Method for transmitting TX waveform distortion to a receiver

Abstract

Various embodiments may use a neural network at a transmitting device to compress transmit (TX) waveform distortion. In various embodiments, the compressed TX waveform distortion information may be communicated to a receiving device. In various embodiments, signaling of TX waveform distortion information from a transmitting device to a receiving device may enable the receiving device to mitigate waveform distortion in a transmit waveform received from the transmitting device. Various embodiments include systems and methods for wireless communication by transmitting a waveform to a receiving device executed by a processor of a transmitting device. Various embodiments include systems and methods for wireless communication by receiving a waveform from a transmitting device executed by a processor of a receiving device.

Description

[Technical Field]

[0001] Related Applications This application claims the benefit of priority to U.S. Provisional Application No. 62 / 980,793, filed February 24, 2020, entitled "Method To Convey The TX Waveform Distortion To The Receiver," the entire contents of which are incorporated herein by reference for all purposes. [Background technology]

[0002] Long Term Evolution (LTE), Fifth Generation (5G) New Radio (NR), and other recently developed communication technologies enable wireless devices to communicate information at data rates several orders of magnitude greater (e.g., gigabits per second) than were available just a few years ago.

[0003] Today's communication networks are also more secure, more resistant to multipath fading, allow for lower network traffic latency, and offer better communication efficiency (e.g., in bits per second per unit of bandwidth used). These and other recent improvements have facilitated the emergence of the Internet of Things (IoT), large-scale machine-to-machine (M2M) communication systems, autonomous vehicles, and other technologies that rely on consistent, secure communications.

[0004] Additionally, the deployment of neural networks, such as deep neural networks, is gaining momentum in today's communication networks. Neural networks can be used for a variety of tasks on computing devices. Neural networks often use multi-layer architectures, where each layer receives input, performs computations on the inputs, and generates outputs. The output of the first layer of nodes often becomes the input to the second layer of nodes, the output of the second layer of nodes becomes the input to the third layer of nodes, and so on. Thus, computations in neural networks are distributed across a collection of processing nodes that make up a computational chain. Summary of the Invention [Means for solving the problem]

[0005] Various aspects include systems and methods for wireless communication by transmitting a waveform to a receiving device, the systems and methods including: obtaining transmit waveform distortion information of the transmitting device; compressing the transmit waveform distortion information of the transmitting device into compressed transmit waveform distortion information using an encoder neural network; and sending the compressed transmit waveform distortion information and one or more decoder neural network weights to the receiving device in a configuration that enables the receiving device to configure a decoder neural network of the receiving device to recover the transmit waveform distortion information of the transmitting device from the compressed transmit waveform distortion information using the one or more decoder neural network weights.

[0006] Some aspects may further include determining a model type of a decoder neural network of the transmitting device, and sending the compressed transmit waveform distortion information and the one or more decoder neural network weights to the receiving device may include sending the compressed transmit waveform distortion information, the one or more decoder neural network weights, and the model type to the receiving device. In some aspects, sending the compressed transmit waveform distortion information and the one or more decoder neural network weights to the receiving device may include sending the compressed transmit waveform distortion information and the one or more decoder neural network weights to the receiving device in control information for each slot transmitted. In some aspects, sending the compressed transmit waveform distortion information and the one or more decoder neural network weights to the receiving device may include sending the compressed transmit waveform distortion information to the receiving device in control information for each slot transmitted, and sending the one or more decoder neural network weights to the receiving device in control information at a period greater than each slot transmitted. In some aspects, the transmitting device may be a user equipment (UE) computing device, and the receiving device may be a base station. In various aspects, the transmitting device may be a base station, and the receiving device may be a UE computing device.

[0007] Some aspects may further include training an encoder neural network to compress the transmit waveform distortion information of the transmitting device into compressed transmit waveform distortion information, and training a decoder neural network of the transmitting device to recover the transmit waveform distortion information of the transmitting device from the compressed transmit waveform distortion information, wherein the one or more decoder neural network weights are weights of the trained decoder neural network of the transmitting device. In some aspects, the encoder neural network and the decoder neural network of the transmitting device may be trained using an unsupervised learning algorithm. In some aspects, training the encoder neural network and training the decoder neural network of the transmitting device may include training the encoder neural network and the decoder neural network of the transmitting device for one transmit antenna of the transmitting device. In some aspects, training the encoder neural network and training the decoder neural network of the transmitting device may include an encoder neural network and a decoder neural network of the transmitting device for each transmit antenna of the transmitting device.

[0008] Further aspects may include systems and methods for wireless communication via reception of a waveform from a transmitting device, executed by a processor of a receiving device. Various aspects may include receiving compressed transmit waveform distortion information of the transmitting device and one or more weights of a trained decoder neural network of the transmitting device, configuring a decoder neural network of the receiving device using the received one or more weights, and recovering the transmit waveform distortion information of the transmitting device from the compressed transmit waveform distortion information of the transmitting device using the configured decoder neural network of the receiving device. Some aspects may further include using the recovered transmit waveform distortion information of the transmitting device to mitigate waveform distortion in a transmit waveform received from the transmitting device.

[0009] In some aspects, receiving the compressed transmit waveform distortion information of the transmitting device and one or more weights of a trained decoder neural network of the transmitting device may include receiving the compressed transmit waveform distortion information of the transmitting device, one or more weights of a trained decoder neural network of the transmitting device, and a model type of the trained decoder neural network of the transmitting device, and configuring the decoder neural network of the receiving device using the received one or more weights may include configuring the decoder neural network of the receiving device using the received one or more weights and the received model type. In some aspects, the recovered transmit waveform distortion information may be a two-dimensional map of distortion errors due to signal clipping of Orthogonal Frequency Division Multiplexing (OFDM) symbols within a slot for one or more antennas of the transmitting device. In some aspects, receiving the compressed transmit waveform distortion information of the transmitting device and one or more weights of a trained decoder neural network of the transmitting device may include receiving the compressed transmit waveform distortion information of the transmitting device and one or more weights of a trained decoder neural network of the transmitting device in control information for each slot to be transmitted. In some aspects, receiving the compressed transmit waveform distortion information of the transmitting device and one or more weights of a trained decoder neural network of the transmitting device may include receiving the compressed transmit waveform distortion information of the transmitting device in control information for each slot transmitted and receiving the one or more weights of the trained decoder neural network of the transmitting device in control information with a periodicity greater than each slot transmitted. In some aspects, the receiving device may be a base station and the transmitting device is a UE computing device. In some aspects, the compressed transmit waveform distortion information and the one or more weights may be received directly from the UE computing device. In some aspects, the compressed transmit waveform distortion information and the one or more weights may be received from a base station other than the transmitting device.

[0010] Further aspects may include a base station or wireless device having a processor configured to perform one or more operations of any of the 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 or wireless device to perform operations of any of the methods summarized above. Further aspects include a base station or wireless device having means for performing several functions of the methods summarized above. Further aspects include a system-on-chip for use in a base station or wireless device, including a processor configured to perform one or more operations of any of the methods summarized above. Further aspects include a system-in-package including two system-on-chips for use in a base station or wireless device, including processors configured to perform one or more operations of any of the methods summarized above.

[0011] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the claims and, together with the general description given above and the detailed description below, serve to explain the features of the claims. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a system block diagram illustrating an example communication system suitable for implementing any of the various embodiments. [Figure 2] FIG. 1 is a component block diagram illustrating an exemplary computing and wireless modem system suitable for implementing any of the various embodiments. [Figure 3] FIG. 1 is a component block diagram illustrating a software architecture including radio protocol stacks for user and control planes in wireless communications suitable for implementing any of the various embodiments. [Figure 4]FIG. 1 is a functional component block diagram illustrating an exemplary software-implemented neural network suitable for implementing any of the various embodiments. [Figure 5] FIG. 1 is a process flow diagram illustrating operations of a method of wireless communication performed by a processor of a transmitting device that transmits a waveform to a receiving device, according to various embodiments. [Figure 6] FIG. 10 is a process flow diagram illustrating operations of a method of wireless communication performed by a processor of a receiving device that receives a waveform from a transmitting device, according to various embodiments. [Figure 7A] 1 is a block diagram illustrating example operations for compressing distortion information using a neural network, according to various embodiments. [Figure 7B] 1 is a block diagram illustrating example operations for compressing distortion information using a neural network, according to various embodiments. [Figure 8] FIG. 1 is a component block diagram of a network computing device suitable for use with various embodiments. [Figure 9] FIG. 1 is a component block diagram of a wireless communication device suitable for use with the various embodiments. DETAILED DESCRIPTION OF THE INVENTION

[0013] Various embodiments will be described in detail with reference to the accompanying drawings. Whenever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts. References made to specific examples and implementations are for illustrative purposes only and do not limit the scope of the claims.

[0014] Various embodiments enable signaling of transmit (TX) waveform distortion information from a transmitting device to a receiving device. As used herein, the term "transmitting device" refers to any device that outputs a TX waveform, and the term "receiving device" refers to any device that attempts to receive the TX waveform. As an example, in uplink (UL) communications in a 3rd Generation Partnership Project (3GPP®) network, the transmitting device may be a user equipment (UE) computing device, and the receiving device may be a base station (e.g., a next-generation Node B (gNB)). As another example, in downlink (DL) communications, the transmitting device may be a base station (e.g., a gNB), and the receiving device may be a UE computing device.

