Abstract: An antenna device is formed in such a manner that reception antennas are arranged at regular intervals between two transmission antennas adjacent to each other among transmission antennas, and a spacing between the transmission antenna and the transmission antenna has a width obtained by adding an integral multiple of a spacing dRx between each two of the reception antennas to a width obtained by dividing the spacing dRx by the number NTx of the transmission antennas.

Abstract: A system for using wireless signals to detect objects of a certain type, such as concealed weapons, leverages the fact that different objects have different resonance characteristics. The system transmits wireless signals of different frequencies, and returns from such signals are measured. The return of a signal from an object at a resonant frequency of the object will be stronger than a return of a signal from the object at a non-resonant frequency. The system analyzes the returns in an effort to determine when a return is sufficiently large to indicate that it was reflected from an object of interest. When an object of interest is detected, the system adjusts a characteristic of the system, such as antenna orientation or pulse shape, based on an estimated location of the object, and then runs a test to confirm the detection of the object, thereby eliminating at least some false positives.

Abstract: A monitoring system for a combine harvester having a header for harvesting a crop and a residue spreading system for spreading a crop residue. The monitoring system includes a sensing system configured to provide one or more measurement waves that intersect a flow of crop residue discharged by the spreading system and receive a plurality of response waves reflected from the crop residue. The system further includes a processing unit configured to receive a response signal of the sensing system; process the response signal; and determine, based on the response signal, a density and velocity distribution of the crop residue across a two-dimensional measurement area. The response signal is representative of the plurality of response waves reflected from the crop residue. The measurement area is at an end of a trajectory of the crop residue towards a deposit area.

Abstract: A method for automatic detection of antenna site conditions, ASC, at an antenna site, AS, of an antenna, A, the method comprising the steps of providing (S1) signal source observations, SSO, derived from signals received by the antenna, A, from at least one signal source, SS, and transforming (S2) the signal source observations, SSO, into images fed to a trained image-processing artificial intelligence, AI, model which calculates antenna site conditions, ASC, at an antenna site, AS, of the respective antenna, A.

Abstract: A method and apparatus for determining a velocity and a direction of arrival (DoA) of an object includes transmitting a signal using a determined signal transmission order of transmission antennas, receiving a reflected signal with respect to the transmitted signal using reception antennas, generating virtual antennas based on an arrangement of the transmission antennas and an arrangement of the reception antennas, generating virtual signals with respect to the virtual antennas based on the reflected signal and the virtual antennas, calculating a first estimated phase and a second estimated phase that are different from each other based on the virtual signals, and determining the velocity and the DoA of the object concurrently based on the first estimated phase and the second estimated phase without performing additional processing.

Abstract: A computing system including a processor configured to train a synthetic aperture radar (SAR) classifier neural network. The SAR classifier neural network is trained at least in part by, at a SAR encoder, receiving training SAR range profiles that are tagged with respective first training labels, and, at an image encoder, receiving training two-dimensional images that are tagged with respective second training labels. Training the SAR classifier neural network further includes, at a shared encoder, computing shared latent representations based on the SAR encoder outputs and the image encoder outputs, and, at a classifier, computing respective classification labels based on the shared latent representations. Training the SAR classifier neural network further includes computing a value of a loss function based on the plurality of first training labels, the plurality of second training labels, and the plurality of classification labels and performing backpropagation based on the value of the loss function.

Abstract: This document describes techniques and systems for a partially-learned model for speed estimates in radar tracking. A radar system is described that determines radial-velocity maps of potential detections in an environment of a vehicle. The model uses a data cube to determine predicted boxes for the potential detections. Using the predicted boxes, the radar system determines Doppler measurements associated with the potential detections that correspond to the predicted boxes. The Doppler measurements are used to determine speed estimates for the predicted boxes based on the corresponding potential detections. These speed estimates may be more accurate than a speed estimate derived from the data cube and the model. Driving decisions supported by the speed estimates may result in safer and more comfortable vehicle behavior.

