WO2021258259A1

WO2021258259A1 - Determining a channel state for wireless communication

Info

Publication number: WO2021258259A1
Application number: PCT/CN2020/097489
Authority: WO
Inventors: Yuwei REN; Liangming WU; Tianyang BAI
Original assignee: Qualcomm Incorporated
Priority date: 2020-06-22
Filing date: 2020-06-22
Publication date: 2021-12-30

Abstract

Embodiment methods for determining a channel state for wireless device-to-base station communications may include the wireless device performing a downlink channel estimation based on a downlink reference signal (DL RS) from the base station, determining channel differential information based on the downlink channel estimation and previous downlink channel estimations, applying a machine learning model to the channel differential information to generate a differential feature, and returning the generated differential feature to enable the base station to determine a downlink channel state and adjust communications with the wireless device. Embodiments also include the base station sending the DL RS, receiving from the wireless device the differential feature of a downlink channel state between the base station and the wireless device, applying a machine learning model to the differential feature to generate a downlink channel state, and adjusting communications with the wireless device based on the determined downlink channel state.

Description

Determining A Channel State For Wireless Communication

BACKGROUND

A base station requires accurate channel downlink information for downlink transmissions to a wireless device. For example, the base station needs information about a downlink channel state to perform procedures such as such as downlink precoding, user scheduling, and beam alignment. A base station may determine the downlink channel state based on feedback information received from a wireless device in an uplink transmission such as a channel quality indicator (CQI) . Advanced wireless communications, such as that provided by Fifth Generation (5G) New Radio (NR) systems, provide high speed communication services through the use of processes such as beamforming and massive Multiple Input Multiple Output (MIMO) communication techniques. The channel feedback from the wireless device required to use such processes can consume a large amount of base station resources. For example, supporting massive MIMO can require up to sixty-four ports to send reference signal (s) and receive related feedback from a wireless device. The measurement feedback from the wireless device also may consume a large amount of uplink communication resources. For example, increasing the density of feedback information and the measurement quantization resolution may rapidly consume uplink communication resources.

SUMMARY

Various aspects include systems and methods of wireless communications performed by a processor of a base station and a wireless device. Various aspects may include a method performed by a processor of a wireless device for determining a channel state for communication with a base station. In various aspects the method may include performing a downlink channel estimation based on a downlink reference signal (DL RS) received from the base station, determining channel differential information based on the downlink channel estimation and a previous downlink channel estimation, applying a machine learning model to the channel differential information to generate a differential feature, and sending to the base station the generated differential feature to enable the base station to determine a downlink channel state and adjust communications with the wireless device.

Various aspects may include a method performed by a processor of a base station for determining a channel state for communication with a wireless device. In various aspects, the method may include sending to the wireless device a downlink reference signal (DL RS) , receiving from the wireless device a differential feature of a downlink channel state between the base station and the wireless device, applying a machine learning model to the differential feature to generate a downlink channel state, and adjusting communications with the wireless device based on the generated downlink channel state.

Further aspects may include a wireless device having a processor configured to perform operations of any of the wireless device methods summarized above. Further aspects may include a non-transitory processor-readable storage medium having stored thereon processor-executable instructions configured to cause a processor of a wireless device to perform operations of any of the wireless device methods summarized above. Further aspects include a wireless device having means for performing functions of any of the methods summarized above. Further aspects include a system on chip for use in a wireless device that includes a processor configured to perform operations of any of the wireless device methods summarized above.

Further aspects may include a base station having a processor configured to perform operations of any of the base station methods summarized above. Further aspects may include a non-transitory processor-readable storage medium having stored thereon processor-executable instructions configured to cause a processor of a base station to perform operations of any of the base station methods summarized above. Further aspects include a base station having means for performing functions of any of the base station methods summarized above. Further aspects include a system on chip for use in a base station that includes a processor configured to perform operations of any of the base station methods summarized above.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying drawings, which are incorporated herein and constitute part of this specification, illustrate exemplary embodiments of the claims, and together with the general description given above and the detailed description given below, serve to explain the features of the claims.

FIG. 1 is a system block diagram illustrating an example communication system suitable for implementing any of the various embodiments.

FIG. 2 is a component block diagram illustrating an example computing and wireless modem system suitable for implementing any of the various embodiments.

FIG. 3 is a component block diagram illustrating a software architecture including a radio protocol stack for the user and control planes in wireless communications suitable for implementing any of the various embodiments.

FIGS. 4A and 4B are component block diagrams illustrating a system 400 configured for determining a channel state for wireless communications between a base station and a wireless device in accordance with various embodiments

FIG. 5 is a process flow diagram illustrating a method that may be performed by a processor of a wireless device for determining a channel state for communication with a base station according to various embodiments

FIGS. 6A–6E are process flow diagrams illustrating operations that may be performed by a processor of a wireless device as part of the method for determining a channel state for communication with a base station according to various embodiments.

FIG. 7 is a process flow diagram illustrating a method that may be performed by a processor of a base station for determining a channel state for communication with a wireless device according to various embodiments.

FIGS. 8A–8D are process flow diagrams illustrating operations that may be performed by a processor of a base station as part of the method for determining a channel state for communication with a wireless device according to various embodiments.

FIG. 9 is a component block diagram of a network computing device suitable for use with various embodiments.

FIG. 10 is a component block diagram of a wireless device suitable for use with various embodiments.

DETAILED DESCRIPTION

Various embodiments will be described in detail with reference to the accompanying drawings. Wherever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts. References made to particular examples and embodiments are for illustrative purposes, and are not intended to limit the scope of the claims.

Various embodiments include systems and methods for determining a channel state for wireless communications between devices such as a base station and a wireless device. Various embodiments improve the function of the wireless device and the base station by enabling a rapid and accurate determination of downlink channel state information while reducing an information density of feedback information sent from the wireless device to the base station. Various embodiments also may reduce a frequency of sending feedback information from the wireless device to the base station. Various embodiments also may improve the function of the wireless device and the base station by reducing the consumption of limited uplink communication resources by such feedback information. Various embodiments also may improve the function of the wireless device and the base station by reducing the consumption of base station resources by receiving and processing such feedback information.

The term “wireless device” is used herein to refer to any one or all of wireless router devices, wireless appliances, cellular telephones, smartphones, portable computing devices, personal or mobile multi-media players, laptop computers, tablet computers, smartbooks, ultrabooks, palmtop computers, wireless electronic mail receivers, multimedia Internet-enabled cellular telephones, medical devices and equipment, biometric sensors/devices, wearable devices including smart watches, smart clothing, smart glasses, smart wrist bands, smart jewelry (e.g., smart rings, smart bracelets, etc. ) , entertainment devices (e.g., wireless gaming controllers, music and video players, satellite radios, etc. ) , wireless-network enabled Internet of Things (IoT) devices including smart meters/sensors, industrial manufacturing equipment, large and small machinery and appliances for home or enterprise use, wireless communication elements within autonomous and semiautonomous vehicles, wireless devices affixed to or incorporated into various mobile platforms, global positioning system devices, and similar electronic devices that include a memory, wireless communication components and a programmable processor.

The term “system on chip” (SOC) is used herein to refer to a single integrated circuit (IC) chip that contains multiple resources and/or processors integrated on a single substrate. A single SOC may contain circuitry for digital, analog, mixed-signal, and radio-frequency functions. A single SOC may also include any number of general purpose and/or specialized processors (digital signal processors, modem processors, video processors, etc. ) , memory blocks (e.g., ROM, RAM, Flash, etc. ) , and resources (e.g., timers, voltage regulators, oscillators, etc. ) . SOCs may also include software for controlling the integrated resources and processors, as well as for controlling peripheral devices.

The term “system in a package” (SIP) may be used herein to refer to a single module or package that contains multiple resources, computational units, cores and/or processors on two or more IC chips, substrates, or SOCs. For example, a SIP may include a single substrate on which multiple IC chips or semiconductor dies are stacked in a vertical configuration. Similarly, the SIP may include one or more multi-chip modules (MCMs) on which multiple ICs or semiconductor dies are packaged into a unifying substrate. A SIP may also include multiple independent SOCs coupled together via high speed communication circuitry and packaged in close proximity, such as on a single motherboard or in a single wireless device. The proximity of the SOCs facilitates high speed communications and the sharing of memory and resources.

A base station needs information about a downlink channel state to perform procedures such as such as downlink precoding, user scheduling, and beam alignment for downlink transmissions to a wireless device. A base station may determine the downlink channel state based on feedback information received from a wireless device in an uplink transmission such as a channel quality indicator (CQI) . Advanced wireless communications, such as that provided by Fifth Generation (5G) New Radio (NR) systems, provide high speed communication services through the use of processes such as beamforming and massive Multiple Input Multiple Output (MIMO) communication techniques. The channel feedback from the wireless device required to use such processes can consume a large amount of base station resources. For example, supporting massive MIMO can require up to sixty-four ports to send reference signal (s) and receive related feedback from a wireless device. Further, the measurement feedback from the wireless device may consume a large amount of uplink communication resources. For example, while a high density or high quantization resolution may improve channel estimation accuracy, such density of feedback and quantization resolution may consume a large amount of uplink communication resources.

