Cooperative bistatic radar sensing using deep neural networks

Abstract

Techniques and apparatuses are described for implementing cooperative bistatic radar sensing using deep neural networks. In particular, a base station (120) operates as a bistatic radar transmitter, and a user equipment (110) operates as a bistatic radar receiver. During radar sensing, the base station (120) and the user equipment (110) use their respective deep neural networks (460 and 420) for signal generation and signal processing. The deep neural networks (460 and 420) also enable the base station (120) and the user equipment (110) to utilize the same hardware for both radar sensing and wireless communication. Cooperative bistatic radar sensing allows the base station (120) and the user equipment (110) to collect explicit information about objects in the operating environment and use this information to improve wireless communication performance.

Description

[Background technology]

[0001] background Evolving wireless communication systems utilize increasingly complex architectures as a means of providing higher performance over predecessor wireless communication systems. In one example, fifth generation new wireless (5G NR) wireless technology transmits data using higher frequency ranges, such as the 6 gigahertz (GHz) band or the terahertz (THz) band, to increase data capacity. However, transmitting and recovering information using these higher frequency ranges is subject to impairments. By way of illustration, higher frequency signals are more susceptible to multipath fading, scattering, atmospheric absorption, diffraction, and interference than lower frequency signals. As another example, hardware capable of transmitting, receiving, routing, and / or using these higher frequencies can be expensive and complex to incorporate into portable electronic devices. Summary of the Invention [Problem to be solved by the invention]

[0002] overview Techniques and apparatuses are described for implementing cooperative bistatic radar sensing using deep neural networks. In particular, a base station and a user equipment include a pair of deep neural networks that are trained together to enable cooperative bistatic radar sensing. The base station acts as a bistatic radar transmitter, and the user equipment acts as a bistatic radar receiver. During cooperative bistatic radar sensing, the base station and the user equipment use their respective deep neural networks for signal generation and signal processing. The deep neural network also enables the base station and the user equipment to utilize the same hardware (e.g., antennas, radio frequency front ends, and transceivers) for both radar sensing and wireless communication. Thus, the base station and the user equipment can perform radar sensing without using a dedicated radar sensor. In some implementations, the deep neural network enables efficient use of the frequency spectrum by simultaneously enabling radar sensing and signal transmission of mobile communication reference and / or data. Cooperative bistatic radar sensing allows base stations and user equipment to gather explicit information about objects in the operating environment and use this information to improve wireless communication performance. [Means for solving the problem]

[0003] In an embodiment, the first device operates as a radar signal receiver of a bistatic radar. In particular, the first device receives a reflected version of a radar signal. The radar signal is transmitted by a second device and reflected from an object. The first device generates radar data by processing the reflected version of the radar signal using a deep neural network. The radar data includes information about the object. In addition, the first device operates as a feedback signal transmitter. In particular, the first device generates a feedback signal using a deep neural network. The feedback signal is based on the radar data. Moreover, the first device transmits the feedback signal to the second device.

[0004] In an embodiment, the first device operates as a radar signal transmitter of a bistatic radar. In particular, the first device generates a radar signal using a deep neural network. The radar signal transmitted by the first device is reflected from an object. In addition, the first device operates as a feedback signal receiver. In particular, the first device receives a feedback signal from the second device. The feedback signal indicates radar data generated by the second device based on the radar signal. Moreover, the first device determines information about the object by processing the radar data using the deep neural network.

[0005] Aspects described below also include a system having a means for cooperative bistatic radar sensing using deep neural networks.

[0006] Apparatus and techniques for implementing cooperative bistatic radar sensing using deep neural networks will now be described with reference to the following drawings, in which like numbers are used to reference like features and components throughout. [Brief description of the drawings]

[0007] [Figure 1] FIG. 1 illustrates an example environment in which various aspects of cooperative bistatic radar sensing using deep neural networks may be implemented. [Diagram 2] FIG. 1 illustrates an example environment in which various aspects of cooperative bistatic radar sensing using deep neural networks may be implemented. [Diagram 3] FIG. 1 illustrates an example signal for cooperative bistatic radar sensing using a deep neural network. [Figure 4] FIG. 1 is an example device diagram of a device capable of implementing various aspects of cooperative bistatic radar sensing using deep neural networks. [Diagram 5]FIG. 1 illustrates an example of generating a configuration of multiple neural network formations with cooperative bistatic radar sensing using deep neural networks. [Figure 6-1] FIG. 1 illustrates an example operation of a radar signal transmitter and a radar signal receiver for cooperative bistatic radar sensing using a deep neural network. [Figure 6-2] FIG. 1 illustrates an example operation of a feedback signal transmitter and a feedback signal receiver for cooperative bistatic radar sensing using a deep neural network. [Figure 7-1] FIG. 1 illustrates an example operation of a multipurpose signal transmitter and a multipurpose signal receiver to simultaneously perform cooperative bistatic radar sensing and wireless communication using a deep neural network. [Figure 7-2] FIG. 13 illustrates another exemplary operation of a multi-purpose feedback signal transmitter and a multi-purpose feedback signal receiver for simultaneously performing bistatic radar sensing and wireless communication using a deep neural network. [Figure 8] FIG. 1 illustrates an example transaction diagram between various network entities that implement cooperative bistatic radar sensing using deep neural networks. [Figure 9] FIG. 1 illustrates an example method for performing cooperative bistatic radar sensing operations using deep neural networks. [Figure 10] FIG. 1 illustrates another example method for performing cooperative bistatic radar sensing operations using deep neural networks. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] Detailed Description overview Channel estimation techniques can improve wireless communication performance in challenging environmental conditions described in the Background of the Invention section. For example, base stations and user equipment can use channel estimation to determine beamforming configurations that enhance signal-to-noise ratios. However, channel estimation techniques provide information about the operating environment in an indirect, synthetic manner. As a result, explicit (e.g., direct) information about the operating environment (e.g., information about objects in the environment) is still unknown.

[0009] In contrast, techniques for cooperative bistatic radar sensing using deep neural networks provide direct information about objects affecting the communication channel. In particular, a base station and a user equipment include a pair of deep neural networks that are trained together to enable cooperative bistatic radar sensing. The base station acts as a bistatic radar transmitter, and the user equipment acts as a bistatic radar receiver. During cooperative bistatic radar sensing, the base station and the user equipment use their respective deep neural networks for signal generation and signal processing. The deep neural network also enables the base station and the user equipment to utilize the same hardware (e.g., antennas, radio frequency front ends, and transceivers) for both radar sensing and wireless communication. Thus, the base station and the user equipment can perform radar sensing without using a dedicated radar sensor. In some implementations, the deep neural network enables efficient use of the frequency spectrum by simultaneously enabling radar sensing and mobile communication data and / or reference signal transmission. Cooperative bistatic radar sensing allows base stations and user equipment to gather explicit information about objects in the operating environment and use this information to improve wireless communication performance.

[0010] Example Environment 1 illustrates an example environment 100 including multiple user equipment 110 (UE 110), shown as UE 111, UE 112, and UE 113. Each user equipment 110 can communicate with one or more base stations 120 (shown as base stations 121 and 122) over one or more wireless communication links 130 (wireless link 130), shown as wireless link 131, wireless link 132, wireless link 133, wireless link 134, wireless link 135, and wireless link 136. User equipment 110 is embodied as a smartphone for simplicity, but may be embodied as any suitable computing or electronic device, such as a mobile communication device, a modem, a mobile phone, a gaming console, a navigation device, a media device, a laptop computer, a desktop computer, a tablet computer, a smart appliance, a vehicle-based communication system, or an Internet of Things (IoT) device, such as a sensor or actuator. The base station 120 (e.g., an Evolved Universal Terrestrial Radio Access Network Node B, E-UTRAN Node B, evolved Node B, eNodeB, eNB, next generation Node B, gNode B, gNB, ng-eNB, etc.) may be implemented as a macro cell, a micro cell, a small cell, a pico cell, a distributed base station, etc., or any combination thereof.

[0011] The base station 120 communicates with the user equipment 110 using a wireless link 130, which may be implemented as any suitable type of wireless link. The wireless link 130 includes communication of control and data, such as a downlink of data and control information communicated from the base station 120 to the user equipment 110, an uplink of other data and control information communicated from the user equipment 110 to the base station 120, or both. The wireless link 130 may include one or more wireless links (e.g., radio links) or bearers implemented using any suitable communication protocol or standard, or combination of communication protocols or standards, such as 3rd Generation Partnership Project Long Term Evolution (3GPP® LTE), 5th Generation New Radio (5G NR), and future developments. The multiple wireless links 130 may be aggregated into carrier aggregation or multiple connection techniques to provide higher data rates to the user equipment 110. The multiple wireless links 130 from multiple base stations 120 may be configured for coordinated multipoint (CoMP) communication with the user equipment 110. In addition, the multiple wireless links 130 may be configured for single radio access technology (RAT) (single-RAT), dual connectivity (single-RAT-DC), or multi-RAT dual connectivity (MR-DC). The wireless links 130 may be subject to permanent or temporary channel impairments such as from buildings, foliage, rainfall, other moving or stationary objects 180, shown as objects 181, 182, and 183.

[0012] The base stations 120 are collectively a radio access network 140 (e.g., RAN, Evolved Universal Terrestrial Radio Access Network, E-UTRAN, 5G NR RAN, or NR RAN). The base stations 121 and 122 in the RAN 140 are connected to the core network 150. The base stations 121 and 122 connect to the core network 150 at the interfaces 102 and 104, respectively, through an NG2 interface for control plane signaling, an NG3 interface for user plane data communication when connecting to a 5G core network, and an S1 interface for control plane signaling and user plane data communication when connecting to an evolved packet core (EPC) network. The base stations 121 and 122 can communicate at the interface 106 using the Xn Application Protocol (XnAP) over the Xn interface or the X2 Application Protocol (X2AP) over the X2 interface to exchange user plane data and control plane data.

[0013] User equipment 110 may connect to a public network, such as the Internet 160, via core network 150 to interact with remote services 170. Remote services 170 represent computing, communication, and storage devices used to provide any of a number of services, including interactive voice or video communication, file transfer, streaming audio or video, and other technical services implemented in any manner, such as voice calling, video calling, website access, messaging services (e.g., text messaging, multimedia messaging), photo file transfer, enterprise software applications, social media applications, video games, streaming video services, and podcasts.

[0014] 2 illustrates an example environment 200 including multiple objects 180 between the base station 120 and the UEs 111, 112, and 113. The multiple objects 180 are illustrated as objects 181, 182, and 183. The objects 180 can be stationary or moving objects. Example stationary objects include buildings, tunnels, bridges, rocks, plants, room walls, glass window frames, and furniture. Example moving objects include humans, animals, water vapor, precipitation, and vehicles.

[0015] At times, the presence and location of objects 180 in the environment 200 may make it difficult for the base station 120 to communicate with the user equipment 111, 112, and 113. For example, the objects 180 may cause reflection, diffraction, or scattering of the wireless communication signal, causing the wireless communication signal to propagate over multiple propagation paths 220. Examples of multiple propagation paths 220 of an omnidirectional signal from the base station 120 include propagation paths 222, 224, 226, and 228. With multiple propagation paths 220, multiple delayed versions of the wireless communication signal may arrive at a receiving entity at different times. This may distort the received wireless communication signal (e.g., by inter-symbol interference (ISI)) or reduce the signal-to-noise ratio. As another example, the objects 180 interfere with direct line-of-sight communication between the base station 120 and the user equipment 110. As shown in FIG. 2, for example, the object 182 blocks the line-of-sight communication between the user equipment 112 and the base station 120.

