Base station-user equipment messaging for deep neural networks

Abstract

Techniques and apparatus for implementing base station-user equipment messaging regarding deep neural networks are described. A network entity (base station 121, core network server 320) determines a neural network formation configuration (architecture and / or parameter configuration 1208) for a deep neural network (deep neural network 604, 608, 612, 616) for processing communications transmitted over a wireless communication system. The network entity (base station 121, core network server 302) transmits the neural network formation configuration to a user equipment (UE 110). The user equipment (UE 110) configures a first neural network (deep neural network 608, 612) based on the neural network formation configuration. In an embodiment, the user equipment (UE 110) uses the first neural network (deep neural network 608, 612) to recover information transmitted over the wireless network. This allows the wireless communication system to adapt to changing operating conditions and improve information recovery.

Description

Background Art

[0001] The development of wireless communication systems generally stems from the demand for data throughput. As an example, as more and more devices gain access to wireless communication systems, the demand for data increases. The evolving devices also execute data-intensive applications that utilize more data than traditional applications such as data-intensive streaming video applications, data-intensive social media applications, data-intensive audio services, etc. This increased demand sometimes exhausts the data resources of the wireless communication system. Thus, in order to accommodate the increased data usage, evolving wireless communication systems utilize increasingly complex architectures as a way to provide more data throughput relative to traditional wireless communication systems.

[0002] As an example, fifth generation (5G) standards and technologies use higher frequency ranges, such as bands above 6 gigahertz (GHz), to transmit data to increase data capacity. However, using these higher frequency ranges to transmit and recover information brings challenges. For illustration, higher frequency signals are more susceptible to multipath fading, scattering, atmospheric absorption, diffraction, interference, etc. relative to lower frequency signals. These signal distortions often lead to errors when the information is recovered at the receiver. As another example, hardware capable of transmitting, receiving, routing and / or otherwise using these higher frequencies can be complex and expensive, which increases the processing cost of wireless networking devices. Summary of the invention

[0003] This document describes techniques and apparatus for base station-user equipment messaging regarding a deep neural network (DNN). In some aspects, a network entity determines a neural network (NN) formation configuration for the DNN to process communications transmitted over a wireless communication system, wherein the NN formation configuration specifies any combination of a neural network architecture configuration and / or a parameter configuration as further described. In an embodiment, the network entity generates a message including an indication of the NN formation configuration for the DNN. In turn, the network entity transmits the message to a user equipment to direct the user equipment to form the DNN using the neural network formation configuration, and to use the DNN to process communications transmitted over the wireless communication system. By doing so, the network entity can monitor changing operating conditions in the wireless communication system, and determine a DNN based on the changing operating conditions, thereby improving information recovery in the wireless communication system.

[0004] Some aspects of transmitting a neural network formation configuration (NN formation configuration) utilize a neural network table. A network entity in a wireless communication system transmits a neural network table including a plurality of NN formation configuration elements to a user device, each of the plurality of NN formation configuration elements configuring at least a portion of a DNN to process communications transmitted through the wireless communication system. The network entity then selects one or more neural network formation configuration elements from the plurality of NN formation configuration elements to create the NN formation configuration, and transmits an indication to the user device to direct the user device to form a deep neural network using the neural network formation configuration and to process communications using the deep neural network. In an exemplary arrangement, the network entity determines an index value of the neural network table corresponding to the selected NN formation configuration element. In turn, the network entity transmits the index value to the user device to direct the user device to form a DNN using the NN formation configuration and to process communications using the DNN. By transmitting the neural network table to the user device, the network entity can quickly reconfigure the DNN at the user device when operating conditions change, such as by transmitting index values ​​to the neural network table instead of transmitting individual parameter configurations, such as, by way of example and not limitation, pooling parameters, kernel parameters, weights and / or layer parameters, etc., to configure the DNN. This reduces data transmission between network entities and user equipment, improves responsiveness of the user equipment, and improves information recovery in the wireless communication system.

[0005] In some aspects, a user device associated with a wireless communication system receives a message indicating a NN formation configuration for a DNN to process communications transmitted through the wireless communication system. In turn, the user device forms the DNN using the NN formation configuration indicated in the message. The user device receives communications from a base station and processes the communications using the DNN to extract information transmitted in the communications. By configuring the DNN based on input from a network entity, the user device can improve information recovery in the wireless communication system.

[0006] In some aspects, a user device in a wireless communication system receives a neural network table including a plurality of NN formation configuration elements, the plurality of NN formation configuration elements providing the user device with the ability to configure a DNN for processing communications transmitted through the wireless communication system. The user device receives a message directing the user device to form a DNN using an NN formation configuration based on one or more of the plurality of NN formation configuration elements. The user device forms a DNN having the NN formation configuration by accessing the neural network table to obtain the one or more NN formation configuration elements, and uses the DNN to process communications transmitted through the wireless communication system. By receiving, storing, and accessing the neural network table to configure the DNN, the user device can quickly respond to guidance to reconfigure the DNN and improve information recovery in the wireless communication system.

[0007] Details of one or more embodiments of base station-user equipment messaging using DNN and transmitting neural network formation configurations are set forth in the accompanying drawings and the following description. Other features and advantages will be apparent from the description and drawings, and from the claims. This summary is provided to introduce the subject matter that is further described in the detailed description and drawings. Therefore, this summary should not be regarded as describing necessary features, nor should it be used to limit the scope of the claimed subject matter. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The following describes details of one or more aspects of base station-user equipment message transmission and transmission neural network formation configuration for deep neural networks. The same reference numerals are used in different instances of the specification and drawings to indicate similar elements:

[0009] Figure 1 An exemplary environment is illustrated in which various aspects of base station-user equipment messaging using deep neural networks can be implemented.

[0010] Figure 2 An exemplary device diagram illustrating devices that can implement various aspects of base station-user equipment messaging using deep neural networks.

[0011] Figure 3 An exemplary device diagram of a device that can implement various aspects of base station-user equipment messaging using a deep neural network is illustrated.

[0012] Figure 4 An exemplary machine learning module that can use a deep neural network to implement various aspects of base station-user equipment messaging is illustrated.

[0013] Figure 5 An exemplary block diagram of a processing chain utilized by a device to process communications transmitted over a wireless communication system is illustrated.

[0014] Figure 6 An exemplary operating environment for utilizing multiple deep neural networks in a wireless communication system is illustrated.

[0015] Figure 7 An exemplary transaction graph between various devices that configure a neural network using a neural network formation configuration is illustrated.

[0016] Figure 8 An exemplary transaction graph between various devices that use a neural network formation configuration to update a neural network is illustrated.

[0017] Figure 9-1 and 9-2 An exemplary transaction graph between various devices that configure a neural network using a neural network formation configuration is illustrated.

[0018] Fig.10 An exemplary method for configuring a neural network for processing communications transmitted over a wireless communication system is illustrated.

[0019] Fig.11 An exemplary method for forming a neural network based on a neural network formation configuration is illustrated.

[0020] Fig.12 An example of generating multiple neural network formation configurations is illustrated.

[0021] Fig.13 An exemplary transaction diagram between various devices for communicating a neural network formation configuration is illustrated.

[0022] Fig.14 An exemplary transaction diagram between various network entities for communicating a neural network formation configuration is illustrated.

[0023] Fig.15 An exemplary transaction diagram between various network entities for communicating a neural network formation configuration is illustrated.

[0024] Figure 16-1 and 16-2 An exemplary environment for delivering a neural network formation configuration using a set of candidate neural network formation configurations is illustrated.

[0025] Fig.17 An exemplary transaction diagram between various network entities for communicating a neural network formation configuration is illustrated.

[0026] Fig.18 An exemplary method for transmitting a neural network formation configuration over a wireless communication system is illustrated.

[0027] Fig.19 An exemplary method for transmitting a neural network formation configuration over a wireless communication system is illustrated. DETAILED DESCRIPTION

[0028] In conventional wireless communication systems, the transmitter and receiver processing chains include complex functions. For example, a channel estimation block in the processing chain estimates or predicts how the transmission environment distorts the signal propagating through the transmission environment. As another example, the channel equalizer block inverts the distortion identified from the signal by the channel estimation block. These complex functions typically become more complex when processing higher frequency ranges such as 5G frequencies at or near the 6 GHz range. For example, the transmission environment adds more distortion to the higher frequency ranges relative to the lower frequency ranges, making information recovery more complex. As another example, hardware capable of processing and routing higher frequency ranges typically increases cost and adds complex physical constraints.

[0029] DNNs provide solutions for complex processing, such as complex functions used in wireless communication systems. By training DNNs on transmitter and / or receiver processing chain operations, DNNs can replace conventional complex functions in a variety of ways, such as by replacing some or all conventional processing blocks used in end-to-end processing of wireless communication signals, replacing individual processing chain blocks, etc. Dynamic reconfiguration of DNNs, such as by modifying individual parameter configurations (e.g., coefficients, layer connections, kernel size), also provides the ability to adapt to changing operating conditions.

[0030] This document describes aspects of base station-user equipment messaging and transport NN formation configuration with respect to a DNN, which can be used to handle communications in a wireless communication system and dynamically reconfigure the DNN as operating conditions change. The various aspects can be implemented by a network entity (e.g., a base station, a core network server, a user equipment) operating in a wireless communication system.

[0031] For example, the network entity determines a NN formation configuration for a DNN to process communications transmitted over a wireless communication system. Here, the phrase "transmitting over" includes generating communications to be transmitted over a wireless communication system (e.g., processing pre-transmission communications) and / or processing communications received over a wireless communication system. Therefore, "processing communications transmitted over a wireless communication system" includes generating transmissions, processing received transmissions, or any combination thereof. In an embodiment, the network entity generates a message including an indication of a NN formation configuration for the DNN. The network entity then transmits the message to a user device to guide the user device to form a DNN using the NN formation configuration, and to process communications transmitted over the wireless communication system using the DNN. By doing so, the network entity can monitor changing operating conditions in the wireless communication system, and update the neural network based on the changing operating conditions, thereby improving information recovery in the wireless communication system.

[0032] In some embodiments, a network entity in a wireless communication system transmits to a user device a neural network table including a plurality of NN formation configuration elements, each of the plurality of NN formation configuration elements configuring at least a portion of a DNN for processing communications transmitted through the wireless communication system. The network entity then selects one or more neural network formation configuration elements from the plurality of neural network formation configuration elements to create a neural network formation configuration, and transmits an instruction to the user device to direct the user device to form a DNN using the NN formation configuration and to process communications using the DNN.

[0033] In an exemplary arrangement, a network entity determines an index value of a neural network table corresponding to a selected NN formation configuration element phase. In turn, the network entity transmits the index value to a user device to direct the user device to form a DNN using the NN formation configuration and to process communications using the DNN. By transmitting the neural network table to the user device, the network entity can quickly reconfigure the DNN at the user device when operating conditions change, such as by transmitting an index value of the neural network table to the user device, rather than transmitting a parameter configuration to reconfigure the DNN. This reduces data transmission between the network entity and the user device, improves responsiveness of the user device, and improves information recovery in the wireless communication system.

[0034] In some aspects, a user device associated with a wireless communication system receives a message indicating a NN formation configuration of a DNN for processing communications transmitted through the wireless communication system. In turn, the user device forms the DNN using the NN formation configuration indicated in the message. When the user device receives communications from a base station, the user device extracts information transmitted in the communication by processing the communication using the DNN. By configuring the DNN based on input from a network entity, the user device can improve information recovery in the wireless communication system.

[0035] In some aspects, a user device in a wireless communication system receives a neural network table including a plurality of NN formation configuration elements, the plurality of NN formation configuration elements providing the user device with the ability to configure a DNN for processing communications transmitted through the wireless communication system. The user device receives a message that, based on one or more of the plurality of NN formation configuration elements, instructs the user device to form a DNN using the NN formation configuration. The user device then forms the DNN by accessing the neural network table to obtain the NN formation configuration (e.g., the NN formation configuration element), and uses the DNN to process communications transmitted through the wireless communication system. By receiving, storing, and accessing the neural network table to configure the DNN, the user device can quickly respond to guidance to reconfigure the DNN and improve information recovery in the wireless communication system.

[0036] Example Environment

[0037] Figure 1An exemplary environment 100 is illustrated, which includes a user equipment 110 (UE 110) that can communicate with a base station 120 (shown as base stations 121 and 122) via one or more wireless communication links 130 (wireless links 130), shown as wireless links 131 and 132. For simplicity, UE 110 is implemented as a smart phone, but can also be implemented as any suitable computing or electronic device, such as a mobile communication device, a modem, a cellular phone, a gaming device, a navigation device, a media device, a laptop, a desktop computer, a tablet, a smart device, a vehicle-based communication system, or an Internet of Things (IoT) device, such as a sensor or actuator. Base station 120 (e.g., an evolved universal terrestrial radio access network node B, E-UTRAN node B, evolved node B, eNodeB, eNB, next generation node B, gNodeB, gNB, ng-eNB, etc.) can be implemented in a macro cell, a micro cell, a small cell, a pico cell, etc., or any combination thereof.

[0038] The base station 120 communicates with the user equipment 110 using radio links 131 and 132, which may be implemented as any suitable type of radio links. The radio links 131 and 132 include control and data communications, such as downlinks of data and control information transmitted from the base station 120 to the user equipment 110, uplinks of other data and control information transmitted from the user equipment 110 to the base station 120, or both. The radio link 130 may include one or more radio links (e.g., radio links) or bearers implemented using any suitable communication protocol or standard, or a combination of communication protocols or standards, such as the Third Generation Partnership Project Long Term Evolution (3GPP LTE), the Fifth Generation New Radio (5G NR), etc. Multiple radio links 130 may be aggregated in carrier aggregation to provide higher data rates for the UE 110. Multiple radio links 130 from multiple base stations 120 may be configured for coordinated multi-point (CoMP) communication with the UE 110.

[0039] The base stations 120 collectively serve as a radio access network 140 (e.g., RAN, Evolved Universal Terrestrial Radio Access Network, E-UTRAN, 5G NR RAN, or NR RAN). The base stations 121 and 122 in the RAN 140 are connected to a core network 150. The base stations 121 and 122 are connected to the core network 150 at 102 and 104, respectively, via an NG2 interface for control plane signaling and an NG3 interface for user plane data communications when connected to a 5G core network, or an S1 interface for control plane signaling and user plane data communications when connected to an evolved packet core (EPC) network. The base stations 121 and 122 may communicate using an Xn application protocol (XnAP) over an Xn interface, or using an X2 application protocol (X2AP) over an X2 interface to exchange user plane and control plane data at 106. The user equipment 110 may be connected to a public network such as the Internet 160 for interacting with a remote service 170 via the core network 150.

[0040] Exemplary Devices

[0041] Figure 2 An exemplary device diagram 200 of user equipment 110 and one of base stations 120 is illustrated. Figure 3 An exemplary device diagram 300 of a core network server 302 is illustrated. For clarity, user equipment 110, base station 120, and / or core network server 302 may include Figure 2 or Figure 3 Additional functions and interfaces omitted in .

[0042] The user equipment 110 includes an antenna 202 for communicating with the base station 120 in the RAN 140, a radio frequency front end 204 (RF front end 204), and a wireless transceiver (e.g., an LTE transceiver 206 and / or a 5G NR transceiver 208). The RF front end 204 of the user equipment 110 can couple or connect the LTE transceiver 206 and the 5G NR transceiver 208 to the antenna 202 to facilitate various types of wireless communications. The antenna 202 of the user equipment 110 may include an array of multiple antennas configured to be similar to each other or different from each other. The antenna 202 and the RF front end 204 may be tuned and / or tunable to one or more frequency bands defined by the 3GPP LTE and 5G NR communication standards and implemented by the LTE transceiver 206 and / or the 5G NR transceiver 208. Additionally, antenna 202, RF front end 204, LTE transceiver 206, and / or 5G NR transceiver 208 may be configured to support beamforming for transmitting and receiving communications with base station 120. By way of example and not limitation, antenna 202 and RF front end 204 may be implemented in sub-gigahertz frequency bands, sub-6 GHZ frequency bands, and / or above 6 GHz frequency bands as defined by 3GPP LTE and 5G NR communication standards.