Various embodiments include systems and methods for determining a channel state for wireless communications between devices such as a base station and a wireless device. Channel estimation relies on feedback from the wireless device because channel quality varies unpredictably in a nonlinear manner.

Various embodiments employ machine learning models to determine nonlinear feature (s) of channel differentials and use the determined nonlinear feature (s) to determine an accurate channel estimation while reducing communication link and processing resource consumption. In various embodiments, a wireless device may be configured with a machine learning model that receives as input channel differential information and provides as output one or more differential features that the wireless device may send to the base station. In various embodiments, the base station may be configured with a separate machine learning model that receives as input the one or more differential features from the wireless device and provides as output a downlink channel state of the downlink between the base station and the wireless device. The base station may adjust communications with the wireless device based on the generated downlink channel state. For example, the base station may perform procedures such as downlink precoding, user scheduling, and beam alignment based on the generated downlink channel state.

Consistent with conventional communication protocols, the base station may send a downlink reference signal (DL RS) to the wireless device. The wireless device may perform a downlink channel estimation based on the DL RS. The wireless device may determine channel differential information based on the downlink channel estimation and a previous downlink channel estimation. In various embodiments, the wireless device may apply a machine learning model (a wireless device machine learning model) to the channel differential information to generate a differential feature, and send to the base station the generated differential feature. In some embodiment, the base station may send to the wireless device an instruction to enable the machine learning model. The base station may receive the differential feature of the downlink channel state from the wireless device. In various embodiments, the base station may apply a machine learning model (a base station machine learning model) to the differential feature to determine or reconstruct a downlink channel state or an estimation of the downlink channel state. The base station may adjust communication with the wireless device based on the determined or reconstructed downlink channel state. In some embodiments, the base station may apply the machine learning model to reconstruct an estimated-quantized channel. In some embodiments, the base station may apply the machine learning model to predict aspects of the downlink channel useful for communications with the wireless device.

In some embodiments, the base station may apply the base station machine learning model to the differential feature and to one or more previously-determined differential features to determine or predict a future downlink channel state. In some embodiments, the base station may store previously-determined differential features in a memory, and may use one or more previously-determined differential features to predict the future downlink channel state. Using a greater number of previously-determined differential features may increase the accuracy of the predicted downlink channel state, but also may consume more base station processing resources.

In some embodiments, the base station may send to the wireless device a structure of the machine learning model. In some embodiments, the structure of the machine learning model may include a number of layers of the machine learning model (e.g., a number of layers of a convolutional neural network) . In some embodiments, the structure of the machine learning model may include an indication of one or more weights to be applied in the wireless device e machine learning model. In some embodiments, the structure of the machine learning model may include an indication of an output feature size. In some embodiments, the structure of the machine learning model may include an indication of a quantization resolution. In some embodiments, the structure of the machine learning model may include a structure of the machine learning model that is based on a capability of the wireless device.

In some embodiments, the base station may send the structure of the machine learning model via signaling such as Radio Resource Control (RRC) signaling, a Media Access Control-Control Element (MAC-CE) , or in Downlink Control Information (DCI) . In some embodiments, the base station may determine the structure of the machine learning model based on one or more capabilities of the wireless device. For example, for a wireless device with a relatively limited processor or memory, the base station may determine a relatively limited number of layers (such as three) for a convolutional neural network.

In some embodiments, the structure of the machine learning model may include an indication of one or more weights to be applied in the wireless device e machine learning model. In some embodiments, the base station may send updated weights to the wireless device periodically (e.g., via RRC signaling, MAC-CE, DCI, etc. ) . In some embodiments, the base station may send updated weights to the wireless device in response to determining that downlink channel performance has dropped below a performance threshold. In some embodiments, the structure of the machine learning model may include an output feature size and a corresponding quantization resolution.

In some embodiments, the wireless device may send the differential feature to the base station in response to determining that a feedback trigger has occurred. In some embodiments, the base station may send an indication of a feedback trigger to the wireless device. In some embodiments, the base station may send one or more feedback trigger conditions to the wireless device. In some embodiments, the base station may receive from the wireless device the differential feature in response to the sending of the indication of the feedback trigger.

In some embodiments, the wireless device may be configured with one or more feedback triggers. In some embodiments, the feedback trigger may include a resource pattern of feedback information that the wireless device may send to the base station from time to time. Such feedback information may include a downlink channel estimation, complete differential information, or partial feedback information. In some embodiments, the feedback trigger may include an occurrence of a number of slots. In this manner, the base station or wireless device may adjust the density of feedback provided to the base station. In some embodiments, the feedback trigger may include a request for feedback sent from the base station to the wireless device.

In some embodiments, the wireless device may determine whether a difference between the downlink channel estimation and the previous downlink channel estimation exceeds a threshold difference, and may send the differential feature to the base station in response to determining that the difference between the downlink channel estimation and the previous downlink channel estimation exceeds the threshold difference. In some embodiments, in response to determining that the difference between the downlink channel estimation and the previous downlink channel estimation does not the threshold difference, the wireless device may determine to not send the differential feature. In that case, the base station may use one or more stored previously-determined differential features to predict a future downlink channel state. In some embodiments, the base station may send information about the feedback trigger to the wireless device via, e.g., RRC signaling, MAC-CE, DCI, etc.

In some embodiments, a channel estimation may be sufficiently inaccurate that the base station may request more information from the wireless device, such as the overall downlink channel estimation or channel differential information (rather than the differential feature) . In some embodiments, the base station may determine whether the determined or reconstructed downlink channel state exceeds an error threshold, and may send to the wireless device a message to send the downlink channel estimation or the channel differential information instead of the differential feature in response to determining that the reconstructed downlink channel state exceeds the error threshold. In response, the wireless device may send to the base station the downlink channel estimation or the channel differential information determined by the wireless device for use by the base station in determining the downlink channel state.

In some embodiments, the base station may provide an updated structure of the machine learning model to the wireless device. In some embodiments, the wireless device may send to the base station a request to update the machine learning model. The base station may update the structure of the machine learning model, and may send the updated structure of the machine learning model to the wireless device. In some embodiments, the base station may update one or more parameters that the wireless device may apply to update the machine learning model.

FIG. 1 is a system block diagram illustrating an example communication system 100 suitable for implementing any of the various embodiments. The communications system 100 may be a 5G New Radio (NR) network, or any other suitable network such as a Long Term Evolution (LTE) network.

The communications system 100 may include a heterogeneous network architecture that includes a core network 140 and a variety of mobile devices (illustrated as wireless device 120a-120e in FIG. 1) . The communications system 100 may also include a number of base stations (illustrated as the BS 110a, the BS 110b, the BS 110c, and the BS 110d) and other network entities. A base station is an entity that communicates with wireless devices (mobile devices) , and also may be referred to as an NodeB, a Node B, an LTE evolved nodeB (eNB) , an access point (AP) , a radio head, a transmit receive point (TRP) , a New Radio base station (NR BS) , a 5G NodeB (NB) , a Next Generation NodeB (gNB) , or the like. Each base station may provide communication coverage for a particular geographic area. In 3GPP, the term “cell” can refer to a coverage area of a base station, a base station subsystem serving this coverage area, or a combination thereof, depending on the context in which the term is used.

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

In some examples, a cell may not be stationary, and the geographic area of the cell may move according to the location of a mobile base station. In some examples, the base stations 110a-110d may be interconnected to one another as well as to one or more other base stations or network nodes (not illustrated) in the communications system 100 through various types of backhaul interfaces, such as a direct physical connection, a virtual network, or a combination thereof using any suitable transport network

The base station 110a-110d may communicate with the core network 140 over a wired or wireless communication link 126. The wireless device 120a-120e may communicate with the base station 110a-110d over a wireless communication link 122.

The wired communication link 126 may use a variety of wired networks (e.g., Ethernet, TV cable, telephony, fiber optic and other forms of physical network connections) that may use one or more wired communication protocols, such as Ethernet, Point-To-Point protocol, High-Level Data Link Control (HDLC) , Advanced Data Communication Control Protocol (ADCCP) , and Transmission Control Protocol/Internet Protocol (TCP/IP) .

The communications system 100 also may include relay stations (e.g., relay BS 110d) . A relay station is an entity that can receive a transmission of data from an upstream station (for example, a base station or a mobile device) and transmit the data to a downstream station (for example, a wireless device or a base station) . A relay station also may be a mobile device that can relay transmissions for other wireless devices. In the example illustrated in FIG. 1, a relay station 110d may communicate with macro the base station 110a and the wireless device 120d in order to facilitate communication between the base station 110a and the wireless device 120d. A relay station also may be referred to as a relay base station, a relay base station, a relay, etc.

