some aspects, the AI-based wireless system provides various benefits. For example, a significant performance benefit of the AI-based wireless system is a complete end-to-end (E2E) auto-encoder, which outperforms conventional, sub-optimal block-wise transceivers. In particular, an AI-based E2E wireless system benefits from data-driven neural network training, which provides improved robustness relative to conventional model-based metrics. In addition, an efficient neural network design exhibits comparable complexity, which is lower than conventional mean average precision (MAP) /machine learning (ML) -based detection and decoding.

FIGURE 1 is a diagram illustrating a network 100 in which aspects of the present disclosure may be practiced. The network 100 may be a 5G or NR network or some other wireless network, such as an LTE network. The wireless network 100 may include a number of BSs 110 (shown as BS 110a, BS 110b, BS 110c, and BS 110d) and other network entities. A BS is an entity that communicates with user equipment (UEs) and may also be referred to as a base station, a NR BS, a Node B, a gNB, a 5G node B (NB) , an access point, a transmit and receive point (TRP) , and/or the like. Each BS may provide communications coverage for a particular geographic area. In 3GPP, the term “cell” can refer to a coverage area of a BS and/or a BS subsystem serving this coverage area, depending on the context in which the term is used.

A BS may provide communications coverage for a macro cell, a pico cell, a femto cell, and/or another type of cell. A macro cell may cover a relatively large geographic area (e.g., several kilometers in radius) and may allow unrestricted access by UEs with service subscription. A pico cell may cover a relatively small geographic area and may allow unrestricted access by UEs with service subscription. A femto cell may cover a relatively small geographic area (e.g., a home) and may allow restricted access by UEs having association with the femto cell (e.g., UEs in a closed subscriber group (CSG) ) . A BS for a macro cell may be referred to as a macro BS. A BS for a pico cell may be referred to as a pico BS. A BS for a femto cell may be referred to as a femto BS or a home BS. In the example shown in FIGURE 1, a BS 110a may be a macro BS for a macro cell 102a, a BS 110b may be a pico BS for a pico cell 102b, and a BS 110c may be a femto BS for a femto cell 102c. A BS may support one or multiple (e.g., three) cells. The terms “eNB, ” “base station, ” “NR BS, ” “gNB, ” “TRP, ” “AP, ” “node B, ” “5G NB, ” and “cell” may be used interchangeably.

In some aspects, a cell may not necessarily be stationary, and the geographic area of the cell may move according to the location of a mobile BS. In some aspects, the BSs may be interconnected to one another and/or to one or more other BSs or network nodes (not shown) in the wireless network 100 through various types of backhaul interfaces such as a direct physical connection, a virtual network, and/or the like using any suitable transport network.

The wireless network 100 may also include relay stations. A relay station is an entity that can receive a transmission of data from an upstream station (e.g., a BS or a UE) and send a transmission of the data to a downstream station (e.g., a UE or a BS) . A relay station may also be a UE that can relay transmissions for other UEs. In the example shown in FIGURE 1, a relay station 110d may communicate with macro BS 110a and a UE 120d in order to facilitate communications between the BS 110a and UE 120d. A relay station may also be referred to as a relay BS, a relay base station, a relay, and/or the like.

The wireless network 100 may be a heterogeneous network that includes BSs of different types, e.g., macro BSs, pico BSs, femto BSs, relay BSs, and/or the like. These different types of BSs may have different transmit power levels, different coverage areas, and different impact on interference in the wireless network 100. For example, macro BSs may have a high transmit power level (e.g., 5 to 40 Watts) whereas pico BSs, femto BSs, and relay BSs may have lower transmit power levels (e.g., 0.1 to 2 Watts) .

A network controller 130 may couple to a set of BSs and may provide coordination and control for these BSs. The network controller 130 may communicate with the BSs via a backhaul. The BSs may also communicate with one another, e.g., directly or indirectly via a wireless or wireline backhaul.

UEs 120 (e.g., 120a, 120b, 120c) may be dispersed throughout the wireless network 100, and each UE may be stationary or mobile. A UE may also be referred to as an access terminal, a terminal, a mobile station, a subscriber unit, a station, and/or the like. A UE may be a cellular phone (e.g., a smart phone) , a personal digital assistant (PDA) , a wireless modem, a wireless communications device, a handheld device, a laptop computer, a cordless phone, a wireless local loop (WLL) station, a tablet, a camera, a gaming device, a netbook, a smartbook, an ultrabook, a medical device or equipment, biometric sensors/devices, wearable devices (smart watches, smart clothing, smart glasses, smart wrist bands, smart jewelry (e.g., smart ring, smart bracelet) ) , an entertainment device (e.g., a music or video device, or a satellite radio) , a vehicular component or sensor, smart meters/sensors, industrial manufacturing equipment, a global positioning system device, or any other suitable device that is configured to communicate via a wireless or wired medium.

Some UEs may be considered machine-type communications (MTC) or evolved or enhanced machine-type communications (eMTC) UEs. MTC and eMTC UEs include, for example, robots, drones, remote devices, sensors, meters, monitors, location tags, and/or the like, that may communicate with a base station, another device (e.g., remote device) , or some other entity. A wireless node may provide, for example, connectivity for or to a network (e.g., a wide area network such as Internet or a cellular network) via a wired or wireless communications link. Some UEs may be considered Internet-of-Things (IoT) devices, and/or may be implemented as NB-IoT (narrowband internet of things) devices. Some UEs may be considered a customer premises equipment (CPE) . UE 120 may be included inside a housing that houses components of UE 120, such as processor components, memory components, and/or the like.

In general, any number of wireless networks may be deployed in a given geographic area. Each wireless network may support a particular RAT and may operate on one or more frequencies. A RAT may also be referred to as a radio technology, an air interface, and/or the like. A frequency may also be referred to as a carrier, a frequency channel, and/or the like. Each frequency may support a single RAT in a given geographic area in order to avoid interference between wireless networks of different RATs. In some cases, NR or 5G RAT networks may be deployed.

In some aspects, two or more UEs 120 (e.g., shown as UE 120a and UE 120e) may communicate directly using one or more sidelink channels (e.g., without using a base station 110 as an intermediary to communicate with one another) . For example, the UEs 120 may communicate using peer-to-peer (P2P) communications, device-to-device (D2D) communications, a vehicle-to-everything (V2X) protocol (e.g., which may include a vehicle-to-vehicle (V2V) protocol, a vehicle-to-infrastructure (V2I) protocol, and/or the like) , a mesh network, and/or the like. In this case, the UE 120 may perform scheduling operations, resource selection operations, and/or other operations described elsewhere as being performed by the base station 110. For example, the base station 110 may configure a UE 120 via downlink control information (DCI) , radio resource control (RRC) signaling, a media access control-control element (MAC-CE) or via system information (e.g., a system information block (SIB) .

As indicated above, FIGURE 1 is provided merely as an example. Other examples may differ from what is described with regard to FIGURE 1.

FIGURE 2 shows a block diagram of a design 200 of the base station 110 and UE 120, which may be one of the base stations and one of the UEs in FIGURE 1. The base station 110 may be equipped with T antennas 234a through 234t, and UE 120 may be equipped with R antennas 252a through 252r, where in general T ≥ 1 and R ≥ 1.