[0043] The user device 110 also includes a processor 210 and a computer readable storage medium 212 (CRM 212). The processor 210 can be a single-core processor or a multi-core processor composed of a variety of materials, such as silicon, polysilicon, high-K dielectrics, copper, etc. The computer readable storage medium described herein does not include propagation signals. The CRM 212 may include any suitable memory or storage device that can be used to store device data 214 of the user device 110, such as a random access memory (RAM), static RAM (SRAM), dynamic RAM (DRAM), non-volatile RAM (NVRAM), read-only memory (ROM), or flash memory. The device data 214 includes user data, multimedia data, beamforming codebooks, applications, neural network tables, and / or an operating system of the user device 110 that can be executed by the processor 210 to enable user plane communications, control plane signaling, and user interaction with the user device 110.

[0044] In some embodiments, the computer-readable storage medium 212 includes a neural network table 216 that stores various architectures and / or parameter configurations for forming a neural network, such as, by way of example and not limitation, parameters specifying a fully connected layer neural network architecture, a convolutional layer neural network architecture, a recursive neural network layer, a plurality of connected hidden neural network layers, an input layer architecture, an output layer architecture, a plurality of nodes utilized by the neural network, coefficients (e.g., weights and biases) utilized by the neural network, kernel parameters, a plurality of filters utilized by the neural network, stride / pooling configurations utilized by the neural network, activation functions for each neural network layer, interconnections between neural network layers, neural network layers to be skipped, etc. Thus, the neural network table 216 includes any combination of NN formation configuration elements (e.g., architecture and / or parameter configurations) that can be used to create a NN formation configuration (e.g., a combination of one or more NN formation configuration elements) that defines and / or forms a DNN. In some embodiments, a single index value of the neural network table 216 maps to a single NN formation configuration element (e.g., a 1:1 correspondence). Alternatively or additionally, a single index value of the neural network table 216 maps to a NN formation configuration (e.g., a combination of NN formation configuration elements). In some embodiments, the neural network table includes input characteristics for each NN formation configuration element and / or NN formation configuration, wherein the input characteristics describe properties of training data used to generate the NN formation configuration element and / or NN formation configuration, as further described.

[0045] In some embodiments, the CRM 212 may also include a user equipment neural network manager 218 (UE neural network manager 218). Alternatively or additionally, the UE neural network manager 218 may be implemented in whole or in part as hardware logic or circuitry that is integrated with or separate from other components of the user equipment 110. The UE neural network manager 218 accesses the neural network table 216, such as by an index value, and forms a DNN using the NN formation configuration elements specified by the NN formation configuration. In an embodiment, the UE neural network manager forms multiple DNNs to process wireless communications (e.g., downlink communications and / or uplink communications exchanged with the base station 120).

[0046] Figure 2The device diagram of the base station 120 shown includes a single network node (e.g., gNode B). The functions of the base station 120 may be distributed across multiple network nodes or devices and may be distributed in any manner suitable for performing the functions described herein. The base station 120 includes an antenna 252, a radio frequency front end 254 (RF front end 254), and one or more wireless transceivers (e.g., one or more LTE transceivers 256 and / or one or more 5G NR transceivers 258) for communicating with the UE 110. The RF front end 254 of the base station 120 may couple or connect the LTE transceiver 256 and the 5G NR transceiver 258 to the antenna 252 to facilitate various types of wireless communications. The antenna 252 of the base station 120 may include an array of multiple antennas that are configured similarly to each other or configured differently from each other. The antenna 252 and the RF front end 254 may be tuned and / or tunable to one or more frequency bands defined by the 3GPP LTE and 5G NR communication standards and implemented by the LTE transceiver 256 and / or the 5G NR transceiver 258. Additionally, the antenna 252 , the RF front end 254 , the LTE transceiver 256 , and / or the 5G NR transceiver 258 may be configured to support beamforming, such as Massive-MIMO, for transmitting and receiving communications with the UE 110 .

[0047] The base station 120 also includes a processor 260 and a computer readable storage medium 262 (CRM 262). The processor 260 may be a single core processor or a multi-core processor composed of a variety of materials such as silicon, polysilicon, high-K dielectrics, copper, etc. The CRM 262 may include any suitable memory or storage device that can be used to store device data 264 of the base station 120, such as a random access memory (RAM), static RAM (SRAM), dynamic RAM (DRAM), non-volatile RAM (NVRAM), read-only memory (ROM), or flash memory. The device data 264 includes network scheduling data, radio resource management data, beamforming codebooks, applications, and / or an operating system of the base station 120 that can be executed by the processor 260 to implement communication with the user equipment 110.

[0048] CRM 262 also includes a base station manager 266. Alternatively or additionally, base station manager 266 can be implemented in whole or in part as hardware logic or circuitry that is integrated or separate from other components of base station 120. In at least some aspects, base station manager 266 configures LTE transceiver 256 and 5G NR transceiver 258 to communicate with user equipment 110 and with a core network such as core network 150.

[0049] The CRM 262 also includes a base station neural network manager 268 (BS neural network manager 268). Alternatively or additionally, the BS neural network manager 268 can be implemented in whole or in part as hardware logic or circuitry that is integrated or separate from other components of the base station 120. In at least some aspects, the BS neural network manager 268 selects a NN formation configuration to be utilized by the base station 120 and / or the UE 110 to configure a deep neural network to process wireless communications, such as by selecting a combination of NN formation configuration elements. In some embodiments, the BS neural network manager receives feedback from the UE 110 and selects the neural network formation configuration based on the feedback. Alternatively or additionally, the BS neural network manager 268 receives a neural network formation configuration indication from a core network 150 element via a core network interface 276 or an inter-base station interface 274 and forwards the neural network formation configuration indication to the UE 110.

[0050] The CRM 262 includes a training module 270 and a neural network table 272. In an embodiment, the base station 120 manages the NN formation configuration and deploys it to the UE 110. Alternatively or additionally, the base station 120 maintains the neural network table 272. The training module 270 teaches and / or trains the DNN using known input data. For example, the training module 270 trains the DNN for different purposes, such as processing communications transmitted on a wireless communication system (e.g., encoding downlink communications, modulating downlink communications, demodulating downlink communications, decoding downlink communications, encoding uplink communications, modulating uplink communications, demodulating uplink communications, decoding uplink communications). This includes training the DNN offline (e.g., when the DNN is not actively involved in processing communications) and / or online (e.g., when the DNN is actively involved in processing communications).

[0051] In an embodiment, the training module 270 extracts learned parameter configurations from the DNN to identify NN formation configuration elements and / or NN formation configurations, and then adds and / or updates the NN formation configuration elements and / or NN formation configurations in the neural network table 272. The extracted parameter configurations include any combination of information that defines the behavior of the neural network, such as node connections, coefficients, active layers, weights, biases, pooling, etc.

[0052] The neural network table 272 stores a plurality of different NN formation configuration elements and / or NN formation configurations generated using the training module 270. In some embodiments, the neural network table includes input characteristics for each NN formation configuration element and / or NN formation configuration, wherein the input characteristics describe properties of the training data used to generate the NN formation configuration element and / or NN formation configuration. For example, the input characteristics include, for example but not limited to, power information, signal-to-interference-and-noise ratio (SINR) information, channel quality indicator (CQI) information, channel state information (CSI), Doppler feedback, frequency band, block error rate (BLER), quality of service (QoS), hybrid automatic repeat request (HARQ) information (e.g., first transmission error rate, second transmission error rate, maximum retransmission), latency, radio link control (RLC), automatic repeat request (ARQ) metrics, received signal strength (RSS), uplink SINR, timing measurements, error metrics, UE capabilities, BS capabilities, power modes, Internet Protocol (IP) layer throughput, end-to-end latency, end-to-end packet loss rate, etc. Thus, the input features sometimes include layer 1, layer 2, and / or layer 3 metrics. In some embodiments, a single index value of the neural network table 272 maps to a single NN formation configuration element (e.g., a 1:1 correspondence). Alternatively or additionally, a single index value of the neural network table 272 maps to a NN formation configuration (e.g., a combination of NN formation configuration elements).

[0053] In an embodiment, the base station 120 synchronizes the neural network table 272 with the neural network table 216 such that NN formation configuration elements and / or input characteristics stored in one neural network table are replicated in the second neural network table. Alternatively or additionally, the base station 120 synchronizes the neural network table 272 with the neural network table 216 such that NN formation configuration elements and / or input characteristics stored in one neural network table represent complementary functions in the second neural network table (e.g., NN formation configuration elements are used in the first neural network table for transmitter path processing and NN formation configuration elements are used in the second neural network table for receiver path processing).

[0054] The base stations 120 also include inter-base station interfaces 274, such as Xn and / or X2 interfaces, which the base station manager 266 configures to exchange user plane, control plane, and other information between other base stations 120 to manage communications between the base stations 120 and the user equipment 110. The base stations 120 include core network interfaces 276, which the base station manager 266 configures to exchange user plane, control plane, and other information with core network functions and / or entities.

[0055] exist Figure 3, the core network server 302 may provide all or part of a function, entity, service, and / or gateway in the core network 150. Each function, entity, service, and / or gateway in the core network 150 may be provided as a service in the core network 150 distributed on multiple servers or embodied on a dedicated server. For example, the core network server 302 may provide all or part of the services or functions of a user plane function (UPF), an access and mobility management function (AMF), a serving gateway (S-GW), a packet data network gateway (P-GW), a mobility management entity (MME), an evolved packet data gateway (ePDG), etc. The core network server 302 is shown as being embodied on a single server including a processor 304 and a computer readable storage medium 306 (CRM306). The processor 304 may be a single-core processor or a multi-core processor composed of a variety of materials, such as silicon, polysilicon, high-K dielectrics, copper, etc. CRM 306 may include any suitable memory or storage device, such as random access memory (RAM), static RAM (SRAM), dynamic RAM (DRAM), non-volatile RAM (NVRAM), read-only memory (ROM), a hard drive, or flash memory, that can be used to store device data 308 for core network server 302. Device data 308 includes data that supports core network functions or entities and / or an operating system of core network server 302 that can be executed by processor 304.

[0056] The CRM 306 also includes one or more core network applications 310, which in one embodiment are embodied on the CRM 306 (as shown). The one or more core network applications 310 can implement functions such as UPF, AMF, S-GW, P-GW, MME, ePDG, etc. Alternatively or additionally, the one or more core network applications 310 can be implemented in whole or in part as hardware logic or circuitry integrated or separated from other components of the core network server 302.

[0057] The CRM 306 also includes a core network neural network manager 312 that manages NN formation configurations for processing communications exchanged between the UE 110 and the base station 120. In some embodiments, the core network neural network manager 312 analyzes various parameters, such as current signal channel conditions (e.g., as reported by the base station 120, as reported by other wireless access points, as reported by the UE 110 (via the base station or other wireless access point)), capabilities at the base station 120 (e.g., antenna configuration, cell configuration, MIMO capabilities, radio capabilities, processing capabilities), capabilities of the UE 110 (e.g., antenna configuration, MIMO capabilities, radio capabilities, processing capabilities), etc. For example, the base station 120 obtains the various parameters during communications with the UE and forwards the parameters to the core network neural network manager 312. Based on the parameters, the core network neural network manager selects a NN formation configuration that improves the accuracy of the DNN processing the communications. Improved accuracy means an improvement in the accuracy of the output generated by the neural network relative to a neural network configured with another NN formation configuration, such as fewer bit errors. The core network neural network manager 312 then transmits the selected NN formation configuration to the base station 120 and / or the UE 110. In an embodiment, the core network neural network manager 312 receives UE and / or BS feedback from the base station 120 and selects an updated NN formation configuration based on the feedback.

[0058] The CRM 306 includes a training module 314 and a neural network table 316. In an embodiment, the core network server 302 manages and deploys NN formation configurations to multiple devices in the wireless communication system, such as UE 110 and base station 120. Alternatively or additionally, the core network server maintains the neural network table 316 external to the CRM 306. The training module 314 teaches and / or trains the DNN using known input data. For example, the training module 314 trains the DNN to process different types of pilot communications transmitted through the wireless communication system. This includes training the DNN offline and / or online. In an embodiment, the training module 314 extracts learned NN formation configurations and / or learned NN formation configuration elements from the DNN and stores the learned NN formation configuration elements in the neural network table 316. Therefore, the NN formation configuration includes any combination of architectural configurations (e.g., node connections, layer connections) and / or parameter configurations (e.g., weights, biases, pooling) that define or affect the behavior of the DNN. In some embodiments, a single index value of the neural network table 316 maps to a single NN formation configuration element (e.g., 1:1 correspondence). Alternatively or additionally, a single index value of the neural network table 316 maps to a NN formation configuration (eg, a combination of NN formation configuration elements).

[0059] In some embodiments, the training module 314 of the core network neural network manager 312 generates NN formation configurations and / or NN formation configuration elements that are complementary to those stored in the neural network table 216 at the UE 110 and / or the neural network table 272 at the base station 121. As an example, the training module 314 generates the neural network table 316 with NN formation configurations and / or NN formation configuration elements that have large changes in architecture and / or parameter configuration relative to the medium and / or low changes used to generate the neural network table 272 and / or the neural network table 216. For example, the NN formation configurations and / or NN formation configuration elements generated by the training module 314 correspond to fully connected layers, full kernel sizes, frequent sampling and / or pooling, high weighted precision, etc. Therefore, the neural network table 316 sometimes includes a high-precision neural network at the expense of increased processing complexity and / or time.

[0060] The NN formation configurations and / or NN formation configuration elements generated by the training module 270 sometimes have more fixed architectures and / or parameter configurations (e.g., fixed connection layers, fixed kernel size, etc.) and are less variable relative to those generated by the training module 314. The training module 270 generates, for example, streamlined NN formation configurations (e.g., faster computation time, less data processing) relative to those generated by the training module 314 to optimize or improve the performance of end-to-end network communications at the base station 121 and / or the UE 110. Alternatively or additionally, the NN formation configurations and / or NN formation configuration elements stored at the neural network table 216 of the UE 110 include more fixed architectures and / or parameter configurations relative to those stored in the neural network table 316 and / or the neural network table 272, which reduce the requirements at the UE 110 (e.g., computation speed, fewer data processing points, fewer computations, less power consumption, etc.) relative to the base station 121 and / or the core network server 302. In an embodiment, changes to the fixed (or flexible) architecture and / or parameter configuration at each neural network are based on the processing resources (e.g., processing power, memory constraints, quantization constraints (e.g., 8-bit vs. 16-bit), fixed-point vs. floating-point computations, power supply) of the device forming the corresponding DNN. Thus, a UE or access point with fewer processing resources relative to a core network server or base station receives a NN formation configuration optimized for the available processing resources.

[0061] The neural network table 316 stores a plurality of different NN formation configuration elements generated using the training module 314. In some embodiments, the neural network table includes input characteristics for each NN formation configuration element and / or NN formation configuration, wherein the input characteristics describe properties of the training data used to generate the NN formation configuration. For example, the input characteristics may include power information, SINR information, CQI, CSI, Doppler feedback, RSS, error metrics, minimum end-to-end (E2E) delay, expected E2E delay, E2E QoS, E2E throughput, E2E packet loss rate, service cost, etc.

[0062] The core network server 302 also includes a core network interface 318 for communicating user plane, control plane, and other information with other functions or entities in the core network 150, the base station 120, or the UE 110. In some embodiments, the core network server 302 uses the core network interface 318 to communicate the NN formation configuration to the base station 120. The core network server 302 uses the core network interface 318 to receive feedback from the base station 120 and / or from the UE 110 through the base station 120, alternatively or additionally.

[0063]

[0066] Considering now a discussion of configurable machine learning modules in accordance with one or more embodiments, an example environment and example devices that may be used to form base station-user equipment messaging using neural networks have been described.

[0064] Configurable machine-learning modules

[0065] Figure 4 An exemplary machine learning module 400 is illustrated. The machine learning module 400 implements a collection of adaptive algorithms that learn and recognize patterns within data. The machine learning module 400 may be implemented using any combination of software, hardware, and / or firmware.