Abstract: Reception antennas include first antennas at positions different in a first direction, second antennas at positions different in a second direction perpendicular to the first direction, and a third antenna different from the first or second antenna. The first and second antennas include one overlapping antenna. The third antenna is arranged at a position different in the second direction from a position of the first antennas. The third antenna is arranged at a position a prescribed spacing apart in the first direction from a position of the second antennas. At least one spacing of the first antennas is the prescribed spacing. Transmission antennas include fourth antennas arranged in the first direction and fifth antennas arranged in the second direction. The fourth antennas and the fifth antennas include one overlapping antenna. A spacing of the fourth antennas is wider in the first direction than an aperture length of the first antennas.

Abstract: Systems and methods for detecting a body part like a human hand near a base station or a user equipment are disclosed. A plurality of radar pulses is transmitted from a communication device in succession and the reflected plurality of radar pulses is received sampled and adaptively processed to remove transmit and receive antenna mutual coupling and clutter from stationary objects near the body part. In one aspect the adaptive processing is accomplished with a single tap adaptive filter. The processed signal may be used to determine if there is a human body part near the communication device allowing the device to determine whether it is safe for the device to transmit a millimeter wave communication signal.

Abstract: A device for a radar sensor is disclosed, the device comprising: transmission circuitry configured to generate transmission signals with a linear frequency chirp modulation in a predetermined frequency band for output to a radar antenna; reception circuitry configured to receive reflection signals corresponding to reflection of the transmitted radar signals from one or more physical objects; and control circuitry configured to select a frequency range within said predetermined frequency band and/or a timing pattern for said transmission signals; wherein said device is configured to: receive a further signal from a further radar sensor; determine, from said further signal, a frequency range and/or timing pattern in use by said further radar sensor for transmission of further transmission signals; and select a frequency range within said predetermined frequency band and/or a timing pattern for said transmission signals which does not conflict with the frequency range and/or timing pattern of said further transm

Abstract: A radar device may include a radar receiver to receive a radio frequency (RF) radar signal and generate a digital signal based on the RF radar signal. The digital signal may comprise a plurality of signal segments. The radar device may include a neural network comprising a plurality of layers to process the plurality of signal segments. Each layer of the plurality of layers may have one or more neurons. The plurality of layers may process the plurality of signal segments using weighting factors having values selected from a predetermined set of discrete values. At least one neuron in an output layer of the plurality of layers may provide an output value that indicates whether a respective signal segment or a sample, associated with the at least one neuron, is overlaid with an interfering signal.

Abstract: The disclosure provides a method for estimating the nondirectional wave spectrum from the sea echoes of multiple HF radar frequencies. The method includes: dividing the radar detection area into a plurality of fan-shaped units at an equal range interval and angle interval according to the distance resolution and the angular resolution of an HF radar; obtaining the Doppler spectrum from the sea echo of a single radar frequency at a fan-shaped unit by performing the first fast Fourier transform (FFT) in distance dimension, the second FFT in Doppler frequency dimension and the digital beamforming; extracting the positive first-order peak and the negative first-order peak from the aforementioned Doppler spectrum by the peak-searching method; and selecting the stronger first-order peak ?R(1)(?); dividing the second-order spectrum on the stronger first-order peak side into an inner second-order spectrum and an outer second-order spectrum.

Abstract: A radar device for automotive applications comprises a radar circuit configured to process a radar signal that has a first signal portion and a second signal portion, wherein the first signal portion occupies a first frequency band and the second signal portion occupies a second frequency band that is separate from the first frequency band. An antenna device of the radar device comprises a first and second antenna element that are both coupled to a common signal port of the radar circuit and the radar device is configured to route both the first signal portion and the second signal portion via the common signal port between the radar circuit and the antenna device. The antenna device is a frequency selective antenna device that transduces the first signal portion via the first antenna element and not via the second antenna element and that transduces the second signal portion at least via the second antenna element.