At the base station 110, a transmit processor 220 may receive data from a data source 212 for one or more UEs, select one or more modulation and coding schemes (MCS) for each UE based at least in part on channel quality indicators (CQIs) received from the UE, process (e.g., encode and modulate) the data for each UE based at least in part on the MCS (s) selected for the UE, and provide data symbols for all UEs. Decreasing the MCS lowers throughput but increases reliability of the transmission. The transmit processor 220 may also process system information (e.g., for semi-static resource partitioning information (SRPI) and/or the like) and control information (e.g., CQI requests, grants, upper layer signaling, and/or the like) and provide overhead symbols and control symbols. The transmit processor 220 may also generate reference symbols for reference signals (e.g., the cell-specific reference signal (CRS) ) and synchronization signals (e.g., the primary synchronization signal (PSS) and secondary synchronization signal (SSS) ) . A transmit (TX) multiple-input multiple-output (MIMO) processor 230 may perform spatial processing (e.g., precoding) on the data symbols, the control symbols, the overhead symbols, and/or the reference symbols, if applicable, and may provide T output symbol streams to T modulators (MODs) 232a through 232t. Each modulator 232 may process a respective output symbol stream (e.g., for OFDM and/or the like) to obtain an output sample stream. Each modulator 232 may further process (e.g., convert to analog, amplify, filter, and upconvert) the output sample stream to obtain a downlink signal. T downlink signals from modulators 232a through 232t may be transmitted via T antennas 234a through 234t, respectively. According to various aspects described in more detail below, the synchronization signals can be generated with location encoding to convey additional information.

At the UE 120, antennas 252a through 252r may receive the downlink signals from the base station 110 and/or other base stations and may provide received signals to demodulators (DEMODs) 254a through 254r, respectively. Each demodulator 254 may condition (e.g., filter, amplify, downconvert, and digitize) a received signal to obtain input samples. Each demodulator 254 may further process the input samples (e.g., for OFDM and/or the like) to obtain received symbols. A MIMO detector 256 may obtain received symbols from all R demodulators 254a through 254r, perform MIMO detection on the received symbols if applicable, and provide detected symbols. A receive processor 258 may process (e.g., demodulate and decode) the detected symbols, provide decoded data for the UE 120 to a data sink 260, and provide decoded control information and system information to a controller/processor 280. A channel processor may determine reference signal received power (RSRP) , received signal strength indicator (RSSI) , reference signal received quality (RSRQ) , channel quality indicator (CQI) , and/or the like. In some aspects, one or more components of the UE 120 may be included in a housing.

On the uplink, at the UE 120, a transmit processor 264 may receive and process data from a data source 262 and control information (e.g., for reports comprising RSRP, RSSI, RSRQ, CQI, and/or the like) from the controller/processor 280. Transmit processor 264 may also generate reference symbols for one or more reference signals. The symbols from the transmit processor 264 may be precoded by a TX MIMO processor 266 if applicable, further processed by modulators 254a through 254r (e.g., for DFT-s-OFDM, CP-OFDM, and/or the like) , and transmitted to the base station 110. At the base station 110, the uplink signals from the UE 120 and other UEs may be received by the antennas 234, processed by the demodulators 254, detected by a MIMO detector 236 if applicable, and further processed by a receive processor 238 to obtain decoded data and control information sent by the UE 120. The receive processor 238 may provide the decoded data to a data sink 239 and the decoded control information to a controller/processor 240. The base station 110 may include communications unit 244 and communicate to the network controller 130 via the communications unit 244. The network controller 130 may include a communications unit 294, a controller/processor 290, and a memory 292.

The controller/processor 240 of the base station 110, the controller/processor 280 of the UE 120, and/or any other component (s) of FIGURE 2 may perform one or more techniques associated with machine learning for communicating of neural network configuration parameters, as described in more detail elsewhere. For example, the controller/processor 240 of the base station 110, the controller/processor 280 of the UE 120, and/or any other component (s) of FIGURE 2 may perform or direct operations of, for example, the processes of FIGURES 8-11, and/or other processes as described.

Memories

242 and 282 may store data and program codes for the base station 110 and UE 120, respectively. A scheduler 246 may schedule UEs for data transmission on the downlink and/or uplink.

In some aspects, the UE 120 may include means for segmenting, means for compressing, means for transmitting, and/or means for reporting. In some aspects, the base station 110 may include means for updating parameters, means for segmenting, means for compressing, and/or means for transmitting. Such means may include one or more components of the UE 120 or base station 110 described in connection with FIGURE 2.

As indicated above, FIGURE 2 is provided merely as an example. Other examples may differ from what is described with regard to FIGURE 2.

In some cases, different types of devices supporting different types of applications and/or services may coexist in a cell. Examples of different types of devices include UE handsets, customer premises equipment (CPEs) , vehicles, Internet of Things (IoT) devices, and/or the like. Examples of different types of applications include ultra-reliable low-latency communications (URLLC) applications, massive machine-type communications (mMTC) applications, enhanced mobile broadband (eMBB) applications, vehicle-to-anything (V2X) applications, and/or the like. Furthermore, in some cases, a single device may support different applications or services simultaneously.

FIGURE 3 illustrates an example implementation of a system-on-a-chip (SOC) 300, which may include a central processing unit (CPU) 302 or a multi-core CPU configured for compressing and segmenting neural signaling of neural network parameter, in accordance with certain aspects of the present disclosure. The SOC 300 may be included in the base station 110 or UE 120. Variables (e.g., neural signals and synaptic weights) , system parameters associated with a computational device (e.g., neural network with weights) , delays, frequency bin information, and task information may be stored in a memory block associated with a neural processing unit (NPU) 308, in a memory block associated with a CPU 302, in a memory block associated with a graphics processing unit (GPU) 304, in a memory block associated with a digital signal processor (DSP) 306, in a memory block 318, or may be distributed across multiple blocks. Instructions executed at the CPU 302 may be loaded from a program memory associated with the CPU 302 or may be loaded from a memory block 318.

The SOC 300 may also include additional processing blocks tailored to specific functions, such as a GPU 304, a DSP 306, a connectivity block 310, which may include fifth generation (5G) connectivity, fourth generation long term evolution (4G LTE) connectivity, Wi-Fi connectivity, USB connectivity, Bluetooth connectivity, and the like, and a multimedia processor 312 that may, for example, detect and recognize gestures. In one implementation, the NPU is implemented in the CPU, DSP, and/or GPU. The SOC 300 may also include a sensor processor 314, image signal processors (ISPs) 316, and/or navigation module 320, which may include a global positioning system.

The SOC 300 may be based on an ARM instruction set. In an aspect of the present disclosure, the instructions loaded into the general-purpose processor 302 may comprise program code to segment a UE neural network configuration message to form a radio resource control (RRC) neural network configuration message; program code to compress the RRC neural network configuration message to form a compressed RRC configuration message; and program code to transmit the compressed RRC neural network configuration message to a base station.

In an aspect of the present disclosure, the instructions loaded into the general-purpose processor 302 may comprise program code to compress an RRC neural network configuration message to form a compressed RRC configuration message; and program code to transmit the compressed RRC configuration message to a base station.

In some aspects of the present disclosure, the instructions loaded into the general-purpose processor 302 may comprise program code to update parameters for configuration of an artificial neural network configuration; program code to segment configuration messages including the updated parameters for configuration of the artificial neural network to form segmented configuration messages; program code to compress the segmented configuration messages to form compressed, segmented configuration messages; and program code to transmit, via radio resource control (RRC) signaling, the compressed, segmented configuration messages to a user equipment (UE) .

Deep learning architectures may perform an object recognition task by learning to represent inputs at successively higher levels of abstraction in each layer, thereby building up a useful feature representation of the input data. In this way, deep learning addresses a major bottleneck of traditional machine learning. Prior to the advent of deep learning, a machine learning approach to an object recognition problem may have relied heavily on human engineered features, perhaps in combination with a shallow classifier. A shallow classifier may be a two-class linear classifier, for example, in which a weighted sum of the feature vector components may be compared with a threshold to predict to which class the input belongs. Human engineered features may be templates or kernels tailored to a specific problem domain by engineers with domain expertise. Deep learning architectures, in contrast, may learn to represent features that are similar to what a human engineer might design, but through training. Furthermore, a deep network may learn to represent and recognize new types of features that a human might not have considered.