[0066] exist Figure 4In the embodiment, the machine learning module 400 includes a deep neural network 402 (DNN 402) having a group of connected nodes (e.g., neurons and / or perceptrons) organized into three or more layers. The nodes between the layers can be configured in a variety of ways, such as a partially connected configuration, in which a first subset of nodes in a first layer is connected to a second subset of nodes in a second layer; a fully connected configuration, in which each node in a first layer is connected to each node in a second layer, and so on. Neurons process input data to produce continuous output values, such as any real number between 0 and 1. In some cases, the output value indicates how close the input data is to the desired category. Perceptrons perform linear classification on the input data, such as binary classification. Whether neurons or perceptrons, these nodes can use various algorithms to generate output information based on adaptive learning. Using DNN, the machine learning module 400 performs a variety of different types of analysis, including single linear regression, multiple linear regression, logistic regression, stepwise regression, binary classification, multi-class classification, multivariate adaptive regression splines, local estimation scatter plot smoothing, etc.

[0067] In some embodiments, the machine learning module 400 learns adaptively based on supervised learning. In supervised learning, the machine learning module receives various types of input data as training data. The machine learning module processes the training data to learn how to map the input to the desired output. As an example, the machine learning module 400 receives a digital sample of a signal as input data, and learns how to map the signal sample to binary data reflecting the information embedded in the signal. As another example, the machine learning module 400 receives binary data as input data, and learns how to map the binary data to a digital sample of a signal, wherein the binary data is embedded in the signal. During the training process, the machine learning module 400 uses labeled or known data as input to the DNN. The DNN uses nodes to analyze the input and generate corresponding outputs. The machine learning module 400 compares the corresponding output with the real data, and adapts the algorithm implemented by the node to improve the accuracy of the output data. Afterwards, the DNN applies the adaptive algorithm to the unlabeled input data to generate the corresponding output data.

[0068] The machine learning module 400 uses statistical analysis and / or adaptive learning to map inputs to outputs. For example, the machine learning module 400 uses characteristics learned from the training data to associate unknown inputs with outputs that are statistically likely to be within a threshold range or value. This allows the machine learning module 400 to receive complex inputs and identify corresponding outputs. Some embodiments train the machine learning module 400 on characteristics of communications transmitted on a wireless communication system (e.g., time / frequency interleaving, time / frequency deinterleaving, convolutional encoding, convolutional decoding, power levels, channel equalization, intersymbol interference, orthogonal amplitude modulation / demodulation, frequency division multiplexing / demultiplexing, transmission channel characteristics). This allows the trained machine learning module 400 to receive samples of a signal as input, such as samples of a downlink signal received at a user device, and to recover information from the downlink signal, such as binary data embedded in the downlink signal.

[0069] exist Figure 4 , the DNN includes an input layer 404, an output layer 406, and one or more hidden layers 408 located between the input layer and the output layer. Each layer has an arbitrary number of nodes, wherein the number of nodes between layers can be the same or different. In other words, the input layer 404 can have the same number and / or a different number of nodes as the output layer 406, the output layer 406 can have the same number and / or a different number of nodes as the hidden layer 408, and so on.

[0070] Node 410 corresponds to one of several nodes included in input layer 404, wherein the nodes perform independent calculations. As further described, the node receives input data and processes the input data using an algorithm to generate output data. Sometimes, the algorithm includes weights and / or coefficients that change based on adaptive learning. Therefore, the weights and / or coefficients reflect the information learned by the neural network. In some cases, each node can determine whether to pass the processed input data to the next node. For illustration, after processing the input data, node 410 can determine whether to pass the processed input data to node 412 and / or node 414 of hidden layer 408. Alternatively or additionally, node 410 passes the processed input data to a node based on a layer connection architecture. This process can be repeated in multiple layers until the DNN generates an output using the nodes of the output layer 406.

[0071] Neural networks may also employ various architectures that determine which nodes within the neural network are connected, how data is advanced and / or retained in the neural network, what weights and coefficients are used to process input data, how data is processed, and so on. These various factors collectively describe a NN formation configuration. To illustrate, a recursive neural network, such as a long short-term memory (LSTM) neural network, forms loops between node connections in order to retain information from a previous portion of an input data sequence. The recursive neural network then uses the retained information for subsequent portions of the input data sequence. As another example, a feedforward neural network passes information to a forward connection without forming a loop to retain information. Although described in the context of node connections, it should be appreciated that a NN formation configuration may include various parameter configurations that affect how a neural network processes input data.

[0072] The NN formation configuration of a neural network can be characterized by various architectures and / or parameter configurations. For illustration, consider an example in which a DNN implements a convolutional neural network. In general, a convolutional neural network corresponds to a type of DNN in which each layer processes data using a convolution operation to filter input data. Accordingly, the convolutional NN formation configuration can be characterized, for example, but not limited to, by pooling parameters, kernel parameters, weights, and / or layer parameters.

[0073] Pooling parameters correspond to parameters that specify a pooling layer within a convolutional neural network, which can reduce the dimensionality of the input data. For illustration, a pooling layer can combine the node outputs of a first layer as node inputs of a second layer. Alternatively or additionally, the pooling parameters specify how and where the neural network pools data in a data processing layer. Pooling parameters indicating "max pooling" configure the neural network to perform pooling by selecting the maximum value from a group of data generated by nodes of the first layer, for example, and use the maximum value as an input to a single node of the second layer. Pooling parameters indicating "average pooling" configure the neural network to generate an average value from a group of data generated by nodes of the first layer, and use the average value as an input to a single node of the second layer.

[0074] The kernel parameters indicate the filter size (e.g., width and height) to be used in processing the input data. Alternatively or additionally, the kernel parameters specify the type of kernel method used in filtering and processing the input data. A support vector machine, for example, corresponds to a kernel method that uses regression analysis to identify and / or classify data. Other types of kernel methods include Gaussian processes, canonical correlation analysis, spectral clustering, etc. Thus, the kernel parameters may indicate the filter size and / or type of kernel method applied in a neural network.

[0075] The weight parameters specify the weights and biases used by the algorithm within the node to classify input data. In an embodiment, the weights and biases are learned parameter configurations, such as those generated by training data.

[0076] The layer parameters specify layer connections and / or layer types, such as a fully connected layer type indicating that each node in a first layer (e.g., output layer 406) is connected to each node in a second layer (e.g., hidden layer 408), a partially connected layer type indicating which nodes in the first layer are disconnected from the second layer, an activation layer type indicating which filters and / or layers are activated within the neural network, etc. Alternatively or additionally, the layer parameters specify a type of node layer, such as a normalization layer type, a convolutional layer type, a pooling layer type, etc.

[0077] Although described in the context of pooling parameters, kernel parameters, weight parameters, and layer parameters, it should be appreciated that other parameter configurations may be used to form a DNN without departing from the scope of the claimed subject matter. Thus, a NN formation configuration may include any other type of configuration parameters that may be applied to a DNN that affect how the DNN processes input data to generate output data.

[0078] Some embodiments configure the machine learning module 400 based on the current operating environment. For illustration, consider a machine learning module that is trained to generate binary data from digital samples of a signal. The transmission environment often modifies the characteristics of the signal propagating in the environment. The transmission environment often changes, which affects how the environment modifies the signal. A first transmission environment, for example, modifies the signal in a first manner, while a second transmission environment modifies the signal in a manner different from the first manner. These differences affect the accuracy of the output results generated by the machine learning module. For example, a neural network configured to process communications transmitted on a first transmission environment may generate errors (e.g., bit errors exceeding a threshold) when processing communications transmitted on a second transmission environment.

[0079] Various embodiments generate and store NN formation configurations and / or NN formation configuration elements (e.g., various architectures and / or parameter configurations) for different transmission environments. For example, the base station 120 and / or the core network server 302 trains the machine learning module 400 using any combination of the BS neural network manager 268, the training module 270, the core network neural network manager 312, and / or the training module 314. Training occurs offline when no active communication exchanges occur, or online during active communication exchanges. For example, the base station 120 and / or the core network server 302 can mathematically generate training data, access files storing training data, obtain real-world communication data, etc. The base station 120 and / or the core network server 302 then extracts the various learned NN formation configurations and stores them in a neural network table. Some embodiments store input characteristics with each NN formation configuration, wherein the input characteristics describe various attributes of the transmission environment corresponding to the corresponding NN formation configuration. In an embodiment, the neural network manager selects a NN formation configuration and / or NN formation configuration element by matching the current transmission environment and / or the current operating environment with the input characteristics.

[0080] Having described a configurable machine learning module, consider now a discussion of deep neural networks in a wireless communication system in accordance with one or more embodiments.

[0081] Deep Neural Networks in Wireless Communication Systems

[0082] Wireless communication systems include various complex components and / or functions, such as reference Figure 1 Example environment 100, Figure 2 An exemplary device diagram 200 and Figure 3 The various devices and modules described in the exemplary device diagram 300. In some embodiments, devices participating in a wireless communication system link a series of functions together to enable information to be exchanged over a wireless connection.

[0083] To demonstrate, now consider Figure 5 , which illustrates an exemplary block diagram 500 and an exemplary block diagram 502, each depicting an exemplary processing chain utilized by a device in a wireless communication system. For simplicity, the block diagrams illustrate high-level functionality, and it should be appreciated that for clarity, the block diagrams may include Figure 5 Additional features omitted in .

[0084] exist Figure 5In the upper half of the block diagram 500, the block diagram 500 includes a transmitter block 504 and a receiver block 506. The transmitter block 504 includes a transmitter processing chain that proceeds from top to bottom. The transmitter processing chain starts with input data, which proceeds to a coding stage, then a modulation stage, followed by a radio frequency (RF) analog transmission (Tx) stage. The coding stage may include any type and number of coding stages employed by a device to transmit data over a wireless communication system.

[0085] For illustration, the exemplary encoding stage receives binary data as input and processes the binary data using various encoding algorithms to append information such as frame information to the binary data. Alternatively or additionally, the encoding stage transforms the binary data, such as by applying forward error correction, which adds redundancy to aid in information recovery at the receiver. As another example, the encoding stage converts the binary data into symbols.

[0086] The exemplary modulation stage receives as input the output generated by the coding stage and embeds the input onto the signal. For example, the modulation stage generates digital samples of the signal embedded into the input from the coding stage. Thus, in the transmitter block 504, the coding stage and the modulation stage represent a high-level transmitter processing chain that typically includes low-level complex functions such as convolutional encoding, serial to parallel conversion, cyclic prefix insertion, channel coding, time / frequency interleaving, etc. The RF analog Tx stage receives the output from the modulation stage, generates an analog RF signal based on the modulation stage output, and transmits the analog RF signal to the receiver block 506.

[0087] Receiver block 506 uses a receiver processing chain that performs complementary processing relative to transceiver block 504. The receiver processing chain shown in receiver block 506 proceeds from top to bottom and includes an RF analog receive (Rx) stage followed by a demodulation stage and a decoding stage.

[0088] The RF analog Rx stage receives the signal transmitted by the transmitter block 504 and generates an input used by the demodulation stage. As an example, the RF analog Rx stage includes a down-conversion component and / or an analog-to-digital converter (ADC) to generate samples of the received signal. The demodulation stage processes the input from the RF analog Rx stage to extract data embedded on the signal (e.g., data embedded by the modulation stage of the transmitter block 504). For example, the demodulation stage recovers symbol and / or binary data.

[0089] The decoding stage receives input from the demodulation stage, such as recovered symbols and / or binary data, and processes the input to recover the transmitted information. For illustration, the decoding stage corrects data errors based on forward error correction applied at the transmitter block, extracts payload data from frames and / or time slots, etc. Thus, the decoding stage generates recovered information.

[0090] As noted, the transmitter and receiver processing chains shown by the transmitter block 504 and the receiver block 506 have been simplified for clarity and may include multiple complex modules. Sometimes, these complex modules are dedicated to specific functions and / or conditions. For example, consider a receiver processing chain that processes orthogonal frequency division modulation (OFDM) transmissions. In order to recover information from OFDM transmissions, the receiver block typically includes multiple processing blocks, each of which is dedicated to a specific function, such as an equalization block that corrects distortion in the received signal, a channel estimation block that estimates transmission channel properties to identify the effects of scattering, power attenuation, etc. on the transmission, etc. At high frequencies such as frequencies in the 6 GHz band, these blocks may be computationally and / or monetaryly expensive (e.g., requiring a large amount of processing power, requiring expensive hardware). In addition, implementing blocks that generate outputs with accuracy within a desired threshold often requires more specific and less flexible components. For illustration, an equalization block used for signals in the 6 GHz band may not be able to perform with the same accuracy in other bands, so different equalization blocks are required for different bands, thereby adding complexity to the corresponding device.

[0091] Some embodiments include DNNs in the transmit and / or receiver processing chains. In block diagram 502, transmitter block 508 includes one or more deep neural networks 510 (DNN 510) in the transmitter processing chain, while receiver block 512 includes one or more deep neural networks 514 (DNN 514) in the receiver processing chain.

[0092] For simplicity, the DNN 510 in the transmitter block 508 corresponds to the coding and modulation stages of the transmitter block 504. However, it should be appreciated that the DNN can perform any high-level and / or low-level operations found in the transmitter processing chain. For example, a first DNN performs low-level transmitter-side forward error correction, a second DNN performs low-level transmitter-side convolutional encoding, and so on. Alternatively or additionally, the DNN 510 performs high-level processing, such as end-to-end processing corresponding to the coding and modulation stages of the transmitter block 508.

[0093] In a similar manner, the DNN 514 in the receiver block 512 performs receiver processing chain functions (e.g., demodulation stage, decoding stage). The DNN 514 can perform any high-level and / or low-level operations found within the receiver processing chain, such as low-level receiver-side error correction, low-level receiver-side symbol recovery, high-level end-to-end demodulation and decoding, etc. Therefore, the DNN in the wireless communication system can be configured to replace high-level operations and / or low-level operations in the transmitter and receiver processing chains. Sometimes, as further described, the DNN that performs high-level operations and / or low-level operations can be configured and / or reconfigured based on the current operating environment. This provides greater flexibility and adaptability to the processing chain relative to more specific and less flexible components.

[0094] Some embodiments use multiple DNNs to process communication exchanges on a wireless communication system, where each DNN has a corresponding purpose (e.g., uplink processing, downlink processing, uplink encoding processing, downlink decoding processing, etc.). For demonstration purposes, consider now an exemplary operating environment 600 including UE 110 and base station 121. Figure 6 UE 110 and base station 121 exchange communications with each other over a wireless communication system by processing the communications using a plurality of DNNs.

[0095] exist Figure 6 , the base station neural network manager 268 of the base station 121 includes a downlink processing module 602 for processing downlink communications, such as for generating downlink communications for transmission to the UE 110. For illustration, as further described, the base station neural network manager 268 forms a deep neural network 604 (DNN 604) in the downlink processing module 602 using a NN formation configuration. In some examples, the DNN 604 corresponds to Figure 5 's DNN 510. In other words, DNN 604 performs some or all of the transmitter processing functions for generating downlink communications.

[0096] Similarly, the UE neural network manager 218 of the UE 110 includes a downlink processing module 606, wherein the downlink processing module 606 includes a deep neural network 608 (DNN 608) for processing (received) downlink communications. In various embodiments, the UE neural network manager 218 forms the DNN 608 using the NN formation configuration. Figure 6 In the example, DNN 608 corresponds to Figure 5 514 of UE 110, wherein the deep neural network 606 of UE 110 performs some or all of the receiver processing functions for the (received) downlink communication. Therefore, DNN 604 and DNN 608 perform processing that complements each other (e.g., encoding / decoding, modulation / demodulation).

[0097] DNN 604 and / or DNN 608 may include multiple deep neural networks, wherein each DNN is dedicated to a corresponding channel, a corresponding purpose, etc. As an example, base station 121 uses a first DNN in DNN 604 to process downlink control channel information, uses a second DNN in DNN 604 to process downlink data channel information, etc. As another example, UE 110 uses a first DNN in DNN 608 to process downlink control channel information, uses a second DNN in DNN 608 to process downlink data channel information, etc.

[0098] Base station 121 and / or UE 110 also uses a DNN to process uplink communications. In environment 600, UE neural network manager 218 includes an uplink processing module 610, wherein the uplink processing module 610 includes a deep neural network 612 (DNN 612) for generating and / or processing uplink communications (e.g., encoding, modulation). In other words, uplink processing module 610 processes pre-transmission communications as part of processing uplink communications. UE neural network manager 218 forms DNN 612, for example, using a NN formation configuration. Sometimes, DNN 612 corresponds to Figure 5 510. Thus, DNN 612 performs some or all of the transmitter processing functions for generating uplink communications transmitted from UE 110 to base station 121.