[0016] Techniques for cooperative bistatic radar sensing using deep neural networks enable base station 120 and user equipment 111, 112, and 113 to directly measure information about objects 180 in the current environment 200. This information may include location information (e.g., distance and / or angle), movement information (e.g., Doppler velocity and / or overall velocity), size information (e.g., width, length, and / or height), material composition information (e.g., reflection coefficient and / or radar cross section), or some combination thereof. Base station 120 can use this information to determine the propagation environment (e.g., estimate propagation path 220) and customize operations to improve wireless communication performance.

[0017] For example, the base station 120 adjusts its beamforming configuration based on the determined propagation environment to facilitate communication with the user equipments 111, 112, and 113. Additionally or alternatively, the base station 120 instructs the user equipments 111, 112, and 113 to utilize specific beamforming configurations. The base station 120 can also instruct the user equipments 111, 112, and 113 to utilize customized plans for transmission and reception (e.g., for beam management). These beamforming configurations can be designed to enhance the signal-to-noise ratio and / or reduce interference at the receiving entity. The base station 120 can dynamically adjust the beamforming configuration in anticipation of changes in the environment 200 based on the measured movement of the object 180. In this manner, the base station 120 proactively plans wireless communications based on knowledge of the environment 200 obtained using cooperative bistatic radar sensing. The base station 120 can also use its knowledge of the propagation environment to counteract interference received from other known propagation paths. This can further improve the signal-to-noise ratio at the base station 120 .

[0018] Consider an example in which the base station 120 customizes the transmission of a downlink signal 230 (DL signal 230), shown as downlink signals 232, 234, and 236, based on knowledge of the environment 200. The base station 120 utilizes line-of-sight propagation 240 to transmit the downlink signal 232 to the user equipment 111. In this case, the base station 120 uses a beamforming configuration to route the downlink signal 232 through a propagation path 222, which is a direct line-of-sight path between the base station 120 and the user equipment 111. The radiation pattern produced by the example beamforming configuration has a narrow beamwidth and has a main lobe that is directed along an angle associated with the propagation path 222. The base station 120 can reduce losses associated with propagation, reflection, and multipath fading by using a narrow beamwidth to direct the main lobe toward the user equipment 111. In this manner, the base station 120 can improve wireless communication performance.

[0019] The base station 120 uses non-line-of-sight (non-LOS) propagation 250 to transmit the downlink signal 234 to the user equipment 112. In particular, the base station 120 uses another beamforming configuration to pass the downlink signal 234 through a propagation path 224 that reflects the downlink signal 234 from the object 183 toward the user equipment 112. In this manner, the downlink signal 234 travels around the object 182. By using the propagation path 224, the base station 120 can overcome the obstacles associated with the object 182 that block line-of-sight communication between the base station 120 and the user equipment 112.

[0020] The base station 120 uses multipath propagation 260 to transmit a downlink signal 236 to the user equipment 113. In this case, the downlink signal 236 travels along propagation paths 226 and 228. The propagation path 226 is a direct path along the line of sight between the base station 120 and the user equipment 113. In contrast, the propagation path 228 is an indirect path that reflects the downlink signal 236 from an object 183 toward the user equipment 113. The base station 120 can utilize the multipath propagation 260 to improve the signal-to-noise ratio and / or increase the channel capacity at the user equipment 113 using MIMO techniques. In general, the base station 120 performs channel planning using direct knowledge of the object 180 obtained from cooperative bistatic radar sensing. The operation of the base station 120 and the user equipment 110 for cooperative bistatic radar sensing using a deep neural network is further described with reference to FIG. 3.

[0021] FIG. 3 illustrates an example signal for cooperative bistatic radar sensing using a deep neural network. In the example environment 300, the base station 120 and the user equipment 110 jointly operate as a bistatic radar to detect an object 180. For example, the base station 120 operates as a bistatic radar transmitter and the user equipment 110 operates as a bistatic radar receiver. In some aspects, the user equipment 110 operates as a second bistatic radar transmitter and the base station 120 operates as a second bistatic radar receiver. Additionally or alternatively, the base station 120 can operate as a monostatic radar. In this situation, the base station 120 operates as both a monostatic radar transmitter and a monostatic radar receiver. In this example, the base station 120 and the user equipment 110 implement a frequency modulated continuous wave radar. However, other types of radar are possible, including pulse Doppler radar, phase modulated spread spectrum radar, impulse radar, radar using Zadov-Chu or Constant Amplitude Zero Autocorrelation (CAZAC) sequences, or MIMO radar.

[0022] The base station 120 transmits a radar signal 310 during radar sensing. The radar signal 310 represents a frequency modulated signal in this example. In other implementations, the radar signal 310 may include a pulsed signal or a phase modulated signal. The example radar signal 310 includes a series of chirps 320, shown as chirps 322, 324, and 326. The chirps 320 may be transmitted in a continuous burst or separated in time. Multiple chirps 320 allow the base station 120 and / or user equipment 110 to make multiple observations of the object 180 over a given period of time.

[0023] The frequency of the chirps 320 can increase and decrease over time. In the illustrated example, the base station 120 employs two cycles of slopes (e.g., triangular wave frequency modulation) for the frequency of each chirp 320 that linearly increases and then linearly decreases over time. The two cycles of slopes allow the base station 120 and user equipment 110 to measure the Doppler frequency shift caused by the motion of the object 180. In general, the base station 120 adjusts the transmission characteristics (e.g., bandwidth, center frequency, duration, and transmit power) of the chirps 320 to achieve a particular detection area, range resolution, or Doppler sensitivity for detecting the object 180.

[0024] Radar signal 310 propagates through space and is reflected off of object 180. The reflected portions of radar signal 310 are represented by reflected radar signals 330 and 332. Reflected radar signal 330 propagates back towards base station 120. Reflected radar signal 332 propagates towards user equipment 110. Like radar signal 310, reflected radar signals 330 and 332 are also composed of chirp 320. As shown, the amplitudes of reflected radar signals 330 and 332 are smaller than that of radar signal 310 due to losses during propagation and reflection.

[0025] For cooperative bistatic radar sensing, the user equipment 110 receives and processes the reflected radar signal 332 to detect the object 180. At the user equipment 110, the reflected radar signal 332 represents a delayed and attenuated version of the radar signal 310. The amount of delay is proportional to the sum of the distance (e.g., straight-line distance) between the base station 120 and the object 180 and the distance between the object 180 and the user equipment 110. In particular, the delay represents the sum of the time it takes for the radar signal 310 to travel from the base station 120 to the object 180 and the time it takes for the reflected radar signal 332 to travel from the object 180 to the user equipment 110. If the object 180, the base station 120, or the user equipment 110 is moving, the reflected radar signal 332 is frequency shifted relative to the radar signal 310 due to the Doppler effect. In other words, certain characteristics of the reflected radar signal 332 depend on the motion of the object 180 , the motion of the base station 120 , and the motion of the user equipment 110 .

[0026] The user equipment 110 analyzes the reflected radar signal 332 to extract radar data containing implicit or explicit information about the object 180. As an example, the radar data includes raw digital samples of the reflected radar signal 332 or processed information about the reflected radar signal 332 (e.g., amplitude and / or phase information, complex range-Doppler maps, or complex interferometric data). The amplitude, phase, and frequency within this example radar data depend on the presence and characteristics of the object 180. As such, the example radar data includes implicit information about the object 180.

[0027] The user equipment 110 (or base station 120) can further process the implicit information related to the object 180 to determine explicit information about the object 180. The explicit information includes location information, movement information, size information, or material or surface composition information (e.g., human tissue, wood, or metal) of the object 180. In another example, the radar data may additionally or instead include explicit information about the object 180.

[0028] The base station 120 can communicate additional information to the user equipment 110 so that the user equipment 110 can determine explicit information about the object 180. For example, the base station 120 communicates its location (e.g., GNSS location) and velocity, the waveform characteristics of the radar signal 310, and / or the timing of the radar signal 310 for synchronization by a control signal transmission separate from the radar signal. In some situations, the user equipment 110 also receives the radar signal 310 through a direct line-of-sight propagation path. As such, the user equipment 110 can directly compare the reflected radar signal 332 with the received radar signal 310 to determine implicit and explicit information about the object 180. In other situations, the user equipment can replicate the radar signal 310 based on the waveform characteristics and timing information provided by the base station 120. In this manner, the user equipment 110 can compare the reflected radar signal 332 with the replicated radar signal 310 to determine implicit and explicit information about the object 180.

[0029] The user equipment 110 transmits a feedback signal 340 to communicate radar data to the base station 120. In some implementations, the user equipment 110 augments the feedback signal 340 to include the location of the user equipment 110 (e.g., a Global Navigation Satellite System (GNSS) position), the velocity of the user equipment 110, the orientation of the user equipment 110, or the antenna configuration of the user equipment 110. The feedback signal 340 propagates through space toward the base station 120, and a logical channel of mobile communications, such as a dedicated control channel (DCCH), may be used to provide this feedback information.

[0030] In some implementations, the base station 120 operates in a full-duplex manner to transmit the radar signal 310 and simultaneously receive the reflected radar signal 330. In this manner, the base station 120 can operate as a continuous wave monostatic radar. In other implementations, the base station 120 operates in a half-duplex manner to transmit the radar signal 310 and receive the reflected radar signal 330. In this manner, the base station 120 can operate as a pulsed Doppler monostatic radar. The base station 120 and the user equipment 110 can use time division duplexing or frequency division duplexing to enable the base station 120 to transmit the radar signal 310 and receive the feedback signal 340.

[0031] For cooperative bistatic radar sensing, the base station 120 processes the feedback signal 340 to determine explicit information about the object 180. In some situations, the base station 120 extracts explicit information about the object 180 provided by the feedback signal 340 from the user equipment 110. The base station 120 often receives multiple feedback signals from multiple user equipment 110 and combines the feedback to assist in determining the explicit information about the object 180, as described further below. In some situations, the base station 120 further processes implicit information about the object 180 provided by the feedback signal 340 from the user equipment 110 to determine the explicit information.

[0032] In another implementation not explicitly shown in FIG. 3, the base station 120 and the user equipment 110 implement a second bistatic radar in which the user equipment 110 transmits a second bistatic radar signal that is received by the base station 120. For this second bistatic radar system, the user equipment 110 transmits a second radar signal (in a manner similar to the radar signal 310) and the base station 120 receives a version of the second radar signal that is a reflected signal from the object 180. This second radar signal may be a reference signal for mobile communications, such as a sounding reference signal (SRS). In this manner, the user equipment 110 acts as a transmitter of the second bistatic radar and the base station 120 acts as a receiver of the second bistatic radar.

[0033] For monostatic radar sensing, the base station 120 processes the reflected radar signal 330 to detect the object 180. At the base station 120, the reflected radar signal 330 represents a delayed and attenuated version of the radar signal 310. The amount of delay is proportional to the distance between the base station 120 and the object 180. In particular, the delay represents the sum of the time it takes for the radar signal 310 to travel from the base station 120 to the object 180 and the time it takes for the reflected radar signal 330 to travel from the object 180 to the base station 120. If the object 180 or the base station 120 is moving, the reflected radar signal 330 will be frequency shifted relative to the radar signal 310 due to the Doppler effect. In other words, certain characteristics of the reflected radar signal 330 depend on the motion of the object 180 and the motion of the base station 120. The base station 120 analyzes the reflected radar signal 330 to detect the object 180 and determine information about the object 180 (eg, its location, movement, size, and / or material composition).