The communications system 100 may be a heterogeneous network that includes base stations of different types, for example, macro base stations, pico base stations, femto base stations, relay base stations, etc. These different types of base stations may have different transmit power levels, different coverage areas, and different impacts on interference in communications system 100. For example, macro base stations may have a high transmit power level (for example, 5 to 40 Watts) whereas pico base stations, femto base stations, and relay base stations may have lower transmit power levels (for example, 0.1 to 2 Watts) .

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

The

wireless devices

120a, 120b, 120c may be dispersed throughout communications system 100, and each wireless device may be stationary or mobile. A wireless device also may be referred to as an access terminal, a terminal, a mobile station, a subscriber unit, a station, etc.

A macro base station 110a may communicate with the communication network 140 over a wired or wireless communication link 126. The

wireless devices

120a, 120b, 120c may communicate with a base station 110a-110d over a wireless communication link 122.

The

wireless communication links

122, 124 may include a plurality of carrier signals, frequencies, or frequency bands, each of which may include a plurality of logical channels. The

wireless communication links

122 and 124 may utilize one or more radio access technologies (RATs) . Examples of RATs that may be used in a wireless communication link include 3GPP LTE, 3G, 4G, 5G (e.g., NR) , GSM, Code Division Multiple Access (CDMA) , Wideband Code Division Multiple Access (WCDMA) , Worldwide Interoperability for Microwave Access (WiMAX) , Time Division Multiple Access (TDMA) , and other mobile telephony communication technologies cellular RATs. Further examples of RATs that may be used in one or more of the various

wireless communication links

122, 124 within the communication system 100 include medium range protocols such as Wi-Fi, LTE-U, LTE-Direct, LAA, MuLTEfire, and relatively short range RATs such as ZigBee, Bluetooth, and Bluetooth Low Energy (LE) .

Certain wireless networks (e.g., LTE) utilize orthogonal frequency division multiplexing (OFDM) on the downlink and single-carrier frequency division multiplexing (SC-FDM) on the uplink. OFDM and SC-FDM partition the system bandwidth into multiple (K) orthogonal subcarriers, which are also commonly referred to as tones, bins, etc. Each subcarrier may be modulated with data. In general, modulation symbols are sent in the frequency domain with OFDM and in the time domain with SC-FDM. The spacing between adjacent subcarriers may be fixed, and the total number of subcarriers (K) may be dependent on the system bandwidth. For example, the spacing of the subcarriers may be 15 kHz and the minimum resource allocation (called a “resource block” ) may be 12 subcarriers (or 180 kHz) . Consequently, the nominal Fast File Transfer (FFT) size may be equal to 128, 256, 512, 1024 or 2048 for system bandwidth of 1.25, 2.5, 5, 10 or 20 megahertz (MHz) , respectively. The system bandwidth may also be partitioned into subbands. For example, a subband may cover 1.08 MHz (i.e., 6 resource blocks) , and there may be 1, 2, 4, 8 or 16 subbands for system bandwidth of 1.25, 2.5, 5, 10 or 20 MHz, respectively.

While descriptions of some embodiments may use terminology and examples associated with LTE technologies, various embodiments may be applicable to other wireless communications systems, such as a new radio (NR) or 5G network. NR may utilize OFDM with a cyclic prefix (CP) on the uplink (UL) and downlink (DL) and include support for half-duplex operation using Time Division Duplexing (TDD) . A single component carrier bandwidth of 100 MHz may be supported. NR resource blocks may span 12 sub-carriers with a sub-carrier bandwidth of 75 kHz over a 0.1 millisecond (ms) duration. Each radio frame may consist of 50 subframes with a length of 10 ms. Consequently, each subframe may have a length of 0.2 ms. Each subframe may indicate a link direction (i.e., DL or UL) for data transmission and the link direction for each subframe may be dynamically switched. Each subframe may include DL/UL data as well as DL/UL control data. Beamforming may be supported and beam direction may be dynamically configured. Multiple Input Multiple Output (MIMO) transmissions with precoding may also be supported. MIMO configurations in the DL may support up to eight transmit antennas with multi-layer DL transmissions up to eight streams and up to two streams per wireless device. Multi-layer transmissions with up to 2 streams per wireless device may be supported. Aggregation of multiple cells may be supported with up to eight serving cells. Alternatively, NR may support a different air interface, other than an OFDM-based air interface.

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

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

In some embodiments, two or more mobile devices 120a-120e (for example, illustrated as the wireless device 120a and the wireless device 120e) may communicate directly using one or more sidelink channels 124 (for example, without using a base station 110a-110d as an intermediary to communicate with one another) . For example, the wireless devices 120a-120e may communicate using peer-to-peer (P2P) communications, device-to-device (D2D) communications, a vehicle-to-everything (V2X) protocol (which may include a vehicle-to-vehicle (V2V) protocol, a vehicle-to-infrastructure (V2I) protocol, or similar protocol) , a mesh network, or similar networks, or combinations thereof. In this case, the wireless device 120a-120e may perform scheduling operations, resource selection operations, as well as other operations described elsewhere herein as being performed by the base station 110a.

FIG. 2 is a component block diagram illustrating an example computing and wireless modem system 200 suitable for implementing any of the various embodiments. Various embodiments may be implemented on a number of single processor and multiprocessor computer systems, including a system-on-chip (SOC) or system in a package (SIP) .

With reference to FIGS. 1 and 2, the illustrated example computing system 200 (which may be a SIP in some embodiments) includes a two

SOCs

202, 204 coupled to a clock 206, a voltage regulator 208, and a wireless transceiver 266 configured to send and receive wireless communications via an antenna (not shown) to/from wireless devices, such as a base station 110a. In some embodiments, the first SOC 202 operate as central processing unit (CPU) of the wireless device that carries out the instructions of software application programs by performing the arithmetic, logical, control and input/output (I/O) operations specified by the instructions. In some embodiments, the second SOC 204 may operate as a specialized processing unit. For example, the second SOC 204 may operate as a specialized 5G processing unit responsible for managing high volume, high speed (e.g., 5 Gbps, etc. ) , and/or very high frequency short wave length (e.g., 28 GHz mmWave spectrum, etc. ) communications.

The first SOC 202 may include a digital signal processor (DSP) 210, a modem processor 212, a graphics processor 214, an application processor 216, one or more coprocessors 218 (e.g., vector co-processor) connected to one or more of the processors, memory 220, custom circuity 222, system components and resources 224, an interconnection/bus module 226, one or more temperature sensors 230, a thermal management unit 232, and a thermal power envelope (TPE) component 234. The second SOC 204 may include a 5G modem processor 252, a power management unit 254, an interconnection/bus module 264, the plurality of mmWave transceivers 256, memory 258, and various additional processors 260, such as an applications processor, packet processor, etc.

Each

processor

210, 212, 214, 216, 218, 252, 260 may include one or more cores, and each processor/core may perform operations independent of the other processors/cores. For example, the first SOC 202 may include a processor that executes a first type of operating system (e.g., FreeBSD, LINUX, OS X, etc. ) and a processor that executes a second type of operating system (e.g., MICROSOFT WINDOWS 10) . In addition, any or all of the

processors

210, 212, 214, 216, 218, 252, 260 may be included as part of a processor cluster architecture (e.g., a synchronous processor cluster architecture, an asynchronous or heterogeneous processor cluster architecture, etc. ) .

The first and

second SOC

202, 204 may include various system components, resources and custom circuitry for managing sensor data, analog-to-digital conversions, wireless data transmissions, and for performing other specialized operations, such as decoding data packets and processing encoded audio and video signals for rendering in a web browser. For example, the system components and resources 224 of the first SOC 202 may include power amplifiers, voltage regulators, oscillators, phase-locked loops, peripheral bridges, data controllers, memory controllers, system controllers, access ports, timers, and other similar components used to support the processors and software clients running on a wireless device. The system components and resources 224 and/or custom circuitry 222 may also include circuitry to interface with peripheral devices, such as cameras, electronic displays, wireless communication devices, external memory chips, etc.

The first and

second SOC

202, 204 may communicate via interconnection/bus module 250. The

various processors

210, 212, 214, 216, 218, may be interconnected to one or more memory elements 220, system components and resources 224, and custom circuitry 222, and a thermal management unit 232 via an interconnection/bus module 226. Similarly, the processor 252 may be interconnected to the power management unit 254, the mmWave transceivers 256, memory 258, and various additional processors 260 via the interconnection/bus module 264. The interconnection/

bus module

226, 250, 264 may include an array of reconfigurable logic gates and/or implement a bus architecture (e.g., CoreConnect, AMBA, etc. ) . Communications may be provided by advanced interconnects, such as high-performance networks-on chip (NoCs) .