[0099] Similarly, the uplink processing module 614 of the base station 121 includes a deep neural network 616 (DNN 616) for processing (received) uplink communications, wherein the base station neural network manager 268 forms the DNN 616 using the NN formation configuration, as further described. In the example, the DNN 616 of the base station 121 corresponds to Figure 5 514 of the uplink processing module 610 and performs some or all of the receiver processing functions for (received) uplink communications, such as uplink communications received from the UE 110. Sometimes, the DNN 612 and the DNN 616 perform complementary functions to each other. Alternatively or additionally, the uplink processing module 610 and / or the uplink processing module 614 include multiple DNNs, each of which has a dedicated purpose (e.g., processing a corresponding channel, performing a corresponding uplink function, etc.). Figure 6 DNNs 604, 608, 612, and 616 are illustrated as residing within respective neural network managers to represent that the neural network managers form the DNNs, and it should be appreciated that the DNNs may be formed external to neural network managers (e.g., UE neural network manager 218 and base station neural network manager 268) within different components, processing chains, modules, etc.

[0100] Having described deep neural networks in wireless communication systems, consider now a discussion of signaling and control transactions on a wireless communication system according to one or more embodiments, which are used to configure a deep neural network for downlink and uplink communications.

[0101] Configuring signaling and control transactions for deep neural networks

[0102] Figure 7 , 89 illustrates an exemplary signaling and control transaction diagram between a base station, a user equipment, and / or a core network server according to one or more aspects of base station-user equipment messaging for a deep neural network. The signaling and control transactions may be performed by Figure 1 The base station 120 and the UE 110, or Figure 3 The core web server 302 uses Figure 1-6 For example, in some embodiments, the core network server 302 performs various signaling and control actions performed by the base station 120, such as Figure 7 and 8 shown.

[0103] A first example of signaling and control transactions for base station-user equipment messaging regarding a deep neural network is given by Figure 7 700 illustrates the signaling and control transactions of the deep neural network. As shown, at 705, the base station 121 determines the neural network formation configuration. In determining the neural network formation configuration, the base station analyzes any combination of information, such as the channel type (e.g., downlink, uplink, data, control, etc.), transmission medium properties (e.g., power measurement, signal-to-interference-plus-noise ratio (SINR) measurement, channel quality indicator (CQI) measurement), coding scheme, UE capabilities, BS capabilities, etc., processed by the deep neural network.

[0104] For example, base station 121 receives a message from UE 110 (not shown) indicating one or more capabilities of the UE, such as, by way of example and not limitation, connectivity information, dual connectivity information, carrier aggregation capability, downlink physical parameter values, uplink physical parameter values, supported downlink / uplink categories, inter-frequency handover, etc. Base station 121 identifies from the message the UE capabilities that affect how the UE processes communications and / or how the base station processes communications from the UE, and selects a neural network formation configuration having improved output accuracy relative to other neural network formation configurations.

[0105] In some embodiments, the base station 121 selects a neural network formation configuration from a plurality of neural network formation configurations. Alternatively or additionally, the base station 121 selects the neural network formation configuration by selecting a subset of neural network architecture formation elements in a neural network table. Sometimes, the base station 121 analyzes a plurality of neural network formation configurations and / or a plurality of neural network formation configuration elements included in the neural network table, and determines the neural network formation configuration by selecting and / or creating a neural network formation configuration consistent with current channel conditions, such as by matching channel type, transmission medium properties, etc. with input characteristics, as further described. Alternatively or additionally, the base station 121 selects the neural network formation configuration based on network parameters, such as scheduling parameters (e.g., scheduling multi-user multiple input multiple output (MU-MIMO) for downlink communications, scheduling MU-MIMO for uplink communications).

[0106] At 710, base station 121 transmits the neural network formation configuration to UE 110. In some embodiments, the base station transmits the neural network formation configuration, such as by transmitting a mapping to a neural network table, such as Figure 2 The message specifying the neural network formation configuration is transmitted by a message with an index value of an entry in the neural network table 216 of the base station. Alternatively or additionally, the base station transmits a message including a neural network parameter configuration (e.g., weight values, coefficient values, number of filters). In some cases, the base station 121 specifies the purpose and / or processing allocation in the message, wherein the processing allocation indicates which channels and / or which position in the processing chain the configured neural network is applied to, such as downlink control channel processing, uplink data channel processing, downlink decoding processing, uplink encoding processing, etc. Thus, the base station can transmit the processing allocation via the neural network formation configuration.

[0107] In some embodiments, the base station 121 transmits multiple neural network formation configurations to the UE 110. For example, the base station transmits a first message instructing the UE to use a first neural network formation configuration for uplink encoding, and a second message instructing the UE to use a second neural network formation configuration for downlink decoding. In some cases, the base station 121 transmits multiple neural network formation configurations and corresponding processing allocations in a single message. As another example, the base station transmits multiple neural network formation configurations using different radio access technologies (RATs). The base station can, for example, transmit a first neural network formation configuration for downlink communication processing to the UE 110 using a first RAT and / or carrier, and transmit a second neural network formation configuration for uplink communication processing to the UE 110 using a second RAT and / or carrier.

[0108] At 715, UE 110 forms a first neural network based on the neural network formation configuration. For example, UE 110 uses the index value transmitted by the base station to access a neural network table to obtain the neural network formation configuration and / or neural network formation configuration elements. Alternatively or additionally, UE 110 extracts the neural network architecture and / or parameter configuration from the message. UE 110 then forms a neural network using the neural network formation configuration, the extracted architecture and / or parameter configuration, etc. In some embodiments, the UE processes all communications using the first neural network, while in other embodiments, the UE processes select communications using the first neural network based on the processing allocation.

[0109] At 720, the base station 121 transmits information based on the neural network formation configuration. Figure 6 , base station 121 processes downlink communications using a second neural network configured with functions complementary to the first neural network. In other words, the second neural network uses a second neural network formation configuration complementary to the neural network formation configuration. Further, at 725, UE 110 recovers information using the first neural network.

[0110] pass Figure 8 The signaling and control transaction diagram 800 of FIG. 1 illustrates a second example of signaling and control transactions for base station-user equipment messaging regarding a deep neural network. In some embodiments, the signaling and control transaction diagram 800 represents Figure 7 Continuation of signaling and control transaction diagram 700 .

[0111] As shown, at 805, base station 121 communicates with UE 110 based on the first neural network formation configuration. Similarly, at 810, UE 110 communicates with base station 121 based on the first neural network formation configuration. For example, base station 121 communicates with UE 110 based on the first neural network formation configuration. Figure 6 The DNN 604 processes one or more downlink communications to communicate with the UE 110, and the UE 110 communicates with the UE 110 by using Figure 6 DNN 608 processes downlink communications received from base station 121 to communicate with base station 121.

[0112] In an embodiment, as described, DNN 604 and DNN 608 are formed based on a first neural network formation configuration. For illustration, DNN 604 and DNN 608 perform complementary functions to each other, wherein the first neural network formation configuration specifies complementary functions for each deep neural network (e.g., the base station forms a first neural network using the first neural network formation configuration, and the UE forms a second neural network complementary to the first neural network by using a neural network formation configuration complementary to the first neural network formation configuration). The complementary functions performed by the deep neural networks allow the two parties exchanging communications to maintain synchronization (e.g., accurately recover information). Therefore, the first neural network formation configuration specifies any combination of a base station side neural network formation configuration, a complementary user device side neural network formation configuration, and / or a common neural network formation configuration used by each device participating in the communication exchange.

[0113] As another example, UE 110 uses Figure 6 DNN 612 processes one or more uplink communications to base station 121, and base station 121 uses Figure 6 The base station 121 and the UE 110 may further include a base station 121 and a base station 110 that form a deep neural network 614 to process uplink communications received from the UE 110. Similar to downlink communications, some embodiments form the DNN 612 and the deep neural network 614 based on a first neural network forming configuration (e.g., the UE forms a first neural network using a first neural network forming configuration, and the base station forms a second neural network using a neural network forming configuration that is complementary to the first neural network forming configuration). Thus, the base station 121 and the UE 110 communicate with each other based on the first neural network forming configuration by forming a deep neural network based on the first neural network forming configuration and processing communications with the deep neural network.

[0114] At 815, the base station generates base station metrics, such as metrics based on uplink communications received from UE 110. Similarly, at 820, UE 110 generates UE metrics, such as metrics based on downlink communications received from base station 121. Any type of metric may be generated by base station 121 and / or UE 110, such as power measurements (e.g., RSS), error metrics, timing metrics, QoS, latency, etc.

[0115] At 825, UE 110 transmits the metric to base station 121. In an embodiment, UE 110 processes the metric communication using a deep neural network based on a first neural network formation configuration. Alternatively or additionally, UE 110 processes the metric communication using a neural network formed using a second neural network formation configuration. Thus, as further described, in some embodiments, UE 110 maintains multiple deep neural networks, where each deep neural network has a designated purpose and / or processing assignment (e.g., a first neural network for downlink control channel processing, a second neural network for downlink data channel processing, a third neural network for uplink control channel processing, a fourth neural network for uplink data channel processing). Sometimes, base station 121 transmits multiple neural network formation configurations for forming multiple deep neural networks to UE 110.

[0116] At 830, the base station 121 identifies a second neural network formation configuration based on the metrics at 830. In some embodiments, the base station 121 identifies the second neural network formation configuration based on UE metrics, base station metrics, or any combination thereof. This includes identifying any combination of architectural changes and / or parameter changes to the neural network formation configuration, as further described, such as a small change to the neural network formation configuration involving updating coefficient parameters to account for a change in a return metric (e.g., SINR change, Doppler feedback change, power level change, BLER change). Alternatively or additionally, identifying the second neural network formation configuration includes a large change based on a metric such as a change in power state (e.g., a transition from a radio resource connected state to an idle state), such as reconfiguring node and / or layer connections.

[0117] In some embodiments, the base station 121 identifies the partial and / or incremental neural network formation configuration as a second neural network formation configuration, wherein the partial and / or incremental neural network formation configuration indicates a change to a complete neural network formation configuration, such as including an architectural configuration and a parameter configuration of the neural network, and the partial and / or incremental neural network formation configuration specifies a change and / or update to the parameter configuration based on using the same architectural configuration indicated in the complete neural network formation configuration.

[0118] In some embodiments, the base station identifies the second neural network formation configuration by identifying a neural network formation configuration in the neural network table, the neural network formation configuration improving the UE's ability and / or the base station's ability to recover data from the communication (e.g., improving the accuracy of the recovered information). Figure 2The base station 121 may further include a base station neural network manager 268 to identify a neural network formation configuration that compensates for the problem identified by the UE measurements and / or the base station measurements. As another example, the base station 121 identifies a neural network formation configuration having one or more input characteristics that are consistent with the changed operating conditions identified by the UE measurements and / or the base station measurements. Alternatively or additionally, such as for the case where the UE moves to a low power state, the base station 121 identifies a neural network formation configuration that produces similar results with less processing.

[0119] At 840, base station 121 directs UE 110 to use the second neural network to form a configuration to update the first neural network. Figure 2 In some embodiments, base station 121 indicates in the message an update message for an index value of a second neural network formation configuration in neural network table 216 of the present invention. In some embodiments, base station 121 indicates in the message a time instance that directs UE 110 as to when to apply the second neural network formation configuration. In other words, the time instance directs UE 110 to switch from processing communications using the first neural network formation configuration to processing communications using the second neural network formation configuration at the time specified in the time instance. In an embodiment, the base station transmits updates to the downlink neural network using a first carrier or RAT and transmits updates to the uplink neural network using a second carrier or RAT.

[0120] At 845, the base station 121 updates the base station neural network based on a second neural network forming configuration, such as a deep neural network formed based on the first neural network forming configuration and used to communicate with the UE 110 at 805 (e.g., a deep neural network for processing downlink communications, a deep neural network for processing uplink communications). Similarly, at 850, the UE 110 updates the user equipment neural network based on a second neural network forming configuration, such as a deep neural network that performs a complementary function to the base station neural network updated at 845. As an example, the UE extracts an index value and / or a time value from the update message transmitted by the base station at 840. The UE 110 obtains the second neural network forming configuration and modifies the user equipment neural network at the time specified in the update message. Thus, the UE 110 uses the first neural network forming configuration for processing communications until the specified time in the update message, at which time the UE 110 switches to using the second neural network forming configuration to process communications.

[0121] In an embodiment, the base station 121 and / or the UE 110 iteratively performs Figure 8 8, the signaling and control transactions depicted in signaling and control transaction diagram 800 are represented by dashed lines. These iterations allow base station 121 and / or UE 110 to dynamically modify the communication processing chain based on changing operating conditions, as further described.

[0122] Figure 9-1 and Figure 9-2 Figure 900 illustrates a third example of signaling and control transactions for base station-user equipment messaging regarding a deep neural network. Figure 9-1 As shown, at 905, the core network server 302 determines the neural network formation configuration based on various metrics and / or parameters, such as metrics from the UE 110, metrics from the base station 121, UE capabilities, etc. For example, the core network server receiver receives any combination of metrics and / or parameters from the base station 121 and / or the UE 110, such as power information, SINR information, CQI, CSI, Doppler feedback, QoS, latency, UE capabilities, base station type (e.g., eNB, gNB, or ng-eNB), protocol version, error metrics, UE capabilities, BS capabilities, power mode, etc. Then, the core network server 302 determines the neural network formation configuration based on the metrics, parameters, etc.

[0123] In some embodiments, the core network server 302 determines a plurality of different neural network formation configurations, each of which is specific to a respective base station and / or respective UE. Alternatively or additionally, the core network server 302 determines a neural network formation configuration used by a plurality of base stations and / or UEs. Sometimes, the core network server 302 determines a default neural network formation configuration used by the base stations and / or UEs to initially connect to each other. As further described, the default neural network formation configuration corresponds to a general neural network formation configuration that configures a deep neural network to process a variety of input data and / or channel conditions with an accuracy within a threshold range or value. However, a dedicated neural network formation configuration corresponds to a deep neural network tuned to a particular type of input data and / or a particular channel condition. When determining a neural network formation configuration, some embodiments of the core network server determine a complementary neural network formation configuration (e.g., a base station side neural network formation configuration that is complementary to a user equipment side neural network formation configuration).

[0124] At 910, the core network server 302 transmits the neural network formation configuration to the base station 121. For example, the core network server 302 transmits the neural network formation configuration to the base station 121. Figure 3 The core network interface 318 transmits the index value to the base station 121, wherein the index value is mapped to Figure 2 Entries in the neural network table 272. Alternatively or additionally, the core network server 302 transmits various parameter configurations, such as coefficients, weights, layer connection values, etc.

[0125] At 915, the base station 121 forwards the neural network formation configuration to the UE 110. For example, the base station wirelessly transmits the index value to the UE, wirelessly transmits the parameter configuration to the UE, etc. When transmitting the neural network formation configuration, the core network server 302 and / or the base station 121 sometimes indicates the allocation of processing (e.g., downlink data channel processing, uplink control channel processing, decoding processing, uplink encoding processing, uplink modulation processing, downlink demodulation processing) for the neural network formation configuration.

[0126] In some embodiments, the base station 121 adds modifications to the neural network formation configuration before forwarding the neural network formation configuration to the UE 110. For example, the base station 121 may have access to more updated information than the core network server, such as through UE capabilities received from the UE 110. Sometimes, the base station 121 updates the neural network formation configuration based on updated information (e.g., UE capabilities specific to the UE 110), such as by removing layers and / or nodes to reduce the corresponding complexity of the deep neural network at the UE 110 based on the available processing power, battery remaining, available radio, etc. at the UE. As another example, the base station 121 adds a convolutional layer to the neural network formation configuration based on the updated information. Thereafter, the base station 121 forwards the modified neural network formation configuration to the UE 110 in place of the neural network formation configuration received from the core network server 302.