[0015] Various embodiments may use a neural network executed within the transmitting device to compress the TX waveform distortion. In various embodiments, the compressed TX waveform distortion information may be communicated to the receiving device. In various embodiments, signaling of the TX waveform distortion information from the transmitting device to the receiving device may enable the receiving device to mitigate waveform distortion in the transmit waveform received from the transmitting device. In various embodiments, the presence of the transmitting device's TX waveform distortion information decoded by the receiving device may enable the transmitting device to efficiently utilize its available transmit power.

[0016] In wireless communications between a transmitting device and a receiving device, distortion of the TX waveform sent from the transmitting device, such as signal clipping, can prevent successful reception of the TX waveform by the receiving device. The distortion of the TX waveform can be caused by various factors, such as components of the transmitting device itself. For example, the power amplifier of the transmitting device can cause distortion of the TX waveform, and the amount of distortion can increase as the power level of the power amplifier increases. Conventional transmitting devices mitigate distortion caused by the power amplifier in the TX waveform by keeping the power level of the power amplifier below a certain level. While this reduction in the power level of the power amplifier of the transmitting device can reduce distortion of the TX waveform, it can also result in a less efficient use of the transmit power (e.g., a less powerful transmit signal than possible).

[0017] Various embodiments enable a neural network at a receiving device to recover TX waveform distortion information of a transmitting device and use the recovered TX waveform distortion information of the transmitting device to mitigate waveform distortion in a TX waveform received from the transmitting device. The presence of the TX waveform distortion information of the transmitting device at the receiving device may enable the transmitting device to efficiently utilize its available transmit power (e.g., by using a power amplifier at full power) because distortion caused by the power amplifier can be mitigated at the receiving device side.

[0018] In various embodiments, a transmitting device may acquire TX waveform distortion information of the transmitting device. In some embodiments, the TX waveform distortion information may be a two-dimensional map of distortion errors due to signal clipping of Orthogonal Frequency Division Multiplexing (OFDM) symbols within a slot for one or more antennas of the transmitting device. The TX waveform distortion information may be configured such that a receiving device may use the TX waveform distortion information of the transmitting device to mitigate waveform distortion of a TX waveform received from the transmitting device. For example, the TX waveform distortion information may be used by the receiving device to mitigate TX waveform distortion of multiple OFDM symbols. Because the data sent may affect the OFDM waveform, the TX waveform distortion information may depend on the data that may be sent on each slot. Therefore, provisioning of TX waveform distortion information per slot may enable a receiving device to mitigate TX waveform distortion per slot.

[0019] In various embodiments, the transmitting device may include a pair of an encoder neural network and a decoder neural network. In various embodiments, the encoder neural network and the decoder neural network may be deep neural networks. In various embodiments, there may be an encoder and decoder pair for each transmit antenna of the transmitting device. The encoder neural network may be configured to compress information. For example, the encoder neural network may be configured to compress TX waveform distortion information into compressed TX waveform distortion information, and the decoder neural network may be configured to decompress the information. For example, the decoder neural network may be configured to recover the TX waveform distortion information from the compressed TX waveform distortion information. In various embodiments, the receiving device may also include a decoder neural network.

[0020] In various embodiments, the transmitting device may train the encoder neural network of the transmitting device to compress the TX waveform distortion information of the transmitting device into compressed TX waveform distortion information of the transmitting device. In various embodiments, the transmitting device may train the decoder neural network of the transmitting device to recover the TX waveform distortion information of the transmitting device from the compressed TX waveform distortion information of the transmitting device. In various embodiments, the encoder neural network of the transmitting device and the decoder neural network of the transmitting device may be trained using an unsupervised learning algorithm. In various embodiments, training the encoder neural network of the transmitting device and training the decoder neural network of the transmitting device may include training the encoder neural network and the decoder neural network for one transmit antenna of the transmitting device. In various embodiments, training the encoder neural network of the transmitting device and training the decoder neural network of the transmitting device may be performed periodically. For example, training may be performed upon initial startup of the transmitting device, daily, upon registration with a new network, etc. In various embodiments, the encoder neural network of the transmitting device and the decoder neural network of the transmitting device may be trained using an unsupervised learning algorithm.

[0021] In various embodiments, the transmitting device may determine the model type of the transmitting device's trained decoder neural network and the weights of the transmitting device's trained decoder neural network.

[0022] The model type may be the structure of the neural network, such as a description of the type, number, and / or interconnections (e.g., layer layout) of nodes in the neural network. The model type may also be an actual representation of the neural network elements themselves and / or descriptors (e.g., model name, model number, model tag) that indicate the structure of the neural network, such as a description of the type, number, and / or interconnections (e.g., layer layout) of nodes in the neural network. In various embodiments, the model type of a decoder neural network of a transmitting device may provide information to a receiving device to reconstruct the particular structure of the decoder neural network trained at the transmitting device.

[0023] The weights of a neural network may be values associated with the interconnections between the nodes of the neural network after training of the neural network. The weights of a trained decoder neural network of a transmitting device may be values associated with the interconnections between the nodes of the decoder neural network after training at the transmitting device. In various embodiments, two decoder neural networks having the same model type may have the same structure such that applying the same weights to the two decoder neural networks provides the same decompressed output of the two decoder neural networks based on the same compressed input to each decoder neural network.

[0024] In various embodiments, a receiving device having the model type of the transmitting device's trained decoder neural network and the weights of the transmitting device's trained decoder neural network may configure the receiving device's decoder neural network to recover the same decompressed output as recovered by the transmitting device's trained decoder neural network from a common compressed input, without having to spend time actually training the receiving device's decoder neural network.

[0025] In various embodiments, the transmitting device may send the transmitting device's compressed TX waveform distortion information, the model type of the transmitting device's trained decoder neural network, and the weights of the transmitting device's trained decoder neural network to the receiving device. The transmitting device's compressed TX waveform distortion information, model type, and weights may be sent to the receiving device in a configuration that allows the receiving device to configure the receiving device's decoder neural network to recover the transmitting device's TX waveform distortion information from the transmitting device's compressed TX waveform distortion information using the model type and weights. In some embodiments, the transmitting device's compressed TX waveform distortion information, model type, and weights may be sent together. In some embodiments, the transmitting device's compressed TX waveform distortion information, model type, and weights may be sent separately. In some embodiments, the transmitting device's compressed TX waveform distortion information, model type, and weights may be sent in overhead information exchanged between the transmitting computing device and the receiving computing device. In some embodiments, the transmitting device's compressed TX waveform distortion information may be sent in control information for each slot to be transmitted. Because the transmitted data may affect the OFDM waveform, the compressed TX waveform distortion information may depend on the data that may be transmitted on each slot. By providing compressed TX waveform distortion information per slot, the receiving device may be able to mitigate TX waveform distortion per slot. In some embodiments, the compressed TX waveform distortion information, model type, and weights may be signaled with different periodicities. In some embodiments, the model type and weights may be signaled less frequently than the compressed TX waveform distortion information of the transmitting device. For example, the compressed TX waveform distortion information may be sent in the control information of each slot, and the weights may be sent with a greater periodicity than each slot (e.g., less frequently than the compressed TX waveform distortion information). The weights and / or model type may be sent on a much larger time scale than the compressed TX waveform distortion information because the encoder and decoder neural networks of the transmitting device may be updated infrequently.In some embodiments, it may not be necessary to send the model type, as the model type may already be known by the receiving device.

[0026] In various embodiments, the receiving device may receive the compressed TX waveform distortion information of the transmitting device, the model type of the trained decoder neural network of the transmitting device, and the weights of the trained decoder neural network of the transmitting device. In some embodiments, the compressed TX waveform distortion information, model type, and weights of the transmitting device may be sent in overhead information. In some embodiments, the compressed TX waveform distortion information of the transmitting device may be received in control information for each slot transmitted. In some embodiments, receiving the model type may not be necessary because the model type may already be known at the receiving device. As one example, a single default model type may be used for both the decoder neural network of the transmitting device and the decoder neural network of the receiving device. As another example, the model type may be known based on the device type, network, or other settings at the receiving device.

[0027] In some embodiments, the compressed TX waveform distortion information, model type, and weights may be received directly from a transmitting device. For example, the transmitting device may be a UE computing device, and the UE computing device may send the compressed TX waveform distortion information, model type, and weights to a base station as part of an initial registration procedure between the UE computing device and the base station to receive service in a cell served by the base station. The base station may be a receiving device, and may utilize the compressed TX waveform distortion information, model type, and weights to mitigate distortion of the TX waveform sent by the UE computing device.

[0028] In some embodiments, the compressed TX waveform distortion information, model type, and weights may be received from a base station other than the transmitting device. For example, the compressed TX waveform distortion information, model type, and weights may be values that are stored and shared among devices in a communication network.

[0029] In some embodiments, a base station's compressed TX waveform distortion information, model type, and / or weights may be shared between devices so that a device can indirectly receive another device's compressed TX waveform distortion information, model type, and weights. As an example, base stations of neighboring cells may share their model types and weights with each other and with UE computing devices within their respective cells to support UE computing device mobility and handoff. As another example, a UE computing device's model type and weights may be centrally stored so that a base station can retrieve a UE computing device's model type and weights upon discovery of the UE computing device without the UE computing device having to directly transmit the model type and weights to a base station, etc. As another example, a base station's compressed TX waveform distortion information, model type, and / or weights may be centrally stored so that a UE computing device can retrieve a next base station's compressed TX waveform distortion information, model type, and / or weights without the UE computing device having to directly receive the compressed TX waveform distortion information, model type, and / or weights from the next base station, etc., before entering that base station's coverage area.