The first and/or

second SOCs

202, 204 may further include an input/output module (not illustrated) for communicating with resources external to the SOC, such as a clock 206 and a voltage regulator 208. Resources external to the SOC (e.g., clock 206, voltage regulator 208) may be shared by two or more of the internal SOC processors/cores.

In addition to the example SIP 200 discussed above, various embodiments may be implemented in a wide variety of computing systems, which may include a single processor, multiple processors, multicore processors, or any combination thereof.

FIG. 3 is a component block diagram illustrating a software architecture 300 including a radio protocol stack for the user and control planes in wireless communications suitable for implementing any of the various embodiments. With reference to FIGS. 1–3, the wireless device 320 may implement the software architecture 300 to facilitate communication between a wireless device 320 (e.g., the wireless device 120a-120e, 200) and the base station 350 (e.g., the base station 110a) of a communication system (e.g., 100) . In various embodiments, layers in software architecture 300 may form logical connections with corresponding layers in software of the base station 350. The software architecture 300 may be distributed among one or more processors (e.g., the

processors

212, 214, 216, 218, 252, 260) . While illustrated with respect to one radio protocol stack, in a multi-SIM (subscriber identity module) wireless device, the software architecture 300 may include multiple protocol stacks, each of which may be associated with a different SIM (e.g., two protocol stacks associated with two SIMs, respectively, in a dual-SIM wireless communication device) . While described below with reference to LTE communication layers, the software architecture 300 may support any of variety of standards and protocols for wireless communications, and/or may include additional protocol stacks that support any of variety of standards and protocols wireless communications.

The software architecture 300 may include a Non-Access Stratum (NAS) 302 and an Access Stratum (AS) 304. The NAS 302 may include functions and protocols to support packet filtering, security management, mobility control, session management, and traffic and signaling between a SIM (s) of the wireless device (e.g., SIM (s) 204) and its core network 140. The AS 304 may include functions and protocols that support communication between a SIM (s) (e.g., SIM (s) 204) and entities of supported access networks (e.g., a base station) . In particular, the AS 304 may include at least three layers (Layer 1, Layer 2, and Layer 3) , each of which may contain various sub-layers.

In the user and control planes, Layer 1 (L1) of the AS 304 may be a physical layer (PHY) 306, which may oversee functions that enable transmission and/or reception over the air interface via a wireless transceiver (e.g., 256) . Examples of such physical layer 306 functions may include cyclic redundancy check (CRC) attachment, coding blocks, scrambling and descrambling, modulation and demodulation, signal measurements, MIMO, etc. The physical layer may include various logical channels, including the Physical Downlink Control Channel (PDCCH) and the Physical Downlink Shared Channel (PDSCH) .

In the user and control planes, Layer 2 (L2) of the AS 304 may be responsible for the link between the wireless device 320 and the base station 350 over the physical layer 306. In the various embodiments, Layer 2 may include a media access control (MAC) sublayer 308, a radio link control (RLC) sublayer 310, and a packet data convergence protocol (PDCP) 312 sublayer, each of which form logical connections terminating at the base station 350.

In the control plane, Layer 3 (L3) of the AS 304 may include a radio resource control (RRC) sublayer 3. While not shown, the software architecture 300 may include additional Layer 3 sublayers, as well as various upper layers above Layer 3. In various embodiments, the RRC sublayer 313 may provide functions INCLUDING broadcasting system information, paging, and establishing and releasing an RRC signaling connection between the wireless device 320 and the base station 350.

In various embodiments, the PDCP sublayer 312 may provide uplink functions including multiplexing between different radio bearers and logical channels, sequence number addition, handover data handling, integrity protection, ciphering, and header compression. In the downlink, the PDCP sublayer 312 may provide functions that include in-sequence delivery of data packets, duplicate data packet detection, integrity validation, deciphering, and header decompression.

In the uplink, the RLC sublayer 310 may provide segmentation and concatenation of upper layer data packets, retransmission of lost data packets, and Automatic Repeat Request (ARQ) . In the downlink, while the RLC sublayer 310 functions may include reordering of data packets to compensate for out-of-order reception, reassembly of upper layer data packets, and ARQ.

In the uplink, MAC sublayer 308 may provide functions including multiplexing between logical and transport channels, random access procedure, logical channel priority, and hybrid-ARQ (HARQ) operations. In the downlink, the MAC layer functions may include channel mapping within a cell, de-multiplexing, discontinuous reception (DRX) , and HARQ operations.

While the software architecture 300 may provide functions to transmit data through physical media, the software architecture 300 may further include at least one host layer 314 to provide data transfer services to various applications in the wireless device 320. In some embodiments, application-specific functions provided by the at least one host layer 314 may provide an interface between the software architecture and the general purpose processor 206.

In other embodiments, the software architecture 300 may include one or more higher logical layer (e.g., transport, session, presentation, application, etc. ) that provide host layer functions. For example, in some embodiments, the software architecture 300 may include a network layer (e.g., Internet Protocol (IP) layer) in which a logical connection terminates at a packet data network (PDN) gateway (PGW) . In some embodiments, the software architecture 300 may include an application layer in which a logical connection terminates at another device (e.g., end user device, server, etc. ) . In some embodiments, the software architecture 300 may further include in the AS 304 a hardware interface 316 between the physical layer 306 and the communication hardware (e.g., one or more radio frequency (RF) transceivers) .

FIGS. 4A and 4B are component block diagrams illustrating a system 400 configured for determining a channel state for wireless communications between a base station and a wireless device in accordance with various embodiments. With reference to FIGS. 1–4B, system 400 may include a base station 402 (e.g., 110a-110d, 200, 350) and a wireless device 404 (e.g., 120a-120e, 200, 320) . The base station 402 and the wireless device 404 may communicate over a wireless communication link 122 that may provide the wireless device 404 with access to a wireless communication network 424 (aspects of which are illustrated in FIG. 1) .

The base station 402 may include one or more processors 428 coupled to electronic storage 426 and a wireless transceiver (e.g., 266) . The wireless transceiver 266 may be configured to receive messages to be sent in downlink transmissions from the processor (s) 428, and to transmit such messages via an antenna (not shown) to the wireless device 404. In some embodiments, the base station 402 may receive message from the wireless communication network 424 for relay to the wireless device 404. Similarly, the wireless transceiver 266 may be configured to receive messages from the wireless device 404 in uplink transmissions and pass the messages (e.g., via a modem (e.g., 252) that demodulates the messages) to the one or more processors 428 for eventual relay to the wireless communication network 424.

The processor (s) 428 may be configured by machine-readable instructions 406. Machine-readable instructions 406 may include one or more instruction modules. The instruction modules may include computer program modules. The instruction modules may include one or more of a transmit and receive (Tx/RX) module 408, a machine learning model module 410, a communication adjustment module 412, or other instruction modules.

The TX/RX module 408 may be configured to enable communication with the wireless device 404, and may send and receive various signals and information as described.

The machine learning model module 410 may be configured to apply a machine learning model to the differential feature to determine or reconstruct a downlink channel state.

The communication adjustment module 412 may be configured to adjust communications with the wireless device based on the determined or reconstructed downlink channel state.

The wireless device 404 may include one or more processors 432 coupled to an electronic storage 430 and a wireless transceiver (e.g., 266) . The wireless transceiver 266 may be configured to receive messages to be sent in uplink transmissions from the processor (s) 432, and to transmit such messages via an antenna (not shown) to the base station 402. Similarly, the wireless transceiver 266 may be configured to receive messages from the base station 402 in downlink transmissions and pass the messages (e.g., via a modem (e.g., 252) that demodulates the messages) to the one or more processors 432.

The processor (s) 432 may be configured by machine-readable instructions 434. Machine-readable instructions 406 may include one or more instruction modules. The instruction modules may include computer program modules. The instruction modules may include one or more of a downlink channel estimation module 436, a channel differential information module 438, a machine learning model module 440, a TX/RX module 442, or other instruction modules.

The downlink channel estimation module 436 may be configured to perform a downlink channel estimation based on a downlink reference signal (DL RS) received from the base station.

The channel differential information module 438 may be configured to determine channel differential information based on the downlink channel estimation and a previous downlink channel estimation.

The machine learning model module 440 may be configured to apply a machine learning model to the channel differential information to generate a differential feature.

The TX/RX module 442 may be configured to enable communication with the base station 402, and may send and receive various signals and information as described.

In some embodiments, the base station 402 and wireless device 404 may be operatively linked via one or more electronic communication links (e.g., wireless communication link 122) . It will be appreciated that this is not intended to be limiting, and that the scope of this disclosure includes embodiments in which the base station 402 and wireless device 404 may be operatively linked via some other communication media.