A deep learning architecture may learn a hierarchy of features. If presented with visual data, for example, the first layer may learn to recognize relatively simple features, such as edges, in the input stream. In another example, if presented with auditory data, the first layer may learn to recognize spectral power in specific frequencies. The second layer, taking the output of the first layer as input, may learn to recognize combinations of features, such as simple shapes for visual data or combinations of sounds for auditory data. For instance, higher layers may learn to represent complex shapes in visual data or words in auditory data. Still higher layers may learn to recognize common visual objects or spoken phrases.

Deep learning architectures may perform especially well when applied to problems that have a natural hierarchical structure. For example, the classification of motorized vehicles may benefit from first learning to recognize wheels, windshields, and other features. These features may be combined at higher layers in different ways to recognize cars, trucks, and airplanes.

Neural networks may be designed with a variety of connectivity patterns. In feed-forward networks, information is passed from lower to higher layers, with each neuron in a given layer communicating to neurons in higher layers. A hierarchical representation may be built up in successive layers of a feed-forward network, as described above. Neural networks may also have recurrent or feedback (also called top-down) connections. In a recurrent connection, the output from a neuron in a given layer may be communicated to another neuron in the same layer. A recurrent architecture may be helpful in recognizing patterns that span more than one of the input data chunks that are delivered to the neural network in a sequence. A connection from a neuron in a given layer to a neuron in a lower layer is called a feedback (or top-down) connection. A network with many feedback connections may be helpful when the recognition of a high-level concept may aid in discriminating the particular low-level features of an input.

The connections between layers of a neural network may be fully connected or locally connected. FIGURE 4A illustrates an example of a fully connected neural network 402. In a fully connected neural network 402, a neuron in a first layer may communicate its output to every neuron in a second layer, so that each neuron in the second layer will receive input from every neuron in the first layer. FIGURE 4B illustrates an example of a locally connected neural network 404. In a locally connected neural network 404, a neuron in a first layer may be connected to a limited number of neurons in the second layer. More generally, a locally connected layer of the locally connected neural network 404 may be configured so that each neuron in a layer will have the same or a similar connectivity pattern, but with connections strengths that may have different values (e.g., 410, 412, 414, and 416) . The locally connected connectivity pattern may give rise to spatially distinct receptive fields in a higher layer, because the higher layer neurons in a given region may receive inputs that are tuned through training to the properties of a restricted portion of the total input to the network.

One example of a locally connected neural network is a convolutional neural network. FIGURE 4C illustrates an example of a convolutional neural network 406. The convolutional neural network 406 may be configured such that the connection strengths associated with the inputs for each neuron in the second layer are shared (e.g., 408) . Convolutional neural networks may be well suited to problems in which the spatial location of inputs is meaningful.

One type of convolutional neural network is a deep convolutional network (DCN) . FIGURE 4D illustrates a detailed example of a DCN 400 designed to recognize visual features from an image 426 input from an image capturing device 430, such as a car-mounted camera. The DCN 400 of the current example may be trained to identify traffic signs and a number provided on the traffic sign. Of course, the DCN 400 may be trained for other tasks, such as identifying lane markings or identifying traffic lights.

The DCN 400 may be trained with supervised learning. During training, the DCN 400 may be presented with an image, such as the image 426 of a speed limit sign, and a forward pass may then be computed to produce an output 422. The DCN 400 may include a feature extraction section and a classification section. Upon receiving the image 426, a convolutional layer 432 may apply convolutional kernels (not shown) to the image 426 to generate a first set of feature maps 418. As an example, the convolutional kernel for the convolutional layer 432 may be a 5x5 kernel that generates 28x28 feature maps. In the present example, because four different feature maps are generated in the first set of feature maps 418, four different convolutional kernels were applied to the image 426 at the convolutional layer 432. The convolutional kernels may also be referred to as filters or convolutional filters.

The first set of feature maps 418 may be subsampled by a max pooling layer (not shown) to generate a second set of feature maps 420. The max pooling layer reduces the size of the first set of feature maps 418. That is, a size of the second set of feature maps 420, such as 14x14, is less than the size of the first set of feature maps 418, such as 28x28. The reduced size provides similar information to a subsequent layer while reducing memory consumption. The second set of feature maps 420 may be further convolved via one or more subsequent convolutional layers (not shown) to generate one or more subsequent sets of feature maps (not shown) .

In the example of FIGURE 4D, the second set of feature maps 420 is convolved to generate a first feature vector 424. Furthermore, the first feature vector 424 is further convolved to generate a second feature vector 428. Each feature of the second feature vector 428 may include a number that corresponds to a possible feature of the image 426, such as “sign, ” “60, ” and “100. ” A softmax function (not shown) may convert the numbers in the second feature vector 428 to a probability. As such, an output 422 of the DCN 400 is a probability of the image 426 including one or more features.

In the present example, the probabilities in the output 422 for “sign” and “60” are higher than the probabilities of the others of the output 422, such as “30, ” “40, ” “50, ” “70, ” “80, ” “90, ” and “100” . Before training, the output 422 produced by the DCN 400 is likely to be incorrect. Thus, an error may be calculated between the output 422 and a target output. The target output is the ground truth of the image 426 (e.g., “sign” and “60” ) . The weights of the DCN 400 may then be adjusted so the output 422 of the DCN 400 is more closely aligned with the target output.

To adjust the weights, a learning algorithm may compute a gradient vector for the weights. The gradient may indicate an amount that an error would increase or decrease if the weight were adjusted. At the top layer, the gradient may correspond directly to the value of a weight connecting an activated neuron in the penultimate layer and a neuron in the output layer. In lower layers, the gradient may depend on the value of the weights and on the computed error gradients of the higher layers. The weights may then be adjusted to reduce the error. This manner of adjusting the weights may be referred to as “back propagation” as it involves a “backward pass” through the neural network.

In practice, the error gradient of weights may be calculated over a small number of examples, so that the calculated gradient approximates the true error gradient. This approximation method may be referred to as stochastic gradient descent. Stochastic gradient descent may be repeated until the achievable error rate of the entire system has stopped decreasing or until the error rate has reached a target level. After learning, the DCN may be presented with new images (e.g., the speed limit sign of the image 426) and a forward pass through the network may yield an output 422 that may be considered an inference or a prediction of the DCN.

Deep belief networks (DBNs) are probabilistic models comprising multiple layers of hidden nodes. DBNs may be used to extract a hierarchical representation of training data sets. A DBN may be obtained by stacking up layers of Restricted Boltzmann Machines (RBMs) . An RBM is a type of artificial neural network that can learn a probability distribution over a set of inputs. Because RBMs can learn a probability distribution in the absence of information about the class to which each input should be categorized, RBMs are often used in unsupervised learning. Using a hybrid unsupervised and supervised paradigm, the bottom RBMs of a DBN may be trained in an unsupervised manner and may serve as feature extractors, and the top RBM may be trained in a supervised manner (on a joint distribution of inputs from the previous layer and target classes) and may serve as a classifier.

Deep convolutional networks (DCNs) are networks of convolutional networks, configured with additional pooling and normalization layers. DCNs have achieved state-of-the-art performance on many tasks. DCNs can be trained using supervised learning in which both the input and output targets are known for many exemplars and are used to modify the weights of the network by use of gradient descent methods.

DCNs may be feed-forward networks. In addition, as described above, the connections from a neuron in a first layer of a DCN to a group of neurons in the next higher layer are shared across the neurons in the first layer. The feed-forward and shared connections of DCNs may be exploited for fast processing. The computational burden of a DCN may be much less, for example, than that of a similarly sized neural network that comprises recurrent or feedback connections.