[0034] In some aspects, base station 120 combines information about object 180 determined using various radar sensing techniques. For example, base station 120 can combine information determined using a first bistatic radar sensing technique in which base station 120 operates as a first bistatic radar transmitter with (i) information determined using a second bistatic radar sensing technique in which base station 120 operates as a second bistatic radar receiver, and / or (ii) information determined using a monostatic radar sensing technique in which base station 120 operates as a monostatic radar.

[0035] By gathering information about the object 180 through bistatic and optional monostatic radar sensing, the base station 120 can gain knowledge about the current operating environment, such as the environment 200 shown in Figure 2. The base station 120 uses this knowledge to map the environment, determine available propagation paths 220, and customize operation for wireless communications accordingly. In general, the cooperative aspect of cooperative bistatic radar sensing involves the base station 120 and the user equipment 110 communicating with each other to enable the determination of explicit information about the object 180.

[0036] The general concept of cooperative bistatic radar sensing may also be applied to multiple user equipment 110 (e.g., user equipment 111, 112, and 113). In some situations, the base station 120 operates in collaboration with multiple user equipment 110 to implement multiple bistatic radars. In other situations, the base station 120 operates in collaboration with multiple user equipment 110 to implement multistatic radars. By employing multiple bistatic or multistatic radars, the base station 120 can determine the angular position of the object 180 using triangulation techniques. This can be advantageous in situations where the user equipment 110 receives reflected radar signals 332 using a single antenna. In both situations, the base station 120 can combine information provided by multiple user equipment 110 to obtain a more accurate estimate of the environment.

[0037] Example Device 4 shows an example device diagram 400 of a user equipment 110 and a base station 120 that can implement various aspects of cooperative bistatic radar sensing. The user equipment 110 and the base station 120 may include additional functionality and interfaces that are omitted from FIG. 4 for clarity.

[0038] The user equipment 110 includes an antenna 402, a radio frequency front end 404 (RF front end 404), and a radio transceiver 406 (e.g., an LTE transceiver and / or a 5G NR transceiver). The antenna 402, the radio frequency front end 404, and the radio transceiver 406 can be used to communicate with the base station 120 in the RAN 140 and cooperative bistatic radar sensing. The radio frequency front end 404 couples or connects the radio transceiver 406 to the antenna 402. The antenna 402 may include an arrangement of multiple antennas of similar or different configurations. The antenna 402 and the radio frequency front end 404 may be tuned and / or tunable to one or more frequency bands defined by the 3GPP LTE and 5G NR communication standards and implemented by the radio transceiver 406. By way of example and not limitation, the antenna 402 and radio frequency front end 404 may be implemented to operate in sub-GHz, sub-6 GHz, and / or above 6 GHz bands (e.g., the GHz band associated with millimeter wavelengths, or the Terahertz (THz) band associated with sub-millimeter wavelengths). Additionally, the antenna 402, radio frequency front end 404, and radio transceiver 406 may be configured to support beamforming for wireless communication and / or cooperative bistatic radar sensing.

[0039] The user equipment 110 also includes at least one processor 408 and at least one computer-readable storage medium 410 (CRM 410). The processor 408 may be a single-core processor or a multi-core processor made of various materials such as silicon, polysilicon, high-K dielectrics, copper, etc. The CRM 410 described herein may include any suitable memory or storage device, such as random access memory (RAM), static RAM (SRAM), dynamic RAM (DRAM), non-volatile RAM (NVRAM), read-only memory (ROM), or flash memory, that can be used to store device data 412 of the user equipment 110, excluding propagating signals. The device data 412 includes user data, multimedia data, beamforming codebooks, applications, neural network (NN) tables, neural network training data, and / or the operating system of the user equipment 110, some of which are executable by the processor 408 to enable user plane data, control plane information, and user interaction with the user equipment 110.

[0040] In an embodiment, CRM 410 includes a neural network table 414, a neural network manager 416, and a training module 418. Alternatively, or in addition, neural network manager 416 and / or training module 418 may be implemented, in whole or in part, as hardware logic or circuitry, integrated into or separate from other components of user equipment 110.

[0041] The neural network table 414 stores various architectures and / or parameter configurations for forming a deep neural network (DNN) 420, such as, for example and not by way of limitation, parameters defining the architecture of a fully connected layer neural network, an architecture of a convolutional layer neural network, a recurrent neural network layer, the number of connected hidden neural network layers, an input layer architecture, an output layer architecture, the number of nodes utilized by the deep neural network 420, the coefficients (e.g., weights and biases) utilized by the deep neural network 420, kernel parameters, the number of filters utilized by the deep neural network 420, the stride / pooling configuration utilized by the deep neural network 420, the activation function of each neural network layer, the interconnections between neural network layers, the neural network layers to skip, etc. Thus, the neural network table 414 includes any combination of neural network formation components (NN formation components). A combination of one or more NN formation components, such as architectures and / or parameter configurations, may be used to generate a neural network formation configuration (NN formation configuration). The configuration of the NN formation defines and / or forms a deep neural network 420.

[0042] In some aspects, a single index value in the neural network table 414 maps to a single NN formation component (e.g., a 1:1 correspondence). Alternatively, or in addition, a single index value in the neural network table 414 maps to a configuration of the NN formation (e.g., a combination of NN formation components). In some implementations, the neural network table 414 includes input characteristics for each NN formation component and / or NN formation configuration, where the input characteristics describe characteristics about the training data used to generate the NN formation component and / or NN formation configuration, as further described with respect to FIG. 5.

[0043] The neural network manager 416 accesses the neural network table 414, such as by an index value, and uses the components of the NN formation defined by the NN formation configuration to form the deep neural network 420. This includes updating the deep neural network 420 with any combination of architectural changes and / or parameter changes. Example changes include minor changes involving parameter updates and / or major changes that reconfigure the node and / or layer connections of the deep neural network 420.

[0044] In response to receiving instructions from the neural network manager 416, the training module 418 provides known input data, such as input data stored as device data 412, to the deep neural network 420. The training module 418 educates and trains the deep neural network 420 using the known input data and / or by providing feedback to the ML algorithm. For example, the training module 418 trains the deep neural network 420 for cooperative bistatic radar sensing (e.g., to process the reflected radar signal 332 to generate the feedback signal 340). The training module 418 can also train the deep neural network 420 for wireless communication (e.g., encoding uplink communications, modulating uplink communications, demodulating downlink communications, or decoding downlink communications). The wireless communication can use any type of radio access technology (RAT), such as 4G LTE, 5G NR, and future developments.

[0045] In an implementation, the training module 418 extracts updated ML information from the deep neural network 420 and forwards it to the neural network manager 416. The extracted updated ML information may include any combination of information such as node connections, coefficients, activation layers, weights, biases, pooling, etc. that defines the behavior of the deep neural network 420.

[0046] The device diagram of the base station 120 shown in FIG. 4 includes a single network node (e.g., gNode B). The functionality of the base station 120 may be distributed across multiple network nodes or devices and may be distributed in any manner suitable for performing the functions described herein. The base station 120 includes an antenna 442, at least one radio frequency front end 444 (RF front end 444), and one or more radio transceivers 446 (e.g., one or more LTE transceivers and / or one or more 5G NR transceivers). The antenna 442, the radio frequency front end 444, and the radio transceiver 446 may be used for communication with the user equipment 110 and for cooperative bistatic radar sensing. In some implementations, the antenna 442, the radio frequency front end 444, and the radio transceiver 446 may also be used for monostatic radar sensing. The radio frequency front end 444 couples or connects the radio transceiver 446 to the antenna 442. The antenna 442 may include an arrangement of multiple antennas of similar or different configurations. The antenna 442 and the radio frequency front end 444 may be tuned and / or tunable to one or more frequency bands defined by 3GPP LTE and 5G NR communications standards and implemented by the radio transceiver 446. By way of example and not limitation, the antenna 442 and the radio frequency front end 444 may be implemented to operate in sub-GHz bands, sub-6 GHz bands, and / or above 6 GHz bands (e.g., GHz bands associated with millimeter wavelengths, or Terahertz (THz) bands associated with sub-millimeter wavelengths). Additionally, the antenna 442, the radio frequency front end 444, and / or the radio transceiver 446 may be configured to support beamforming, such as massive MIMO, for wireless communications and radar sensing.

[0047] The base station 120 also includes at least one processor 448 and at least one computer-readable storage medium 450 (CRM 450). The processor 448 may be a single-core processor or a multi-core processor made of various materials such as silicon, polysilicon, high-K dielectrics, copper, etc. The CRM 450 may include any suitable memory or storage device such as random access memory (RAM), static RAM (SRAM), dynamic RAM (DRAM), non-volatile RAM (NVRAM), read-only memory (ROM), or flash memory that can be used to store device data 452 of the base station 120. The device data 452 includes network scheduling data, radio resource management data, beamforming codebooks, applications, and / or operating systems of the base station 120 that are executable by the processor 448 to enable cooperative bistatic radar sensing or wireless communication with the user equipment 110.

[0048] The CRM 450 includes a neural network table 454, a neural network manager 456, and a training module 458. Alternatively, or in addition, the neural network manager 456 and / or the training module 458 may be implemented, in whole or in part, as hardware logic or circuitry, integrated into or separate from other components of the base station 120.

[0049] The neural network table 454 stores various architectures and / or parameter configurations forming the deep neural network 460. The neural network table 454 of the base station 120, like the neural network table 414 of the user equipment 110, includes any combination of components of a NN formation and / or configurations of a NN formation generated using the training module 458. In some implementations, a single index value of the neural network table 454 maps to a single component of a NN formation (e.g., a 1:1 correspondence). Alternatively, or in addition, a single index value of the neural network table 454 maps to a configuration of a NN formation (e.g., a combination of components of a NN formation).

[0050] The training module 458 uses known input data to teach and / or train the deep neural network 460. For example, the training module 458 trains the deep neural network 460 for cooperative bistatic radar sensing (e.g., generating the radar signal 310, processing the feedback signal 340, processing a reflected version of a second radar signal transmitted by the user equipment 110, processing the reflected radar signal 330, modeling the propagation path 220). The training module 458 can also train the deep neural network 460 for wireless communications of any particular RAT (e.g., encoding downlink communications, modulating downlink communications, demodulating uplink communications, or decoding uplink communications).

[0051] In an implementation, the training module 458 extracts learned parameter configurations from the deep neural network 460 to identify components of the NN formation and / or configurations of the NN formation, and then adds to and / or updates the components of the NN formation and / or configurations of the NN formation in the neural network table 454. The extracted parameter configurations include any combination of information, such as node connections, coefficients, activation layers, weights, biases, or pooling, that defines the behavior of the deep neural network 460.

[0052] In some cases, the base station 120 synchronizes the neural network table 454 with the neural network table 414 of the user equipment 110 such that components and / or input characteristics of the NN formation stored in one neural network table are replicated in the second neural network table. Alternatively, or in addition, the base station 120 synchronizes the neural network table 454 with the neural network table 414 such that components and / or input characteristics of the NN formation stored in one neural network table represent complementary functions in the second neural network table (e.g., components of the NN formation for transmit side path processing in the first neural network table, components of the NN formation for receive side path processing in the second neural network table).

[0053] The supervised joint training of the deep neural networks 420 and 460 may be performed offline (e.g., when the deep neural networks 420 and 460 are not actively participating in cooperative bistatic radar sensing or wireless communication). Training data for offline training may include simulated data for various operating environments and radar waveforms and ground truth data that defines the expected output when this data is input to the deep neural networks 420 and 460. Additionally or alternatively, the supervised joint training of the deep neural networks 420 and 460 may be performed online (e.g., when the deep neural networks 420 and 460 are actively participating in cooperative bistatic radar sensing or wireless communication). In this case, the training data may be supplied by a user or a sensor. In general, the training data includes explicit information about the object 180 (e.g., location information, movement information, size information, or material composition information). The training modules 418 and 458 can use gradient descent techniques to train the deep neural networks 420 and 460 to reduce errors.