[0127] At 920, base station 121 forms a first deep neural network using the neural network formation configuration, such as by identifying the base station side neural network formation configuration and forming the first deep neural network using the base station side neural network formation configuration. For illustration, base station 121 accesses Figure 2 Similarly, at 925, UE 110 uses the neural network formation configuration to form a second deep neural network, such as by identifying a complementary and / or user device side neural network formation configuration and forming the second neural network using the complementary and / or user device side neural network formation configuration. In some embodiments, UE 110 accesses the second neural network using an index value. Figure 2 The neural network table 216 is used to obtain a complementary and / or user device side neural network formation configuration. Accordingly, the first deep neural network and the second deep neural network are synchronized neural networks based on the neural network formation configuration determined and transmitted by the core network server at 905 and 910.

[0128] exist Figure 9-2, at 930, base station 121 communicates with UE 110 using a first deep neural network, such as by generating and / or processing downlink communications to UE 110, by receiving and / or processing uplink communications from UE 110, etc. Similarly, at 935, UE 110 communicates with base station 121 using a second deep neural network at 935. In other words, as further described, UE 110 communicates with base station 121 using a complementary deep neural network (e.g., a second deep neural network) configured based on a neural network formation.

[0129] At 940, the base station generates base station metrics, wherein the base station 121 generates the metrics based on the communication at 930 and / or 935. Thus, the base station metrics may be based on uplink communications received from the UE 110. For example, the base station 121 generates uplink received power, uplink SINR, uplink packet errors, uplink throughput, timing measurements, etc. In some embodiments, the base station includes base station capabilities (BS capabilities) in the base station metrics, such as processing capabilities (e.g., macro base station, small cell base station), power status, etc.

[0130] Similarly, at 945 , UE 110 generates UE metrics (e.g., power information, SINR information, CQI, CSI, Doppler feedback, QoS, latency) based on downlink communications from base station 121 , and at 950 , transmits the UE metrics to base station 121 .

[0131] At 955, base station 121 forwards the metrics to core network server 302, such as via core network interface 318. This includes any combination of base station metrics and UE metrics generated at 940 and / or 945. Thereafter, core network server 302 determines updates to the neural network formation configuration at 960. This may include any combination of architectural changes (e.g., reconfiguring node connections, reconfiguring active / inactive layers), changing applied processing parameters (e.g., coefficients, kernels), etc. Thus, core network server 302 sometimes identifies, such as at Figure 8 At 830, small changes and / or large changes are described with reference to base station 121. By receiving the metrics generated by UE 110 via base station 121, core network server 302 receives feedback about the communication channel between the base station and the UE and / or an indication of how the neural network is processing communications transmitted on the communication channel. Core network server 302 analyzes the feedback to identify adjustments to the neural network formation configuration that, when applied to the neural network, improve the accuracy of the way the deep neural network processes communications (e.g., more accurate data recovery, less erroneous data) and / or the efficiency with which the neural network processes communications (e.g., selecting a configuration that reduces the processing time of the deep neural network).

[0132] In an embodiment, the core network server 302, the base station 121 and / or the UE 110 iteratively executes Figure 9-1 and 9-2 The signaling and control transactions depicted in Figure 900 are represented by dashed lines returning from 960 to 910. These iterations allow the core network server 302, base station 121, and / or UE 110 to dynamically modify the communication processing chain based on changing operating conditions, as further described.

[0133] In some embodiments, the core network server 302 receives feedback from multiple UEs and / or base stations in the wireless communication system. This provides the core network server with more insights into how the wireless communication system is performing, what devices are communicating in the wireless communication system, how the devices are communicating, etc. In various embodiments, the core network server 302 determines updates to the neural network formation configuration based on optimizing communication for multiple devices and / or overall system performance rather than optimizing communication for a specific device.

[0134] Having described signaling and control transactions that may be used to configure a neural network for handling communications, consider now some example methods in accordance with one or more implementations.

[0135] Exemplary Methods

[0136] According to one or more aspects of base station-user equipment messaging for deep neural networks, reference Fig.10 and 11 Describe exemplary methods 1000 and 1100. The order in which the method boxes are described is not intended to be construed as limiting, and any number of the method boxes may be skipped or combined in any order to implement a method or an alternative method. Generally, any of the components, modules, methods, and operations described herein may be implemented using software, firmware, hardware (e.g., fixed logic circuits), manual processing, or any combination thereof. Some operations of the exemplary methods may be described in the general context of executable instructions stored on a computer-readable memory locally and / or remotely of a computer processing system, and implementations may include software applications, programs, functions, and the like. Alternatively or additionally, any of the functions described herein may be performed, at least in part, by one or more hardware logic components, such as, but not limited to, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), an application specific standard product (ASSP), a system on a chip (SoC), a complex programmable logic device (CPLD), and the like.

[0137] Fig.10An exemplary method 1000 for configuring a neural network to process communications exchanged over a wireless communication system is illustrated. In some implementations, the operations of method 1000 are performed by a network entity such as base station 120 or core network server 302.

[0138] At 1005, a network entity determines a neural network formation configuration for a deep neural network to process communications transmitted over a wireless communication system. For example, a network entity (e.g., base station 121, core network server 302) determines the neural network formation configuration based at least in part on metrics, feedback, and / or other types of information from a user device (e.g., UE 110). For illustration, base station 121 receives a message from UE 110 indicating one or more capabilities of UE 110. Then, base station 121 determines the neural network formation configuration based at least in part on the capabilities received from UE 110. Alternatively or additionally, base station 121 forwards the UE capabilities to core network server 302, and core network server 302 determines the neural network formation configuration based on the capabilities of UE 110. As another example, base station 121 determines the neural network formation configuration based on scheduled MU-MIMO downlink and / or uplink transmissions in the wireless communication system. When determining the neural network formation configuration, the network selects any combination of neural network formation configuration elements for a user device-side deep neural network, a base station-side deep neural network, and / or a deep neural network corresponding to both a user device-side deep neural network and a base station-side deep neural network.

[0139] When determining the neural network formation configuration, the network entity (e.g., base station 121, core network server 302) sometimes selects a default neural network formation configuration. Alternatively or additionally, the network entity analyzes multiple neural network formation configurations and selects a neural network formation configuration from the multiple neural network formation configurations that is consistent with current channel conditions, capabilities of a particular UE, etc.

[0140] At 1010, a network entity generates a message including an indication of a neural network formation configuration for a deep neural network. The network entity (e.g., base station 121, core network server 302), for example, generates a message including an index value that maps to one or more entries of a neural network table (e.g., neural network table 216, neural network table 272, neural network table 316). In some embodiments, the network entity includes an indication of a processing allocation for the deep neural network, wherein the processing allocation specifies a processing chain function for a deep neural network formed by the neural network formation configuration. Alternatively or additionally, the network entity specifies a time instance in the message that indicates a time and / or location to begin processing communications using the deep neural network.

[0141] At 1015, the network entity transmits a message to the user equipment to direct the user equipment to form a deep neural network using the neural network formation configuration, and to process communications transmitted over the wireless communication system using the deep neural network. The network entity (e.g., core network server 302) transmits the message to the user equipment (e.g., UE 110), for example, by transmitting the message to base station 121, which transmits the message to the user equipment. Alternatively or additionally, the network entity (e.g., base station 121) transmits the message to the user equipment (e.g., UE 110). In an embodiment, the deep neural network corresponds to a user equipment side deep neural network.

[0142] In some embodiments, a network entity (e.g., base station 121) transmits a message using a specific RAT and / or carrier based on the processing allocation of the deep neural network. For example, base station 121 transmits a first neural network formation configuration having a first processing allocation (e.g., downlink data channel processing) using a first RAT and / or carrier, and transmits a second neural network formation configuration having a second processing allocation (e.g., downlink control channel processing) using a second RAT and / or carrier.

[0143] In some implementations, the network entity analyzes the feedback received from the user equipment at 1020. As an example, the network entity (e.g., base station 121) receives metrics from the user equipment (e.g., UE 110) and analyzes the metrics to determine whether to update and / or reconfigure the neural network with a different neural network formation configuration. Alternatively or additionally, the base station forwards the feedback to the network entity (e.g., core network server 302), and the network entity analyzes the feedback to determine whether to update and / or reconfigure the neural network with a different neural network formation configuration.

[0144] At 1025, the network entity updates the deep neural network based on the feedback. For example, the network entity (e.g., base station 121, core network server 302) analyzes a plurality of neural network formation configurations included in a neural network table (e.g., neural network table 216, neural network table 272, neural network table 316), and selects a second neural network formation configuration consistent with the new channel condition indicated by the feedback. The network entity generates a second message including an indication of the second neural network formation configuration, and transmits the second message to the user equipment. This includes configuring and transmitting the second message as described at 1010 and / or 1015. In an embodiment, the network entity iteratively performs various operations of method 1000, represented herein by a dashed line returning from 1025 to 1010.

[0145] Fig.11An exemplary method 1100 for forming a neural network based on a neural network formation configuration is illustrated. In some aspects, the operations of method 1100 are implemented by a UE, such as UE 110.

[0146] At 1105, a user equipment receives a message indicating a neural network formation configuration for a deep neural network to process communications transmitted on a wireless communication system. The user equipment (e.g., UE 110) receives a message, for example, from a base station (e.g., base station 121), indicating a neural network formation configuration and / or a processing allocation for a deep neural network formed using the neural network formation configuration. In some embodiments, the user equipment receives multiple messages, each message belonging to a different neural network formation configuration (e.g., a first message indicating a first neural network formation configuration for a first deep neural network to process downlink communications, a second message indicating a second neural network formation configuration for a second deep neural network to process uplink communications). In an embodiment, the user equipment receives the message as a broadcast message from a base station, or as a UE-specific message.

[0147] At 1110, the user device forms a deep neural network using the neural network formation configuration indicated in the message. For illustration, the user device (e.g., UE 110) extracts parameter configurations, such as coefficients, weights, layer connections, etc., from the message and forms the deep neural network using the extracted parameter configurations. As another example, the user device (e.g., UE 110) extracts an index value from the message and obtains the parameter configuration from a neural network table, as further described. Alternatively or additionally, the user device extracts a time instance from the message, wherein the time instance indicates a time and / or location to begin processing communications using the deep neural network formed by the neural network formation configuration.

[0148] At 1115, the user equipment receives a communication from the base station. For example, the user equipment (e.g., UE 110) receives a downlink data channel communication from the base station (e.g., base station 121), receives a downlink control channel communication from the base station, etc. At 1120, the user equipment processes the communication using a deep neural network to extract information transmitted in the communication. As an example, the user equipment (e.g., UE 110) processes the communication using a deep neural network based on the processing allocation included in the message, such as by demodulating and / or decoding downlink communications from the base station (e.g., base station 121), encoding and / or modulating uplink communications to the base station, etc. using a deep neural network.

[0149] In some implementations, the user equipment optionally transmits feedback based on the communication at 1125. For example, the user equipment (e.g., UE 110) generates metrics (e.g., error metrics, SINR information, CQI, CSI, Doppler feedback) based on the communication and transmits the metrics as feedback to a base station (e.g., base station 121).

[0150] Having described some example methods that may be used to implement various aspects of base station-user equipment messaging using deep neural networks, consider now a discussion of generating and transmitting neural network formation configurations in accordance with one or more embodiments.

[0151] Generate and transmit neural network formation configuration

[0152] In supervised learning, a machine learning module processes labeled training data to generate an output. The machine learning module receives feedback about the accuracy of the generated output and modifies processing parameters to improve the accuracy of the output. Fig.12 An example 1200 is illustrated that describes aspects of generating multiple NN formation configurations. Sometimes, various aspects of example 1200 are described by Figure 2 and Figure 3 This may be implemented by any combination of the training module 270, the base station neural network manager 268, the core network neural network manager 312 and / or the training module 314.

[0153] Fig.12 The upper part includes Figure 4 The machine learning module 400 of FIG. 400. In an embodiment, the neural network manager determines to generate different NN formation configurations. For illustration, consider a case where the base station neural network manager 268 determines to generate a NN formation configuration by selecting a combination of NN formation configuration elements from a neural network table, wherein the NN formation configuration corresponds to a UE decoding and / or demodulating downlink communications. In other words, the NN formation configuration (through a combination of NN formation configuration elements) forms a DNN that processes downlink communications received by the UE. However, transmission channel conditions often change, which in turn affects the characteristics of downlink communications. For example, a first transmission channel distorts downlink communications by introducing a frequency offset, a second transmission channel distorts downlink communications by introducing a Doppler effect, a third transmission channel distorts downlink communications by introducing a multipath channel effect, and so on. In order to accurately process downlink communications (e.g., reduce bit errors), various embodiments select multiple NN formation configurations, wherein each NN formation configuration (and the associated combination of NN formation configuration elements) corresponds to a corresponding input condition, such as a first transmission channel, a second transmission channel, and the like.

[0154] Training data 1202 represents exemplary input to machine learning module 400. Fig.12, the training data represents data corresponding to the downlink communication. The training data 1202 may include, for example, digital samples of the downlink communication signal, recovered symbols, recovered frame data, etc. In some embodiments, the training module generates the training data mathematically or accesses a file storing the training data. At other times, the training module obtains real-world communication data. Therefore, the training module can use mathematically generated data, static data, and / or real-world data to train the machine learning module. Some embodiments generate input characteristics 1204 that describe various qualities of the training data, such as transmission channel metrics, UE capabilities, UE speed, etc.

[0155] The machine learning module 400 analyzes the training data and generates an output 1206, represented here as binary data. Some embodiments use the same set of training data and / or additional training data with the same input characteristics to iteratively train the machine learning module 400 to improve the accuracy of the machine learning module. During training, the machine learning module modifies some or all of the architecture and / or parameter configurations of the neural network included in the machine learning module, such as node connections, coefficients, kernel size, etc. At certain moments in training, the training module determines to extract the architecture and / or parameter configuration 1208 (e.g., pooling parameters, kernel parameters, layer parameters, weights) of the neural network, such as when the training module determines that the accuracy meets or exceeds the expected threshold, the training process meets or exceeds the number of iterations, etc. The training module then extracts the architecture and / or parameter configuration from the machine learning module for use as a NN formation configuration and / or NN formation configuration element. The architecture and / or parameter configuration may include any combination of fixed architecture and / or parameter configuration and / or variable architecture and / or parameter configuration.

[0156] Fig.12 The lower half of includes a neural network table 1212 representing a set of NN formation configuration elements, such as Figure 2 and Figure 3 1202. The neural network table 1212 stores various combinations of architecture configurations, parameter configurations, and input characteristics, although alternative embodiments exclude input characteristics from the table. Various embodiments update and / or maintain the NN formation configuration elements and / or input characteristics as the machine learning module learns additional information. For example, at index 1214, the neural network manager and / or training module updates the neural network table 1212 to include the architecture and / or parameter configuration 1208 generated by the machine learning module 400 when analyzing the training data 1202.

[0157] The neural network manager and / or training module may alternatively or additionally add input characteristics 1204 to the neural network table and link the input characteristics to the architecture and / or parameter configuration 1208. This allows the input characteristics to be obtained simultaneously with the architecture and / or parameter configuration, such as by using an index value introduced to the neural network table (e.g., reference to a NN formation configuration, reference to a NN formation configuration element). In some embodiments, the neural network manager selects a NN formation configuration by matching the input characteristics to the current operating environment, such as by matching the input characteristics to current channel conditions, UE capabilities, UE characteristics (e.g., speed, location, etc.), etc.

[0158] Having described generating and transmitting a neural network formation configuration, consider now a discussion of signaling and control transactions over a wireless communication system that may be used to transmit a neural network formation configuration in accordance with one or more embodiments.

[0159] Transmits signaling and control transactions for neural network formation configuration

[0160] Figure 13-15 FIG. 1 illustrates an exemplary signaling and control transaction diagram between a base station, a user equipment, and / or a core network server according to one or more aspects of a transport neural network formation configuration, such as a transport NN formation configuration. In an embodiment, the signaling and control transactions may be performed by Figure 1 Base station 120 and UE 110, or Figure 3 Any one of the core network servers 302 uses Figure 1-12 Execute the elements of .