The processing of each layer of a convolutional network may be considered a spatially invariant template or basis projection. If the input is first decomposed into multiple channels, such as the red, green, and blue channels of a color image, then the convolutional network trained on that input may be considered three-dimensional, with two spatial dimensions along the axes of the image and a third dimension capturing color information. The outputs of the convolutional connections may be considered to form a feature map in the subsequent layer, with each element of the feature map (e.g., 220) receiving input from a range of neurons in the previous layer (e.g., feature maps 218) and from each of the multiple channels. The values in the feature map may be further processed with a non-linearity, such as a rectification, max (0, x) . Values from adjacent neurons may be further pooled, which corresponds to down sampling, and may provide additional local invariance and dimensionality reduction. Normalization, which corresponds to whitening, may also be applied through lateral inhibition between neurons in the feature map.

The performance of deep learning architectures may increase as more labeled data points become available or as computational power increases. Modern deep neural networks are routinely trained with computing resources that are thousands of times greater than what was available to a typical researcher just fifteen years ago. New architectures and training paradigms may further boost the performance of deep learning. Rectified linear units may reduce a training issue known as vanishing gradients. New training techniques may reduce over-fitting and thus enable larger models to achieve better generalization. Encapsulation techniques may abstract data in a given receptive field and further boost overall performance.

FIGURE 5 is a block diagram illustrating a deep convolutional network 550. The deep convolutional network 550 may include multiple different types of layers based on connectivity and weight sharing. As shown in FIGURE 5, the deep convolutional network 550 includes the convolution blocks 554A, 554B. Each of the convolution blocks 554A, 554B may be configured with a convolution layer (CONV) 356, a normalization layer (LNorm) 558, and a max pooling layer (MAX POOL) 560.

The convolution layers 556 may include one or more convolutional filters, which may be applied to the input data to generate a feature map. Although only two of the convolution blocks 554A, 554B are shown, the present disclosure is not so limiting, and instead, any number of the convolution blocks 554A, 554B may be included in the deep convolutional network 550 according to design preference. The normalization layer 558 may normalize the output of the convolution filters. For example, the normalization layer 558 may provide whitening or lateral inhibition. The max pooling layer 560 may provide down sampling aggregation over space for local invariance and dimensionality reduction.

The parallel filter banks, for example, of a deep convolutional network may be loaded on a CPU 302 or GPU 304 of an SOC 300 to achieve high performance and low power consumption. In alternative embodiments, the parallel filter banks may be loaded on the DSP 306 or an ISP 316 of an SOC 300. In addition, the deep convolutional network 550 may access other processing blocks that may be present on the SOC 300, such as sensor processor 314 and navigation module 320, dedicated, respectively, to sensors and navigation.

The deep convolutional network 550 may also include one or more fully connected layers 562 (FC1 and FC2) . The deep convolutional network 550 may further include a logistic regression (LR) layer 564. Between each

layer

556, 558, 560, 562, 564 of the deep convolutional network 550 are weights (not shown) that are to be updated. The output of each of the layers (e.g., 556, 558, 560, 562, 564) may serve as an input of a succeeding one of the layers (e.g., 556, 558, 560, 562, 564) in the deep convolutional network 550 to learn hierarchical feature representations from input data 552 (e.g., images, audio, video, sensor data and/or other input data) supplied at the first of the convolution blocks 554A. The output of the deep convolutional network 550 is a classification score 566 for the input data 552. The classification score 566 may be a set of probabilities, where each probability is the probability of the input data, including a feature from a set of features.

As indicated above, FIGURES 3-5 are provided as examples. Other examples may differ from what is described with respect to FIGURES 3-5.

As noted above, an artificial intelligence (AI) -based end-to-end (E2E) wireless system incorporates neural networks within a transmitter (Tx) and a receiver (Rx) , according to aspects of the present disclosure. In one configuration, a neural network (e.g., a transmitter neural network) replaces encoding, modulation, and/or precoding modules in a transmitter. In addition, a neural network (e.g., a receiver neural network) replaces synchronization, channel estimation, detection, demodulation, and/or decoding modules in the receiver. In some aspects, a neural network (s) replaces one, some, or all transmitting/receiving modules of the AI-based E2E wireless system. According to this aspect of the present disclosure, offline training and online refinement configure the transmitter neural network and the receiver neural network.

Unfortunately, a parameter size for configuration of the neural networks of an AI-based E2E wireless system may be up to tens of megabytes. In current 5G NR systems, a radio resource control (RRC) message is only a few bytes (and not more than a few kilobytes) for an uplink or downlink transmission. Aspects of the present disclosure are directed to segmenting and compressing of signals for configuration of the neural network parameters of an AI-based wireless system in a 5G NR system.

In aspects of the present disclosure, UE parameters of an AI-based wireless system may be configured by a base station (e.g., gNB) . In some aspects, refinement of one or multiple parameters of the neural networks of the AI-based wireless system is performed for improved end-to-end transmission. The neural networks of the AI-based wireless system may be composed of one or any combination of kernels and/or coefficients of the kernels. The combination of kernels and/or coefficients of the kernels may be for a particular convolutional layer, a locally-connected layer, or parameters for a particular dense layer. In one aspect, the signaling configuration from the base station can be based on an explicit indication, a selection from multiple preconfigured choices, and/or implicitly determined from other signaling configuration parameters.

FIGURE 6 is a block diagram illustrating an artificial intelligence (AI) -based end-to-end (E2E) wireless system 600, incorporating neural networks within a transmitter (Tx) 610 and/or a receiver (Rx) 650, according to aspects of the present disclosure. In this example, the AI-based E2E wireless system 600 shows the transmitter 610 of a base station 602 and the receiver 650 of a user equipment (UE) 640. The base station 602 includes a radio resource module 620 communicably coupled to the UE 640 through a wireless channel 630. In this configuration, a neural network (e.g., a transmitter neural network 612) replaces encoding, modulation, and/or precoding modules in the transmitter 610. In addition, a neural network (e.g., a receiver neural network 652) replaces synchronization, channel estimation, detection, demodulation, and/or decoding components in the receiver 650. The transmitter 610 is communicably coupled to the receiver 650 through the wireless channel 630.

Aspects of the present disclosure are directed to configuration of the neural network parameters of the transmitter 610 and the receiver 650 of the AI-based wireless system 600, for example in a 5G NR system. In some aspects, the transmitter neural network 612 and the receiver neural network 652 replace one, some, or all transmitting and receiving modules of the AI-based wireless system 600.

FIGURE 7 is a block diagram illustrating an encoded radio resource control (RRC) message 700. The 5G NR standard (e.g., 3GPP Release-16) specifies a capability for RRC level segmenting of downlink and uplink configuration messages. According to this 5G NR standard, segmenting is limited to the following uplink and downlink RRC configuration messages: (1) a UECapabilityInformation message; (2) an RRCReconfiguration message; and (3) an RRCResume message. As specified by the 5G NR standard, uplink RRC message segmenting is enabled and disabled by an RRC parameter (e.g., rrc-SegAllowed) , when the encoded RRC message 700 is larger than a maximum supported size of a packet data convergence protocol (PDCP) service data unit (SDU) .

Table I--Uplink Dedicated Message Segment Format

Table II--Downlink Dedicated Message Segment Format

Tables I and II illustrate an uplink dedicated message segment format and a downlink dedicated message segment format. In operation, UE message segmenting for each uplink dedicated control channel (DCCH) message is as follows: First, the UE 640 sets a segmentNumber field to 0 for the first message segment and increments the segmentNumber field for each subsequent RRC message segment, as shown in Table 1 and FIGURE 7. Next, the UE 640 sets the rrc-MessageSegmentType field to lastSegment or notLastSegment depending on whether the segment is last. According to the 5G NR standard, the UE 640 is specified to reduce the number of segments and set the segmented uplink RRC message into an ULDedicatedMessageSegment field.