[0054] The base station 120 also includes a core network interface 462, configured by the base station 120 to exchange user plane data, control plane information, and / or other data / information with core network functions and / or entities. The base station 120 further includes an inter-base station interface 464, such as an Xn interface and / or an X2 interface, configured by the base station 120 to exchange user plane data, control plane information, and / or other data / information with other base stations to manage communications between the base station 120 and the user equipment 110.

[0055] The components of the user equipment 110 and the components of the base station 120 may implement a radar signal transmitter, a radar signal receiver, a feedback signal transmitter, a feedback signal receiver, or some combination thereof to support cooperative bistatic radar sensing, as further described with reference to Figures 6-1 and 6-2. Additionally or alternatively, the components of the user equipment 110 and the components of the base station 120 may implement a multipurpose signal transmitter, a multipurpose signal receiver, a multipurpose feedback signal transmitter, a multipurpose feedback signal receiver, or some combination thereof to support simultaneous cooperative bistatic radar sensing and wireless communication, as further described with reference to Figures 7-1 to 7-2.

[0056] FIG. 5 illustrates an embodiment of generating multiple NN formation configurations to support cooperative bistatic radar sensing and wireless communication. The top portion of FIG. 5 includes deep neural network 420 or 460. In an implementation, neural network manager 416 or 456 generates different NN formation configurations, such as NN formation configurations for cooperative bistatic radar sensing, based on different locations of user equipment 110 or base station 120 or different capabilities of user equipment 110 or base station 120. Alternatively, or in addition, neural network manager 416 or 456 generates NN formation configurations based on different environmental and / or channel conditions.

[0057] The training data 504 represents an example input for the deep neural network 420 or 460. For the deep neural network 420, the training data 504 represents a reflected radar signal 332 received from a particular object 180 and / or a particular propagation environment. For the deep neural network 460, the training data 504 represents radar waveform characteristics for generating the radar signal 310. In some implementations, the training module 418 or 458 mathematically generates the training data 504 using a simulation or accesses a file that stores the training data 504. The training module 418 or 458 obtains real radar data at another time. Thus, the training module 418 or 458 can train the deep neural networks 420 and 460 using mathematically generated data, static data, and / or real data. Some implementations generate input characteristics 506 that describe various qualities of the training data 504, such as operating configuration, transmission channel metrics, base station 120 and / or user equipment 110 capabilities, base station 120 and / or user equipment speed, base station 120 and / or user equipment 110 location, or some combination thereof.

[0058] The deep neural network 420 or 460 analyzes the training data 504 to generate an output 508, represented here as binary data. Some implementations repeatedly train the deep neural network 420 or 460 using the same set of training data 504 and / or additional training data with the same input characteristics to improve the accuracy of the machine learning module. During training, the neural network manager 416 or 456 changes some or all of the architecture and / or parameter configurations, such as node connections, coefficients, or kernel sizes, of the deep neural network 420 or 460. The training module 418 or 458 decides at some point during training to extract an architecture and / or parameter configuration 510 (e.g., pooling parameters, kernel parameters, layer parameters, or weights) of the deep neural network 420 or 460. This may occur when the training module 418 or 458 determines that the accuracy has met or exceeded a desired threshold (i.e., the output 508 from the deep neural network 420 or 460 is deemed to have sufficiently converged to the (known) ground truth for the set of training data 504) or when the training process has met or exceeded a number of iterations. The training module 418 or 458 then extracts architectures and / or parameter configurations for use as configurations and / or components of the NN formation. The architectures and / or parameter configurations may include any combination of immutable architectures and / or parameter configurations and / or variable architectures and / or parameter configurations.

[0059] The bottom of FIG. 5 includes a neural network table 414 or 454 that represents a collection of components of a NN formation. The neural network table 414 or 454 stores various combinations of architecture configurations, parameter configurations, and input characteristics, although alternative implementations omit the input characteristics from the table. Various implementations update and / or maintain the components and / or input characteristics of the NN formation as the deep neural network 420 or 460 learns additional information. For example, in index 514, the neural network manager 416 or 456 and / or the training module 418 or 458 update the neural network table 414 or 454 to include the architecture and / or parameter configurations 510 that the deep neural network 420 or 460 generated while analyzing the training data 504. At a later point in time, the neural network manager 416 or 456 selects one or more configurations of the NN formation from the neural network table 414 or 454 by matching the input characteristics with the current operating environment and / or configuration.

[0060] The neural network manager 416 of the user equipment 110 and the neural network manager 456 of the base station 120 select the configuration of the NN formation to be utilized by the deep neural networks 420 and 460 for cooperative bistatic radar sensing. In some implementations, the neural network manager 456 receives an indication of the configuration of the NN formation from the core network 150 via the core network interface 462 or the inter-base station interface 464 and forwards it to the user equipment 110.

[0061] In some implementations, the neural network manager 416 forms multiple deep neural networks 420 for cooperative bistatic radar sensing. As an example, the deep neural network 420 includes a first deep neural network using a first NN formation configuration to process the reflected radar signal 332 and a second deep neural network using a second NN formation configuration to generate the feedback signal 340. In some cases, the first NN formation configuration and the second NN formation configuration support simultaneous cooperative bistatic radar sensing and wireless communication. For example, the first NN formation configuration enables the first deep neural network to extract a downlink signal, such as a synchronization signal or a reference signal, to support wireless communication. Exemplary synchronization signals include a primary synchronization signal (PSS) or a secondary synchronization signal (SSS) used in mobile communication, or a preamble training sequence used in wireless local area network (WLAN) communication. Exemplary mobile communication reference signals include a demodulation reference signal (DM-RS), a phase tracking reference signal (PT-RS), a channel state information reference signal (CSI-RS), or a tracking reference signal (TRS). The second NN formation configuration enables the second deep neural network to generate an uplink reference signal (e.g., SRS) to support wireless communication.

[0062] In other implementations, the neural network manager 416 configures the deep neural network 420 to use configurations of the NN formation for both processing the reflected radar signal 332 and generating the feedback signal 340. The deep neural network 420 can also include other deep neural networks dedicated to wireless communications.

[0063] Additionally or alternatively, neural network manager 456 forms multiple deep neural networks 460 for cooperative bistatic radar sensing tailored to the example implementation of base station 120. For example, deep neural network 460 includes a first deep neural network that uses a first NN formation configuration to generate radar signal 310 and a second deep neural network that uses a second NN formation configuration to process some combination of reflected radar signal 330, a reflected version of a second radar signal transmitted from user equipment 110, or feedback signal 340. In some cases, the first NN formation configuration and the second NN formation configuration support simultaneous cooperative bistatic radar sensing and wireless communication.

[0064] In another example implementation of the base station, the neural network manager 416 configures the deep neural network 460 to use a configuration of NN formations to both generate the radar signal 310 and process the feedback signal 340. The deep neural network 460 can also include other deep neural networks dedicated to wireless communication. For example, a first configuration of NN formations allows the first deep neural network to transmit downlink data for wireless communication, and a second configuration of NN formations allows the second deep neural network to extract uplink data for wireless communication.

[0065] Cooperative bistatic radar sensing using deep neural networks FIG. 6-1 illustrates an example operation of a radar signal transmitter 602 and a radar signal receiver 604 for cooperative bistatic radar sensing. The radar signal transmitter 602 and the radar signal receiver 604 are located at separate locations and together form a bistatic radar. The radar signal transmitter 602 includes a deep neural network 606, a wireless transceiver 608 (transceiver 608), a radio frequency front end 610, and one or more antennas 612. The radar signal receiver 604 includes a deep neural network 614, a wireless transceiver 616 (transceiver 616), a radio frequency front end 618, and one or more antennas 620. In FIG. 6-1, deep neural network 606 of radar signal transmitter 602 and deep neural network 614 of radar signal receiver 604 represent a pair of deep neural networks that enable both transmission and reception of radar signal 310 for cooperative bistatic radar sensing.

[0066] 3, the base station 120 operates as a radar signal transmitter 602 and the user equipment 110 operates as a radar signal receiver 604 to form a first bistatic radar. As such, the deep neural network 606 represents the deep neural network 460 of the base station 120, the transceiver 608 represents the transceiver 446 of the base station 120, the radio frequency front end 610 represents the radio frequency front end 444 of the base station 120, and the antenna 612 represents the antenna 442 of the base station 120. Also, the deep neural network 614 represents the deep neural network 420 of the user equipment 110, the transceiver 616 represents the transceiver 406 of the user equipment 110, the radio frequency front end 618 represents the radio frequency front end 404 of the user equipment 110, and the antenna 620 represents the antenna 402 of the user equipment 110.

[0067] In another example implementation, the user equipment 110 operates as a radar signal transmitter 602 and the base station 120 operates as a radar signal receiver 604 to form a second bistatic radar. As such, the deep neural network 606 represents the deep neural network 420 of the user equipment 110, the transceiver 608 represents the transceiver 406 of the user equipment 110, the radio frequency front end 610 represents the radio frequency front end 404 of the user equipment 110, and the antenna 612 represents the antenna 402 of the user equipment 110. Also, the deep neural network 614 represents the deep neural network 460 of the base station 120, the transceiver 616 represents the transceiver 446 of the base station 120, the radio frequency front end 618 represents the radio frequency front end 444 of the base station 120, and the antenna 620 represents the antenna 442 of the base station 120.

[0068] The radar signal transmitter 602 transmits the radar signal 310 using a deep neural network 606, a transceiver 608, a radio frequency front end 610, and one or more of the antennas 612. A neural network manager (e.g., neural network manager 456 or 416) configures the deep neural network 606 according to a transmit (TX) radar and neural network (NN) formation configuration 622 (configuration 622), which may be stored by a neural network table (e.g., neural network table 454 or 414). The deep neural network 606 uses the configuration 622 to perform signal generation for cooperative bistatic radar sensing. In general, the deep neural network 606 performs some or all of the functionality of a bistatic radar transmitter to generate the radar signal 310. For example, the deep neural network 606 may implement a waveform generation stage, a coherence stage, a modulation stage, or a digital beamforming stage to generate a digital representation of the radar signal 310.

[0069] The transceiver 608 is coupled to the deep neural network 606 and performs operations such as digital-to-analog conversion, up-conversion, amplification, and / or filtering. A radio frequency front end 610 is coupled between the transceiver 608 and an antenna 612. The radio frequency front end 610 performs operations such as amplification, filtering, and / or phase shifting.

[0070] The radar signal receiver 604 receives the reflected radar signal 332 using a deep neural network 614, a transceiver 616, a radio frequency front end 618, and one or more of the antennas 620. A neural network manager (e.g., neural network manager 416 or 456) configures the deep neural network 614 according to a receive (RX) radar and neural network (NN) formation configuration 624 (configuration 624), which may be stored by a neural network table (e.g., neural network table 414 or 454). The deep neural network 614 uses the configuration 624 to perform signal processing for cooperative bistatic radar sensing. In general, the deep neural network 614 performs some or all of the functionality of a bistatic radar receiver to process the radar signal 310. For example, the deep neural network 614 may implement a demodulation stage, a clutter tracker, an interference mitigation stage, a Fourier transform stage (e.g., a fast Fourier transform (FFT) stage), a threshold detection stage, a range processing stage, a Doppler processing stage, or a digital beamforming stage to analyze the reflected radar signal 332 and determine implicit or explicit information about the object 180.

[0071] The transceiver 616 is coupled to the deep neural network 614 and performs operations such as analog-to-digital conversion, down-conversion, amplification, and / or filtering. A radio frequency front end 618 is coupled between the transceiver 616 and the antenna 620. The radio frequency front end 618 performs operations such as amplification, filtering, and / or phase shifting.