[0030] In various embodiments, the receiving device may use the received model type and weights to configure its decoder neural network. The model type may be the structure of the neural network, such as a description of the type, number, and / or interconnections (e.g., layer layout) of nodes in the neural network. The model type may also be an actual representation of the neural network elements themselves and / or descriptors (e.g., model name, model number, model tag) that indicate the structure of the neural network, such as a description of the type, number, and / or interconnections (e.g., layer layout) of nodes in the neural network. The neural network weights may be values associated with the interconnections between the nodes of the neural network after training of the neural network. In various embodiments, a receiving device having the model type of the transmitting device's trained decoder neural network and the weights of the transmitting device's trained decoder neural network may configure the receiving device's decoder neural network to recover the same decompressed output as recovered by the transmitting device's trained decoder neural network from the transmitting device's compressed TX waveform distortion information received from the transmitting device. In this way, the receiving device can configure its decoder neural network as a trained neural network without having to spend time training the decoder neural network.

[0031] In various embodiments, the receiving device may recover the TX waveform distortion information of the transmitting device from the compressed TX waveform distortion information of the transmitting device using a configured decoder neural network of the receiving device. In some embodiments, the recovered TX waveform distortion information may include a two-dimensional map of distortion errors due to signal clipping of OFDM symbols within slots for one or more antennas of the transmitting device.

[0032] In various embodiments, the receiving device may use the transmitting device's recovered TX waveform distortion information to mitigate waveform distortion in a TX waveform received from the transmitting device. Mitigating waveform distortion may include using the recovered TX waveform distortion information to reconstruct the original TX waveform signal at the receiving device. By mitigating waveform distortion, the receiving device may compensate for any distortion caused by the transmitting device itself, such as distortion caused by the transmitting device's power amplifier.

[0033] The terms “wireless device” and “UE computing device” are used interchangeably herein to refer to any one or all of wireless router devices, wireless appliances, cellular telephones, smartphones, portable computing devices, personal or mobile multimedia players, laptop computers, tablet computers, smartbooks, ultrabooks, palmtop computers, wireless email receivers, multimedia Internet-enabled cellular telephones, medical devices and equipment, biometric sensors / devices, wearable devices such as smart watches, smart wear, smart glasses, smart wristbands, smart jewelry (e.g., smart rings, smart bracelets, etc.), entertainment devices (e.g., wireless game controllers, music and video players, satellite radio, etc.), smart meters / sensors, industrial manufacturing equipment, wireless network-enabled Internet of Things (IoT) devices including large and small machines and appliances for home and business use, wireless communication elements in autonomous and semi-autonomous vehicles, wireless devices fixed to or embedded in various mobile platforms, global positioning system devices, and similar electronic devices that include memory, wireless communication components, and programmable processors.

[0034] The term "system on a 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 include circuits for digital, analog, mixed-signal, and radio frequency functions. A single SOC may also include any number of general-purpose and / or special-purpose processors (such as 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.). A SOC may also include software for controlling the integrated resources and processors, as well as for controlling peripheral devices.

[0035] The term "system in package" (SIP) may be used herein to refer to a single module or package containing multiple resources, computing 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, a SIP may include one or more multi-chip modules (MCMs) on which multiple ICs or semiconductor dies are packaged in a unifying substrate. A SIP may also include multiple independent SOCs packaged in close proximity and coupled to each other via high-speed communication circuits, such as on a single motherboard or within a single wireless device. The proximity of the SOCs facilitates high-speed communication and sharing of memory and resources.

[0036] 1 is a system block diagram illustrating an exemplary communication system 100 suitable for implementing any of the various embodiments. The communication system 100 may be a 5G New Radio (NR) network or any other suitable network, such as a Long Term Evolution (LTE) network.

[0037] Communications system 100 may include a heterogeneous network architecture including a core network 140 and various mobile devices (depicted in FIG. 1 as wireless devices 120a-120e). Communications system 100 may also include several base stations (depicted as BS 110a, BS 110b, BS 110c, and BS 110d) and other network entities. A base station is an entity that communicates with wireless devices (mobile devices or UE computing devices) and may also be referred to as a Node B, Node B, LTE evolved Node B (eNB), access point (AP), radio head, transmit receive point (TRP), new radio base station (NR BS), 5G Node B (NB), next generation Node B (gNB), etc. Each base station may provide communication coverage for a particular geographic area. In 3GPP, the term “cell” can refer to a base station coverage area, a base station subsystem serving this coverage area, or a combination thereof, depending on the context in which the term is used.

[0038] Base stations 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 (e.g., a few kilometers in radius) and may allow unrestricted access by mobile devices with service subscriptions. A pico cell may cover a relatively small geographic area and may allow unrestricted access by mobile devices with service subscriptions. A femto cell may cover a relatively small geographic area (e.g., a home) and may allow restricted access by mobile devices that have an association with the femto cell (e.g., 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. 1, base station 110a may be a macro BS for macro cell 102a, base station 110b may be a pico BS for pico cell 102b, and base station 110c may be a femto BS for femto cell 102c. Base stations 110a-110d may support one or more (e.g., three) cells. The terms “eNB,” “base station,” “NR BS,” “gNB,” “TRP,” “AP,” “Node B,” “5G NB,” and “cell” may be used interchangeably herein.

[0039] In some examples, the cells may not be stationary and the geographic area of the cells may move according to the location of the mobile base station. In some examples, the base stations 110a-110d may be interconnected to each other and to one or more other base stations or network nodes (not shown) in the communication system 100 through various types of backhaul interfaces, such as direct physical connections, virtual networks, or combinations thereof, using any suitable transport network.

[0040] The base stations 110a-110d may communicate with the core network 140 over wired or wireless communication links 126. The wireless devices 120a-120e (UE computing devices) may communicate with the base stations 110a-110d over wireless communication links 122.

[0041] The wired communication link 126 may use various wired networks (such as Ethernet, TV cable, telephony, fiber optics, 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).

[0042] Communications system 100 may also include relay stations (such as relay BS 110d). A relay station is an entity that can receive data transmissions from an upstream station (e.g., a base station or a mobile device) and transmit the data to a downstream station (e.g., a wireless device or a base station). A relay station may also be a mobile device that can relay transmissions for other wireless devices. In the example shown in FIG. 1, relay station 110d may communicate with base station 110a and wireless device 120d to facilitate communication between base station 110a and wireless device 120d. A relay station may also be referred to as a relay base station, a repeater, etc.

[0043] Communications system 100 may be a heterogeneous network including different types of base stations, e.g., 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 may have different impacts on interference in communications system 100. For example, macro base stations may have high transmit power levels (e.g., 5-40 watts), while pico base stations, femto base stations, and relay base stations may have lower transmit power levels (e.g., 0.1-2 watts).

[0044] Network controller 130 may couple to a set of base stations and provide coordination and control for these base stations. Network controller 130 may communicate with the base stations via a backhaul. The base stations may also communicate with each other directly or indirectly, e.g., via wireless or wireline backhaul.

[0045] Wireless devices (UE computing devices) 120a, 120b, 120c may be dispersed throughout communication system 100, and each wireless device may be fixed or mobile. A wireless device may also be called an access terminal, terminal, mobile station, subscriber unit, station, etc.

[0046] The 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 the base stations 110a-110d over a wireless communication link 122.

[0047] The wireless communication links 122, 124 may include multiple carrier signals, frequencies, or frequency bands, each of which may include multiple logical channels. The wireless communication links 122, 124 may utilize one or more radio access technologies (RATs). Examples of RATs that may be used in the wireless communication links include 3GPP LTE, 3G, 4G, 5G (such as 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 technology 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).

[0048] Some 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. Generally, 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 depend on the system bandwidth. For example, the subcarrier spacing may be 15 kHz, and the minimum resource allocation (called a "resource block") may be 12 subcarriers (or 180 kHz). Thus, the nominal fast file transfer (FFT) size may be equal to 128, 256, 512, 1024, or 2048 for system bandwidths 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 bandwidths of 1.25, 2.5, 5, 10, or 20 MHz, respectively.

[0049] While the description of some embodiments may use terminology and examples related to LTE technology, various embodiments may be applicable to other wireless communication systems, such as New Radio (NR) or 5G networks. NR may utilize OFDM with cyclic prefix (CP) on the uplink (UL) and downlink (DL) and may include support for half-duplex operation using time division duplexing (TDD). A single component carrier bandwidth of 100 MHz may be supported. An NR resource block may span 12 subcarriers with a subcarrier bandwidth of 75 kHz over a duration of 0.1 milliseconds (ms). 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 contain DL / UL data as well as DL / UL control data. Beamforming may be supported, and the beam direction may be dynamically configured. Multiple-input multiple-output (MIMO) transmission with precoding may also be supported. MIMO configurations in the DL may support up to eight transmit antennas with multi-layer DL transmission of up to eight streams and up to two streams per wireless device. Multi-layer transmission with up to two streams per wireless device may be supported. Aggregation of multiple cells may be supported with up to eight serving cells. Alternatively, NR may support an air interface other than an OFDM-based air interface.