The

electronic storage

426, 430 may include non-transitory storage media that electronically stores information. The electronic storage media of

electronic storage

426, 430 may include one or both of system storage that is provided integrally (i.e., substantially non-removable) with the base station 402 or wireless device 404 and/or removable storage that is removably connectable to the base station 402 or wireless device 404 via, for example, a port (e.g., a universal serial bus (USB) port, a firewire port, etc. ) or a drive (e.g., a disk drive, etc. ) .

Electronic storage

426, 430 may include one or more of optically readable storage media (e.g., optical disks, etc. ) , magnetically readable storage media (e.g., magnetic tape, magnetic hard drive, floppy drive, etc. ) , electrical charge-based storage media (e.g., EEPROM, RAM, etc. ) , solid-state storage media (e.g., flash drive, etc. ) , and/or other electronically readable storage media.

Electronic storage

426, 430 may include one or more virtual storage resources (e.g., cloud storage, a virtual private network, and/or other virtual storage resources) .

Electronic storage

426, 430 may store software algorithms, information determined by processor (s) 428, 432, information received from the base station 402 or wireless device 404, or other information that enables the base station 402 or wireless device 404 to function as described herein.

Processor (s) 428, 432 may be configured to provide information processing capabilities in the base station 402. As such, the processor (s) 428, 432 may include one or more of a digital processor, an analog processor, a digital circuit designed to process information, an analog circuit designed to process information, a state machine, and/or other mechanisms for electronically processing information. Although the processor (s) 428, 432 are illustrated as single entities, this is for illustrative purposes only. In some embodiments, the processor (s) 428, 432 may include a plurality of processing units and/or processor cores. The processing units may be physically located within the same device, or processor (s) 428, 432 may represent processing functionality of a plurality of devices operating in coordination. The processor (s) 428, 432 may be configured to execute modules 408–414 and modules 436–440 and/or other modules by software; hardware; firmware; some combination of software, hardware, and/or firmware; and/or other mechanisms for configuring processing capabilities on processor (s) 428, 432. As used herein, the term “module” may refer to any component or set of components that perform the functionality attributed to the module. This may include one or more physical processors during execution of processor readable instructions, the processor readable instructions, circuitry, hardware, storage media, or any other components.

The description of the functionality provided by the different modules 408–414 and modules 436–442 described below is for illustrative purposes, and is not intended to be limiting, as any of modules 408–414 and modules 436–442 may provide more or less functionality than is described. For example, one or more of the modules 408–414 and modules 436–442 may be eliminated, and some or all of its functionality may be provided by other modules 408–414 and modules 436–442. As another example, the processor (s) 428, 432 may be configured to execute one or more additional modules that may perform some or all of the functionality attributed below to one of the modules 408–414 and modules 436–442.

FIG. 5 is a process flow diagram illustrating a method 500 that may be performed by a processor of a wireless device for determining a channel state for communication with a base station according to various embodiments. With reference to FIGS. 1–5, the method 500 may be implemented by a processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) of a wireless device (e.g., the wireless device 120a–130e, 320, 404) .

In block 502, the processor may perform a downlink channel estimation based on a downlink reference signal (DL RS) received from the base station. Means for performing functions of the operations in block 502 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) and a wireless transceiver (e.g., 266) .

In block 504, the processor may determine channel differential information based on the downlink channel estimation and a previous downlink channel estimation. Means for performing functions of the operations in block 504 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) .

In block 506, the processor may apply a machine learning model to the channel differential information to generate a differential feature. Means for performing functions of the operations in block 506 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) .

In block 508, the processor may send the generated differential feature to the base station to enable the base station to reconstruct the downlink channel state or predict a future downlink channel state. Means for performing functions of the operations in block 508 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) and a wireless transceiver (e.g., 266) .

The method 500 may be repeated continuously or periodically as the processor may again perform the operations of block 502 as described.

FIGS. 6A–6E are process flow diagrams illustrating operations 600a–600e that may be performed by a processor of a wireless device as part of the method 500 for determining a channel state for communication with a base station according to various embodiments. With reference to FIGS. 1–6E, the operations 600a–600e may be implemented by a processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) of a wireless device wireless device (e.g., 120a–120e, 200, 320, 404) .

Referring to FIG. 6A, the processor may receive from the base station a structure of the machine learning model in block 602. In some embodiments, the processor may receive from the base station one or more parameters that the processor may apply to generate or construct the machine learning model. In some embodiments, the processor may receive an indication of a number of layers of the machine learning model. In some embodiments, the processor may receive an indication of one or more weights to be applied in the machine learning model. In some embodiments, the processor may receive an indication of an output feature size. In some embodiments, the processor may receive an indication of a corresponding quantization resolution. In some embodiments, the base station may determine the structure of the machine learning model based on one or more capabilities of the wireless device. For example, for a wireless device with a relatively limited processor or memory, the base station may determine a relatively limited number of layers (such as three) for a convolutional neural network. Means for performing functions of the operations in block 602 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) and a wireless transceiver (e.g., 266) .

In block 604, the processor may structure the machine learning model according to the received structure. Means for performing functions of the operations in block 602 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) .

In block 605, the processor may receive from the base station an instruction to enable the machine learning model. In some embodiments, the processor may receive the structure of the machine learning model and may structure the machine learning model according to the received structure, but may not apply the machine learning model until instructed to enable the machine learning model by the base station.

The processor may then proceed to perform the operations of block 504 of the method 500 as described.

Referring to FIG. 6B, following the performance of the operations of block 506 of the method 500, the processor may determine that a feedback trigger has occurred in block 606. In some embodiments, the processor may determine that the feedback trigger has occurred based on a resource pattern of feedback information, such as a pattern of reports or reporting opportunities of complete feedback information, or of differential feedback information, from the wireless device to the base station. In some embodiments, the processor may determine that a number of slots has occurred. In some embodiments, the processor may receive a request for feedback received from the base station. In some embodiments, the processor may determine that a difference between the downlink channel estimation and the previous downlink channel estimation exceeds a threshold difference. Means for performing functions of the operations in block 606 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) .

In block 608, the processor may send to the base station the generated differential feature in response to determining that the feedback trigger has occurred. Means for performing functions of the operations in block 608 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) and a wireless transceiver (e.g., 266) .

The processor may then proceed to perform the operations of block 502 of the method 500 as described.

Referring to FIG. 6C, following the performance of the operations of block 508 of the method 500, the processor may receive from the base station an indication that a downlink channel estimation performed by the base station using the generated differential feature exceeds an error threshold in block 610. Means for performing functions of the operations in block 602 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) and the wireless transceiver (e.g., 266) .

In block 612, the processor may send the downlink channel estimation to the base station in response to the received indication. Means for performing functions of the operations in block 606 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) .

Referring to FIG. 6D, following the performance of the operations of block 508 of the method 500, the processor may determine to disable use of the machine learning model in block 614 (i.e., determine that the processor should stop applying the machine learning model to the determined channel differential information in block 506) . Means for performing functions of the operations in block 614 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) .

In block 616, the processor may send a request to the base station to disable use of the machine learning model. Means for performing functions of the operations in block 616 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) and the wireless transceiver (e.g., 266) .

In block 618, the processor may receive an indication from the base station to disable use of the machine learning model in block 618. Means for performing functions of the operations in block 618 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) and the wireless transceiver (e.g., 266) .

In block 620, the processor may disable use of the machine learning model. In some embodiments, the processor may disable the use the machine learning model in response to an indication from the base station to disable the use of the machine learning model. Means for performing functions of the operations in block 620 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) .

Referring to FIG. 6E, following the performance of the operations of block 508 of the method 500, the processor may send a request to the base station to update the machine learning model in block 622. Means for performing functions of the operations in block 622 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) and the wireless transceiver (e.g., 266) .

In block 624, the processor may receive from the base station an updated structure of the machine learning model. In some embodiments, the processor may receive from the base station one or more parameters that the processor may apply to update the machine learning model. Means for performing functions of the operations in block 624 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) and the wireless transceiver (e.g., 266) .

In block 626, the processor may update structure the machine learning model according to the received updated structure. Means for performing functions of the operations in block 626 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 432) .

FIG. 7 is a process flow diagram illustrating a method 700 that may be performed by a processor of a base station for determining a channel state for communication with a wireless device according to various embodiments. With reference to FIGS. 1–7, the method 700 may be implemented by a processor (e.g., 210, 212, 214, 216, 218, 252, 260, 428) of a base station (e.g., the base station 110a–110d, 350, 402) .

In block 702, the processor may send to the wireless device a downlink reference signal (DL RS) . Means for performing functions of the operations in block 702 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 428) and a wireless transceiver (e.g., 266) .

In block 704, the processor may receive from the wireless device a differential feature of a downlink channel state between the base station and the wireless device. Means for performing functions of the operations in block 704 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 428) .