[0072] During cooperative bistatic radar sensing, the radar signal transmitter 602 transmits the radar signal 310. In particular, the deep neural network 606 receives the radar waveform characteristics 626 and generates a digital radar signal 628 based on the radar waveform characteristics 626. The radar waveform characteristics 626 may define a center frequency, a bandwidth, a pulse repetition frequency (PRF), a beamforming configuration, and / or a modulation type of the radar signal 310. In this manner, the radar waveform characteristics 626 are provided as an input to the deep neural network 606, which outputs digital samples that are used to generate the radar signal 310.

[0073] In some cases, the radar waveform characteristics 626 may be determined based on a tracking module of the radar signal transmitter 602. The tracking module may be implemented using an alpha-beta tracker, a Kalman filter, a multi-hypothesis tracker, or another deep neural network. The tracking module adjusts the radar waveform characteristics 626 for the estimated position and movement of the object 180, taking into account previous measurements of the object 180's position and movement. In this manner, the radar signal transmitter 602 may improve measurement accuracy and increase the probability of detecting the object 180. For example, the radar signal transmitter 602 may adjust a beamforming configuration to direct the main lobe of the radiation pattern along an expected angle to the object 180, or adjust the pulse repetition frequency to enhance separation between the object 180 and surrounding clutter in the Doppler region.

[0074] The transceiver 608 receives and converts the digital radar signal 628 to an analog signal. The transceiver 608 and radio frequency front end 610 further condition the analog signal for transmission. For example, the transceiver 608 upconverts the analog signal from a baseband frequency to a radio frequency. The radio frequency front end 610 amplifies the analog signal and applies a phase shift for analog beamforming or filters the analog signal to remove spurious frequencies. The antenna 612 transmits the analog signal representing the radar signal 310.

[0075] The radar signal receiver 604 receives the reflected radar signal 332 using one or more of the antennas 620. The antennas 620 pass the reflected radar signal 332 to a radio frequency front end 618. The radio frequency front end 618 and the transceiver 616 can further condition the reflected radar signal 332 for processing. For example, the radio frequency front end 618 amplifies the reflected radar signal 332 and applies a phase shift for analog beamforming or filters the reflected radar signal 332 to remove interference. The transceiver 616 downconverts the reflected radar signal 332 from a radio frequency to a baseband frequency. The transceiver 616 also performs an analog-to-digital conversion to generate a digital representation of the reflected radar signal 332, shown as a digital reflected radar signal 630.

[0076] The deep neural network 614 receives and processes the digital reflected radar signal 630 to generate radar data 632. Thus, digital samples of the digital reflected radar signal 630 are provided as input to the deep neural network 614, which outputs radar data 632 that includes implicit and / or explicit information about the object 180. In some situations, the radar signal receiver 604 provides the radar data 632 to an application running on a device that includes the radar signal receiver 604. As an example, the device can use the radar data 632 to perform gesture recognition, collision avoidance, or health monitoring. In some situations, the radar signal receiver 604 also receives and processes the radar signal 310 to enable bistatic radar techniques.

[0077] Sometimes a device (e.g., user equipment 110 or base station 120) includes both a radar signal transmitter 602 and a radar signal receiver 604. In this case, the device can use the radar signal transmitter 602 and the radar signal receiver 604 for monostatic radar sensing. In particular, the radar signal receiver 604 processes the reflected radar signal 330 to determine explicit information about the object 180.

[0078] Although not shown in the figure, at least some of the components of the radar signal transmitter 602 and / or the radar signal receiver 604 may also be used for wireless communication, such as data signal transmission for mobile communication or WLAN signal transmission. In this situation, the radar signal transmitter 602 connects the transceiver 608 to another deep neural network associated with the wireless communication (or reconfigures the deep neural network 606 for wireless communication). Similarly, the radar signal receiver 604 may connect the transceiver 608 to another deep neural network associated with the wireless communication (or reconfigures the deep neural network 614). Further operations for cooperative bistatic radar sensing are further described with reference to FIG. 6-2.

[0079] FIG. 6-2 illustrates an example operation of a feedback signal transmitter 634 and a feedback signal receiver 636 for cooperative bistatic radar sensing. The feedback signal transmitter 634 is co-located with the radar signal receiver 604, and the feedback signal receiver 636 is co-located with the radar signal transmitter 602. The feedback signal transmitter 634 includes a deep neural network 638, a wireless transceiver 640 (transceiver 640), a radio frequency front end 642, and one or more antennas 644. The feedback signal receiver 636 includes a deep neural network 646, a wireless transceiver 648 (transceiver 648), a radio frequency front end 650, and one or more antennas 652. In FIG. 6-2, the deep neural network 638 of the feedback signal transmitter 634 and the deep neural network 646 of the feedback signal receiver 636 represent a pair of deep neural networks that enable both transmission and reception of the feedback signal 340 for cooperative bistatic radar sensing.

[0080] In some implementations, the feedback signal transmitter 634 uses different components than the radar signal receiver 604. For example, the deep neural network 638 of the feedback signal transmitter 634 is different (e.g., separate) from the deep neural network 614 of the radar signal receiver 604. In this case, the deep neural network 638 is designed and trained to generate the feedback signal, and the deep neural network 614 is designed and trained to process the radar signal. Similarly, the feedback signal receiver 636 can use different components than the radar signal transmitter 602. In particular, the deep neural network 646 of the feedback signal receiver 636 is different from the deep neural network 606 of the radar signal transmitter 602. In this case, the deep neural network 646 is designed and trained to process the feedback signal, and the deep neural network 606 is designed and trained to generate the radar signal.

[0081] In other implementations, the feedback signal transmitter 634 uses the same components as the radar signal receiver 604 uses. For example, the deep neural network 638 of the feedback signal transmitter 634 and the deep neural network 614 of the radar signal receiver 604 can be merged to form a single deep neural network that serves both the feedback signal transmitter 634 and the radar signal receiver 604. The feedback signal transmitter 634 and the radar signal receiver 604 can also share the same transceiver, radio frequency front end, antenna, or some combination thereof. The feedback signal receiver 636 can also use the same components as the radar signal transmitter 602 uses. For example, the deep neural network 646 of the feedback signal receiver 636 and the deep neural network 606 of the radar signal transmitter 602 can be the same deep neural network that reuses the same NN nodes with different configurations. The feedback signal receiver 636 and the radar signal transmitter 602 may also share the same transceiver, radio frequency front end, antenna, or some combination thereof.

[0082] When the base station 120 and the user equipment 110 form a first bistatic radar as described above with reference to FIG. 6-1, the user equipment 110 operates as a feedback signal transmitter 634 and the base station 120 operates as a feedback signal receiver 636. Thus, the deep neural network 638 represents the deep neural network 420 of the user equipment 110, the transceiver 640 represents the transceiver 406 of the user equipment 110, the radio frequency front end 642 represents the radio frequency front end 404 of the user equipment 110, and the antenna 644 represents the antenna 402 of the user equipment 110. Additionally, deep neural network 646 represents deep neural network 460 of base station 120, transceiver 648 represents transceiver 446 of base station 120, radio frequency front end 650 represents radio frequency front end 444 of base station 120, and antenna 652 represents antenna 442 of base station 120.

[0083] In the case where the base station 120 and the user equipment 110 form a second bistatic radar, the base station 120 operates as a feedback signal transmitter 634 and the user equipment 110 operates as a feedback signal receiver 636. Thus, the deep neural network 638 represents the deep neural network 460 of the base station 120, the transceiver 640 represents the transceiver 446 of the base station 120, the radio frequency front end 642 represents the radio frequency front end 444 of the base station 120, and the antenna 612 represents the antenna 442 of the base station 120. Also, the deep neural network 646 represents the deep neural network 420 of the user equipment 110, the transceiver 648 represents the transceiver 406 of the user equipment 110, the radio frequency front end 650 represents the radio frequency front end 404 of the user equipment 110, and the antenna 652 represents the antenna 402 of the user equipment 110.

[0084] A neural network manager (e.g., neural network manager 416 or 456) configures a deep neural network 638 according to a transmit (TX) feedback (FB) neural network (NN) formation configuration 654 (configuration 654), which may be stored by a neural network table (e.g., neural network table 414 or 454). The deep neural network 638 uses the configuration 654 to perform feedback signal generation for cooperative bistatic radar sensing. In general, the deep neural network 638 performs some or all of the processing functions of the transmitter to generate the feedback signal 340. In particular, the deep neural network 638 may perform encoding and modulation stages to generate a digital representation of the feedback signal 340.

[0085] A neural network manager (e.g., neural network manager 456 or 416) configures deep neural network 646 according to a configuration 656 (configuration 656) of a receive (RX) feedback neural network (NN) formation, which may be stored by a neural network table (e.g., neural network table 454 or 414). Deep neural network 646 uses configuration 656 to perform feedback signal processing for cooperative bistatic radar sensing. In general, deep neural network 646 performs some or all of the processing functions of a receiver to process feedback signal 340. In particular, deep neural network 646 may implement demodulation and decoding stages to extract radar data 632 provided by feedback signal 340.

[0086] During cooperative bistatic radar sensing, feedback signal transmitter 634 transmits feedback signal 340. In particular, deep neural network 638 receives radar data 632 (of FIG. 6-1 ) and generates digital feedback signal 658 based on radar data 632. Thus, radar data 632 is provided as an input to deep neural network 638, which outputs digital samples that are used to generate feedback signal 340.

[0087] The transceiver 640 receives and converts the digital feedback signal 658 to an analog signal. The transceiver 640 and the radio frequency front end 642 further condition the analog signal for transmission. For example, the transceiver 640 upconverts the analog signal from a baseband frequency to a radio frequency. The radio frequency front end 642 amplifies the analog signal and applies a phase shift for analog beamforming or filters the analog signal to remove spurious frequencies. The antenna 644 transmits the analog signal representing the feedback signal 340.

[0088] The feedback signal receiver 636 receives the feedback signal 340 using one or more of the antennas 652. The antennas 652 pass the feedback signal 340 to the radio frequency front end 650. The radio frequency front end 650 and the transceiver 648 can further condition the feedback signal 340 for processing. For example, the radio frequency front end 650 amplifies the feedback signal 340, applies a phase shift for analog beamforming, and / or filters the feedback signal 340 to remove interference. The transceiver 648 downconverts the feedback signal 340 from radio frequencies to baseband frequencies. The transceiver 648 also performs analog-to-digital conversion to generate a digital representation of the feedback signal 340, shown as a digital feedback signal 660.

[0089] The deep neural network 646 receives and processes the digital feedback signal 660 to generate object data 662. Thus, digital samples of the digital feedback signal 660 are provided as input to the deep neural network 646, which outputs the object data 662. At times, the feedback signal receiver 636 collects object data 662 from radar data 632 transmitted by one or more feedback signal transmitters 634, as well as object data determined using monostatic radar sensing. The feedback signal receiver 636 can use the object data 662 to model the environment and estimate the propagation path 220 to improve wireless communication performance.

[0090] The operation for cooperative bistatic radar sensing can operate according to a frequency division duplexing (FDD) system or a time division duplexing (TDD) system. To implement an FDD system, the radar signal 310 and the feedback signal 340 utilize separate frequency bands. In this manner, the radar signal transmitter 602 can simultaneously transmit the radar signal 310 and receive the feedback signal 340 and / or the reflected radar signal 330. To implement a TDD system, the radar signal transmitter 602 transmits the radar signal 310 during a designated time slot, and the feedback signal receiver 636 receives the feedback signal 340 during another designated time slot. In this case, the radar signal 310 and the feedback signal 340 can utilize the same frequency band. In some cases, cooperative bistatic radar sensing and wireless communication can be performed simultaneously, as further described with reference to FIGS. 7-1 and 7-2.