[0161] Fig.13 The signaling and control transaction diagram 1300 of FIG. 1 illustrates a first example of signaling and control transactions for transmitting a neural network formation configuration. In an embodiment, some or all of the signaling and control transactions described with reference to the signaling and control transaction diagram correspond to the signaling and control transactions described with reference to FIG. Figure 7 and / or Figure 8 Describes the signaling and control transactions.

[0162] As shown, at 1305, the base station 121 maintains a neural network table. For example, a base station neural network manager and / or a training module (e.g., the base station neural network manager 268, the training module 270) of the base station 121 generates and / or maintains a neural network table (e.g., the neural network table 272) using any combination of mathematically generated training data, data extracted from real communications, files, etc. In various embodiments, the base station 121 maintains a plurality of neural network tables, wherein each neural network table includes a plurality of neural network formation configurations and / or neural network formation configuration elements for a specified purpose, such as a first neural network table specified for data channel communications, a second neural network table specified for control channel communications, and so on.

[0163] At 1310, the base station 121 transmits a neural network table to the UE 110. As an example, the base station transmits the neural network table using layer 3 messaging (e.g., a radio resource control (RRC) message). In transmitting the neural network table, the base station transmits any combination of an architecture and / or parameter configuration that can be used to form a deep neural network, examples of which are provided in the present disclosure. Alternatively or additionally, the base station transmits an indication of a processing allocation specified for the neural network table with the neural network table. Thus, the base station transmits multiple neural network tables to the UE, specifying a corresponding processing allocation for each neural network table. In some embodiments, the base station 121 broadcasts the neural network table to a group of UEs. At other times, the base station 121 transmits a UE-specific neural network table to the UE 110.

[0164] At 1315, base station 121 identifies the neural network formation configuration used in processing the communication. For example, the base station determines the neural network formation configuration used in processing the communication by selecting a neural network formation configuration such as in Figure 7 The neural network described at 705 of the present invention forms a combination of architectural elements, by analyzing information such as channel type (e.g., downlink, uplink, data, control, etc.), transmission medium properties (e.g., power measurement, signal-to-interference-plus-noise ratio (SINR) measurement, channel quality indicator (CQI) measurement), coding scheme, UE capabilities, BS capabilities, etc. processed by the deep neural network. Therefore, in some embodiments, the base station 121 can be based on any combination of information such as in Figure 8 The various metrics and / or parameters described at 815, 820, and 830 of the present invention are received to identify the neural network formation configuration. In some embodiments, base station 121 identifies a default neural network formation configuration to use as the neural network formation configuration. Thus, identifying the neural network formation configuration may include identifying the default neural network formation configuration and / or an update to the neural network formation configuration.

[0165] In identifying the neural network formation configuration, the base station 121 finds the neural network formation configuration in the neural network table that corresponds to the determined neural network formation configuration. In other words, the base station 121 identifies the neural network formation configuration, such as by correlating and / or matching input characteristics. Figure 2 The base station identifies the neural network formation configuration and / or the neural network formation configuration element that is consistent with the determined neural network formation configuration in the neural network table 272 and / or the neural network table 216. When identifying the neural network formation configuration and / or the neural network formation configuration element in the neural network table, the base station identifies the index value of the neural network formation configuration and / or the neural network formation configuration element.

[0166] At 1320, base station 121 transmits an instruction directing UE 110 to form a deep neural network using a neural network formation configuration from the neural network table. For example, similar to Figure 7 710 and / or Figure 8 As described at 840 of , the base station 121 transmits the index value to the UE 110 and directs the UE 110 to form a deep neural network using the neural network formation configuration indicated by the index value. The base station may transmit the indication to the UE in any suitable manner. As an example, the base station uses a layer 2 message (e.g., a radio link control (RLC) message, a medium access control (MAC) control element) to transmit an index value corresponding to the neural network formation configuration. In some embodiments, the base station compares the current operating environment (e.g., channel conditions, UE capabilities, BS capabilities, metrics) with the input characteristics stored in the neural network table, and identifies the stored input characteristics that are consistent with the current operating environment. In turn, the base station obtains an index value of the stored input characteristic, which in turn provides an index value of the neural network formation configuration and / or an index value of the neural network formation configuration element. The base station 121 then transmits the index value as an indication. Sometimes, the base station 121 includes a processing allocation to indicate a location in the processing chain where the deep neural network is applied. In some embodiments, the base station uses a downlink control channel to transmit the index value and / or the processing allocation.

[0167] Sometimes, the base station transmits to the UE a rule specifying an operating parameter related to applying a neural network formation configuration. In one example, the rule includes a time instance indicating when to process a communication with a deep neural network formed using the neural network formation configuration. Alternatively or additionally, the rule specifies a time threshold that directs the UE to use a default neural network formation configuration instead of a specified neural network formation configuration when the data channel and the control channel are within the time threshold. Additionally or alternatively, when the data channel and the control channel are within the time threshold, the rule may direct the user equipment to use the same neural network formation configuration for data channel communication and control channel communication. For illustration, consider an example in which the UE processes data channel communication using a first deep neural network (formed using a first neural network formation configuration) and processes control channel communication using a second deep neural network (formed using a second neural network formation configuration). If the data channel communication and the control channel communication fall within the time threshold specified by the time instance, the UE processes both channels using a default deep neural network (formed by the default neural network formation configuration) and / or the same deep neural network because there may not be enough time to switch between the first deep neural network and the second deep neural network.

[0168] Sometimes, the base station transmits a default neural network formation configuration using a downlink control channel to specify a default neural network formation configuration to the UE, wherein the default neural network formation configuration forms a deep neural network that processes various input data. In some embodiments, the default neural network formation configuration forms a deep neural network that processes various input data with an accuracy within a threshold range. The default neural network formation configuration may include a general neural network formation configuration.

[0169] To illustrate, some embodiments generate or select neural network formation configurations for specific operating conditions, such as a first neural network formation configuration specific to UE downlink control channel processing (e.g., demodulation and / or decoding) having a current operating environment “X,” a second neural network formation configuration specific to UE downlink control channel processing having a current operating environment “Y,” etc. For example, the first neural network formation configuration may be associated with a current operating environment in which a detected interference level is high, the second neural network formation configuration may be associated with a current operating environment in which a detected interference level is low, the third neural network formation configuration may be associated with a current operating environment in which a connected UE appears stationary, the fourth neural network formation configuration may be associated with a current operating environment in which a connected UE appears to be moving at a particular speed, and so on.

[0170] Forming a deep neural network using a neural network formation configuration for a particular operating condition (relative to forming a deep neural network by other neural network formation configurations) improves the accuracy of outputs generated by the deep neural network when processing input data corresponding to the particular operating condition. However, this introduces a tradeoff because the deep neural network formed by the neural network formation configuration for the particular operating condition generates outputs with lower accuracy when processing inputs associated with other operating conditions. In contrast, the default neural network formation configuration corresponds to a neural network formation configuration that processes a variety of inputs, such as a variety of inputs across more operating conditions. In other words, a deep neural network configured with a default neural network formation configuration processes a wider variety of communications relative to a neural network formation configuration for a particular operating condition.

[0171] At 1325, UE 110 forms a deep neural network using the neural network formation configuration. As an example, the UE extracts the index value transmitted by base station 121 and obtains the neural network formation configuration and / or the neural network formation configuration element by accessing the neural network table using the index value. Alternatively or additionally, UE 110 extracts the processing allocation and forms the deep neural network in the processing chain as specified by the processing allocation.

[0172] At 1330, base station 121 transmits a communication, such as a downlink data channel communication, to UE 110. At 1335, UE 110 processes the communication using the deep neural network. For example, UE 110 processes the downlink data channel communication using the deep neural network to recover the data. As another example, processing the communication includes processing a reply to the communication, wherein UE 110 processes the uplink communication using the deep neural network in the reply to the downlink communication.

[0173] Fig.14 1400 illustrates a second example of signaling and control transactions for communicating a neural network formation configuration. As shown, at 1405, the core network server 302 maintains a neural network table. The core network neural network manager 312 and / or the training module 314 of the core network server 302 generates and / or maintains the neural network table 316, for example, using any combination of mathematically generated training data, data extracted from real-world communications, files, etc. In various embodiments, the core neural network server 302 maintains a plurality of neural network tables, wherein each neural network table includes a plurality of neural network formation configuration elements for specifying processing assignments (e.g., a first neural network table specified for data channel communications, a second neural network table specified for control channel communications).

[0174] At 1410, the core network server 302, such as by using Figure 3 The core network interface 318 transmits the neural network table to the base station 121. In some embodiments, the core network server transmits multiple neural network tables to the base station. At 1415, the base station 121 transmits the neural network table to the UE 110, such as by transmitting the neural network table using layer 3 messaging, transmitting the neural network table using a downlink control channel, by broadcasting the neural network table, transmitting the neural network table using a UE-specific message, etc.

[0175] At 1420, the core network server 302 selects a neural network formation configuration. As an example, the core network server 302 compares the current operating environment to the input characteristics stored within the neural network table and identifies a stored input characteristic that is consistent with the current operating environment (e.g., channel conditions, UE capabilities, BS capabilities, metrics). The core network server then obtains an index value for the consistent input characteristic, which in turn provides an index value for the neural network formation configuration and / or a neural network formation configuration element. The core network server 302 then transmits the selected neural network formation configuration to the base station at 1425, such as by transmitting the index value using the core network interface 318. In some embodiments, the core network server transmits the processing assignment with the neural network formation configuration.

[0176] At 1430, base station 121 forwards the neural network formation configuration to UE 110. As an example, base station 121 transmits an index value to UE 110, such as via layer 2 messaging (e.g., RLC message, MAC control element), to instruct the UE to use the neural network formation configuration to form a deep neural network, such as Fig.13 In an embodiment, the base station 121 additionally forms a complementary deep neural network based on a neural network formation configuration (not shown here), such as in Figure 6 middle, Figure 7 720 places, Figure 8 845 and / or Figure 9-1 Sometimes, the base station transmits the processing assignment received from the core network server together with the index value. In response to receiving the neural network formation configuration, UE 110 forms a deep neural network at 1435 and uses the deep neural network to process the communication received from base station 121 at 1440, such as in Figure 7 725 of Figure 8 810 places Figure 9-2 935 and / or Fig.13 As described in 1335.

[0177] In some embodiments, when similarities exist in the wireless communication system, the UE derives the first neural network forming configuration from the second neural network forming configuration. For illustration, consider the example of a quasi-communicating channel. A quasi-communicating channel is a channel in a wireless communication system that has shared or identical properties, such as the same delay spread, the same Doppler spread, the same spatial signature, the same spatial beam, the same spatial direction, the same data rate, etc. Alternatively or additionally, the quasi-communicating channels have related physical properties within a threshold range or value. In various embodiments, in response to identifying these similarities, the UE derives the first neural network forming configuration from the second neural network forming configuration.

[0178] For illustration, now consider Fig.15 A third example of signaling and control transactions for transmitting a neural network formation configuration is illustrated by signaling and control transaction diagram 1500. In the third example, the UE derives the neural network formation configuration based on the identified similarities. Although signaling and control transaction diagram 1500 illustrates Figure 1 Base station 121 and UE 110, alternative embodiments include a core network server such as Figure 3 The core network server 302 performs some or all of the functions performed by the base station 121.

[0179] At 1505, the base station 121 identifies two or more channels as quasi-communication channels. The base station 121 or the core network server 302 (using Figure 2The base station neural network manager 268 or Figure 3 The core network neural network manager 312) compares various beam characteristics of two or more channels, such as direction, strength, divergence, profile, quality, etc. When any number of attributes match and / or correlate with each other within a predefined threshold, the base station 121 determines that the two or more channels are in communication.

[0180] At 1510, the base station 121 transmits an indication of a quasi-communicating channel to the UE 110, such as by transmitting an indication of the quasi-communicating channel using layer 2 messaging, layer 3 messaging, and / or layer 1 signaling. The indication indicates that any number of channels are in quasi-communicating relation with each other, where the channels can be physical channels and / or logical channels. As an example, the indication indicates that a downlink control channel is in quasi-communicating relation with a downlink data channel when the channels have similar data rates. As another example, the indication indicates that two physical channels at different carrier frequencies are in quasi-communicating relation based on the physical channels having similar spatial beams.

[0181] At 1515, base station 121 transmits an indication of a neural network formation configuration for one of the quasi-communication channels. For example, the base station transmits an index value and processing assignment specifying a formed deep neural network for processing a first channel, such as a downlink control channel.

[0182] At 1520, UE 110 forms a first deep neural network using the neural network formation configuration for a first channel (e.g., a downlink control channel). At 1525, UE 110 identifies the first channel as one of the quasi-communication channels. The UE compares the first channel to a quasi-communication channel identified by and received from base station 121, for example. In response to identifying the first channel as one of the quasi-communication channels, the UE determines to apply the neural network formation configuration to other channels that are in quasi-communication with the first channel. Therefore, at 1530, UE 110 uses the neural network formation configuration to form a deep neural network for the other quasi-communication channels. By identifying the channel as a quasi-communication channel, only the neural network formation configuration needs to be changed for one of the quasi-communication channels, which will result in changing the neural network formation configuration for all quasi-communication channels without having to change the neural network formation configuration for each other quasi-communication channel individually.

[0183] Having described signaling and control transactions that may be used to communicate a neural network formation configuration, consider now an example environment that may be used to communicate a neural network formation configuration using a set of candidate neural network formation configurations in accordance with one or more implementations.

[0184] Transmit candidate neural networks to form a set of configurations

[0185] In some embodiments, the network entity transmits a set of candidate NN formation configuration elements to the UE, and the UE selects a NN formation configuration from the set of candidate NN formation configuration elements. Figure 16-1 and 16-2 , which illustrates an exemplary environment 1600 for transmitting a neural network formation configuration using a set of candidate neural network formation configuration elements. Figure 16-1 and 16-2 An exemplary environment 1600 is illustrated at different points in time, labeled as environment 1600-1, environment 1600-2, environment 1600-3, and environment 1600-4, respectively. Thus, environment 1600-1, environment 1600-2, environment 1600-3, and environment 1600-4 collectively represent environment 1600 in a series of events.

[0186] In environment 1600-1, the base station neural network manager 268 of the base station 121 receives Fig.12 At least one candidate NN is selected from the neural network table 1212 to form a set of configuration elements. For example, the base station neural network manager 268 receives various metrics, parameters and / or other types of feedback, such as in Figure 8 815 and 825. The base station neural network manager 268 then analyzes the neural network table 1212 to identify NN formation configuration elements consistent with the feedback, correct problems identified by the feedback, and so on.

[0187] In an embodiment, the base station neural network manager identifies a plurality of sets of NN formation configuration elements, such as by identifying NN formation configurations having input characteristics that fall within a threshold range. In environment 1600-1, the base station neural network manager 268 identifies and selects three sets of NN formation configuration elements: NN formation configuration element set 1602, NN formation configuration element set 1604, and NN formation configuration element set 1606. In some embodiments, each NN formation configuration element set corresponds to a respective NN formation configuration.

[0188] In environment 1600-1, NN formation configuration element set 1602 includes index values ​​1608, 1610, and 1612, NN formation configuration element set 1604 includes index values ​​1614, 1616, and 1618, and NN formation configuration element set 1606 includes index values ​​1620, 1622, and 1624. Although environment 1600-1 illustrates that the base station neural network manager 268 of the base station 121 selects the set of candidate NN formation configuration elements, alternative embodiments use the core network neural network manager 312 of the core network server 302, or both, to select the set of candidate NN formation configuration elements. Alternatively or additionally, the base station neural network manager and / or the core network neural network manager selects a single index value that maps to the group of NN formation configuration elements.

[0189] In environment 1600-2, base station 121 transmits an indication 1608 to UE 110, which refers to each set of candidate NN formation configuration elements (e.g., NN formation configuration element set 1602, NN formation configuration element set 1604, NN formation configuration element set 1606). For example, the indication includes index values ​​1602, 1604, 1606, 1608, 1610, 1612, 1614, 1616, and 1618 to transmit three candidate NN formation configuration element sets to UE 110. Alternatively or additionally, the indication includes an instruction to select a NN formation configuration from the candidate NN formation configuration sets. Although environment 1600-1 and environment 1600-2 illustrate NN formation configuration element sets corresponding to candidate NN formation configurations, alternative embodiments select a single index mapped to a group of elements for each candidate NN formation configuration, as further described.