[0050] Some mobile devices may be considered machine-type communication (MTC) or evolved or extended machine-type communication (eMTC) mobile devices. MTC and eMTC mobile devices include, for example, a robot, a drone, a remote device, a sensor, a meter, a monitor, a location tag, etc. that may communicate with a base station, another device (e.g., a remote device), or some other entity. A wireless node may provide, for example, connectivity for or to a network (e.g., a wide area network such as the 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. Wireless devices 120a-120e may be included within a housing that houses components of the wireless device, such as a processor component, a memory component, similar components, or a combination thereof.

[0051] In general, any number of communication systems and any number of wireless networks may be deployed in a given geographic area. Each communication system and wireless network may support a particular radio access technology (RAT) and may operate on one or more frequencies. A RAT may also be referred to as a radio technology, air interface, etc. A frequency may also be referred to as a carrier, frequency channel, etc. Each frequency may support a single RAT in a given geographic area to avoid interference between communication systems of different RATs. In some cases, NR or 5G RAT networks may be deployed.

[0052] In some embodiments, two or more mobile devices 120a-e (e.g., shown as wireless device 120a and wireless device 120e) may communicate directly (e.g., without using base station 110 as an intermediary to communicate with each other) using one or more sidelink channels 124. For example, wireless devices 120a-120e may communicate using peer-to-peer (P2P) communication, device-to-device (D2D) communication, vehicle-to-everything (V2X) protocols (which may include vehicle-to-vehicle (V2V) protocols, vehicle-to-infrastructure (V2I) protocols, or similar protocols), mesh networks, or similar networks, or combinations thereof. In this case, wireless devices 120a-120e may perform scheduling operations, resource selection operations, and other operations described elsewhere herein as being performed by base station 110a.

[0053] 2 is a component block diagram illustrating an exemplary computing and wireless modem system 200 suitable for implementing any of the various embodiments. The various embodiments may be implemented on a number of single-processor and multi-processor computer systems, including systems-on-chips (SOCs) or systems-in-packages (SIPs).

[0054] 1 and 2 , an exemplary computing system 200 (which may be a SIP in some embodiments) is shown including two SOCs 202, 204 coupled to a clock 206, a voltage regulator 208, and one or more wireless transceivers 266 configured to transmit and receive wireless communications via one or more antennas 267 to and from a wireless device, such as a base station 110a. In some embodiments, the first SOC 202 operates as a central processing unit (CPU) of the wireless device, executing instructions of a software application program by performing arithmetic, logic, control, and input / output (I / O) operations specified by the instructions. In some embodiments, the second SOC 204 may operate as a dedicated processing unit. For example, the second SOC 204 may operate as a dedicated 5G processing unit responsible for managing high-capacity, high-speed (e.g., 5 Gbps, etc.), and / or very short-wavelength (e.g., 28 GHz mmWave spectrum, etc.) communications.

[0055] 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 (such as a vector coprocessor) connected to one or more of the processors, memory 220, custom circuitry 222, system components and resources 224, an interconnect / bus module 226, one or more temperature sensors 230, a thermal management unit 232, and thermal power envelope (TPE) components 234. The second SOC 204 may include a 5G modem processor 252, a power management unit 254, an interconnect / bus module 264, multiple mmWave transceivers 256, memory 258, and various additional processors 260, such as application processors, packet processors, etc. The multiple mmWave transceivers 256 may be connected to one or more antennas 268 and may be configured to send and receive wireless communications via the one or more antennas 268 to and from wireless devices, such as the base station 110a.

[0056] 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, a first SOC 202 may include a processor that runs a first type of operating system (e.g., FreeBSD, LINUX, OS X, etc.) and a processor that runs a second type of operating system (e.g., MICROSOFT WINDOWS 10). Additionally, 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.).

[0057] The first SOC 202 and the second SOC 204 may include various system components, resources, and custom circuitry for managing sensor data, analog-to-digital conversion, wireless data transmission, and 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 processors and software clients running on the wireless device. The system components and resources 224 and / or custom circuitry 222 may also include circuitry for interfacing with peripheral devices, such as cameras, electronic displays, wireless communication devices, external memory chips, etc.

[0058] The first SOC 202 and the second SOC 204 may communicate via an interconnect / 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, as well as a thermal management unit 232, via an interconnect / bus module 226. Similarly, the processor 252 may be interconnected to a power management unit 254, an mmWave transceiver 256, memory 258, and various additional processors 260 via an interconnect / bus module 264. The interconnect / bus modules 226, 250, 264 may include arrays of reconfigurable logic gates and / or implement a bus architecture (e.g., CoreConnect, AMBA, etc.). Communication may occur through advanced interconnects such as a high-performance network-on-chip (NoC).

[0059] The first SOC 202 and / or the second SOC 204 may further include input / output modules (not shown) for communicating with resources external to the SOC, such as a clock 206 and a voltage regulator 208. Resources external to the SOC (such as the clock 206, the voltage regulator 208) may be shared by two or more of the internal SOC processors / cores.

[0060] In addition to the exemplary SIP 200 described above, various embodiments may be implemented in a wide variety of computing systems, which may include a single processor, multiple processors, multi-core processors, or any combination thereof.

[0061] 3 is a software architecture diagram illustrating a software architecture 300 including radio protocol stacks for user and control planes in wireless communications suitable for implementing any of the various embodiments. With reference to FIGS. 1-3, a wireless device (UE computing device) 320 may implement the software architecture 300 to facilitate communications between the wireless device 320 (e.g., wireless devices 120a-120e, 200) and a base station 350 (e.g., base station 110a) of a communications system (e.g., 100). In various embodiments, layers in the software architecture 300 may form logical connections with corresponding layers in the software of the base station 350. The software architecture 300 may be distributed among one or more processors (e.g., processors 212, 214, 216, 218, 252, 260, etc.). Although illustrated with respect to one radio protocol stack in a multi-SIM (Subscriber Identity Module) wireless device, software architecture 300 may include multiple protocol stacks, each associated with a different SIM (such as two protocol stacks associated with two SIMs in a dual-SIM wireless communication device). Although described below with respect to an LTE communication layer, software architecture 300 may support any of a variety of standards and protocols for wireless communication and / or may include additional protocol stacks supporting any of a variety of standards and protocols for wireless communication.

[0062] 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 for packet filtering, security management, mobility control, session management, and supporting traffic and signaling between a wireless device's SIM (such as SIM 204) and its core network 140. The AS 304 may include functions and protocols that support communication between a SIM (such as SIM 204) and supported access network entities (such as base stations). In particular, the AS 304 may include at least three layers (Layer 1, Layer 2, and Layer 3), each of which may include various sublayers.

[0063] In the user and control plane, Layer 1 (L1) of the AS 304 may be the physical layer (PHY) 306, which may oversee functions that enable transmission and / or reception over the air interface. Examples of such physical layer 306 functions may include cyclic redundancy check (CRC) attachment, coding blocks, scrambling and descrambling, modulation and demodulation, signal measurement, MIMO, etc. The physical layer may include various logical channels, including a physical downlink control channel (PDCCH) and a physical downlink shared channel (PDSCH).

[0064] In the user and control plane, Layer 2 (L2) of the AS 304 may carry the link between the wireless device 320 and the base station 350 on the physical layer 306. In various embodiments, Layer 2 may include a medium access control (MAC) sublayer 308, a radio link control (RLC) sublayer 310, and a packet data convergence protocol (PDCP) sublayer 312, each of which forms a logical connection that terminates at the base station 350.

[0065] In the control plane, Layer 3 (L3) of the AS 304 may include a radio resource control (RRC) sublayer 3. Although 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 RRC signaling connections between the wireless device 320 and the base station 350.

[0066] In various embodiments, the PDCP sublayer 312 may provide uplink functions including multiplexing between different radio bearers and logical channels, sequence numbering, handover data processing, integrity protection, ciphering, and header compression. In the downlink, the PDCP sublayer 312 may provide functions including in-order delivery of data packets, duplicate data packet detection, integrity verification, decryption, and header recovery.

[0067] 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, 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.

[0068] In the uplink, the MAC sublayer 308 may provide functions including multiplexing between logical and transport channels, random access procedures, logical channel priorities, and hybrid ARQ (HARQ) operations. In the downlink, MAC layer functions may include channel mapping within a cell, demultiplexing, discontinuous reception (DRX), and HARQ operations.

[0069] While the software architecture 300 may provide functionality for transmitting data over a physical medium, the software architecture 300 may further include at least one host layer 314 for providing data transfer services to various applications in the wireless device 320. In some embodiments, the application-specific functionality provided by the at least one host layer 314 may provide an interface between the software architecture and the general-purpose processor 206.

[0070] In other embodiments, software architecture 300 may include one or more upper logical layers (e.g., transport, session, presentation, application, etc.) that provide host layer functionality. For example, in some embodiments, software architecture 300 may include a network layer (e.g., Internet Protocol (IP) layer) in which logical connections terminate at a packet data network (PDN) gateway (PGW). In some embodiments, software architecture 300 may include an application layer in which logical connections terminate at another device (e.g., an end-user device, a server, etc.). In some embodiments, software architecture 300 may further include a hardware interface 316 between physical layer 306 and communications hardware (e.g., one or more radio frequency (RF) transceivers) within AS 304.