[0091] FIG. 7-1 illustrates an example operation of a multipurpose signal transmitter 702 and a multipurpose signal receiver 704 for simultaneously performing cooperative bistatic radar sensing and wireless communication. The multipurpose signal transmitter 702 and the multipurpose signal receiver 704 are located at separate locations and together form a bistatic radar. The multipurpose signal transmitter 702 includes a deep neural network 706, a wireless transceiver 708 (transceiver 708), a radio frequency front end 710, and one or more antennas 712. The multipurpose signal receiver 704 includes a deep neural network 714, a wireless transceiver 716 (transceiver 716), a radio frequency front end 718, and one or more antennas 720. In FIG. 7-1 , deep neural network 706 of multipurpose signal transmitter 702 and deep neural network 714 of multipurpose signal receiver 704 represent a pair of deep neural networks that enable transmission of a multipurpose signal 722 and reception of a reflected version of the multipurpose signal 722 (e.g., a reflected multipurpose signal 724) to enable simultaneous cooperative bistatic radar sensing and wireless communication.

[0092] 3, the base station 120 operates as a multi-purpose signal transmitter 702 and the user equipment 110 operates as a multi-purpose signal receiver 704 to form a first bistatic radar. Thus, the deep neural network 706 represents the deep neural network 460 of the base station 120, the transceiver 708 represents the transceiver 446 of the base station 120, the radio frequency front end 710 represents the radio frequency front end 444 of the base station 120, and the antenna 712 represents the antenna 442 of the base station 120. Also, the deep neural network 714 represents the deep neural network 420 of the user equipment 110, the transceiver 716 represents the transceiver 406 of the user equipment 110, the radio frequency front end 718 represents the radio frequency front end 404 of the user equipment 110, and the antenna 720 represents the antenna 402 of the user equipment 110.

[0093] In another example implementation, the user equipment 110 operates as a multi-purpose signal transmitter 702 and the base station 120 operates as a multi-purpose signal receiver 704 to form a second bistatic radar. Thus, the deep neural network 706 represents the deep neural network 420 of the user equipment 110, the transceiver 708 represents the transceiver 406 of the user equipment 110, the radio frequency front end 710 represents the radio frequency front end 404 of the user equipment 110, and the antenna 712 represents the antenna 402 of the user equipment 110. Also, the deep neural network 714 represents the deep neural network 460 of the base station 120, the transceiver 716 represents the transceiver 446 of the base station 120, the radio frequency front end 718 represents the radio frequency front end 444 of the base station 120, and the antenna 720 represents the antenna 442 of the base station 120.

[0094] A neural network manager (e.g., neural network manager 456 or 416) configures the deep neural network 706 according to the transmit (TX) radar and a configuration 726 (configuration 726) of a communication neural network (NN) formation that may be stored by a neural network table (e.g., neural network table 454 or 414). The deep neural network 706 uses the configuration 726 to perform signal generation for simultaneous cooperative bistatic radar sensing and wireless communication. In general, the deep neural network 706 performs some or all of the functionality of the radar signal transmitter 602 for generating the radar signal 310 and the processing functions of the transmitter for transmitting wireless communication data. For example, the deep neural network 706 can implement a waveform generation stage, a coherence stage, a modulation stage, a digital beamforming stage, an encoding stage, or a multiplexing stage.

[0095] A neural network manager (e.g., neural network manager 416 or 456) configures deep neural network 714 according to the receiving (RX) radar and a configuration 728 (configuration 728) of a communication neural network (NN) formation that may be stored by a neural network table (e.g., neural network table 414 or 454). Deep neural network 714 uses configuration 728 to perform signal processing for cooperative bistatic radar sensing and wireless communication. In general, deep neural network 714 performs some or all of the functionality of radar signal receiver 604 for processing radar signal 310 and the processing functions of the receiver for extracting wireless communication data. For example, deep neural network 420 can implement a demodulation stage, a clutter tracker, an interference mitigation stage, a Fourier transform stage (e.g., a fast Fourier transform (FFT) stage), a threshold detection stage, a range processing stage, a Doppler processing stage, a digital beamforming stage, a decoding stage, or a demultiplexing stage.

[0096] During simultaneous cooperative bistatic radar sensing and wireless communication, the multipurpose signal transmitter 702 transmits a multipurpose signal 722 representing a combination of the radar signal 310 and the wireless communication signal. The wireless communication signal may be a mobile communication reference signal, such as a standalone reference signal, a reference preamble preceding a data signal, and / or a reference signal interleaved in time and frequency with the data signal. In particular, the deep neural network 706 receives the radar waveform characteristics 626 and the reference signal parameters 730 and generates a digital multipurpose signal 732 based on the radar waveform characteristics 626 and the reference signal parameters 730. In one aspect, the deep neural network 706 generates the digital radar signal 628 (of FIG. 6-1) and multiplexes the reference signal parameters 730 onto the digital radar signal 628 (e.g., using PHY-layer multiplexing or another radio bearer). In another aspect, the deep neural network 706 modulates a reference signal onto the digital radar signal 628 (or modulates the digital radar signal 628 onto the reference signal). The reference signal, when associated with an uplink reference signal, may be a sounding reference signal (SRS). Alternatively, the reference signal, when associated with a downlink reference signal or synchronization signal, may be a demodulation reference signal, a phase tracking reference signal, a channel state information reference signal, a tracking reference signal, a primary synchronization signal, or a secondary synchronization signal. The resulting signal transmitted by the antenna 712 is a multi-purpose signal 722 and may optionally be used for cooperative bistatic radar sensing and / or wireless communication.

[0097] Multi-purpose signal receiver 704 receives a reflected version of multi-purpose signal 722, shown as reflected multi-purpose signal 724. Deep neural network 714 receives a digital reflected multi-purpose signal 734, which is a digital representation of reflected multi-purpose signal 734. Deep neural network 714 processes digital reflected multi-purpose signal 734 to reconstruct reference signal parameters 730 as reference signal parameters 736 to generate radar data 632.

[0098] Sometimes, a device (e.g., user equipment 110 or base station 120) includes both a multipurpose signal transmitter 702 and a multipurpose signal receiver 704. In this case, the device can use the multipurpose signal transmitter 702 and the multipurpose signal receiver 704 for monostatic radar sensing. In particular, the multipurpose signal receiver 704 processes another reflected version of the multipurpose signal 722 to determine explicit information about the object 180. Further operations for simultaneous cooperative bistatic radar sensing and wireless communication are further described with reference to FIG. 7-2.

[0099] FIG. 7-2 illustrates an example operation of a multipurpose feedback signal transmitter 738 and a multipurpose feedback signal receiver 740 for simultaneous cooperative bistatic radar sensing and wireless communication. The multipurpose feedback signal transmitter 738 is co-located with the multipurpose signal receiver 704, and the multipurpose feedback signal receiver 740 is co-located with the multipurpose signal transmitter 702. The multipurpose feedback signal transmitter 738 includes a deep neural network 742, a wireless transceiver 744 (transceiver 744), a radio frequency front end 746, and one or more antennas 748. The multipurpose feedback signal receiver 740 includes a deep neural network 750, a wireless transceiver 752 (transceiver 752), a radio frequency front end 754, and one or more antennas 756. In FIG. 7-2, the deep neural network 742 of the multi-purpose feedback signal transmitter 738 and the deep neural network 750 of the multi-purpose feedback signal receiver 740 represent a pair of deep neural networks that enable both transmission and reception of a multi-purpose feedback signal 758 for simultaneous cooperative bistatic radar sensing and wireless communication.

[0100] In some implementations, the multi-purpose feedback signal transmitter 738 uses different components than the multi-purpose signal receiver 704. For example, the deep neural network 742 of the multi-purpose feedback signal transmitter 738 is different from the deep neural network 714 of the multi-purpose signal receiver 704. In this case, the deep neural network 742 is designed and trained to generate a multi-purpose feedback signal, and the deep neural network 714 is designed and trained to process a multi-purpose signal. Similarly, the multi-purpose feedback signal receiver 740 can use different components than the multi-purpose signal transmitter 702. In particular, the deep neural network 750 of the multi-purpose feedback signal receiver 740 is different from the deep neural network 706 of the multi-purpose signal transmitter 702. In this case, the deep neural network 750 is designed and trained to process a multi-purpose feedback signal, and the deep neural network 706 is designed and trained to generate a multi-purpose signal.

[0101] In other implementations, the multipurpose feedback signal transmitter 738 uses the same components as the multipurpose signal receiver 704 uses. For example, the deep neural network 742 of the multipurpose feedback signal transmitter 738 and the deep neural network 714 of the multipurpose signal receiver 704 can be integrated to form a single deep neural network that serves both the multipurpose feedback signal transmitter 738 and the multipurpose signal receiver 704. The multipurpose feedback signal transmitter 738 and the multipurpose signal receiver 704 can also share the same transceiver, radio frequency front end, antenna, or some combination thereof. The multipurpose feedback signal receiver 740 can also use the same components as the multipurpose signal transmitter 702 uses. For example, the deep neural network 750 of the multipurpose feedback signal receiver 740 and the deep neural network 706 of the multipurpose signal transmitter 702 can be the same deep neural network that reuses the same NN nodes with different configurations. The multipurpose feedback signal receiver 740 and the multipurpose signal transmitter 702 may also share the same transceiver, radio frequency front end, antenna, or some combination thereof.

[0102] When the base station 120 and the user equipment 110 form a first bistatic radar as described above with reference to FIG. 7-1, the user equipment 110 operates as a multi-purpose feedback signal transmitter 738 and the base station 120 operates as a multi-purpose feedback signal receiver 740. Thus, the deep neural network 742 represents the deep neural network 420 of the user equipment 110, the transceiver 744 represents the transceiver 406 of the user equipment 110, the radio frequency front end 746 represents the radio frequency front end 404 of the user equipment 110, and the antenna 748 represents the antenna 402 of the user equipment 110. Additionally, deep neural network 750 represents deep neural network 460 of base station 120, transceiver 752 represents transceiver 446 of base station 120, radio frequency front end 754 represents radio frequency front end 444 of base station 120, and antenna 756 represents antenna 442 of base station 120.

[0103] In the case where the base station 120 and the user equipment 110 form a second bistatic radar, the base station 120 operates as a multi-purpose feedback signal transmitter 738 and the user equipment 110 operates as a multi-purpose feedback signal receiver 740. Thus, the deep neural network 742 represents the deep neural network 460 of the base station 120, the transceiver 744 represents the transceiver 446 of the base station 120, the radio frequency front end 746 represents the radio frequency front end 444 of the base station 120, and the antenna 748 represents the antenna 442 of the base station 120. Also, the deep neural network 750 represents the deep neural network 420 of the user equipment 110, the transceiver 752 represents the transceiver 406 of the user equipment 110, the radio frequency front end 754 represents the radio frequency front end 404 of the user equipment 110, and the antenna 756 represents the antenna 402 of the user equipment 110.

[0104] A neural network manager (e.g., neural network manager 416 or 456) configures deep neural network 742 according to a configuration 760 (configuration 760) for forming a transmit (TX) feedback neural network (NN), which may be stored by a neural network table (e.g., neural network table 414 or 454). Deep neural network 742 uses configuration 760 to perform signal generation for simultaneous cooperative bistatic radar sensing and wireless communication. In general, deep neural network 742 performs some or all of the processing functions of a transmitter to generate feedback signal 340 and transmit wireless communication data. In particular, deep neural network 742 may implement encoding, multiplexing, and modulation stages.