[0190] exist Figure 16-2In the environment 1600-3 of , the UE neural network manager 218 of the UE 110 analyzes the set of candidate NN formation configuration elements to determine which of the set is used to process a specific communication (e.g., for processing a transmission communication, for processing a received communication). In one example, the UE neural network manager forms a first candidate deep neural network 1628 (candidate DNN 1628) based on the first NN formation configuration element set 1602 (e.g., index values ​​1608, 1610, 1612), forms a second candidate deep neural network 1630 (candidate DNN 1630) based on the second NN formation configuration element set 1604 (e.g., index values ​​1608, 1610, 1612), forms a third candidate deep neural network 1632 (candidate DNN 1632) based on the third NN formation configuration element set 1606 (e.g., index values ​​1614, 1616, 1618), and so on. In other words, the UE neural network manager 218 uses the index value to obtain each NN formation configuration from the neural network table (e.g., neural network table 216), and uses the set of candidate NN formation configuration elements to form at least one candidate deep neural network.

[0191] The UE neural network manager 218 then obtains one or more metrics associated with the set of candidate deep neural networks and selects a neural network formation configuration based on the one or more metrics. Some embodiments obtain a corresponding error metric for each candidate neural network formation configuration in the set of candidate neural network formation configurations to generate a set of error metrics. The UE neural network manager then compares each error metric in the set of error metrics to a threshold value and selects a neural network formation configuration based on the set of error metrics. For illustration, in some examples, the UE neural network manager identifies a specific error metric (e.g., CRC pass) in the set of error metrics that indicates fewer errors relative to other error metrics in the set of error metrics, and selects a specific candidate neural network formation configuration corresponding to the specific error metric from the set of candidate neural network formation configurations as the neural network formation configuration. In an exemplary embodiment, the UE neural network manager 218 analyzes each candidate DNN, such as by providing known inputs 1634 to each candidate DNN and comparing corresponding outputs (e.g., outputs 1636, outputs 1638, outputs 1640) from each candidate deep neural network. Alternatively or additionally, the UE neural network manager provides other types of inputs, examples of which are provided herein.

[0192] For illustration, consider an example where each candidate DNN corresponds to generating transmitter communications. The UE neural network manager may compare various transmission metrics of outputs 1636, outputs 1638, outputs 1640, such as output power levels, modulation quality metrics, intermodulation product metrics, etc. As another example, consider a case where each candidate DNN corresponds to processing receiver communications. The UE neural network manager may then compare various receiver metrics of outputs 1636, outputs 1638, outputs 1640, such as CRC, signal-to-noise ratio (SNR), adjacent / alternate channel metrics, BER, inter-symbol interference (ISI) metrics, etc. The UE neural network manager 218 may then select one of the candidate neural network formation configurations based on the analysis.

[0193] In environment 1600-4, UE 110 uses a deep neural network 1644 (DNN 1644) formed by the selected candidate neural network formation configuration to process communications 1642. In environment 1600-4, the DNN (which corresponds to candidate DNN 1632) processes downlink communications received from base station 121, but in an alternative embodiment, DNN 1644 processes uplink communications, such as by generating transmission communications that UE 110 transmits to base station 121.

[0194] Now consider Fig.17 , which illustrates the signaling and control transactions in Figure 1700. Fig.17 1700 for transmitting a neural network formation configuration. In the fourth example, the base station provides a plurality of candidate neural network formation configurations to the UE, and the UE selects one of the candidate neural network formation configurations to form a deep neural network for processing communications. Although the signaling and control transaction diagram 1700 illustrates Figure 1 1 and UE 110, but alternative embodiments include a core network server such as Figure 3 The core network server 302 performs some or all of the functions performed by the base station 121.

[0195] At 1705, the base station 121 determines a candidate neural network to form a configuration set. For example, the base station 121 or the core network server 302 (using Figure 2 The base station neural network manager 268 or Figure 3The core network neural network manager 312 of the base station analyzes the neural network table to identify candidate neural network formation configurations (e.g., a set of neural network structure configuration elements) that are linked to input characteristics within a specified range and / or threshold. For illustration, the base station analyzes the input characteristics to identify candidate neural network formation configurations having operating conditions that fall within an "X" + / - threshold. In identifying the candidate neural network formation configurations, the base station may alternatively or additionally identify an index value for each candidate neural network formation configuration element selected for the corresponding candidate neural network formation configuration, such as a reference Figure 16-1 as described in environment 1600-1 (e.g., NN forms a set of configuration elements 1602, NN forms a set of configuration elements 1604, NN forms a set of configuration elements 1606).

[0196] At 1710, base station 121 transmits an indication of the candidate neural network forming a configuration set to UE 110. For example, referring to Figure 16-1 In environment 1600-2, the base station transmits each index value associated with the candidate neural network formation configuration set to the UE (e.g., a set of index values ​​corresponding to a set of neural network formation configurations and / or configuration elements). In some cases, the base station transmits the index values ​​in a single message, while in other cases, the base station transmits each index value in a corresponding message.

[0197] At 1715, UE 110 analyzes the set of candidate neural network formation configurations based on communications transmitted over the wireless communication system, such as communications transmitted by base station 121. In some embodiments, UE 110 processes the communication by forming a corresponding deep neural network with each candidate neural network formation configuration, and processing the communication with each corresponding deep neural network, such as extracting a corresponding NN formation configuration element from a neural network table. As an example, the UE performs blind identification of the neural network formation configuration by processing the communication using each of the corresponding deep neural networks to attempt to decode an expected data pattern within the communication, such as referring to Figure 16-2 Then, the UE analyzes each corresponding deep neural network, such as by analyzing the output of each corresponding deep neural network and generating a corresponding metric (e.g., accuracy, bit error, etc.).

[0198] At 1720, UE 110 selects a candidate neural network formation configuration from the set of candidate neural network formation configurations. For example, UE selects a candidate neural network formation configuration that forms a corresponding deep neural network that decodes the desired data pattern with minimal bit errors relative to other candidate neural network formation configurations. At 1725, UE 110 uses the selected candidate neural network formation configuration to form a deep neural network and processes the communication with the deep neural network, such as with reference to Figure 16-2 as described in environment 1600-4.

[0199] Having described an example environment and signaling and control transactions in which neural network formation configurations may be communicated using a set of candidate neural network formation configurations, consider now some example methods in accordance with one or more implementations.

[0200] Exemplary Methods

[0201] According to one or more aspects of the transmission neural network formation configuration, reference Fig.18 and Fig.19 Describe exemplary methods 1800 and 1900. The order in which the method boxes are described is not intended to be construed as limiting, and any number of the method boxes may be skipped or combined in any order to implement a method or an alternative method. In general, any of the components, modules, methods, and operations described herein may be implemented using software, firmware, hardware (e.g., fixed logic circuits), manual processing, or any combination thereof. Some operations of the exemplary methods may be described in the general context of executable instructions stored on a computer-readable memory local and / or remote to a computer processing system, and implementations may include software applications, programs, functions, and the like. Alternatively or additionally, any of the functions described herein may be performed, at least in part, by one or more hardware logic components, such as, but not limited to, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), an application specific standard product (ASSP), a system on a chip (SoC), a complex programmable logic device (CPLD), and the like.

[0202] Fig.18 An exemplary method 1800 for transmitting a neural network formation configuration to process communications transmitted over a wireless communication system is illustrated. In some implementations, the operations of method 1800 are performed by a network entity such as one of base station 120 and / or core network server 302.

[0203] At 1805, a network entity transmits a neural network table including a plurality of neural network forming configuration elements, wherein each of the plurality of neural network forming configuration elements configures at least a portion of a deep neural network to process communications transmitted via a wireless communication system. For example, a network entity (e.g., base station 121) transmits a neural network table (e.g., neural network table 272, neural network table 316) to a UE (e.g., UE 110) using a broadcast or multicast message to a group of UEs. As another example, a network entity (e.g., base station 121) transmits the neural network table to a UE (e.g., UE 110) using a UE-specific message. In some embodiments, a network entity (e.g., core network server 302) transmits the neural network table to a base station (e.g., base station 121) and instructs the base station to transmit the neural network table to the UE (e.g., UE 110). Sometimes, the network entity transmits the neural network table using layer 3 messaging. Alternatively or additionally, the network entity transmits a plurality of neural network tables, wherein each neural network table has a specified processing assignment, as further described.

[0204] At 1810, the network entity selects one or more neural network formation configuration elements from a plurality of neural network formation configuration elements to create a neural network formation configuration. For example, the network entity (e.g., base station 121, core network server 302) selects the neural network formation configuration element by comparing the current operating conditions with the input characteristics stored in the neural network table, and selecting the neural network formation configuration element by associating or matching the input characteristics with the current operating conditions. In some embodiments, the network entity selects a set of candidate neural network formation configurations.

[0205] At 1815, the network entity transmits an indication to the user equipment to direct the user equipment to form a deep neural network using the neural network formation configuration and to process communications using the deep neural network. As an example, the network entity (e.g., base station 121) determines an index value of a neural network table corresponding to the neural network formation configuration and / or a set of neural network formation configuration elements, and transmits the index value to UE 110, such as by transmitting the index value using a downlink control channel, by transmitting the index value in a layer 2 message, etc. As another example, the network entity (e.g., core network server 302) transmits an indication to a base station (e.g., base station 121) and instructs the base station to transmit the indication to the user equipment (e.g., UE 110). The network entity may alternatively or additionally indicate a processing allocation for a deep neural network formed by the neural network formation configuration, such as by indicating a processing allocation to direct the user equipment to process downlink control channel communications with a deep neural network. In some embodiments, the network entity indicates a set of candidate neural network formation configurations, such as a reference Figure 16-1 and16-2 described.

[0206] When transmitting the index value, the network entity sometimes specifies rules regarding when to process communications with the corresponding deep neural network. As an example, the network entity (e.g., base station 121, core network server 302) determines a time threshold between data channel communications and control channel communications. When transmitting the index value to guide a user device (e.g., UE 110) to form a deep neural network, the network entity transmits the time value threshold and a rule that guides the user device to form a deep neural network using a default formation configuration when the timing between the data channel communication and the control channel communication is below the time threshold. Additionally or alternatively, the rule may guide the user device to use the same neural network formation configuration for data channel communications and control channel communications when both the data channel and the control channel are within the time threshold, and thus there is not enough time to switch between different DNNs.

[0207] Fig.19 An exemplary method 1900 is illustrated for communicating a neural network formation configuration to process communications transmitted via a wireless communication system, such as by processing a set of candidate neural network formation configurations and selecting a particular candidate neural network formation configuration, such as with reference to Fig.17 In some implementations, the operations of method 1900 are performed by a user equipment such as UE 110.

[0208] At 1905, a user device receives a neural network table that includes a plurality of neural network forming configuration elements that provide the user device with the ability to configure a deep neural network to process communications transmitted over a wireless communication system. The user device (e.g., UE 110) receives the neural network table (e.g., neural network table 272) from a base station (e.g., base station 121), for example, in layer 3 messaging. As another example, the user device receives the neural network table in a multicast or broadcast message. Alternatively or additionally, the user device receives the neural network table in a UE-specific message. In some cases, the user device receives multiple neural network tables, each of which has a specified processing assignment.

[0209] At 1910, a user device receives a message directing the user device to form a deep neural network using a neural network formation configuration based on one or more neural network formation configuration elements in a plurality of neural network formation configuration elements. For example, a user device (e.g., UE 110) receives in the message a set of index values ​​pointing to a neural network table, wherein the set of index values ​​corresponds to a set of candidate neural network formation configurations and / or configuration elements, such as Figure 16-1 Communication 1626 in environment 1600-2, Fig.17In some embodiments, the message includes an indication of a processing allocation for a deep neural network formed by the neural network formation configuration, such as a communication channel processing allocation, a processing chain processing allocation, etc.

[0210] At 1915, the user device forms a deep neural network using the neural network formation configuration by accessing a neural network table to obtain a neural network formation configuration element. In one example, the user device (e.g., UE 110) accesses the neural network table using an index value. Alternatively or additionally, the user device obtains the set of candidate neural network formation configurations by accessing the neural network table using a set of index values, and forms a candidate deep neural network using each candidate neural network formation configuration, such as a reference neural network. Figure 16-1 Environment 1600-2 and Figure 16-2 For example, UE 110 uses each candidate neural network to form a configuration and / or configuration element to form a corresponding candidate deep neural network, and processes communications with each candidate corresponding deep neural network to obtain a corresponding metric about each candidate deep neural network, such as Fig.17 1715, and selecting a neural network formation configuration based on one or more metrics, such as Fig.17 As described in 1720.

[0211] As an example, a user equipment (e.g., UE 110) obtains an error metric for each candidate neural network formation configuration in the set of candidate neural network formation configurations to generate an error metric set such as a CRC metric. The user equipment compares each CRC metric in the set of CRC metric to a threshold. In turn, the user equipment identifies a specific CRC metric in the set of CRC metric that exceeds the threshold, and selects a specific candidate neural network formation configuration corresponding to the specific CRC metric from the set of candidate neural network formation configurations as the neural network formation configuration. Although described in the context of a CRC metric, other error metrics may also be utilized, such as BER, ARQ, HARQ, frame error rate (FER), etc.

[0212] In some embodiments, when forming the deep neural network, the user device (e.g., UE 110) determines to process the first communication channel using a first deep neural network formed with the neural network formation configuration, and identifies a second communication channel that is quasi-communicative with the first communication channel. In response to identifying that the second communication channel is quasi-communicative, the user device (e.g., UE 110) determines to process the second communication channel using a second deep neural network formed with the neural network formation configuration.

[0213] In response to forming the deep neural network, the user equipment uses the deep neural network to process communications transmitted over the wireless communication system at 1920. For example, the user equipment (e.g., UE 110) uses the deep neural network to process a downlink communication channel.

[0214] Although various aspects of base station-user equipment messaging with respect to neural networks and transmitting neural network formation configurations have been described in language specific to features and / or methods, the subject matter of the appended claims is not necessarily limited to the specific features or methods described. Rather, the specific features and methods are disclosed as exemplary implementations of base station-user equipment messaging with respect to neural networks and transmitting neural network formation configurations, and other equivalent features and methods are intended to fall within the scope of the appended claims. Furthermore, various different aspects are described, and it should be appreciated that each described aspect may be implemented independently or in combination with one or more other described aspects.

[0215] In the following, several examples are described.

[0216] Example 1: A method performed by a network entity associated with a wireless communication system, the method comprising: determining a neural network formation configuration for a deep neural network to process communications transmitted on the wireless communication system; generating a message including an indication of the neural network formation configuration for the deep neural network; and transmitting the message to a user device to direct the user device to use the neural network formation configuration to form the deep neural network and use the deep neural network to process the communications transmitted on the wireless communication system.

[0217] Example 2: The method of Example 1, wherein generating the message further comprises including a processing allocation for the deep neural network in the message.

[0218] Example 3: The method of Example 1 or Example 2, wherein generating the message further comprises specifying a time instance in the message, the time instance indicating a time to start processing the communication using the deep neural network having the neural network formation configuration.

[0219] Example 4: A method as described in any of the preceding examples, wherein the neural network formation configuration includes a first neural network formation configuration, the indication includes a first indication, the message includes a first message, and the method further includes: receiving feedback from the user device, the feedback providing one or more metrics associated with the communication; determining a second neural network formation configuration for the deep neural network based at least in part on the feedback; generating a second message, the second message including a second indication of the second neural network formation configuration for the deep neural network; and transmitting the second message to the user device to direct the user device to update the deep neural network with the second neural network formation configuration.

[0220] Example 5: The method of Example 4, wherein the second neural network formation configuration comprises an incremental neural network formation configuration.

[0221] Example 6: A method as described in any of the preceding examples, wherein determining the neural network formation configuration includes determining the neural network formation configuration based at least in part on scheduling multi-user, multi-input, multi-output downlink transmissions in the wireless communication system.

[0222] Example 7: A method as described in any one of Examples 1 to 6, wherein the neural network formation configuration includes a first neural network formation configuration, the deep neural network includes a first deep neural network, the communication includes downlink communication, and the method further includes: determining a second neural network formation configuration for a second deep neural network to process uplink communication from the user device; and transmitting the second neural network formation configuration to the user device and directing the user device to use the second neural network formation configuration to form the second deep neural network, and use the second deep neural network to process the uplink communication.