In block 706, the processor may apply a machine learning model to the differential feature to reconstruct a downlink channel state. In some embodiments, the processor may apply the machine learning model to the differential feature and to one or more previously-determined differential features to predict a future downlink channel state. Means for performing functions of the operations in block 706 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 428) .

In block 708, the processor may adjust communications with the wireless device based on the generated downlink channel state. Means for performing functions of the operations in block 708 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 428) and a wireless transceiver (e.g., 266) .

The method 700 may be repeated continuously or periodically as the processor may again perform the operations of block 702 as described.

FIGS. 8A–8C are process flow diagrams illustrating operations 800a–800c that may be performed by a processor of a base station as part of the method 700 for determining a channel state for communication with a wireless device according to various embodiments. With reference to FIGS. 1–8C, the operations 800a–800c may be implemented by a processor (e.g., 210, 212, 214, 216, 218, 252, 260, 428) of a base station (e.g., the base station 110a–110d, 350, 402) .

Referring to FIG. 8A, following the performance of the operations of block 702 of the method 700, the processor may send to the wireless device a structure of a wireless device machine learning model in block 802. In some embodiments, the processor may send an indication of a number of layers of the wireless device machine learning model. In some embodiments, the processor may send an indication of one or more weights to be applied in the wireless device machine learning model. In some embodiments, the processor may send an indication of an output feature size. In some embodiments, the processor may send an indication of a quantization resolution. In some embodiments, the processor may send a structure of the machine learning model that is based on a capability of the wireless device. Means for performing functions of the operations in block 802 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 428) and a wireless transceiver (e.g., 266) .

In block 803, the processor may send to the wireless device an instruction to enable the machine learning model. Means for performing functions of the operations in block 802 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 428) and the wireless transceiver (e.g., 266) .

In block 804, the processor may receive the differential feature based on the structure of the wireless device machine learning model. Means for performing functions of the operations in block 804 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 428) and the wireless transceiver (e.g., 266) .

The processor may proceed to perform the operations of block 706 of the method 700 as described.

Referring to FIG. 8B, following the performance of the operations of block 702 of the method 700, the processor may send to the wireless device an indication of a feedback trigger in block 806. In some embodiments, the processor may send an indication that the feedback trigger comprises a resource pattern of feedback information, such as a pattern of reports or reporting opportunities of complete feedback information, or of differential feedback information, from the wireless device to the base station. In some embodiments, the processor may send an indication that the feedback trigger comprises an occurrence of a number of slots. In some embodiments, the processor may send to the wireless device a request for feedback. Means for performing functions of the operations in block 806 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 428) and the wireless transceiver (e.g., 266) .

In block 808, the processor may receive from the wireless device a differential feature of a downlink channel state between the base station and the wireless device in response to sending the indication of the feedback trigger.

The processor may then perform the operations of block 706 of the method 700 as described.

Referring to FIG. 8C, following the performance of the operations of block 702 of the method 700, the processor may determine whether the reconstructed downlink channel state exceeds an error threshold in determination block 810.

In response to determining that the reconstructed downlink channel state does not exceed the error threshold (i.e., determination block 810 = “No” ) , the processor may perform the operations in block 708 of the method 700 as described.

In response to determining that the reconstructed downlink channel state exceeds the error threshold (i.e., determination block 810 = “Yes” ) , the processor may send a message to the wireless device to send the overall channel estimation made by the wireless device based on the DL RS in block 814. Means for performing functions of the operations in block 814 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 428) and the wireless transceiver (e.g., 266) .

In block 816, the processor may receive from the wireless device a downlink channel estimation or channel differential information determined by the wireless device. Means for performing functions of the operations in block 816 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 428) and the wireless transceiver (e.g., 266) .

In block 818, the processor may reconstruct the downlink channel state using the downlink channel estimation or channel differential estimation received from the wireless device. This reconstruction may use conventional channel state reconstruction methods. Means for performing functions of the operations in block 706 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 428) .

The processor may then perform the operations of block 708 of the method 700 as described.

Referring to FIG. 8D, following the performance of the operations of block 708 of the method 700, the processor may receive a request to update the machine learning model from the wireless device in block 820. Means for performing functions of the operations in block 820 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 428) and the wireless transceiver (e.g., 266) .

In block 822, the processor may update the structure of the machine learning model. In some embodiments, the processor may update one or more parameters that the wireless device may apply to update the machine learning model. In some embodiments, the update to the structure of the machine learning model may be based on one or more differential features received from the wireless device. In some embodiments, the update to the structure of the machine learning model may be based on other information received from the wireless device, such as a downlink channel estimation or channel differential information. Means for performing functions of the operations in block 822 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 428) .

In block 824, the processor may send the updated structure of the machine learning model to the wireless device. Means for performing functions of the operations in block 824 may include the processor (e.g., 210, 212, 214, 216, 218, 252, 260, 428) and the wireless transceiver (e.g., 266) .

Various embodiments, including the method 700 and the operations 800a–800c, may be performed in a variety of network computing devices (e.g., in a base station) , an example of which is illustrated in FIG. 9 that is a component block diagram of a network computing device 900 suitable for use with various embodiments. Such network computing devices may include at least the components illustrated in FIG. 9. With reference to FIGS. 1–9, a network computing device 900 may include a processor 901 coupled to volatile memory 902 (e.g., 426) and a large capacity nonvolatile memory, such as a disk drive 903. The network computing device 900 may also include a peripheral memory access device such as a floppy disc drive, compact disc (CD) or digital video disc (DVD) drive 906 coupled to the processor 901. The network computing device 900 may also include network access ports 904 (or interfaces) coupled to the processor 901 for establishing data connections with a network, such as the Internet and/or a local area network coupled to other system computers and servers. The network computing device 900 may be connected to one or more antennas for sending and receiving electromagnetic radiation that may be connected to a wireless communication link. The network computing device 900 may include additional access ports, such as USB, Firewire, Thunderbolt, and the like for coupling to peripherals, external memory, or other devices.

Various embodiments, including the methods 500 and operations 600a–600c, may be performed in a variety of wireless devices (e.g., the wireless device 120a-120e, 200, 320, 402) , an example of which is illustrated in FIG. 10 that is a component block diagram of a wireless device 10 suitable for use with various embodiments. With reference to FIGS. 1–10, a wireless device 1000 may include a first SOC 202 (e.g., a SOC-CPU) coupled to a second SOC 204 (e.g., a 5G capable SOC) . The first and

second SOCs

202, 204 may be coupled to

internal memory

430, 1016, a display 1012, and to a speaker 1014. Additionally, the wireless device 1000 may include an antenna 1004 for sending and receiving electromagnetic radiation that may be connected to a wireless data link and/or cellular telephone transceiver 266 coupled to one or more processors in the first and/or

second SOCs

202, 204. The wireless device 1000 may also include menu selection buttons or rocker switches 1020 for receiving user inputs.

The wireless device 1000 also may include a sound encoding/decoding (CODEC) circuit 1010, which digitizes sound received from a microphone into data packets suitable for wireless transmission and decodes received sound data packets to generate analog signals that are provided to the speaker to generate sound. Also, one or more of the processors in the first and

second SOCs

202, 204, wireless transceiver 266 and CODEC 1010 may include a digital signal processor (DSP) circuit (not shown separately) .

The processors of the network computing device 1000 and the wireless device 1000 may be any programmable microprocessor, microcomputer or multiple processor chip or chips that can be configured by software instructions (applications) to perform a variety of functions, including the functions of the various embodiments described below. In some mobile devices, multiple processors may be provided, such as one processor within an SOC 204 dedicated to wireless communication functions and one processor within an SOC 202 dedicated to running other applications. Software applications may be stored in the

memory

426, 430, 1016 before they are accessed and loaded into the processor. The processors may include internal memory sufficient to store the application software instructions.

As used in this application, the terms “component, ” “module, ” “system, ” and the like are intended to include a computer-related entity, such as, but not limited to, hardware, firmware, a combination of hardware and software, software, or software in execution, which are configured to perform particular operations or functions. For example, a component may be, but is not limited to, a process running on a processor, a processor, an object, an executable, a thread of execution, a program, and/or a computer. By way of illustration, both an application running on a wireless device and the wireless device may be referred to as a component. One or more components may reside within a process and/or thread of execution and a component may be localized on one processor or core and/or distributed between two or more processors or cores. In addition, these components may execute from various non-transitory computer readable media having various instructions and/or data structures stored thereon. Components may communicate by way of local and/or remote processes, function or procedure calls, electronic signals, data packets, memory read/writes, and other known network, computer, processor, and/or process related communication methodologies.