[0105] A neural network manager (e.g., neural network manager 456 or 416) configures deep neural network 750 according to a configuration 762 (configuration 762) for forming a receive (RX) feedback neural network (NN), which may be stored by a neural network table (e.g., neural network table 454 or 414). Deep neural network 750 uses configuration 762 to perform signal processing for simultaneous cooperative bistatic radar sensing and wireless communication. In general, deep neural network 750 performs some or all of the processing functions of a receiver to process feedback signal 340 and extract wireless communication data. In particular, deep neural network 750 may implement a decoding stage, a demultiplexing stage, and a demodulation stage.

[0106] During the simultaneous cooperative bistatic radar sensing and wireless communication, the multi-purpose feedback signal transmitter 738 transmits a multi-purpose feedback signal 758 representing a combination of the feedback signal 340 and the wireless communication signal. As previously described, the wireless communication signal can be a reference signal for mobile communication. In particular, the deep neural network 742 receives the radar data 632 and the reference signal parameters 764 and generates a digital multi-purpose feedback signal 766 based on the radar data 632 and the reference signal parameters 764. In one aspect, the deep neural network 742 generates the digital feedback signal 658 (of FIG. 6-2) and multiplexes the reference signal parameters 764 onto the digital feedback signal 658 (e.g., using PHY-layer multiplexing or another radio bearer). In another aspect, the deep neural network 742 modulates a reference signal onto the digital feedback signal 658 (or modulates the digital feedback signal 658 onto the reference signal). The reference signal can be a sounding reference signal when associated with an uplink reference signal. Alternatively, the reference signal, when associated with a downlink reference signal or synchronization signal, may be a demodulation reference signal, a phase tracking reference signal, a channel state information reference signal, a tracking reference signal, a primary synchronization signal, or a secondary synchronization signal. The resulting signal transmitted by antenna 748 is a multi-purpose feedback signal 758, which may optionally be used for cooperative bistatic radar sensing and / or wireless communication.

[0107] The multi-purpose feedback signal receiver 740 receives the multi-purpose feedback signal 758. The deep neural network 750 receives a digital multi-purpose feedback signal 768, which is a digital representation of the multi-purpose feedback signal 758. The multi-purpose feedback signal receiver 740 processes the digital multi-purpose feedback signal 768 to reconstruct the reference signal parameters 764 as reference signal parameters 770 to generate the object data 662.

[0108] 8 shows an example transaction diagram 800 between a first device 802 and a second device 804 for cooperative bistatic radar sensing. In this case, the first device 802 operates as a radar signal transmitter 602 and a feedback signal receiver 636. The second device 804 operates as a radar signal receiver 604 and a feedback signal transmitter 634. In this example, the first device 802 is a base station 120 and the second device 804 is a user equipment 110. In another example, the first device 802 is a user equipment 110 and the second device 804 is a base station 120. In this case, the second device 815 (e.g., the base station 120) can perform operations at 805 and 815 on behalf of the first device 802, and the first device 802 (e.g., the user equipment 110) can perform operations at 810 on behalf of the second device 804.

[0109] At 805, the first device 802 configures the deep neural network of the second device 804 for cooperative bistatic radar sensing. In particular, the first device 802 transmits a configuration message 806 to the second device 804. The configuration message 806 may include components of a particular NN formation or an index to a neural network table (e.g., an index to the configuration 624 of the receive radar NN formation and the configuration 654 of the transmit feedback NN formation).

[0110] The configuration message 806 may also include information about the first device 802 or about the upcoming radar signal 310. For example, the information about the first device 802 may include the location, movement, or antenna configuration of the first device 802. The information about the upcoming radar signal 310 may include radar waveform characteristics 626 or timing information. In some aspects, the first device 802 communicates the configuration message 806 using a frequency band different from the frequency band used for cooperative bistatic radar sensing. As an example, the frequency band may be associated with a side channel.

[0111] At 810, the second device 804 communicates its availability to the first device 802. In particular, the second device 804 transmits an availability message 812 that provides information regarding the second device 804's ability to support cooperative bistatic radar sensing. The availability message 812 may include information regarding the second device 804's battery level, processing capabilities, antenna configuration, and transmit power level. The operations of 805 and 810 may occur simultaneously or in different orders.

[0112] At 815, the first device 802 activates one or more second devices 804 for cooperative bistatic radar sensing. For example, the first device 802 transmits an activation message 816 to the second device 804 to cause the second device 804 to support cooperative bistatic radar sensing.

[0113] In some cases, the first device 802 may select a group of second devices 804. This selection may be designed to obtain additional information about a particular object 180 in the environment based on the relative proximity of the second devices 804 in the group or the position of the second devices 804 along a particular angle. Typically, the first device 802 selects multiple second devices 804 that are likely to observe radar signal 310 reflections from the same object 180. The activation message 816 may be implemented as a beam-specific paging message.

[0114] At 820, the first device 802 performs radar signal generation using transmit deep neural network processing. For example, the deep neural network 606 of the radar signal transmitter 602 uses the transmit radar NN formation configuration 622 as described above with reference to FIG. 6-1 to generate a digital radar signal 628 based on the radar waveform characteristics 626. Sometimes, the first device 802 supports simultaneous cooperative bistatic radar sensing and wireless communication. In this situation, the deep neural network 706 of the multipurpose signal transmitter 702 uses the transmit radar and communication NN formation configuration 726 as described above with reference to FIG. 7-1 to generate a digital multipurpose signal 732 based on the radar waveform characteristics 626 and reference signal parameters 730.

[0115] At 825, the first device 802 transmits a radar signal 310, as described above with reference to Figures 3 and 6-1. The radar signal 310 is reflected from the object 180. The second device 804 receives a reflected version of the radar signal 310 (e.g., the reflected radar signal 332).

[0116] At 830, the second device 804 performs radar signal processing using receive deep neural network processing. For example, the deep neural network 614 of the radar signal receiver 604 uses the receive radar NN formation configuration 624 as described above with reference to FIG. 6-1 to generate radar data 632 based on the digital reflected radar signal 630. Sometimes, the second device 804 supports simultaneous cooperative bistatic radar sensing and wireless communication. In this situation, the deep neural network 714 of the multi-purpose signal receiver 704 uses the receive radar and communication NN formation configuration 728 as described above with reference to FIG. 7-1 to generate the radar data 632 and recover the reference signal parameters 730 as reference signal parameters 736.

[0117] At 835, the second device 804 performs signal generation using a transmit deep neural network process. For example, the deep neural network 638 of the feedback signal transmitter 634 uses the transmit feedback NN formation configuration 654 as described above with reference to FIG. 6-2 to generate a digital feedback signal 658 based on the radar data 632. Sometimes, the second device 804 supports simultaneous cooperative bistatic radar sensing and wireless communication. In this situation, the deep neural network 742 of the multi-purpose feedback signal transmitter 738 uses the transmit feedback NN formation configuration 760 as described above with reference to FIG. 7-2 to generate a digital multi-purpose feedback signal 766 based on the radar data 632 and the reference signal parameters 764.

[0118] At 840, the second device 804 transmits the feedback signal 340, as described above with reference to Figures 3 and 6-2. The second device 804 may transmit the feedback signal 340 using the same frequency band or a different frequency band than that of the radar signal 310. The first device 802 receives the feedback signal 340.

[0119] At 845, the first device 802 performs signal processing using a receiving deep neural network processing. For example, the deep neural network 646 of the feedback signal receiver 636 uses the receiving feedback NN formation configuration 656 to generate object data 662 based on the digital feedback signal 660. Sometimes, the first device 802 supports simultaneous cooperative bistatic radar sensing and wireless communication. In this situation, the deep neural network 750 of the multi-purpose feedback signal receiver 740 uses the receiving feedback NN formation configuration 762 to generate the object data 662 and recovers the reference signal parameters 764 as the reference signal parameters 770.

[0120] The control and feedback signals (e.g., configuration message 806, availability message 812, activation message 816, and feedback signal 340) may occur in a frequency band different from the frequency band associated with the radar signal 310. For example, the frequency band for the control and feedback signals may be a sub-6 GHz frequency band, while the frequency band for the radar signal 310 may be in a frequency band associated with millimeter wavelengths (e.g., including frequencies higher than 6 GHz, 30 GHz, or 100 GHz).

[0121] Example Method 9 and 10 depict example methods 900 and 1000 for performing cooperative bistatic radar sensing operations using deep neural networks. In some of the following discussion, the environment 200 of FIG. 2 and the environment 300 of FIG. 3, as well as the entities detailed in FIG. 1, FIG. 4, or FIG. 8 are referenced for illustrative purposes only. The techniques are not limited to performance by one entity or multiple entities operating in one device. For purposes of this example, a base station performs the radar signal transmitter and feedback signal receiver method, and one or more user equipment performs the radar signal receiver and feedback signal transmitter method. However, in other implementations, the user equipment performs the radar signal transmitter and feedback signal receiver method, and at least one base station performs the radar signal receiver and feedback signal transmitter method.

[0122] At 902 in FIG. 9, the user equipment receives a reflected version of a radar signal. The radar signal is transmitted by a base station and reflected from an object. For example, the user equipment 110 receives a reflected radar signal 332, which represents a reflected version of the radar signal 310, as shown in FIG. 3. The radar signal 310 is reflected from the object 180. As an example, the user equipment 110 can represent the user equipment 112 or 113 shown in FIG. 2, and the object 180 can represent the object 183. In this case, the radar signal 310 follows a propagation path 224 or 228. The user equipment 110 can also receive the radar signal 310 through line-of-sight propagation.

[0123] At 904, the user equipment generates radar data by processing the received radar signal using a deep neural network. The radar data includes information about the object. For example, the user equipment 110 generates radar data 632 by processing the reflected radar signal 332 using a deep neural network 614 of the radar signal receiver 604 as shown in FIG. 6-1. The user equipment 110 can also process the radar signal 310 directly and demodulate the reflected radar signal 332 using the line-of-sight radar signal 310.

[0124] The radar data 632 may include implicit or explicit information about the object 180. For example, the radar data 632 may include raw digital samples of the digital reflected radar signal 630, which includes a phase based on the position of the digital object 180, a frequency based on the movement of the object 180, and an amplitude based on both the position of the object 180 and the material composition of the object 180. As another example, the radar data 632 may include processed information, such as amplitude and / or phase information, complex range-Doppler maps, or interferometric data, about the digital reflected radar signal 630. Additionally or instead, the radar data 632 may include explicit information, such as location information, movement information, size information, or material composition information, about the object 180.

[0125] At 906, the user equipment generates a feedback signal using a deep neural network. The feedback signal is based on the radar data. For example, the user equipment 110 generates the feedback signal 340 using a deep neural network 638 of the feedback signal transmitter 634 as shown in FIG. 6-2. The deep neural network 638 used to generate the feedback signal 340 may be the same or a different deep neural network than the deep neural network 614 used to process the reflected radar signal 332. The feedback signal 340 is based on the radar data 632. In some cases, the feedback signal 340 may include the radar data 632.

[0126] At 908, the user equipment transmits a feedback signal to the base station. For example, the user equipment 110 transmits a feedback signal 340 to the base station 120 as shown in FIG.

[0127] At 1002 in Figure 10, a base station generates a radar signal using a deep neural network. For example, the base station 120 generates a digital radar signal 628 using the deep neural network 606 of the radar signal transmitter 602 as shown in Figure 6-1. In particular, the base station 120 generates the digital radar signal 628 based on the radar waveform characteristics 626.