[0223] Example 8: A method as described in Example 7, wherein transmitting the message to guide the user device to use the first neural network formation configuration to form the deep neural network includes transmitting the message using a first radio access technology, and wherein transmitting the second neural network formation configuration includes transmitting the second neural network formation configuration using a second radio access technology different from the first radio access technology.

[0224] Example 9: A method as described in any one of Examples 1 to 6, wherein the neural network formation configuration includes a first neural network formation configuration, the deep neural network includes a first deep neural network, the communication includes downlink communication, and the method further includes: forming a second deep neural network; and using the second deep neural network to process the uplink communication to decode the uplink communication.

[0225] Example 10: A method as described in any one of Examples 1 to 6, wherein the neural network formation configuration includes a first neural network formation configuration, the deep neural network includes a first deep neural network, the communication includes downlink communication, and the method further includes: forming a second deep neural network; and using the second deep neural network to process the downlink communication to encode the downlink communication.

[0226] Example 11: A method as described in any of the preceding examples, wherein the message includes a first message, and wherein determining the neural network formation configuration to process the communication includes: receiving a second message from the user device indicating one or more capabilities of the user device; and determining the neural network formation configuration based at least in part on the one or more capabilities of the user device.

[0227] Example 12: A method performed by a user device associated with a wireless communication system, the method comprising: receiving, by the user device, a message indicating a neural network formation configuration for a deep neural network to process communications associated with the wireless communication system; forming the deep neural network using the neural network formation configuration indicated in the message; receiving the communications from a base station; and processing the communications using the deep neural network to extract information from the communications.

[0228] Example 13: The method of Example 12, wherein receiving a message indicating the neural network forming a configuration comprises: receiving a broadcast message from the base station; and extracting the neural network forming a configuration from the broadcast message.

[0229] Example 14: The method of Example 12 or Example 13, wherein forming the deep neural network using the neural network formation configuration includes extracting a time instance from the message indicating a time to begin using the neural network formation configuration to process the communication, and wherein processing the communication using the deep neural network includes processing the communication using the deep neural network based on the time indicated by the time instance.

[0230] Example 15: A method as described in any of Examples 12 to 14, wherein the message includes a first message, the deep neural network includes a first deep neural network, the communication includes downlink communication, processing the communication includes using the first deep neural network to decode information from the downlink communication, and the method further includes: receiving a second message by the user equipment, the second message indicating a second neural network formation configuration for a second deep neural network to process uplink communication transmitted on the wireless communication system; forming the second deep neural network using the second neural network formation configuration; and processing the uplink communication using the second deep neural network to encode information about the uplink communication.

[0231] Example 16: A method as described in any of the preceding examples, wherein the neural network formation configuration includes a first neural network formation configuration, the message includes a first message, the indication includes a first indication, and the method further includes: receiving a second message, the second message including a second indication of a second neural network formation configuration for the deep neural network; and using the second neural network formation configuration to form the deep neural network.

[0232] Example 17: The method of Example 16, wherein the second neural network formation configuration comprises an incremental neural network formation configuration.

[0233] Example 18: A method as described in any of Examples 11 to 17, wherein receiving a message indicating the neural network forming a configuration includes receiving an indication of a processing assignment for the deep neural network in the message, and wherein forming the deep neural network includes forming the deep neural network in a communication processing chain specified in the processing assignment.

[0234] Example 19: A network entity comprising: a wireless transceiver; a processor; and a computer-readable storage medium, the computer-readable storage medium comprising instructions for implementing a deep neural network manager module, the deep neural network manager module, in response to being executed by the processor, directing the network entity to perform any one of the methods described in Examples 1 to 11.

[0235] Example 20: A user device comprising: a wireless transceiver; a processor; and a computer-readable storage medium, the computer-readable storage medium comprising instructions for implementing a deep neural network manager module, the deep neural network manager module, in response to being executed by the processor, directing the user device to perform any one of the methods described in Examples 12 to 18.

[0236] Example 21: A method performed by a network entity associated with a wireless communication system, the method comprising: transmitting, by the network entity to a user device, a neural network table comprising a plurality of neural network formation configuration elements, each of the plurality of neural network formation configuration elements configuring at least a portion of a deep neural network for processing communications transmitted through the wireless communication system; selecting one or more neural network formation configuration elements from the plurality of neural network formation configuration elements to create a neural network formation configuration; and transmitting an indication to the user device to guide the user device to use the neural network formation configuration to form a deep neural network and use the deep neural network to process the communications.

[0237] Example 22: The method of Example 21, wherein transmitting the indication comprises transmitting at least one index value mapped to the one or more neural network formation configuration elements in the neural network table.

[0238] Example 23: The method of Example 1, wherein selecting the neural network formation configuration comprises selecting a set of candidate neural network formation configuration elements, and wherein transmitting the indication comprises transmitting an indication of the set of candidate neural network formation configuration elements.

[0239] Example 24: The method of Example 3, wherein transmitting the indication of the set of candidate neural network forming configuration elements comprises transmitting a set of index values ​​mapped to entries in the neural network table.

[0240] Example 25: A method as described in any of the preceding examples, wherein the indication includes a first indication, the deep neural network includes a first deep neural network, the neural network formation configuration includes a first neural network formation configuration, and the method further includes: directing the user device to use the first deep neural network to process control channel communications transmitted through the wireless communication system; determining a second neural network formation configuration; and transmitting a second indication to the user device to direct the user device to use the second neural network formation configuration to form a second deep neural network, and to use the second deep neural network to process data channel communications transmitted on the wireless communication system.

[0241] Example 26: The method as described in Example 25 further includes: determining a time threshold between the data channel communication and the control channel communication; and when the timing between the data channel communication and the control channel communication is lower than the time threshold, guiding the user device to form a second deep neural network using a default formation configuration.

[0242] Example 27: A method as described in any of the preceding examples, wherein transmitting the neural network table includes transmitting the neural network table in a multicast message.

[0243] Example 28: A method as described in any of the preceding examples, wherein the neural network formation configuration includes a first neural network formation configuration, the indication includes a first indication, and the method further includes: receiving feedback from the user device, the feedback providing one or more metrics associated with the communication; determining a second neural network formation configuration for the deep neural network based at least in part on the feedback; and transmitting a second indication to the user device to direct the user device to update the deep neural network with the second neural network formation configuration.

[0244] Example 29: The method of Example 28, wherein the second neural network formation configuration comprises an incremental neural network formation configuration.

[0245] Example 30: A method performed by a user device associated with a wireless communication system, the method comprising: receiving, by the user device, a neural network table, the neural network table comprising a plurality of neural network formation configuration elements, the plurality of neural network formation configuration elements providing the user device with the ability to configure a deep neural network for processing communications transmitted via the wireless communication system; receiving, by the user device, a message instructing the user device to form the deep neural network using a neural network formation configuration based on one or more of the plurality of neural network formation configuration elements; forming the deep neural network using the neural network formation configuration by accessing the neural network table to obtain the one or more neural network formation configuration elements; and using the deep neural network to process communications transmitted on the wireless communication system.

[0246] Example 31: A method as described in Example 30, wherein receiving a message instructing the user device to use the neural network formation configuration to form the deep neural network further includes: extracting a processing assignment from the message; and wherein forming the deep neural network with the neural network formation configuration includes: forming the deep neural network in a communication processing chain specified in the processing assignment.

[0247] Example 32: A method as described in Example 30 or Example 31, wherein receiving a message instructing the user device to use the neural network formation configuration to form the deep neural network further includes: receiving at least one index value mapped to an entry in the neural network table, and wherein forming the deep neural network using the neural network formation configuration includes using the index value to access the neural network table.

[0248] Example 33: A method as described in any of the preceding examples, wherein the deep neural network includes a first deep neural network, and wherein forming the deep neural network with the neural network configuration further includes: determining to process a first communication channel by using the first deep neural network formed with the neural network configuration; identifying that a second communication channel is in quasi-communication with the first communication channel; and based on the identification, determining to process the second communication channel using a second deep neural network formed with the neural network configuration.

[0249] Example 34: The method of Example 33, wherein identifying that the second communication channel is in quasi-communication with the first communication channel comprises: receiving an indication that the second communication channel is in quasi-communication with the first communication channel.

[0250] Example 35: A method as described in Example 30, wherein receiving the message instructing the user device to use the neural network formation configuration to form the deep neural network further includes: receiving, by the user device, a set of index values ​​corresponding to candidate neural network formation configuration elements to be used as the neural network formation configuration, and wherein obtaining the neural network formation configuration by accessing the neural network table to form the deep neural network with the neural network formation configuration includes: accessing the neural network table using the index value set to obtain the candidate neural network formation configuration element; using the candidate neural network formation configuration element to form a candidate deep neural network set to process the communication; obtaining one or more metrics associated with the candidate deep neural network set; and selecting the neural network formation configuration based on the one or more metrics.

[0251] Example 36: A method as described in Example 35, wherein obtaining the one or more metrics includes: obtaining a corresponding error metric for each candidate deep neural network in the set of candidate deep neural networks to generate an error metric set; and comparing each error metric in the set of error metric with a threshold, and wherein selecting the neural network to form a configuration includes: identifying a specific error metric in the set of error metric, the specific error metric indicating less error relative to other error metrics in the set of error metric; and selecting a specific candidate deep neural network corresponding to the specific error metric from the set of candidate deep neural networks as the neural network to form the configuration.

[0252] Example 37: A method as described in any of the preceding examples, wherein the message includes a first message, the deep neural network includes a first deep neural network, the neural network formation configuration includes a first neural network formation configuration, the first message includes a first processing assignment instructing the user equipment to use the first deep neural network to process control channel communications, and the method further includes: receiving a second message instructing the user equipment to use a second neural network formation configuration to form a second deep neural network, the second message including a second processing assignment for the second deep neural network; forming the second deep neural network with the second neural network formation by accessing the neural network table to obtain the second neural network formation configuration; and based on the processing assignment, using the second deep neural network to process data channel communications transmitted on the wireless communication system.

[0253] Example 38: A method as described in any of the preceding examples, wherein receiving the neural network table includes receiving the neural network table in a multicast message.

[0254] Example 39: A network entity comprising: a wireless transceiver; a processor; and a computer-readable storage medium comprising instructions for implementing a deep neural network manager module, wherein the deep neural network manager module, in response to being executed by the processor, directs the network entity to perform any of the methods described in Examples 21 to 29.

[0255] Example 40: A user device comprising: a wireless transceiver; a processor; and a computer-readable storage medium comprising instructions, the instructions being used to direct the user device to perform one of the methods described in Examples 30 to 38 in response to being executed by the processor.

Claims

1. A method performed by a network entity associated with a wireless communication system, the method comprising: determining a neural network formation configuration for a deep neural network to process communications transmitted over the wireless communication system; generating a message comprising: an indication of the neural network formation configuration for the deep neural network and an indication of a time instance to begin using the deep neural network having the neural network formation configuration; as well as The message is transmitted to a user device to direct the user device to use the neural network formation configuration to form the deep neural network and to use the deep neural network to process the communication transmitted over the wireless communication system.

2. The method of claim 1, wherein: Generating the message further comprises including in the message a processing allocation for the deep neural network.

3. The method according to claim 1, wherein: The neural network formation configuration includes a first neural network formation configuration, the message includes a first message, and the method further includes: receiving feedback from the user device, the feedback providing one or more metrics associated with the communication; determining a second neural network formation configuration for the deep neural network based at least in part on the feedback; generating a second message comprising an indication of the second neural network forming configuration for the deep neural network; and The second message is transmitted to the user device to direct the user device to update the deep neural network with the second neural network forming configuration for the deep neural network.

4. The method of claim 3, wherein: The second neural network formation configuration for the deep neural network includes an incremental neural network formation configuration.

5. The method of claim 1, wherein: Determining the neural network formation configuration includes determining the neural network formation configuration based at least in part on scheduling multi-user, multiple-input, multiple-output downlink transmissions in the wireless communication system.

6. The method according to any one of claims 1 to 5, wherein: The neural network formation configuration includes a first neural network formation configuration, the deep neural network includes a first deep neural network, the communication includes downlink communication, and the method further includes: determining a second neural network formation configuration for a second deep neural network to process uplink communications from the user equipment; and The second neural network formation configuration for the second deep neural network is transmitted to the user device, and the user device is directed to use the second neural network formation configuration for the second deep neural network to form the second deep neural network, and to use the second deep neural network to process the uplink communication.

7. The method of claim 6, wherein: Transmitting the message to direct the user device to form the deep neural network using the first neural network formation configuration comprises transmitting the message using a first radio access technology, and wherein transmitting the second neural network forming configuration for the second deep neural network comprises transmitting the second neural network forming configuration for the second deep neural network using a second radio access technology different from the first radio access technology.

8. The method according to any one of claims 1 to 5, wherein: The neural network formation configuration includes a first neural network formation configuration, the deep neural network includes a first deep neural network, the communication includes downlink communication, and the method further includes: forming a second deep neural network; and The uplink communication is processed using the second deep neural network to decode the uplink communication.

9. The method according to any one of claims 1 to 5, wherein: The neural network formation configuration includes a first neural network formation configuration, the deep neural network includes a first deep neural network, the communication includes downlink communication, and the method further includes: forming a second deep neural network; and The downlink communication is processed using the second deep neural network to encode the downlink communication.

10. The method of claim 1, wherein: The message comprises a first message, and wherein determining that the neural network is configured to process the communication comprises: receiving, from the user equipment, a second message indicating one or more capabilities of the user equipment; and The neural network formation configuration is determined based at least in part on the one or more capabilities of the user device.

11. A method performed by a user equipment associated with a wireless communication system, the method comprising: The user equipment receives a message including: an indication for a deep neural network to form a neural network configuration to process communications associated with the wireless communication system; as well as an indication of a time instance to begin using the deep neural network having the neural network formation configuration; forming the deep neural network using the neural network formation configuration indicated in the message; receiving the communication from a base station; as well as The communications are processed using the deep neural network to extract information in the communications.

12. The method of claim 11, wherein: Receiving the message includes: receiving a broadcast message from the base station; and The neural network formation configuration and the indication of the time instance are extracted from the broadcast message.

13. The method of claim 11, wherein: forming the deep neural network using the neural network formation configuration includes extracting from the message the time instance indicating a time to begin processing the communication using the neural network formation configuration, and Wherein processing the communication using the deep neural network comprises processing the communication using the deep neural network based on the time indicated by the time instance.

14. The method of claim 11, wherein: The message comprises a first message, the deep neural network comprises a first deep neural network, the communication comprises a downlink communication, processing the communication comprises using the first deep neural network to decode information from the downlink communication, and the method further comprises: receiving, by the user equipment, a second message indicating a second neural network forming configuration for a second deep neural network to process uplink communications transmitted over the wireless communication system; forming the second deep neural network using the second neural network forming configuration; and The uplink communication is processed using the second deep neural network to encode information about the uplink communication.

15. The method of claim 11, wherein: The neural network formation configuration includes a first neural network formation configuration, the message includes a first message, the indication includes a first indication, and the method further includes: receiving a second message comprising a second indication of a second neural network forming configuration for the deep neural network; and The deep neural network is formed using the second neural network forming configuration.

16. The method of claim 15, wherein: The second neural network formation configuration includes an incremental neural network formation configuration.

17. The method according to any one of claims 11 to 16, wherein: Receiving the message includes receiving in the message an indication of a processing assignment for the deep neural network, wherein forming the deep neural network includes forming the deep neural network in a communication processing chain specified in the processing assignment.

18. A network entity, comprising: Wireless transceiver; processor; as well as A non-transitory computer-readable storage medium comprising instructions for implementing a deep neural network manager module, wherein the deep neural network manager module, in response to being executed by the processor, directs the network entity to perform any of the methods described in claims 1 to 10.

19. A user equipment, comprising: Wireless transceiver; processor; as well as A non-transitory computer-readable storage medium comprising instructions for implementing a deep neural network manager module, wherein the deep neural network manager module, in response to being executed by the processor, directs the user device to perform any of the methods described in claims 11 to 17.

Images