A number of different cellular and mobile communication services and standards are available or contemplated in the future, all of which may implement and benefit from the various embodiments. Such services and standards include, e.g., third generation partnership project (3GPP) , long term evolution (LTE) systems, third generation wireless mobile communication technology (3G) , fourth generation wireless mobile communication technology (4G) , fifth generation wireless mobile communication technology (5G) , global system for mobile communications (GSM) , universal mobile telecommunications system (UMTS) , 3GSM, general packet radio service (GPRS) , code division multiple access (CDMA) systems (e.g., cdmaOne, CDMA1020TM) , enhanced data rates for GSM evolution (EDGE) , advanced mobile phone system (AMPS) , digital AMPS (IS-136/TDMA) , evolution-data optimized (EV-DO) , digital enhanced cordless telecommunications (DECT) , Worldwide Interoperability for Microwave Access (WiMAX) , wireless local area network (WLAN) , Wi-Fi Protected Access I &II (WPA, WPA2) , and integrated digital enhanced network (iDEN) . Each of these technologies involves, for example, the transmission and reception of voice, data, signaling, and/or content messages. It should be understood that any references to terminology and/or technical details related to an individual telecommunication standard or technology are for illustrative purposes only, and are not intended to limit the scope of the claims to a particular communication system or technology unless specifically recited in the claim language.

Various embodiments illustrated and described are provided merely as examples to illustrate various features of the claims. However, features shown and described with respect to any given embodiment are not necessarily limited to the associated embodiment and may be used or combined with other embodiments that are shown and described. Further, the claims are not intended to be limited by any one example embodiment. For example, one or more of the operations of the methods or operations disclosed herein may be substituted for or combined with one or more operations of the methods or operations disclosed herein.

The foregoing method descriptions and the process flow diagrams are provided merely as illustrative examples and are not intended to require or imply that the operations of various embodiments must be performed in the order presented. As will be appreciated by one of skill in the art the order of operations in the foregoing embodiments may be performed in any order. Words such as “thereafter, ” “then, ” “next, ” etc. are not intended to limit the order of the operations; these words are used to guide the reader through the description of the methods. Further, any reference to claim elements in the singular, for example, using the articles “a, ” “an, ” or “the” is not to be construed as limiting the element to the singular.

Various illustrative logical blocks, modules, components, circuits, and algorithm operations described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and operations have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such embodiment decisions should not be interpreted as causing a departure from the scope of the claims.

The hardware used to implement various illustrative logics, logical blocks, modules, and circuits described in connection with the embodiments disclosed herein may be implemented or performed with a general purpose processor, a digital signal processor (DSP) , an application specific integrated circuit (ASIC) , a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general-purpose processor may be a microprocessor, but, in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of receiver smart objects, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration. Alternatively, some operations or methods may be performed by circuitry that is specific to a given function.

In one or more embodiments, the functions described may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, the functions may be stored as one or more instructions or code on a non-transitory computer-readable storage medium or non-transitory processor-readable storage medium. The operations of a method or algorithm disclosed herein may be embodied in a processor-executable software module or processor-executable instructions, which may reside on a non-transitory computer-readable or processor-readable storage medium. Non-transitory computer-readable or processor-readable storage media may be any storage media that may be accessed by a computer or a processor. By way of example but not limitation, such non-transitory computer-readable or processor-readable storage media may include RAM, ROM, EEPROM, FLASH memory, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage smart objects, or any other medium that may be used to store desired program code in the form of instructions or data structures and that may be accessed by a computer. Disk and disc, as used herein, includes compact disc (CD) , laser disc, optical disc, digital versatile disc (DVD) , floppy disk, and Blu-ray disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasers. Combinations of the above are also included within the scope of non-transitory computer-readable and processor-readable media. Additionally, the operations of a method or algorithm may reside as one or any combination or set of codes and/or instructions on a non-transitory processor-readable storage medium and/or computer-readable storage medium, which may be incorporated into a computer program product.

The preceding description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the claims. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the scope of the claims. Thus, the present disclosure is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the following claims and the principles and novel features disclosed herein.

Claims

A method performed by a processor of a wireless device for determining a channel state for communication with a base station, comprising:

performing a downlink channel estimation based on a downlink reference signal (DL RS) received from the base station;

determining channel differential information based on the downlink channel estimation and a previous downlink channel estimation;

applying a machine learning model to the channel differential information to generate a differential feature; and

sending to the base station the generated differential feature to enable the base station to determine a downlink channel state and adjust communications with the wireless device.
The method of claim 1, further comprising:

receiving from the base station a structure of the machine learning model; and

structuring the machine learning model according to the received structure.
The method of claim 2, further comprising receiving from the base station an instruction to enable the machine learning model.
The method of claim 2, wherein receiving from the base station the structure of the machine learning model comprises receiving an indication of a number of layers of the machine learning model.
The method of claim 2, wherein receiving from the base station the structure of the machine learning model comprises receiving an indication of one or more weights to be applied in the machine learning model.
The method of claim 2, wherein receiving from the base station the structure of the machine learning model comprises receiving an indication of an output feature size.
The method of claim 2, wherein receiving from the base station the structure of the machine learning model comprises receiving an indication of a quantization resolution.
The method of claim 2, wherein receiving from the base station the structure of the machine learning model comprises receiving from the base station a structure of the machine learning model that is based on a capability of the wireless device.
The method of claim 1, wherein sending to the base station the generated differential feature to enable the base station to determine a downlink channel state and adjust communications with the wireless device based on the determined downlink channel state comprises:

determining that a feedback trigger has occurred; and

sending to the base station the generated differential feature in response to determining that the feedback trigger has occurred.
The method of claim 9, wherein the feedback trigger comprises a resource pattern of feedback information.
The method of claim 9, wherein the feedback trigger comprises an occurrence of a number of slots.
The method of claim 9, wherein the feedback trigger comprises a request for feedback received from the base station.
The method of claim 9, wherein determining that a feedback trigger has occurred comprises determining that a difference between the downlink channel estimation and the previous downlink channel estimation exceeds a threshold difference.
The method of claim 1, further comprising:

receiving from the base station an indication that a downlink channel estimation performed by the base station using the generated differential feature exceeds an error threshold; and

sending the downlink channel estimation or the channel differential information to the base station in response to the received indication.
The method of claim 1, further comprising disabling the use of the machine learning model in response to an indication from the base station to disable use of the machine learning model.
The method of claim 1, further comprising:

sending to the base station a request to update the machine learning model;

receiving an updated structure of the machine learning model; and

structuring the machine learning model according to the received updated structure.
A method performed by a processor of a base station for determining a channel state for communication with a wireless device, comprising:

sending to the wireless device a downlink reference signal (DL RS) ;

receiving from the wireless device a differential feature of a downlink channel state between the base station and the wireless device;

applying a machine learning model to the differential feature to determine a downlink channel state; and

adjusting communications with the wireless device based on the reconstructed downlink channel state.
The method of claim 17, wherein applying the machine learning model to the differential feature to determine a downlink channel state comprises applying the machine learning model to the differential feature and to one or more previously-determined differential features to predict a future downlink channel state.
The method of claim 17, further comprising sending to the wireless device a structure of a wireless device machine learning model,

wherein receiving from the wireless device a differential feature of a downlink channel state between the base station and the wireless device comprises receiving the differential feature based on the structure of the wireless device machine learning model.
The method of claim 19, further comprising sending to the wireless device an instruction to enable a machine learning model on the wireless device.
The method of claim 19, wherein sending to the wireless device a structure of a wireless device machine learning model comprises sending an indication of a number of layers of the wireless device machine learning model.
The method of claim 19, wherein sending to the wireless device a structure of a wireless device machine learning model comprises sending an indication of one or more weights to be applied in the wireless device e machine learning model.
The method of claim 19, wherein sending to the wireless device a structure of a wireless device machine learning model comprises sending an indication of an output feature size.
The method of claim 19, wherein sending to the wireless device a structure of a wireless device machine learning model comprises sending an indication of a quantization resolution.
The method of claim 19, wherein sending to the wireless device a structure of a wireless device machine learning model comprises sending a structure of the machine learning model that is based on a capability of the wireless device.
The method of claim 17, wherein receiving from the wireless device a differential feature of a downlink channel state between the base station and the wireless device comprises:

sending to the wireless device an indication of a feedback trigger; and

receiving from the wireless device a differential feature of a downlink channel state between the base station and the wireless device in response to sending the indication of the feedback trigger.
The method of claim 26, wherein sending to the wireless device an indication of a feedback trigger comprises sending an indication of a resource pattern of feedback information.
The method of claim 26, wherein sending to the wireless device an indication of a feedback trigger comprises sending an indication that the feedback trigger comprises an occurrence of a number of slots.
The method of claim 26, wherein sending to the wireless device an indication of a feedback trigger comprises sending to the wireless device a request for feedback.
The method of claim 17, further comprising:

determining whether the determined downlink channel state exceeds an error threshold;

sending to the wireless device a message to transmit a downlink channel estimation or channel differential information in response to determining that the determined downlink channel state exceeds the error threshold; and

receiving from the wireless device a downlink channel estimation performed by the wireless device.
The method of claim 19, further comprising:

receiving from the wireless device a request to update the machine learning model;

updating the structure of the machine learning model; and

sending the updated structure of the machine learning model to the wireless device.
A wireless device, comprising:

a wireless transceiver; and

a processor coupled to the wireless transceiver and configure to perform operations comprising:

performing a downlink channel estimation based on a downlink reference signal (DL RS) received from a base station;

determining channel differential information based on the downlink channel estimation and a previous downlink channel estimation;

applying a machine learning model to the channel differential information to generate a differential feature; and

sending to the base station the generated differential feature to enable the base station to determine a downlink channel state and adjust communications with the wireless device.
The wireless device of claim 32, wherein the processor is configured with processor-executable instructions to perform operations further comprising:

receiving from the base station a structure of the machine learning model; and

structuring the machine learning model according to the received structure.
The wireless device of claim 33, wherein the processor is configured with processor-executable instructions to perform operations further comprising receiving from the base station an instruction to enable the machine learning model.
The wireless device of claim 33, wherein the processor is configured with processor-executable instructions to perform operations such that receiving from the base station the structure of the machine learning model comprises receiving an indication of a number of layers of the machine learning model.
The wireless device of claim 33, wherein the processor is configured with processor-executable instructions to perform operations such that receiving from the base station the structure of the machine learning model comprises receiving an indication of one or more weights to be applied in the machine learning model.
The wireless device of claim 33, wherein the processor is configured with processor-executable instructions to perform operations such that receiving from the base station the structure of the machine learning model comprises receiving an indication of an output feature size.
The wireless device of claim 33, wherein the processor is configured with processor-executable instructions to perform operations such that receiving from the base station the structure of the machine learning model comprises receiving an indication of a quantization resolution.
The wireless device of claim 33, wherein the processor is configured with processor-executable instructions to perform operations such that receiving from the base station the structure of the machine learning model comprises receiving from the base station a structure of the machine learning model that is based on a capability of the wireless device.
The wireless device of claim 33, wherein the processor is configured with processor-executable instructions to perform operations such that sending to the base station the generated differential feature to enable the base station to determine a downlink channel state and adjust communications with the wireless device based on the determined downlink channel state comprises:

determining that a feedback trigger has occurred; and

sending to the base station the generated differential feature in response to determining that the feedback trigger has occurred.
The wireless device of claim 40, wherein the processor is configured with processor-executable instructions to perform operations such that the feedback trigger comprises a resource pattern of feedback information.
The wireless device of claim 40, wherein the processor is configured with processor-executable instructions to perform operations such that the feedback trigger comprises an occurrence of a number of slots.
The wireless device of claim 40, wherein the processor is configured with processor-executable instructions to perform operations such that the feedback trigger comprises a request for feedback received from the base station.
The wireless device of claim 40, wherein the processor is configured with processor-executable instructions to perform operations such that determining that a feedback trigger has occurred comprises determining that a difference between the downlink channel estimation and the previous downlink channel estimation exceeds a threshold difference.
The wireless device of claim 32, wherein the processor is configured with processor-executable instructions to perform operations further comprising:

receiving from the base station an indication that a downlink channel estimation performed by the base station using the generated differential feature exceeds an error threshold; and

sending the downlink channel estimation or the channel differential information to the base station in response to the received indication.
The wireless device of claim 32, wherein the processor is configured with processor-executable instructions to perform operations further comprising disabling the use of the machine learning model in response to an indication from the base station to disable use of the machine learning model.
The wireless device of claim 32, wherein the processor is configured with processor-executable instructions to perform operations further comprising:

sending to the base station a request to update the machine learning model;

receiving an updated structure of the machine learning model; and

structuring the machine learning model according to the received updated structure.
A base station, comprising:

a processor configured with processor-executable instructions to perform operations comprising:

sending to a wireless device a downlink reference signal (DL RS) ;

receiving from the wireless device a differential feature of a downlink channel state between the base station and the wireless device;

applying a machine learning model to the differential feature to determine a downlink channel state; and

adjusting communications with the wireless device based on the reconstructed downlink channel state.
The base station of claim 48, wherein the processor is configured with processor-executable instructions to perform operations such that applying the machine learning model to the differential feature to determine a downlink channel state comprises applying the machine learning model to the differential feature and to one or more previously-determined differential features to predict a future downlink channel state.
The base station of claim 48, wherein the processor is configured with processor-executable instructions to perform operations further comprising sending to the wireless device a structure of a wireless device machine learning model, and

wherein the processor is configured with processor-executable instructions to perform operations such that receiving from the wireless device a differential feature of a downlink channel state between the base station and the wireless device comprises receiving the differential feature based on the structure of the wireless device machine learning model.
The base station of claim 50, wherein the processor is configured with processor-executable instructions to perform operations further comprising sending to the wireless device an instruction to enable a machine learning model on the wireless device.
The base station of claim 50, wherein the processor is configured with processor-executable instructions to perform operations such that sending to the wireless device a structure of a wireless device machine learning model comprises sending an indication of a number of layers of the wireless device machine learning model.
The base station of claim 50, wherein the processor is configured with processor-executable instructions to perform operations such that sending to the wireless device a structure of a wireless device machine learning model comprises sending an indication of one or more weights to be applied in the wireless device e machine learning model.
The base station of claim 50, wherein the processor is configured with processor-executable instructions to perform operations such that sending to the wireless device a structure of a wireless device machine learning model comprises sending an indication of an output feature size.
The base station of claim 50, wherein the processor is configured with processor-executable instructions to perform operations such that sending to the wireless device a structure of a wireless device machine learning model comprises sending an indication of a quantization resolution.
The base station of claim 50, wherein the processor is configured with processor-executable instructions to perform operations such that sending to the wireless device a structure of a wireless device machine learning model comprises sending a structure of the machine learning model that is based on a capability of the wireless device.
The base station of claim 48, wherein the processor is configured with processor-executable instructions to perform operations such that receiving from the wireless device a differential feature of a downlink channel state between the base station and the wireless device comprises:

sending to the wireless device an indication of a feedback trigger; and

receiving from the wireless device a differential feature of a downlink channel state between the base station and the wireless device in response to sending the indication of the feedback trigger.
The base station of claim 57, wherein the processor is configured with processor-executable instructions to perform operations such that sending to the wireless device an indication of a feedback trigger comprises sending an indication of a resource pattern of feedback information.
The base station of claim 57, wherein the processor is configured with processor-executable instructions to perform operations such that sending to the wireless device an indication of a feedback trigger comprises sending an indication that the feedback trigger comprises an occurrence of a number of slots.
The base station of claim 57, wherein the processor is configured with processor-executable instructions to perform operations such that sending to the wireless device an indication of a feedback trigger comprises sending to the wireless device a request for feedback.
The base station of claim 48, wherein the processor is configured with processor-executable instructions to perform operations further comprising:

determining whether the determined downlink channel state exceeds an error threshold;

sending to the wireless device a message to transmit a downlink channel estimation or channel differential information in response to determining that the determined downlink channel state exceeds the error threshold; and

receiving from the wireless device a downlink channel estimation performed by the wireless device.
The base station of claim 50, wherein the processor is configured with processor-executable instructions to perform operations further comprising:

receiving from the wireless device a request to update the machine learning model;

updating the structure of the machine learning model; and

sending the updated structure of the machine learning model to the wireless device.
A non-transitory processor-readable medium having stored thereon processor-executable instructions configured to cause a processor of a wireless device to perform operations comprising:

performing a downlink channel estimation based on a downlink reference signal (DL RS) received from a base station;

determining channel differential information based on the downlink channel estimation and a previous downlink channel estimation;

applying a machine learning model to the channel differential information to generate a differential feature; and

sending to the base station the generated differential feature to enable the base station to determine a downlink channel state and adjust communications with the wireless device.
The non-transitory processor-readable medium of claim 63, wherein the stored processor-executable instructions are configured to cause the processor of the wireless device to perform operations further comprising:

receiving from the base station a structure of the machine learning model; and

structuring the machine learning model according to the received structure.
A non-transitory processor-readable medium having stored thereon processor-executable instructions configured to cause a processor of a base station to perform operations comprising:

sending to a wireless device a downlink reference signal (DL RS) ;

receiving from the wireless device a differential feature of a downlink channel state between the base station and the wireless device;

applying a machine learning model to the differential feature to determine a downlink channel state; and

adjusting communications with the wireless device based on the reconstructed downlink channel state.
The non-transitory processor-readable medium of claim 63, wherein the stored processor-executable instructions are configured to cause the processor of the wireless device to perform operations further comprising sending to the wireless device a structure of a wireless device machine learning model, and

wherein the processor is configured with processor-executable instructions to perform operations such that receiving from the wireless device a differential feature of a downlink channel state between the base station and the wireless device comprises receiving the differential feature based on the structure of the wireless device machine learning model.