According to aspects of the present disclosure, the UE 640 of FIGURE 6 receives the encoded RRC message 700, including parameters of an artificial neural network. For example, the UE 640 receives the encoded RRC message 700 via RRC signaling from the base station 602 of FIGURE 6. In this example, the encoded RRC message 700 is composed of message segments (e.g., 710, 720, 730, 740) , in which a size of the message segments equals or is less than a maximum packet data convergence protocol (PDCP) service data unit (SDU) size. In this example, RRC message segmenting is enabled to generate a first message segment 710, a second message segment 720, a third message segment 730, and a fourth message segment 740, in which the segment numbers (Seg#) are incremented starting from zero to the last message segment (e.g., the fourth message segment 740) .

Unfortunately, the PDCP SDU size may be insufficient for communicating configuration parameters of the transmitter neural network 612 and the receiver neural network 652 of the AI-based wireless system 600. In particular, a parameter size for configuration of the transmitter neural network 612 and the receiver neural network 652 of the AI-based wireless system 600 may be very large. Aspects of the present disclosure are directed to compressing and segmenting the neural network parameters of the AI-based wireless system 600 to enable communication within a 5G NR system.

In practice, each packet data convergence protocol (PDCP) entity is configured to use either uplink data compression (UDC) or header compression. Uplink data compression provides several benefits. For example, uplink data compression results in the transmission of a reduced number of bits in the uplink channel. This reduced number of bits increases uplink network capacity, while reducing interference--especially at a cell edge. In addition, uplink data compression is transparent to both lower layers and applications.

In operation, the network configures uplink data compression (UDC) for each bearer of the UE. This configuration includes: (1) enable/disable/reset UDC; (2) a dictionary (e.g., an algorithm version to be used for compression) ; and (3) a buffer size (e.g., a memory size per bearer, such as 2k, 4k, or 8k bytes) . The dictionary may be a standard (session information protocol (sip) -session description protocol (SDP) ) dictionary or an operator defined public land mobile network (PLMN) based dictionary.

Uplink data compression may be implemented using a DEFLATE algorithm (e.g., IETF RFC 1951) , which is standardized (e.g., 3GPP Rel-15) using Static Huffman coding to implement the DEFLATE algorithm. The DEFLATE algorithm is applied to uplink channel data to form uplink data compression (UDC) packets. For example, one UDC packet is associated with one PDCP SDU, for example, as shown in FIGURE 7. In this example, a UDC packet contains one byte of UDC header information plus deflated UDC data. It is noted that the current uplink data compression is only employed for LTE and is not employed for 5G NR communications. Moreover, the compression is for data channels, rather than signaling channels. Finally, the compression is currently limited to uplink communications.

Referring again to FIGURE 6, a parameter size for configuration of the transmitter neural network 612 and the receiver neural network 652 of the AI-based wireless system 600 may be up to tens of megabytes. According to the 5G NR specification, the RRC message segments (e.g., 710, 720, 730, 740 of FIGURE 7) are usually just a few bytes (and not more than a few kilobytes) of an uplink/downlink transmission. Unfortunately, this PDCP SDU size may be insufficient for communicating configuration parameters of the transmitter neural network 612 and the receiver neural network 652 of the AI-based wireless system 600.

As noted, conventional uplink data compression is limited to user plane data associated with a dedicated radio bearer (DRB) . That is, conventional uplink data compression and header compression are not applicable to radio resource control (RRC) signaling of neural network configuration parameters having a size of up to tens of megabytes. Aspects of the present disclosure are directed to compression and segmenting of neural network configuration parameters of the AI-based wireless systems to enable 5G NR communication .

FIGURE 8 is a call flow diagram 800 illustrating radio resource control (RRC) message segmenting and compression for uplink communication of neural network configuration parameters, according to aspects of the present disclosure. In some aspects, a user equipment (UE) receives configuration messages, including parameters of an artificial neural network, via radio resource control (RRC) signaling. In the call flow diagram 800, a base station (e.g., gNB) 850 transmits an enquiry message 810 to a UE 802. The enquiry message 810 may seek capability information of the UE 802 (e.g., a UECapabilityEnquiry message) . The UE 802 responds, with a response message 812, including UE capability information (e.g., a UECapabilityInformation message) . For example, the response message 812 may indicate whether the UE 802 supports radio resource control (RRC) segmenting (see FIGURE 7) and/or radio resource control (RRC) compression. The UE capability information also indicates whether PDCP level compression is supported.

In some aspects, the base station 850 transmits a UE neural network (NN) configuration enquiry message 814 to the UE 802. In this example, the UE 802 decodes the UE neural network configuration enquiry message 814 at block 820. Based on the decoding, the UE 802 may report the updated neural configuration parameters to the network. For example, in response to the message decoding at block 820, the UE 802 may generate a UE neural network configuration response message at block 822, based on the decoding at block 820. The UE neural network configuration response message generated at block 822 may include information regarding a UE neural network configuration update based on the message decoding at block 820.

In conventional operation, the UE 802 transmits the UE neural network configuration response message of block 822 using an uplink RRC signaling message of the neural network configuration parameters. As noted above, however, RRC signaling messages usually consume just a few bytes in an uplink or downlink channel of current 5G NR systems. In one aspect of the present disclosure, an RRC message segmenting is described for configuring uplink RRC messages for communicating updated neural network configuration parameters.

In some aspects, a new signaling radio bearer (for example, SRB4) is introduced for specific RRC messages with a large size for supporting neural network configuration. In this aspect of the present disclosure, the proposed RRC segmenting automatically occurs for uplink transmission using the new signaling radio bearer, without any configuration. In some aspects, however, the network may prefer a selective segmenting mechanism. According to this alternative configuration, the network independently enables and disables the uplink RRC segmenting for communicating the neural network configuration parameters.

In one aspect, a semi-persistent configuration enables and disables uplink RRC segmenting. For example, semi-persistent configuration may rely on a new information element (e.g., rrc-SegAllowed2) , which is similar to the legacy information element rrc-SegAllowed contained in the UECapabilityEnquiry message 810. According to aspects of the present disclosure, semi-persistent configuration enables a one-time UE neural network configuration. In another aspect, a dynamic configuration mechanism is provided using new RRC dedicated signaling. In a further aspect, a legacy information element (e.g., rrc-SegAllowed) is reused from the UECapabilityEnquiry message 810.

Referring again to FIGURE 8, at block 824, the UE segments at the RRC layer and the packet data convergence protocol (PDCP) compresses the UE neural network configuration response message of block 822. Once generated, a compressed neural network configuration response message of block 830 is transmitted on an uplink channel to the base station 850. In some aspects, a legacy uplink data compression (UDC) mechanism compresses uplink RRC messages.

For example, the legacy UDC mechanism may be applied to specific uplink RRC messages to perform PDCP compression, such as the UE neural network configuration response message of block 830. In some aspects, the legacy UDC mechanism is applied to compress the RRC messages transmitted in the new signaling radio bearer (e.g., SRB4) . If uplink data compression is not configured, however, a PDCP entity submits the PDCP SDU of the RRC message, without compression, to a lower layer of a 5G NR communication stack for transmission.