Abstract: A radar device may include a memory to store a program associated with operating the radar device. The radar device may include a decoder to read the program from the memory, and generate a control value and a timestamp based at least in part on the program. The control value may be a value to be provided as an input to a component of the radar device at a time indicated by the timestamp. The radar device may include a first-in first-out (FIFO) buffer to store at least the control value and provide the control value as the input to the component of the radar device at the time indicated by the timestamp.

Abstract: Disclosed is a method and a corresponding distance-measuring device for measuring a distance to an object using FMCW radar. The method includes the frequency-dependent determination of the amplitude of the radar signal, i.e. the frequency response in the output path and in the input path of the distance-measuring device. The standard windowing of the evaluation signal can be corrected using a correction factor dependent on the frequency responses. Thus the frequency dependence of the radar signal is compensated independently of device-internal or external interferences by adapting the window function. The result is more accurate and reliable distance measurement using FMCW radar. Because the distance can be determined by the disclosed method very accurately and without distortion, it is advantageous to use the distance-measuring device as a fill-level measuring device to measure the fill level of a filling material in a container.

Abstract: The present disclosure relates to a method for safe and exact ascertaining of fill level of a fill substance located in a container by means of an ultrasonic, or radar-based, fill level measuring device. In such case, the method is distinguished by the feature that the evaluation curve created based on the reflected received signal is differently greatly smoothed as a function of measured distance. To achieve this, the evaluation curve can be specially filtered, depending on the application. In this way, noise fractions and disturbance echoes can be efficiently suppressed, without unnecessarily limiting the accuracy of the fill level measurement.

Abstract: A radar system with on-system calibration for cross-coupling and gain/phase variations includes capabilities for radar detection and correction for system impairments to improve detection performance. The radar system is equipped with pluralities of transmit antennas and pluralities of receive antennas. The radar system uses a series of calibration measurements of a known object to estimate the system impairments. A correction is then applied to the beamforming weights to mitigate the effect of these impairments on radar detection. The estimation and correction requires no external measurement equipment and can be computed on the radar system itself.

Abstract: Disclosed are techniques for employing deep learning to analyze radar signals. In an aspect, an on-board computer of a host vehicle receives, from a radar sensor of the vehicle, a plurality of radar frames, executes a neural network on a subset of the plurality of radar frames, and detects one or more objects in the subset of the plurality of radar frames based on execution of the neural network on the subset of the plurality of radar frames. Further, techniques for transforming polar coordinates to Cartesian coordinates in a neural network are disclosed. In an aspect, a neural network receives a plurality of radar frames in polar coordinate space, a polar-to-Cartesian transformation layer of the neural network transforms the plurality of radar frames to Cartesian coordinate space, and the neural network outputs the plurality of radar frames in the Cartesian coordinate space.

Abstract: A method of determining a filling level of a product in a tank having a tank wall, the method comprising: generating and transmitting an electromagnetic first transmit signal; propagating the first transmit signal through a tank atmosphere towards a corner reflector formed by the surface of the product and the tank wall where the surface of the product meets the tank wall, the corner reflector being at a known horizontal distance from the radar level gauge system; receiving an electromagnetic first reflection signal resulting from reflection of the first transmit signal at the corner reflector; determining a measure indicative of a propagation direction of the first reflection signal; and determining the filling level based on the measure indicative of the propagation direction of the first reflection signal and the known horizontal distance between the radar level gauge system and the corner reflector.

Abstract: A radar device may include a radar transmitter to output a radio frequency (RF) transmission signal including a plurality of frequency-modulated chirps. The radar device may include a radar receiver to receive an RF radar signal, and generate, based on the RF radar signal, a dataset including a set of digital values, the dataset being associated with a chirp or a sequence of successive chirps. The radar device may include a neural network to filter the dataset to reduce an interfering signal included in the dataset, the neural network being a convolutional neural network. At least one layer of the neural network may be a complex-valued neural network layer includes complex-valued weighting factor, where the complex-valued neural network layer is configured to perform one or more operations according to a complex-valued computation.