[0071] FIG. 4 illustrates an exemplary neural network 400 that may be implemented in a computing device for implementing any of the various embodiments. Referring to FIGS. 1-4 , any device in a communication system (e.g., 100), such as a wireless device (UE computing device) (e.g., wireless device 120a-120e, 200, 320) and / or a base station (e.g., base station 110a, 350, etc.), may implement the neural network 400. The neural network 400 may be any purpose-built neural network, such as an encoder neural network, a decoder neural network, etc. As an example, the neural network 400 may be a feed-forward deep neural network. The neural network 400 may be distributed among one or more processors (e.g., processors 212, 214, 216, 218, 252, 260, etc.).

[0072] The neural network 400 may include an input layer 402, a hidden layer 404, and an output layer 406. Each of the layers 402, 404, 406 may include one or more processing nodes that receive input values, perform calculations based on the input values, and propagate results (activations) to the next layer. The structure of the neural network may be a description of the type, number, and / or interconnections of the nodes in the neural network 400 (e.g., the layer 402, 404, 406 layout). The model type may be an actual representation of the neural network elements themselves (e.g., the layers 402, 404, 406 and the nodes therein) and / or descriptors (e.g., model name, model number, model tag) that indicate the structure of the neural network 400, such as a description of the type, number, and / or interconnections of the nodes in the neural network 400 (e.g., the layer 402, 404, 406 layout). 4 may have an input layer 402 with a single input node X, an intermediate layer 404 with four nodes Y1, Y2, Y3, and Y4, and an output layer 406 with a single output node Z. Although shown with particular layers 402, 404, 406 and nodes X, Y1, Y2, Y3, Y4, and Z, neural network 400 may include additional layers, nodes, and / or interconnections therebetween, and neural network 400 may be any type of neural network 400.

[0073] In a feedforward neural network such as neural network 400, all computations are performed as a series of operations on the output of the previous layer. The final set of operations generates the output of the neural network, such as compressed information in an encoder neural network or decompressed information in a decoder neural network. The weights of neural network 400 may be values associated with the interconnections between nodes of neural network 400 after neural network 400 has been trained. For example, weight W11 shown in FIG. 4 is the value associated with the interconnections between the X node and Y nodes Y1, Y2, Y3, and Y4 of neural network 400 after neural network 400 has been trained. The final output of neural network 400 may correspond to a task that neural network 400 may perform, such as operating as an encoder to compress TX waveform distortion information or operating as a decoder to recover TX waveform distortion information from compressed TX waveform information.

[0074] In the neural network 400, learning may be achieved during a training process in which the values of the weights in each layer 402, 404, 406 are determined. After the training process is complete, the neural network 400 may successfully perform its intended purpose. For example, a trained encoder network may successfully compress information, such as successfully compressing TX waveform distortion information into compressed TX waveform distortion information. As another example, a trained decoder neural network may successfully decompress information, such as successfully decompressing compressed TX waveform distortion information into TX waveform distortion information (also referred to as recovered TX waveform distortion information).

[0075] Training the neural network 400 may involve having the neural network 400 perform a task for which the expected / desired output is known and comparing the output produced by the neural network 400 to the expected / desired output. Training may be supervised or unsupervised. During training, the weights of the neural network 400 may be updated until the output of the neural network 400 matches the expected / desired output. The weights of the trained decoder neural network of the transmitting device may be values associated with the interconnections between each node of the decoder neural network after training at the transmitting device. For example, the weights of the trained neural network 400 as shown in FIG. 4 may include a weight W11, which is a value associated with the interconnections between the X node and the Y nodes Y1, Y2, Y3, and Y4 of the neural network 400 after training of the neural network 400. For example, in the example shown in FIG. 4, Y may be related to X by the equation Y=W*X, where Y=[Y, Y, Y, Y], and W is a 1×4 matrix such that W is a weight; similarly, Z may be related to Y by other weights.

[0076] Another instance of the same neural network 400 may be created by providing the model of neural network 400 and the weights of the trained neural network 400. In various embodiments, two decoder neural networks having the same model type may have the same structure such that applying the same weights to the two decoder neural networks provides each decoder neural network with the same output (e.g., when neural network 400 is a decoder neural network, the same decompressed output) of the two neural networks based on the same input (e.g., the same compressed input, such as compressed TX waveform distortion information). In this way, a device having the model type of trained neural network 400 and the weights of trained neural network 400 may configure a second neural network to correspond to neural network 400 (e.g., configure the second neural network to be a copy of trained neural network 400) to produce the same output as that produced by trained neural network 400 for the same input, without having to spend time actually training the second neural network.

[0077] 5 is a process flow diagram of an example wireless communication method 500 performed by a processor of a transmitting device that transmits a waveform to a receiving device, according to various embodiments. Referring to FIGS. 1-5 , method 500 may be implemented by a processor (e.g., 212, 216, 252, or 260) of a transmitting device, such as a wireless device (UE computing device) (e.g., wireless device 120a-120e, 200, 320), and / or a base station (e.g., base station 110a, 350, etc.) implementing one or more neural networks (e.g., neural network 400). As an example, in UL communication in a 3GPP network, the transmitting device may be a UE computing device and the receiving device may be a base station (e.g., a gNB). As another example, in DL communication in a 3GPP network, the transmitting device may be a base station (e.g., a gNB) and the receiving device may be a UE computing device.

[0078] In block 502, the processor of the transmitting device may perform operations including obtaining TX waveform distortion information of the transmitting device. In some embodiments, the TX waveform distortion information may be a two-dimensional map of distortion errors due to signal clipping of Orthogonal Frequency Division Multiplexing (OFDM) symbols within a slot for one or more antennas of the transmitting device. The TX waveform distortion information may be configured such that the receiving device may use the TX waveform distortion information of the transmitting device to mitigate waveform distortion of a TX waveform received from the transmitting device. For example, the TX waveform distortion information may be used by the receiving device to mitigate TX waveform distortion of multiple OFDM symbols.

[0079] At block 504, the processor of the transmitting device may perform operations including training an encoder neural network to compress the TX waveform distortion information of the transmitting device into compressed TX waveform distortion information of the transmitting device.

[0080] At block 506, the processor of the transmitting device may perform operations including training a decoder neural network to recover the TX waveform distortion information of the transmitting device from the compressed TX waveform distortion information of the transmitting device.

[0081] In various embodiments, the transmitting device may include a pair of an encoder neural network and a decoder neural network. In some embodiments, the encoder neural network and the decoder neural network may be deep neural networks. In some embodiments, there may be an encoder and decoder pair for each transmit antenna of the transmitting device. The encoder neural network may be configured to compress information. For example, the encoder neural network may be configured to compress TX waveform distortion information into compressed TX waveform distortion information. The decoder neural network may be configured to decompress the information. For example, the decoder neural network may be configured to recover TX waveform distortion information from the compressed TX waveform distortion information.

[0082] In some embodiments, the encoder neural network of the transmitting device and the decoder neural network of the transmitting device may be trained using an unsupervised learning algorithm. In some embodiments, training the encoder neural network of the transmitting device and training the decoder neural network of the transmitting device may include training the encoder neural network and the decoder neural network for one transmit antenna of the transmitting device. In some embodiments, training the encoder neural network of the transmitting device and training the decoder neural network of the transmitting device may occur periodically. For example, training may occur upon initial startup of the transmitting device, daily, upon registration with a new network, etc. In some embodiments, the encoder neural network of the transmitting device and the decoder neural network of the transmitting device may be trained using an unsupervised learning algorithm.

[0083] At block 508, the processor of the transmitting device may perform operations including determining a model type of the trained decoder neural network of the transmitting device and one or more weights of the trained decoder neural network of the transmitting device. The model type may be the structure of the neural network, such as a description of the type, number, and / or interconnections (e.g., layer layout) of nodes in the neural network. The model type may also be an actual representation of the neural network elements themselves and / or descriptors (e.g., model name, model number, model tag) indicating the structure of the neural network, such as a description of the type, number, and / or interconnections (e.g., layer layout) of nodes in the neural network. In some embodiments, the model type of the decoder neural network of the transmitting device may provide information to the receiving device to reconstruct the specific structure of the decoder neural network trained at the transmitting device. The weights of the neural network may be values associated with the interconnections between nodes of the neural network after training of the neural network. The weights of the trained decoder neural network of the transmitting device may be values associated with the interconnections between nodes of the decoder neural network after training at the transmitting device. In some embodiments, two decoder neural networks having the same model type may have the same structure such that applying the same weights to the two decoder neural networks will provide each decoder neural network with the same decompressed output of the two decoder neural networks based on the same compressed input. In this way, a receiving device having the model type of the trained decoder neural network of the transmitting device and the weights of the trained decoder neural network of the transmitting device may configure the decoder neural network of the receiving device to recover the same decompressed output as recovered by the trained decoder neural network of the transmitting device from a common compressed input, without having to spend time actually training the decoder neural network of the receiving device.

[0084] At block 509, the processor of the transmitting device may perform operations including compressing the transmit waveform distortion information of the transmitting device into compressed transmit waveform distortion information using an encoder neural network. In some embodiments, the compression of the transmit waveform distortion information of the transmitting device into compressed transmit waveform distortion information may be performed after the encoder neural network has been trained to achieve a selected compression level of the transmit waveform distortion information. For example, the trained encoder neural network may compress the obtained TX waveform distortion information to a selected compression level for sending to the receiving device.