FIGURE 9 is a call flow diagram 900 illustrating radio resource control (RRC) message segmenting and packet data convergence protocol (PDCP) compression for downlink communication of neural network configuration parameters, according to aspects of the present disclosure. In some aspects, a UE 902 receives configuration messages, including parameters of an artificial neural network, via radio resource control (RRC) signaling. In the call flow diagram 900, a base station 950 transmits an enquiry message 910 to the UE 902. The enquiry message 910 may seek capability information of the UE 902 (e.g., a UECapabilityEnquiry message) . The UE 902 responds, with a response message 912, including UE capability information (e.g., a UECapabilityInformation message) . For example, the response message 912 may indicate whether the UE 902 supports radio resource control segmenting (see FIGURE 7) downlink signaling compression (DSC) and/or PDCP compression.

In some aspects, the network updates UE neural network configuration parameters. For example, the network (re) configures the UE neural network and communicates the update using a specific downlink RRC message, such as a UE neural network configuration update message 916. In this example, the network updates the parameters of a UE neural network configuration based on a UE request, such as the UE neural network configuration request message 914 from the UE 902. Alternatively, the network updates the parameters of the UE neural network configuration in response to triggering of a network (re) configuration event.

In some aspects, the network performs segmenting and PDCP compression of downlink RRC messages for communicating the neural network configuration parameters. In one aspect, downlink signaling compression (DSC) compresses the downlink RRC messages. For example, the downlink signaling compression protocol may be implemented by reusing a legacy uplink data compression (UDC) algorithm, such as the DEFLATE algorithm, for compressing the downlink RRC message including the neural network configuration parameters. In one aspect, downlink signaling compression (DSC) compresses messages on the new signaling radio bearer (e.g., SRB4) , which is dedicated to communication of downlink RRC messages containing neural network configuration parameters.

Similar to the UDC protocol, the DSC protocol generates DSC packets, each associated with one PDCP SDU. The PDCP SDU contains either a completed downlink RRC message or a downlink RRC message segment. The PDCP entity associated with the new signaling radio bearer (SRB) for a specific downlink RRC message (e.g., parameters of a neural network (NN) configuration/indication) can be configured by an upper layer to use DSC. The network should enable the DSC protocol before transmitting the compressed PDCP SDU. The UE reports whether it supports the DSC protocol in a UE capability message, before the network configures the UE with the DSC protocol.

Aspects of the present disclosure are directed to RRC level compression of specific uplink and downlink RRC messages (e.g. neural network configuration parameters) . For example, RRC level compression may be performed by reusing the well-known DEFLATE algorithm. Other compression algorithms are also contemplated according to aspects of the present disclosure.

In some aspects, RRC message compression is enabled and disabled for the noted new signaling radio bearer (for example, SRB4) for each UE. For example, the new signaling radio bearer may be associated with a new information element (IE) indicating whether compression is enabled for the signaling radio bearer. According to this aspect, each RRC messages transmitted in the new signaling radio bearer is compressed when compression is enabled.

FIGURE 10 is a block diagram illustrating an encoded radio resource control (RRC) message, according to aspects of the present disclosure. In this example, an encoded RRC message 1000 is composed of message segments (e.g., 1010, 1020, 1030, 1040) . In this example, segmenting of the encoded RRC message 1000 is enabled to generate a first message segment 1010, a second message segment 1020, a third message segment 1030, and a fourth message segment 1040. In one aspect, segmenting of the encoded RRC message 1000 includes incrementing the segment numbers (Seg#) starting from zero to the last message segment (e.g., the fourth message segment 1040) .

In some aspects, the RRC level compression of specific uplink and downlink RRC messages, such as the encoded RRC message 1000, is selectively indicated for each RRC message. To enable the message-by-message compression, in one aspect, each RRC message includes a compression indication. For example, the first message segment 1010 includes a compression field 1012 ( ‘Co’ ) . For example, if ‘Co’ is set to one (1) , the compression field 1012 indicates the entire RRC message is compressed. Conversely, if ‘Co’ is set to zero (1) , the compression field 1012 indicates the entire RRC message is not compressed. In addition, the second message segment 1020 includes a compression field 1022. The third message segment 1030 includes a compression field 1032. Similarly, the fourth message segment 1040 includes a compression field 1042. In this example, RRC level compression is selectively enabled and disabled for the message segments of the encoded RRC message 1000 using the compression fields (e.g., 1012, 1022, 1032, 1042) .

In one aspect, a modification is performed to an RRC segment type information element field (e.g., rrc-MessageSegmentType-r16) . In this aspect of the present disclosure, the RRC segment type information element field is modified to become a two-bit enumerated field. A pseudo code modification is as follows: 2-bit ENUMERATED {notLastSegment, notCompressed, Compressed, lastSegment} . In this example, the first message segment 1010 includes a segment type field 1014, and the second message segment 1020 includes a segment type field 1024. Similarly, the third message segment 1030 includes a segment type field 1034, and the fourth message segment 1040 includes a segment type field 1044. In this example, RRC level compression is selectively enabled and disabled for the message segments of the encoded RRC message 1000 using segment type fields (e.g., 1014, 1024, 1034, 1044) . For example, if the entire RRC message is compressed, the segment type field (e.g., 1014, 1024, 1034) is set to ‘Compressed. ’ By contrast, if the entire RRC message is not compressed, the segment type field (e.g., 1014, 1024, 1034) is set to ‘notCompressed. ’ In this example, the segment type field 1044 is set to ‘lastSegment. ’

In other aspects, a modification is performed to an RRC segment number field (e.g., segmentNumber) . In this aspect of the present disclosure, a special value is provided in the segment number field. For example, a zero may be assigned for not compressed, and a maximum value may be assigned to indicate the message is compressed. In this example, the first message segment 1010 includes a segment number field 1016, and the second message segment 1020 includes a segment number field 1026. Similarly, the third message segment 1030 includes a segment number field 1036, and the fourth message segment 1040 includes a segment number field 1046. In one aspect, RRC level compression is selectively enabled and disabled for the entire RRC message. For example, For example, the special value (e.g., zero/Maximum value) is used in the first message segment 1010 to indicate whether the entire RRC message is compressed or not. In this example, the second message segment 1020 starts numbering from one, and so on, for the third message segment 1030 and the fourth message segment 1040.

FIGURE 11 is a flow chart illustrating a method for compressing and segmenting of radio resource control (RRC) messages for uplink communication of neural network configuration parameters, according to aspects of the present disclosure. A method 1100 begins at block 1102, in which a UE determines whether RRC level compression is configured. For example, as shown FIGURE 10, RRC level compression may be selectively enabled and disabled for the message segments of the encoded RRC message 1000 using the compression fields (e.g., 1012, 1022, 1032, 1042) or with some other indicator. When RRC level compression is configured, RRC message compression is performed at block 1104. Otherwise, control flow branches to block 1106. At block 1106, the UE determines whether RRC message segmenting is configured. When configured, RRC message segmenting is performed, as shown in FIGURE 10 at block 1108. At block 1110, a segmented and compressed RRC message is submitted to a lower layer for wireless transmission.

FIGURE 12 is a diagram illustrating a process 1200 performed, for example, by a UE, in accordance with various aspects of the present disclosure. The process 1200 is an example of a 5G new radio (NR) UE enhancement of segmenting and compressing signals for communicating parameters of a neural network configuration.

As shown in FIGURE 12, in some aspects, the process 1200 includes segmenting a UE neural network configuration message to form a radio resource control (RRC) neural network configuration message (block 1202) . For example, the UE (e.g., using the controller/processor 280, and/or the memory 282) can segment the UE neural network configuration message to form the RRC neural network configuration message. In some aspects, the process 1200 also includes compressing the RRC neural network configuration message to form a compressed RRC configuration message (block 1204) . For example, the UE (e.g., using the controller/processor 280, and/or the memory 282) can compress the RRC neural network configuration message to form the compressed RRC configuration message.