[0128] At 1004, the base station transmits a radar signal, which is reflected from an object. For example, as shown in Figures 3, 6-1, and 8, the base station 120 transmits a radar signal 310. The radar signal 310 can be a frequency modulated signal or a phase modulated signal. The radar signal 310 is reflected from the object 180 as shown in Figure 3.

[0129] At 1006, the base station receives a feedback signal from the user equipment. The feedback signal is based on radar data generated by the user equipment processing the radar signal. For example, the base station 120 receives a feedback signal 340 from the user equipment 110. The feedback signal 340 includes radar data 632 generated by the user equipment 110 based on the radar signal 310 (e.g., the reflected radar signal 332). The radar data 632 includes implicit or explicit information about the object 180.

[0130] At 1008, the base station determines information about the object by processing the radar data using a deep neural network. For example, the base station 120 uses the deep neural network 646 of the feedback signal receiver 636 to process the radar data 632 in the feedback signal 340 to generate object data 662. The object data 662 includes explicit information about the object 180, such as location information, movement information, size information, and / or material composition information.

[0131] Methods 900 and 1000 are depicted as a set of operations (or acts) to be performed, but are not necessarily limited to the order or combination of operations depicted herein. Additionally, any one or more of these operations may be repeated, combined, rearranged, skipped, or linked to provide a variety of additional and / or alternative methods.

[0132] conclusion Although the apparatus including and techniques for using cooperative bistatic radar sensing using deep neural networks have been described in feature and / or method specific language, it is understood that the subject matter of the appended claims is not necessarily limited to the particular features or methods described. Rather, the specific features and methods are disclosed as example implementations of cooperative bistatic radar sensing using deep neural networks.

[0133] Some examples are provided below. Example 1: A method performed by a first device, the method comprising: and operating as a radar signal receiver for a bistatic radar, the operating as a radar signal receiver comprising: receiving a reflected version of a radar signal transmitted from a second device associated with the bistatic radar transmitter, the radar signal being transmitted by the second device associated with the bistatic radar transmitter and reflected from an object; and operating as a radar signal receiver comprising: generating radar data including information about the object by processing a reflected version of the radar signal using a deep neural network; The method further includes operating as a feedback signal transmitter, the operating as a feedback signal transmitter comprising: generating a feedback signal based on the radar data using a deep neural network; and transmitting a feedback signal to the second device.

[0134] Example 2: The method according to example 1, the deep neural network includes a first deep neural network and a second deep neural network; generating the radar data includes generating the radar data using a first deep neural network; The method, wherein generating the feedback signal includes generating the feedback signal using a second deep neural network.

[0135] Example 3: The method of example 1 or example 2, wherein generating the feedback signal comprises: receiving radar data by a deep neural network; generating, by a deep neural network, digital samples related to the feedback signal based on the radar data.

[0136] Example 4: The method according to any one of the preceding examples, wherein the radar data is: Unprocessed digital samples of the reflected version of the radar signal, Range Doppler Map, or The method includes at least one of:

[0137] Example 5: The method according to any one of the preceding examples, wherein the radar data is Location information associated with the object; Movement information associated with the object; Size information relating to the object, or The method includes at least one of material composition information associated with the object.

[0138] Example 6: The method of any one of the preceding examples, comprising: The feedback signal is The location of the first device, The speed of the first device, the orientation of the first device, or The method further indicates at least one of the antenna configurations of the first device.

[0139] Example 7: The method of any one of the preceding examples, comprising: modulating a reference signal onto the feedback signal; or The method further includes modulating the feedback signal onto the reference signal.

[0140] Example 8: The method according to Example 7, The method wherein the reference signal includes an uplink reference signal or a downlink reference signal.

[0141] Example 9: The method according to Example 8, the reference signal comprises an uplink reference signal; The method includes a method in which the uplink reference signal includes a sounding reference signal.

[0142] Example 10: The method according to Example 8, the reference signal comprises a downlink reference signal; The method is such that the downlink reference signal includes a primary synchronization signal, a secondary synchronization signal, a demodulation reference signal, a phase tracking reference signal, a channel state information reference signal, or a tracking reference signal.

[0143] Example 11: The method of any one of the preceding examples, comprising: receiving a configuration message from a second device; and modifying a configuration of the neural network formation of the deep neural network based on the configuration message.

[0144] Example 12: The method of any one of the preceding examples, comprising: The radar signal is associated with a first frequency band; The method, wherein transmitting the feedback signal includes transmitting the feedback signal using a second frequency band different from the first frequency band.

[0145] Example 13: The method according to any one of the preceding examples, comprising: The method wherein transmitting the feedback signal includes transmitting the feedback signal using the assigned time slot.

[0146] Example 14: A method performed by a first device, the method comprising: and operating as a radar signal transmitter of a bistatic radar, the operating as a radar signal transmitter comprising: generating a radar signal using a deep neural network; transmitting a radar signal, the radar signal being reflected from an object; The method further includes operating as a feedback signal receiver, the operating as a feedback signal receiver comprising: The method further includes receiving a feedback signal from a second device, the feedback signal being based on radar data generated by the second device by processing the radar signal, and operating as a feedback signal receiver further includes: and processing the radar data using a deep neural network to determine information about the object.

[0147] Example 15: The method according to example 14, the deep neural network includes a first deep neural network and a second deep neural network; generating a radar signal includes generating the radar signal using a first deep neural network; The method, wherein determining information about the object includes determining the information by processing the radar data using a second deep neural network.

[0148] Example 16: The method according to example 14 or 15, The step of generating a radar signal includes: receiving, by a deep neural network, at least one radar waveform characteristic; generating digital samples based on at least one radar waveform characteristic by a deep neural network; and generating a radar signal using the digital samples.

[0149] Example 17: The method according to example 16, The at least one radar waveform characteristic includes: center frequency, Bandwidth, Pulse repetition frequency, Beamforming configuration, or at least one of the modulation types.

[0150] Example 18: The method according to any one of Examples 14 to 17, modeling a propagation path within an operating environment based on information about the object; and adjusting a beamforming configuration associated with the wireless communication based on the modeled propagation paths.

[0151] Example 19: The method according to any one of Examples 14 to 18, modulating a reference signal onto the radar signal; or The method further includes modulating the radar signal onto a reference signal.

[0152] Example 20: The method according to Example 19, The method wherein the reference signal includes an uplink reference signal or a downlink reference signal.

[0153] Example 21: The method according to example 20, the reference signal comprises an uplink reference signal; The method includes a method in which the uplink reference signal includes a sounding reference signal.

[0154] Example 22: The method according to example 20, the reference signal comprises a downlink reference signal; The method is such that the downlink reference signal includes a primary synchronization signal, a secondary synchronization signal, a demodulation reference signal, a phase tracking reference signal, a channel state information reference signal, or a tracking reference signal.

[0155] Example 23: The method according to any one of Examples 14 to 22, selecting a plurality of devices including a second device; and transmitting an activation message to the plurality of devices to enable the plurality of devices to receive a reflected version of the radar signal and to transmit respective feedback signals.

[0156] Example 24: The method according to any one of Examples 14 to 23, receiving a reflected version of the radar signal; and determining additional information about the object by processing a reflected version of the radar signal using a deep neural network.

[0157] Example 25: The method according to any one of Examples 14 to 24, receiving a reflected version of a feedback signal; and determining other additional information about the object by processing a reflected version of the feedback signal using a deep neural network.

[0158] Example 26: A device, At least one antenna; At least one transceiver; At least one processor; and at least one computer-readable storage medium, the at least one computer-readable storage medium including instructions, in response to being executed by a processor, for directing a device to perform a method as described in any one of Examples 1 to 25.

[0159] Example 27: The device of example 26, A user device is provided.

[0160] Example 28: The device of example 26, Equipped with a base station.

[0161] Example 29: A computer-readable storage medium, comprising: The method includes instructions that, in response to being executed by a processor, cause an apparatus having the processor to perform a method as described in any one of Examples 1 to 25.

Claims

1. 1. A method performed by a first device, the method comprising: and operating as a radar signal receiver for a bistatic radar, the operating as the radar signal receiver comprising: The method includes the steps of: receiving a reflected version of a radar signal, the radar signal being transmitted by a second device associated with the bistatic radar transmitter and reflected from an object; and operating as the radar signal receiver includes: processing the reflected version of the radar signal using a deep neural network to generate radar data that includes information about the object; The method further includes operating as a feedback signal transmitter, the operating as a feedback signal transmitter comprising: using the deep neural network to generate a feedback signal based on the radar data; transmitting the feedback signal to the second device; the radar signal is associated with a first frequency band; The step of transmitting the feedback signal comprises:

4. The method of claim 3, further comprising transmitting the feedback signal using a second frequency band different from the first frequency band.

2. the deep neural network includes a first deep neural network and a second deep neural network; generating the radar data includes generating the radar data using the first deep neural network; The method of claim 1 , wherein generating the feedback signal comprises generating the feedback signal using the second deep neural network.

3. The step of generating a feedback signal comprises: receiving the radar data by the deep neural network; and generating, by the deep neural network, digital samples related to the feedback signal based on the radar data.

4. The radar data is unprocessed digital samples of the reflected version of the radar signal; Range Doppler Map, or at least one of: interferometric data; The information in the radar data is: location information relating to said object; movement information relating to said object; size information associated with the object; or The method of claim 1 , further comprising at least one of material composition information associated with the object.

5. receiving a configuration message from the second device; and modifying a configuration of a neural network formation of the deep neural network based on the configuration message.

6. The step of transmitting the feedback signal comprises: The method of claim 1 , further comprising the step of transmitting the feedback signal using an assigned time slot.

7. 1. A method performed by a first device, the method comprising: and operating as a radar signal transmitter of a bistatic radar, the operating as a radar signal transmitter comprising: generating a radar signal using a deep neural network; transmitting the radar signal, the radar signal being reflected from an object; The method further includes operating as a feedback signal receiver, the operating as a feedback signal receiver comprising: The method further comprises the step of receiving a feedback signal from a second device, the feedback signal being based on radar data generated by the second device processing the radar signal, and operating as the feedback signal receiver further comprising: determining information about the object by processing the radar data using the deep neural network; modeling a propagation path within an operating environment based on the information about the object; and adjusting a beamforming configuration associated with wireless communication based on the modeled propagation paths.

8. the deep neural network includes a first deep neural network and a second deep neural network; generating the radar signal includes generating the radar signal using the first deep neural network; The method of claim 7 , wherein determining information about the object comprises determining the information by processing the radar data using the second deep neural network.

9. The step of generating a radar signal includes: receiving, by the deep neural network, at least one radar waveform characteristic; generating, by the deep neural network, digital samples based on the at least one radar waveform characteristic; and generating the radar signal using the digital samples.

10. The at least one radar waveform characteristic is: center frequency, Bandwidth, Pulse repetition frequency, Beamforming configuration, or 10. The method of claim 9, comprising at least one of a modulation type.

11. selecting a plurality of devices including the second device; and transmitting an activation message to the plurality of devices to enable the plurality of devices to receive a reflected version of the radar signal and to transmit respective feedback signals.

12. receiving a reflected version of the radar signal or a reflected version of the feedback signal; and determining additional information about the object by processing the reflected version of the radar signal or the reflected version of the feedback signal using the deep neural network.

13. A device, comprising: At least one antenna; At least one transceiver; At least one processor; and at least one computer-readable storage medium, the at least one computer-readable storage medium comprising instructions, in response to being executed by the at least one processor, for directing the device to perform the method of any one of claims 1 to 12.

14. User equipment, or The device of claim 13 comprising a base station.

15. A computer program comprising instructions which, when executed by a processor, cause an apparatus comprising said processor to perform a method according to any one of claims 1 to 12.

Images