[0085] At block 510, the processor of the transmitting device may perform operations including sending the transmitting device's compressed TX waveform distortion information, one or more weights, and / or model type to the receiving device in a configuration that allows the receiving device to configure a decoder neural network of the receiving device to recover the transmitting device's TX waveform distortion information from the compressed TX waveform distortion information using the one or more weights and / or model type. The transmitting device's compressed TX waveform distortion information, model type, and weights may be sent to the receiving device in a configuration that allows the receiving device to configure a decoder neural network of the receiving device to recover the transmitting device's TX waveform distortion information from the transmitting device's compressed TX waveform distortion information using the model type and weights. In some embodiments, the transmitting device's compressed TX waveform distortion information, model type, and weights may be sent together. In some embodiments, the transmitting device's compressed TX waveform distortion information, model type, and weights may be sent separately. In some embodiments, the transmitting device's compressed TX waveform distortion information, model type, and weights may be sent in overhead information exchanged between the transmitting computing device and the receiving computing device. In some embodiments, the transmitting device's compressed TX waveform distortion information may be sent in control information for each slot to be transmitted. Because the data being sent may affect the OFDM waveform, the compressed TX waveform distortion information may depend on the data that may be sent on each slot. By providing compressed TX waveform distortion information per slot, the receiving device may be able to mitigate TX waveform distortion per slot. In some embodiments, the compressed TX waveform distortion information, model type, and weights may be signaled with different periodicities. In some embodiments, the model type and weights may be signaled less frequently than the compressed TX waveform distortion information of the transmitting device. For example, the compressed TX waveform distortion information may be sent in the control information of each slot, and the weights may be sent with greater periodicity than each slot (e.g., less frequently than the compressed TX waveform distortion information).The weights and / or model types may be sent on a much larger time scale than the compressed TX waveform distortion information, since the encoder and decoder neural networks of the transmitting device may be updated infrequently.

[0086] FIG. 6 is a process flow diagram of an example method 600 of wireless communication performed by a processor of a receiving device that receives a waveform from a transmitting device, in accordance with various embodiments. With reference to FIGS. 1-6 , method 600 may be implemented by a processor (e.g., 212, 216, 252, or 260) of a receiving device, such as a wireless device (UE computing device) (e.g., wireless device 120a-120e, 200, 320), and / or a base station (e.g., base station 110a, 350, etc.) implementing one or more neural networks (e.g., neural network 400). The operations of method 600 may be performed in conjunction with the operations of method 500 (FIG. 5). As an example, in UL communications in a 3GPP network, the receiving device may be a base station (e.g., a gNB), and the transmitting device may be a UE computing device. As another example, in DL communications in a 3GPP network, the receiving device may be a UE computing device, and the transmitting device may be a base station (e.g., a gNB).

[0087] At block 602, the processor of the receiving device may perform operations including receiving compressed TX waveform distortion information of the transmitting device, one or more weights of a trained decoder neural network of the transmitting device, and / or a model type of the trained decoder neural network of the transmitting device. For example, the compressed TX waveform distortion information, model type, and weights may be the compressed TX waveform distortion information, model type, and weights of a transmitting device performing the operations of method 500 of FIG. 5. In some embodiments, the compressed TX waveform distortion information, model type, and weights of the transmitting device may be received in overhead information. In some embodiments, the compressed TX waveform distortion information of the transmitting device may be received in control information for each slot to be transmitted.

[0088] In some embodiments, the model type and weights may be received directly from the transmitting device. For example, the transmitting device may be a UE computing device, and the UE computing device may send the model type and weights to the base station as part of an initial registration procedure between the UE computing device and the base station to receive service in a cell served by the base station. The base station may be a receiving device, and may utilize the model type and weights to mitigate distortion of the TX waveform sent by the UE computing device. In some embodiments, the model type and weights may be received from a base station other than the transmitting device. For example, the model type and weights may be values stored and shared among devices in a communication network. The model type and weights of a base station may be shared between devices so that a device can indirectly receive the model type and weights of another device. As an example, base stations of neighboring cells may share their model types and weights with each other and with UE computing devices in their respective cells to support UE computing device mobility and handoff. As another example, the UE computing model type and weights may be stored centrally so that a base station can retrieve the model type and weights of a UE computing device upon discovery of the UE computing device without the UE computing device having to transmit the model type and weights directly to a base station or the like.

[0089] In some embodiments, a base station's compressed TX waveform distortion information, along with the model type and / or weight, may be centrally stored so that a UE computing device can retrieve a next base station's compressed TX waveform distortion information, model type, and / or weight before entering that base station's coverage area without the UE computing device having to receive the compressed TX waveform distortion information, model type, and / or weight directly from the next base station, etc. Storing compressed TX waveform distortion information may be useful in situations where a network-side transmitter sends the same data multiple times (e.g., a service description sent periodically) and the UE computing device is pre-set up to receive the same data at the next cell using compressed TX waveform distortion information from the next cell.

[0090] At block 604, the processor of the receiving device may perform operations including configuring a decoder neural network of the receiving device using the received one or more weights and / or the received model type. The model type may be the structure of the neural network, such as a description of the type, number, and / or interconnections (e.g., layer layout) of nodes in the neural network. The model type may also be an actual representation of the neural network elements themselves and / or descriptors (e.g., model name, model number, model tag) indicating the structure of the neural network, such as a description of the type, number, and / or interconnections (e.g., layer layout) of nodes in the neural network. The weights of the neural network may be values associated with the interconnections between nodes of the neural network after training of the neural network. In some embodiments, a receiving device having the model type of the transmitting device's trained decoder neural network and the weights of the transmitting device's trained decoder neural network may configure the receiving device's decoder neural network to recover the same decompressed output as recovered by the transmitting device's trained decoder neural network from the transmitting device's compressed TX waveform distortion information received from the transmitting device. In this way, the receiving device can configure the decoder neural network as a trained neural network without having to spend time actually training the decoder neural network.

[0091] At block 606, the processor of the receiving device may perform operations including recovering TX waveform distortion information of the transmitting device from the compressed TX waveform distortion information using the configured decoder neural network of the receiving device. In some embodiments, the TX waveform distortion information may include a two-dimensional map of distortion errors due to signal clipping of OFDM symbols within a slot for one or more antennas of the transmitting device. The receiving device may recover the same two-dimensional map of distortion originally created at the transmitting device.

[0092] At block 608, the processor of the receiving device may perform operations including using the TX waveform distortion information of the transmitting device to mitigate waveform distortion in a TX waveform received from the transmitting device. Mitigating waveform distortion may include using the TX waveform distortion information to reconstruct the original TX waveform signal at the receiving device. In this manner, by mitigating waveform distortion, the receiving device may compensate for any distortion caused by the transmitting device itself, such as distortion caused by a power amplifier of the transmitting device.

[0093] 7A illustrates an example interaction between a receiving device and a transmitting device that performs the operations of methods 500 and 600 to provide distortion information from the transmitting device to the receiving device utilizing a neural network. With reference to FIGS. 1-7A, the illustrated and discussed interaction may be implemented by a processor (e.g., 212, 216, 252, or 260) of a transmitting device 750 of a wireless device (UE computing device) (e.g., wireless device 120a-120e, 200, 320, etc.) and / or base station (e.g., base station 110a, 350) that implements one or more neural networks (e.g., neural network 400) transmitting a TX waveform to a receiving device 752 of a wireless device (UE computing device) (e.g., wireless device 120a-120e, 200, 320, etc.) and / or base station (e.g., base station 110a, 350) that implements one or more neural networks (e.g., neural network 400).

[0094] A transmitting device 750 may obtain a two-dimensional distortion map 700 of distortion errors due to signal clipping for each transmit antenna of the transmitting device 750. The two-dimensional distortion map 700 is shown as an error map for a single transmit antenna for ease of illustration. Each cell in the grid represents a time-domain sample in an OFDM symbol, and the shaded blocks 702, 704, 706, 708, and 710 represent different levels of transmit waveform distortion, typically expressed as complex numbers.

[0095] The transmitting device 750 may input the two-dimensional distortion map 700 to an encoder neural network 714 (e.g., neural network 400) of the transmitting device 750. The output of the encoder neural network 714 may be compressed TX waveform distortion information 716. The compressed TX waveform distortion information 716 may be output by the transmitting device 750 to a decoder neural network 718 (e.g., neural network 400) of the transmitting device 750 as an input to the decoder neural network 718.

[0096] The transmitting device 750 may train its encoder neural network 714 and decoder neural network 718 until a selected compression level is achieved by the encoder neural network 714 and a two-dimensional distortion map 700 is correctly output by the decoder neural network 718. For example, with output compressed TX waveform distortion information 716, being the compressed TX waveform distortion map 700 provided as input to the decoder neural network 718 of the transmitting device 750, the decoder neural network 718 of the transmitting device 750 may be considered trained when it causes the decoder neural network 718 to output a correct copy of the original distortion map 700.

[0097] In response to the decoder neural network 718 of the transmitting device 750 being trained, the transmitting device 750 may determine a model type of the trained decoder neural network 718 of the transmitting device 750 and weights of the trained decoder neural network 718 of the transmitting device 750. Once the encoder neural network 714 and the decoder neural network 718 are trained, the transmitting device 750 may send the compressed TX waveform distortion information 716 and the model type and weights of the trained decoder neural network 718 to the receiving device 752.