In some aspects, the process 1200 further includes transmitting the compressed RRC neural network configuration message to a base station (block 1206) . For example, the UE (e.g., using the antenna 252, the DEMOD/MOD 254, the MIMO detector 258, the receive processor 258, the controller/processor 280, and/or the memory 282) can transmit the compressed RRC neural network configuration message to the base station. In some aspects, the process 1200 further includes reporting a capability to decompress in accordance with a downlink signaling protocol.

In some aspects, the process 1200 also includes receiving an indication to operate in accordance with the downlink signaling protocol. In some aspects, the process 1200 further includes receiving a compressed downlink RRC message for neural network configuration, the compression in accordance with the downlink signaling compression protocol.

In some aspects, the process 1200 further includes transmitting an indication to enable RRC segmentation according to a field within the PDCP SDU payload. In some aspects, the process 1200 also includes receiving an indicator for enabling compression, within a compression field of an RRC message segment information element. In some aspects, the process 1200 further includes receiving an indicator for enabling compression, within a three value RRC segment type information element.

FIGURE 13 is a diagram illustrating an example process 1300 performed, for example, by a base station, in accordance with various aspects of the present disclosure. The example process 1300 is an example of a 5G new radio (NR) base station enhancement of segmenting and compressing signals for communicating parameters of a neural network configuration.

As shown in FIGURE 13, in some aspects, the process 1300 includes updating parameters for configuration of an artificial neural network (block 1302) . For example, the base station (e.g., using the controller/processor 240, and/or the memory 242) can update the parameters for configuration of the artificial neural network. In some aspects, the process 1300 also includes segmenting configuration messages including the updated parameters for configuration of the artificial neural network to form segmented configuration messages (block 1304) . For example, the base station (e.g., using the controller/processor 240, and/or the memory 242) can segment the configuration messages including the updated parameters of the artificial neural network.

In some aspects, the process 1300 further includes compressing the segmented configuration messages to form compressed, segmented configuration messages (block 1306) . For example, the base station (e.g., using the controller/processor 240, and/or the memory 242) can compress the segmented configuration messages including the updated parameters of the artificial neural network. In some aspects, the process 1300 also includes transmitting, via radio resource control (RRC) signaling, the compressed, segmented configuration messages to a user equipment (UE) (block 1308) . For example, the base station (e.g., using the antenna 234, the TX MIMO processor 230, the transmit processor 220, the controller/processor 240, and/or the memory 242) can transmit the compressed, segmented configuration messages to the UE.

The foregoing disclosure provides illustration and description, but is not intended to be exhaustive or to limit the aspects to the precise form disclosed. Modifications and variations may be made in light of the above disclosure or may be acquired from practice of the aspects.

As used, the term “component” is intended to be broadly construed as hardware, firmware, and/or a combination of hardware and software. As used, a processor is implemented in hardware, firmware, and/or a combination of hardware and software.

Some aspects are described in connection with thresholds. As used, satisfying a threshold may, depending on the context, refer to a value being greater than the threshold, greater than or equal to the threshold, less than the threshold, less than or equal to the threshold, equal to the threshold, not equal to the threshold, and/or the like.

It will be apparent that systems and/or methods described may be implemented in different forms of hardware, firmware, and/or a combination of hardware and software. The actual specialized control hardware or software code used to implement these systems and/or methods is not limiting of the aspects. Thus, the operation and behavior of the systems and/or methods were described without reference to specific software code-it being understood that software and hardware can be designed to implement the systems and/or methods based, at least in part, on the description.

Even though particular combinations of features are recited in the claims and/or disclosed in the specification, these combinations are not intended to limit the disclosure of various aspects. In fact, many of these features may be combined in ways not specifically recited in the claims and/or disclosed in the specification. Although each dependent claim listed below may directly depend on only one claim, the disclosure of various aspects includes each dependent claim in combination with every other claim in the claim set. A phrase referring to “at least one of” a list of items refers to any combination of those items, including single members. As an example, “at least one of: a, b, or c” is intended to cover a, b, c, a-b, a-c, b-c, and a-b-c, as well as any combination with multiples of the same element (e.g., a-a, a-a-a, a-a-b, a-a-c, a-b-b, a-c-c, b-b, b-b-b, b-b-c, c-c, and c-c-c or any other ordering of a, b, and c) .

No element, act, or instruction used should be construed as critical or essential unless explicitly described as such. Also, as used, the articles “a” and “an” are intended to include one or more items, and may be used interchangeably with “one or more. ” Furthermore, as used, the terms “set” and “group” are intended to include one or more items (e.g., related items, unrelated items, a combination of related and unrelated items, and/or the like) , and may be used interchangeably with “one or more. ” Where only one item is intended, the phrase “only one” or similar language is used. Also, as used, the terms “has, ” “have, ” “having, ” and/or the like are intended to be open-ended terms. Further, the phrase “based on” is intended to mean “based, at least in part, on” unless explicitly stated otherwise.

Claims

A method of wireless communication by a user equipment (UE) , comprising:

segmenting a UE neural network configuration message to form a radio resource control (RRC) neural network configuration message;

compressing the RRC neural network configuration message to form a compressed RRC configuration message; and

transmitting the compressed RRC neural network configuration message to a base station.
The method of claim 1, further comprising transmitting the compressed RRC neural network configuration message in a signaling radio bearer (SRB) dedicated to neural network configuration.
The method of claim 2, in which the segmenting is performed in the SRB.
The method of claim 1, in which the compressed RRC neural network configuration message comprises a plurality of message segments, each segment within a PDCP (packet data convergence protocol) SDU (service data unit) payload.
The method of claim 1, comprising receiving an indication from the base station to enable RRC segmenting.
The method of claim 5, in which the indication to enable RRC segmenting is a dynamic configuration.
The method of claim 5, in which the indication to enable RRC segmenting is a semi-persistent configuration.
The method of claim 1, further comprising:

reporting a capability to decompress in accordance with a downlink signaling compression protocol;

receiving an indication to operate in accordance with the downlink signaling protocol; and

receiving a compressed downlink RRC message for neural network configuration, the compression in accordance with the downlink signaling compression protocol.
A method of wireless communication by a user equipment (UE) , comprising:

compressing an RRC neural network configuration message to form a compressed RRC configuration message; and

transmitting the compressed RRC configuration message to a base station.
The method of claim 9, further comprising segmenting the compressed RRC configuration message into PDCP SDUs prior to transmitting.
The method of claim 9, in which the compression is enabled or disabled for an SRB.
The method of claim 9, in which the compression is enabled or disabled for each RRC message.
The method of claim 12, further comprising receiving an indicator for enabling compression, within a compression field of an RRC message segment information element.
The method of claim 12, further comprising receiving an indicator for enabling compression, within a three value RRC segment type information element.
The method of claim 12, further comprising receiving an indicator for enabling compression, as a special value in a segment number field of an SDU.
A method of wireless communication by a base station, comprising:

updating parameters for configuration of an artificial neural network;

segmenting configuration messages including the updated parameters for configuration of the artificial neural network to form segmented configuration messages;

compressing the segmented configuration messages to form compressed, segmented configuration messages; and

transmitting, via radio resource control (RRC) signaling, the compressed, segmented configuration messages to a user equipment (UE) .
The method of claim 16, further comprising transmitting the compressed, segmented configuration messages in a signaling radio bearer (SRB) dedicated to neural network configuration.
The method of claim 17, further comprising transmitting the SRB over a dedicated control channel (DCCH) logical channel.
The method of claim 16, in which segmenting comprising segmenting the compressed, segmented configuration messages into a plurality of message segments, each segment within a PDCP (packet data convergence protocol) SDU (service data unit) payload.
The method of claim 19, comprising transmitting an indication to enable RRC segmentation according to a field within the PDCP SDU payload.
An apparatus for wireless communication by a user equipment (UE) , comprising:

means for segmenting a UE neural network configuration message to form a radio resource control (RRC) neural network configuration message;

means for compressing the RRC neural network configuration message to form a compressed RRC configuration message; and

means for transmitting the compressed RRC neural network configuration message to a base station.
The apparatus of claim 21, further comprising means for transmitting the compressed RRC neural network configuration message in a signaling radio bearer (SRB) dedicated to neural network configuration.
The apparatus of claim 22, in which the means for segmenting is performed in the SRB.
The apparatus of claim 21, in which the compressed RRC neural network configuration message comprises a plurality of message segments, each segment within a PDCP (packet data convergence protocol) SDU (service data unit) payload.
The apparatus of claim 21, comprising means for receiving an indication from the base station to enable RRC segmenting.
The apparatus of claim 25, in which the indication to enable RRC segmenting is a dynamic configuration.
The apparatus of claim 25, in which the indication to enable RRC segmenting is a semi-persistent configuration.
The apparatus of claim 21, further comprising:

means for reporting a capability to decompress in accordance with a downlink compression signaling protocol;

means for receiving an indication to operate in accordance with the downlink signaling protocol; and

means for receiving a compressed downlink RRC message for neural network configuration, the compression in accordance with the downlink signaling compression protocol.
A user equipment (UE) , comprising:

a processor;

a memory coupled with the processor; and

instructions stored in the memory and operable, when executed by the processor, to cause the UE:

to segment a UE neural network configuration message to form a radio resource control (RRC) neural network configuration message;

to compress the RRC neural network configuration message to form a compressed RRC configuration message; and

to transmit the compressed RRC neural network configuration message to a base station.
The UE of claim 29, in which the instructions further cause the UE to transmit the compressed RRC neural network configuration message in a signaling radio bearer (SRB) dedicated to neural network configuration.
The UE of claim 30, in which the instructions to segment are performed in the SRB.
The UE of claim 29, in which the compressed RRC neural network configuration message comprises a plurality of message segments, each segment within a PDCP (packet data convergence protocol) SDU (service data unit) payload.
The UE of claim 29, in which the instructions further cause the UE to receive an indication from the base station to enable RRC segmenting.
The UE of claim 33, in which the indication to enable RRC segmenting is a dynamic configuration.
The UE of claim 33, in which the indication to enable RRC segmenting is a semi-persistent configuration.
The UE of claim 29, in which the instructions further cause the UE:

to report a capability to decompress in accordance with a downlink compression signaling protocol;

to receive an indication to operate in accordance with the downlink signaling protocol; and

to receive a compressed downlink RRC message for neural network configuration, the compression in accordance with the downlink signaling compression protocol.
A non-transitory computer-readable medium having program code recorded thereon, the program code executed by a processor and comprising:

program code to segment a UE neural network configuration message to form a radio resource control (RRC) neural network configuration message;

program code to compress the RRC neural network configuration message to form a compressed RRC configuration message; and

program code to transmit the compressed RRC neural network configuration message to a base station.
A apparatus for wireless communication by a user equipment (UE) , comprising:

means for compressing an RRC neural network configuration message to form a compressed RRC configuration message; and

means for transmitting the compressed RRC configuration message to a base station.
The apparatus of claim 38, further comprising means for segmenting the compressed RRC configuration message into PDCP SDUs prior to transmitting.
The apparatus of claim 38, in which the compression is enabled or disabled for an SRB.
The apparatus of claim 38, in which the compression is enabled or disabled for each RRC message.
The apparatus of claim 41, further comprising means for receiving an indicator for enabling compression, within a compression field of an RRC message segment information element.
The apparatus of claim 41, further comprising means for receiving an indicator for enabling compression, within a three value RRC segment type information element.
The apparatus of claim 41, further comprising means for receiving an indicator for enabling compression, as a special value in a segment number field of an SDU.
A user equipment (UE) , comprising:

a processor;

a memory coupled with the processor; and

instructions stored in the memory and operable, when executed by the processor, to cause the UE:

to compress an RRC neural network configuration message to form a compressed RRC configuration message; and

to transmit the compressed RRC configuration message to a base station.
The UE of claim 45, in which the instructions further cause the UE to segment the compressed RRC configuration message into PDCP SDUs prior to transmitting.
The UE of claim 45, in which the compression is enabled or disabled for an SRB.
The UE of claim 45, in which the compression is enabled or disabled for each RRC message.
The UE of claim 48, in which the instructions further cause the UE to receive an indicator for enabling compression, within a compression field of an RRC message segment information element.
The UE of claim 48, in which the instructions further cause the UE to receive an indicator for enabling compression, within a three value RRC segment type information element.
The UE of claim 48, in which the instructions further cause the UE to receive an indicator for enabling compression, as a special value in a segment number field of an SDU.
A non-transitory computer-readable medium having program code recorded thereon, the program code executed by a processor and comprising:

program code to compress an RRC neural network configuration message to form a compressed RRC configuration message; and

program code to transmit the compressed RRC configuration message to a base station.
An apparatus for wireless communication by a base station, comprising:

means for updating parameters for configuration of an artificial neural network;

means for segmenting configuration messages including the updated parameters for configuration of the artificial neural network to form segmented configuration messages;

means for compressing the segmented configuration messages to form compressed, segmented configuration messages; and

means for transmitting, via radio resource control (RRC) signaling, the compressed, segmented configuration messages to a user equipment (UE) .
The apparatus of claim 53, further comprising means for transmitting the compressed, segmented configuration messages in a signaling radio bearer (SRB) dedicated to neural network configuration.
The apparatus of claim 54, further comprising means for transmitting the SRB over a dedicated control channel (DCCH) logical channel.
The apparatus of claim 53, in which the means for segmenting comprising means for segmenting the compressed, segmented configuration messages into a plurality of message segments, each segment within a PDCP (packet data convergence protocol) SDU (service data unit) payload.
The apparatus of claim 56, further comprising means for transmitting an indication to enable RRC segmentation according to a field within the PDCP SDU payload.
A base station, comprising:

a processor;

a memory coupled with the processor; and

instructions stored in the memory and operable, when executed by the processor, to cause the base station:

to update parameters for configuration of an artificial neural network;

to segment configuration messages including the updated parameters for configuration of the artificial neural network to form segmented configuration messages;

to compress the segmented configuration messages to form compressed, segmented configuration messages; and

to transmit, via radio resource control (RRC) signaling, the compressed, segmented configuration messages to a user equipment (UE) .
The base station of claim 58, in which the instructions further cause the base station to transmit the configuration messages in a signaling radio bearer (SRB) dedicated to neural network configuration.
The base station of claim 59, in which the instructions further cause the base station to transmit the SRB over a dedicated control channel (DCCH) logical channel.
The base station of claim 58, in which the instructions to segment further causes the base station to segment the configuration messages into a plurality of message segments, each segment within a PDCP (packet data convergence protocol) SDU (service data unit) payload.
The base station of claim 61, in which the instructions further cause the base station to transmit an indication to enable RRC segmentation according to a field within the PDCP SDU payload.
A non-transitory computer-readable medium having program code recorded thereon, the program code executed by a processor and comprising:

program code to update parameters for configuration of an artificial neural network;

program code to segment configuration messages including the updated parameters for configuration of the artificial neural network to form segmented configuration messages;

program code to compress the segmented configuration messages to form compressed, segmented configuration messages; and

program code to transmit, via radio resource control (RRC) signaling, the compressed, segmented configuration messages to a user equipment (UE) .