[0098] The receiving device 752 may use the received model type and weights to configure its decoder neural network 720. In this way, the decoder neural network 720 may effectively be configured to be a copy of the trained decoder neural network 718 without having to actually train the decoder neural network 720.

[0099] The receiving device 752 may input the compressed TX waveform distortion information 716 into its decoder neural network 720, and the output may be the original distortion map 700. The receiving device 752 may use the distortion map 700 to mitigate TX waveform distortion in the transmit waveform received from the transmitting device 750.

[0100] FIG. 7B illustrates an additional example interaction of the receiving device 752 and transmitting device 750 of FIG. 7A performing the operations of methods 500 and 600 to provide distortion information from the transmitting device 750 to the receiving device 752 utilizing a neural network. 1-7B, the interactions shown and discussed may be implemented by a processor (e.g., 212, 216, 252, or 260) of a transmitting device 750 of a wireless device (UE computing device) (e.g., wireless device 120a-120e, 200, 320) and / or base station (e.g., base station 110a, 350, etc.) implementing one or more neural networks (e.g., neural network 400) transmitting a TX waveform to a receiving device 752 of a wireless device (UE computing device) (e.g., wireless device 120a-120e, 200, 320) and / or base station (e.g., base station 110a, 350, etc.) implementing one or more neural networks (e.g., neural network 400).

[0101] FIG. 7B shows that the OFDM symbols 764 may be output as a digital representation of a transmit waveform (x) that is passed to a digital-to-analog converter (D2A) 751 and a power amplifier 753 for transmission in a TX waveform 763 via an antenna 760 of the transmitting device 750. The TX waveform 763 may be an analog signal sent over the air to the receiving device 752. The operation of transmitting the TX waveform 763, such as power amplification, may cause signal distortion that may be measured by tapping the signal output to the antenna 760 and passing that output to an analog-to-digital converter (A2D) 754. The output of the A2D 754 may be a digital representation of a distorted TX waveform (y). The distortion present in the distorted TX waveform (y) may be a function of distortion caused and / or experienced in the transmit chain of the transmitting device 750. The digital representation of the transmit waveform (x) and the digital representation of the distorted TX waveform (y) may be passed to a comparator 756. The comparator 756 may determine a difference between the two waveforms, such as yx, which may be TX waveform distortion information of the transmitting device 750. The TX waveform distortion information (e.g., yx) of the transmitting device 750 may be provided to the encoder neural network 714 of the transmitting device 750. The encoder neural network 714 may compress the TX waveform distortion information of the transmitting device 750 into compressed TX waveform distortion information 716. Additionally, the TX waveform distortion information (e.g., yx) of the transmitting device 750 may be provided to a training module 780 configured to control training of the encoder neural network 714 and the decoder neural network 718. For example, the training module 780 may apply one or more loss functions used to train the encoder neural network 714 and / or the decoder neural network 718, such as a mean squared error (MSE) loss function.

[0102] In a training mode of operation, the compressed TX waveform distortion information 716 may be passed to a decoder neural network 718 of the transmitting device 750, which may recover TX waveform distortion information (e.g., yx) from the compressed TX waveform distortion information 716. The output of the decoder neural network 718 may be output to a training module 780 and compared with the TX waveform distortion information input to the encoder neural network 714. The training module 780 of the transmitting device 750 may train its encoder neural network 714 and decoder neural network 718 until a selected compression level is achieved by the encoder neural network 714 and the TX waveform distortion information (e.g., yx) of the transmitting device 750 is correctly output by the decoder neural network 718. The training may include applying a loss function, such as an MSE loss, between the input to the encoder neural network 714 and the output of the decoder neural network 718. For example, the decoder neural network 718 of the transmitting device 750 may be considered trained when the output compressed TX waveform distortion information 716, which is the compressed TX waveform distortion information 716 provided as input to the decoder neural network 718 of the transmitting device 750, causes the decoder neural network 718 to output a correct copy of the original TX waveform distortion information of the transmitting device 750 (e.g., a correct copy of yx).

[0103] In response to the decoder neural network 718 of the transmitting device 750 being trained, the transmitting device 750 may determine a model type of the trained decoder neural network 718 of the transmitting device 750 and / or one or more weights of the trained decoder neural network 718 of the transmitting device 750. Once the encoder neural network 714 and the decoder neural network 718 are trained, the transmitting device 750 may send the compressed TX waveform distortion information 716, the model type of the trained decoder neural network 718, and / or the one or more weights of the trained decoder neural network 718 to the receiving device 752. The compressed TX waveform distortion information 716, the model type of the trained decoder neural network 718, and / or the one or more weights of the trained decoder neural network 718 may be sent in various manners, such as via overhead signaling, out-of-band signaling, etc.

[0104] The receiving device 752 may use the received weight(s) and / or the received model type to configure its decoder neural network 720. In this way, the decoder neural network 720 may effectively be configured to be a copy of the trained decoder neural network 718 without having to actually train the decoder neural network 720.

[0105] The receiving device 752 may input the compressed TX waveform distortion information 716 to its decoder neural network 720, and the output may be the original TX waveform distortion information of the transmitting device 750 (e.g., a recovered copy of yx). The receiving device 752 may pass this TX waveform distortion information of the transmitting device 750 (e.g., a recovered copy of yx) to a receiver 757 to mitigate TX waveform distortion in a transmit waveform 763 received from the transmitting device 750 via an antenna 761 of the receiving device 752. The receiver 757 of the receiving device 752 may use the recovered TX waveform distortion information of the transmitting device 750 (e.g., a recovered copy of yx) to mitigate the waveform distortion and receive the OFDM symbols 764 transmitted on the transmit waveform 763.

[0106] Various embodiments may be implemented on various wireless network devices, an example of which is shown in FIG. 8 in the form of a wireless network computing device 800 that functions as a network element of a communications network, such as a base station. Such a network computing device may include at least the components shown in FIG. 8. Referring to FIGS. 1-8, the network computing device 800 may typically include a processor 801 coupled to volatile memory 802 and mass non-volatile memory, such as a disk drive 803. The network computing device 800 may also include a peripheral memory access device, such as a floppy disk drive, compact disk (CD), or digital video disk (DVD) drive 806, coupled to the processor 801. The network computing device 800 may also include a network access port 804 (or interface) coupled to the processor 801 for establishing a data connection with a network, such as the Internet and / or a local area network coupled to other system computers and servers. The network computing device 800 may be coupled to one or more antennas for sending and receiving electromagnetic radiation to establish a wireless communications link. The network computing device 800 may include additional access ports, such as USB, Firewire, Thunderbolt, etc., for coupling to peripherals, external memory, or other devices.

[0107] Various embodiments may be implemented on various wireless devices (e.g., wireless devices 120a-120e, 200, 320), an example of which is shown in FIG. 9 in the form of a smartphone 900. The smartphone 900 may include a first SOC 202 (e.g., a SOC-CPU) coupled to a second SOC 204 (e.g., a 5G-enabled SOC). The first SOC 202 and the second SOC 204 may be coupled to internal memory 906, 916, a display 912, and a speaker 914. Additionally, the smartphone 900 may include an antenna 904 for sending and receiving electromagnetic radiation, which may be connected to a wireless data link and / or a cellular telephone transceiver 908, coupled to one or more processors in the first SOC 202 and / or second SOC 204. The smartphone 900 also typically includes menu selection buttons or rocker switches 920 for receiving user input.

[0108] The typical smartphone 900 also includes a voice encoding / decoding (codec) circuit 910 that digitizes sound received from the microphone into data packets suitable for wireless transmission and decodes the received sound data packets to generate analog signals that are provided to a speaker to generate sound. One or more of the processors, wireless transceiver 908, and codec 910 in the first SOC 202 and second SOC 204 may also include digital signal processor (DSP) circuitry (not separately shown).

[0109] The processors of the wireless network computing device 800 and the smartphone 900 may be any programmable microprocessor, microcomputer, or one or more multi-processor chips that can be configured by software instructions (applications) to perform various functions, including those of the various embodiments described below. In some mobile devices, multiple processors may be provided, such as one processor in the SOC 204 dedicated to wireless communication functions and one processor in the SOC 202 dedicated to running other applications. Typically, software applications may be stored in memory 906, 916 before being accessed and loaded into the processor. The processors may include sufficient internal memory to store application software instructions.

[0110] As used herein, terms such as “component,” “module,” and “system” are intended to include, but are not limited to, computer-related entities, such as hardware, firmware, a combination of hardware and software, software, or software in execution, 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 example, 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 reside on one processor or core and / or be distributed among 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 via 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 methods.

[0111] Several different cellular and mobile communication services and standards are available or are contemplated in the future, all of which may implement and benefit from various embodiments, such as 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 (such as cdmaOne, CDMA1020™), 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), and others. Telecommunications standards include, for example, Wi-Fi (Wireless Telecommunications), 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. Any reference to terminology and / or technical details regarding particular telecommunications standards or technologies is for illustrative purposes only and is not intended to limit the scope of the claims to any particular communications system or technology unless specifically recited in the claim language.

[0112] The various embodiments shown 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 with or combined with other embodiments shown and described. Furthermore, the claims are not limited by any single exemplary embodiment. For example, one or more of the operations of methods 500 and / or 600 may be substituted for or combined with one or more operations of methods 500 and / or 600.

[0113] The above-described method descriptions and process flow diagrams are provided as illustrative examples only and do not require or imply that the operations of the various embodiments must be performed in the order presented. As will be appreciated by one of ordinary skill in the art, the order of operations in the above-described embodiments may be performed in any order. Terms such as "then," "then," and "next" do not limit the order of operations; these terms are used to guide the reader through the method descriptions. Additionally, any reference to claim elements in the singular, for example, using the articles "a," "an," or "the," should not be construed as limiting the element to the singular.

[0114] The various illustrative logical blocks, modules, components, circuits, and algorithmic operations described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or a combination of both. To clearly illustrate this interchangeability of hardware and software, the 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 on the particular application and design constraints imposed on the overall system. Those skilled in the art may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the claims.

[0115] The hardware used to implement the various exemplary logic, logic blocks, modules, and circuits described in connection with the embodiments disclosed herein may be implemented or performed using general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, 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 alternatively, 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, multiple 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 specific to a given function.

[0116] 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 or 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. A non-transitory computer-readable or processor-readable storage medium may be any storage medium that can be accessed by a computer or processor. By way of example and not limitation, such non-transitory computer-readable or processor-readable storage medium 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. As used herein, disk and disc include compact discs (CDs), laser discs, optical discs, digital versatile discs (DVDs), floppy disks, and Blu-ray discs, where disks typically reproduce data magnetically and discs reproduce data optically using lasers. Combinations of the above are also included within the scope of non-transitory computer-readable medium and non-transitory processor-readable medium. Furthermore, the operations of a method or algorithm may reside as one or any combination or set of code 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.

[0117] The previous 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 general 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. [Explanation of symbols]

[0118] 100 Communication Systems 102a Macrocell 102b Picocell 102c Femtocell 110a~110d Base station 120a~120e Wireless Devices 122 Wireless Communication Links 124 Wireless Communication Links 126 Communication Links 130 Network Controller 140 Core Network 200 Computing Systems 200 Wireless Modem System 202 First SOC 204 Second SOC 206 Clock 208 Voltage Regulator 210 Digital Signal Processor (DSP) 212 modem processor 214 graphics processor 216 Application Processors 218 Coprocessor 220 memory 222 Custom Circuits 224 System Components and Resources 226 Interconnect / Bus Modules 230 Temperature Sensor 232 Thermal Management Unit 234 Thermal Power Envelope Components 250 Interconnect / Bus Modules 252 5G modem processor 254 Power Management Unit 256 mmWave transceivers 258 memory 260 additional processors 264 Interconnect / Bus Module 266 Wireless Transceiver 268 Antenna 267 Antenna 268 Antenna 300 Software Architecture 302 Non-Access Layer (NAS) 304 Access Layer (AS) 306 Physical Layer (PHY) 308 Medium Access Control (MAC) Sublayer 310 Radio Link Control (RLC) Sublayer 312 Packet Data Convergence Protocol (PDCP) Sublayer 313 Radio Resource Control (RRC) sublayer 314 Host Layer 316 Hardware Interface 320 Wireless Devices 350 base station 400 Neural Networks 402 Input Layer 404 Intermediate Layer 406 Output Layer 700 2D distortion maps 714 Encoder Neural Network 716 Compressed TX Waveform Distortion Information 718 Decoder Neural Network 720 Decoder Neural Network 750 sending device 751 Digital / Analog Converter (D2A) 752 receiving devices 753 Power Amplifier 754 Analog-to-Digital Converter (A2D) 756 Comparator 757 receiver 760 Antenna 761 Antenna 763 TX waveform 764 OFDM symbols 780 Training Module 800 Network Computing Devices 801 processor 802 Volatile Memory 803 disk drive 804 Network Access Port 806 Digital Video Disc (DVD) Drive 900 smartphones 904 Antenna 906 Internal Memory 908 Cellular Telephone Walkie-Talkie 910 Voice coding / decoding (codec) circuit 912 Display 914 Speaker 916 Internal Memory 920 Rocker Switch

Claims

1. 1. A method of wireless communication performed by a processor of a transmitting device that transmits a waveform to a receiving device, comprising: acquiring transmission waveform distortion information of the transmitting device; compressing the transmit waveform distortion information of the transmitting device into compressed transmit waveform distortion information using an encoder neural network, the encoder neural network being trained to compress the transmit waveform distortion information of the transmitting device into the compressed transmit waveform distortion information; sending the compressed transmit waveform distortion information and the one or more decoder neural network weights to the receiving device, such that the receiving device can configure a decoder neural network of the receiving device to recover the transmit waveform distortion information of the transmitting device from the compressed transmit waveform distortion information using one or more decoder neural network weights, wherein the decoder neural network has been trained to recover the transmit waveform distortion information of the transmitting device from the compressed transmit waveform distortion information, and the one or more decoder neural network weights are weights of the trained decoder neural network of the transmitting device; A method comprising:

2. determining a model type of a decoder neural network of the transmitting device; and wherein sending the compressed transmit waveform distortion information and the one or more decoder neural network weights to the receiving device comprises sending the compressed transmit waveform distortion information, the one or more decoder neural network weights, and the model type to the receiving device. The method of claim 1.

3. 2. The method of claim 1, wherein sending the compressed transmit waveform distortion information and the one or more decoder neural network weights to the receiving device comprises sending the compressed transmit waveform distortion information and the one or more decoder neural network weights to the receiving device in control information for each slot transmitted.

4. the transmitting device is a user equipment (UE) computing device and the receiving device is a base station; or The method of claim 1 , wherein the transmitting device is a base station and the receiving device is a user equipment (UE) computing device.

5. the encoder neural network and the decoder neural network of the transmitting device are trained using an unsupervised learning algorithm; or training the encoder neural network and training the decoder neural network of the transmitting device includes training the encoder neural network and the decoder neural network of the transmitting device for one transmit antenna of the transmitting device; or 2. The method of claim 1 , wherein training the encoder neural network and training the decoder neural network of the transmitting device comprises training the encoder neural network and the decoder neural network of the transmitting device for each transmit antenna of the transmitting device.

6. 1. A method of wireless communication performed by a processor of a receiving device configured to receive a waveform from a transmitting device, the method comprising: receiving compressed transmit waveform distortion information of the transmitting device and one or more weights of a trained decoder neural network of the transmitting device; configuring a decoder neural network of the receiving device using the received one or more weights; recovering transmit waveform distortion information of the transmitting device from the compressed transmit waveform distortion information of the transmitting device using the configured decoder neural network of the receiving device; A method comprising:

7. using the recovered transmit waveform distortion information of the transmitting device to mitigate waveform distortion in a transmit waveform received from the transmitting device.

7. The method of claim 6, further comprising:

8. receiving the compressed transmit waveform distortion information of the transmitting device and the one or more weights of the trained decoder neural network of the transmitting device comprises receiving the compressed transmit waveform distortion information of the transmitting device, the one or more weights of the trained decoder neural network of the transmitting device, and a model type of the trained decoder neural network of the transmitting device; configuring the decoder neural network of the receiving device using the received one or more weights includes configuring the decoder neural network of the receiving device using the received one or more weights and the received model type. The method of claim 6.

9. 7. The method of claim 6, wherein the recovered transmit waveform distortion information comprises a two-dimensional map of distortion errors due to signal clipping of Orthogonal Frequency Division Multiplexing (OFDM) symbols within slots for one or more antennas of the transmitting device.

10. 7. The method of claim 6, wherein receiving the compressed transmit waveform distortion information of the transmitting device and one or more weights of a trained decoder neural network of the transmitting device comprises receiving the compressed transmit waveform distortion information of the transmitting device and one or more weights of a trained decoder neural network of the transmitting device in control information for each slot to be transmitted.

11. The method of claim 6 , wherein the receiving device is a base station and the transmitting device is a user equipment (UE) computing device.

12. The method of claim 11 , wherein the compressed transmit waveform distortion information and the one or more weights are received directly from the UE computing device.

13. The method of claim 6 , wherein the compressed transmit waveform distortion information and the one or more weights are received from a base station other than the transmitting device.

14. 1. A transmitting device, comprising: a processor configured with processor-executable instructions for performing operations, said operations comprising: acquiring transmission waveform distortion information of the transmitting device; compressing the transmit waveform distortion information of the transmitting device into compressed transmit waveform distortion information using an encoder neural network, the encoder neural network being trained to compress the transmit waveform distortion information of the transmitting device into the compressed transmit waveform distortion information; sending the compressed transmit waveform distortion information and one or more decoder neural network weights to a receiving device in a configuration that enables the receiving device to configure a decoder neural network of the receiving device to recover the transmit waveform distortion information of the transmitting device from the compressed transmit waveform distortion information using the one or more decoder neural network weights, wherein the decoder neural network has been trained to recover the transmit waveform distortion information of the transmitting device from the compressed transmit waveform distortion information, and the one or more decoder neural network weights are weights of the trained decoder neural network of the transmitting device; Including, Sending device.

15. a receiving device, a processor configured with processor-executable instructions for performing operations, said operations comprising: receiving compressed transmit waveform distortion information of a transmitting device and one or more weights of a trained decoder neural network of said transmitting device; configuring a decoder neural network of the receiving device using the received one or more weights; recovering transmit waveform distortion information of the transmitting device from the compressed transmit waveform distortion information of the transmitting device using the configured decoder neural network of the receiving device; Including, Receiving device.

Images