Abstract: Methods, systems, and computer-readable storage media for a machine learning (ML) system for matching a query entity to one or more target entities, the ML system that reducing a number of query-target entity pairs from consideration as potential matches during inference.

Abstract: A radar sensor that includes a circuit board, a waveguide structure having at least one waveguide channel, and at least one pin that is pressed into the circuit board. The pin connects the circuit board to the at least one waveguide channel of the waveguide structure in such a way that radar signals may be coupled into the at least one waveguide channel and/or decoupled from the at least one waveguide channel.

Abstract: An electronic component for high-frequency applications, in which an integrated circuit with a chip for processing high-frequency signals are arranged together with at least one signal coupling element or launcher for coupling and/or decoupling high-frequency signals in a common housing substrate. A land grid array (LGA) structure is provided on an outside of the housing substrate. An electrically conductive border is provided around the respective launcher on the surface of the housing substrate.

Abstract: Computer-readable media, methods, and systems are disclosed for feature attribution in a machine learning model. Samples may be generated for a machine learning model based on a normalized probability distribution. The samples may be used to determine a weight for features and feature pairs for the machine learning model. The weights of the features and feature pairs may be used to determine which features are significant for predictions within the machine learning model.

Abstract: A radar system, which includes an antenna for sending and/or receiving radar signals and a circuit board including at least one high-frequency component and an element for sending and/or receiving high-frequency signals. The antenna includes at least one press-in pin, which is introduced into the antenna. At least one sleeve is situated on/at the circuit board, into which the press-in pin is inserted. The at least one press-in pin cooperates with the at least one sleeve in such a way that, when the antenna is being pushed onto the circuit board, the press-in pin is pressed into the at least one sleeve and deforms, providing fixation.

Abstract: An antenna device. The antenna device includes a carrier element having at least one first strip conductor and having a second strip conductor; at least one fastening structure, which is formed in or on the carrier element; at least one antenna element, which is arranged or fastened on or in the fastening structure and is connected to the strip conductor; a transmitter device, which is arranged on the carrier element and is connected to the second strip conductor, and is designed to transmit a transmitter signal to the at least one antenna element and/or to receive a transmitter signal from the at least one antenna element.

Abstract: Methods, systems, and computer-readable storage media for receiving a set of inference results generated by a ML model, the inference results including a set of query entities and a set of target entities, each query entity having one or more target entities matched thereto by the ML model, processing the set of inference results to generate a set of matched sub-sets of target entities by executing a search over target entities in the set of target entities based on constraints, for each problem in a set of problems, providing the problem as a tuple including an index value representative of a target entity in the set of target entities and a value associated with the query entity, the value including a constraint relative to the query entity, and executing at least one task in response to one or more matched sub-sets in the set of matched sub-sets.

Abstract: Methods, systems, and computer-readable storage media for receiving a set of inference results generated by a ML model, the inference results including a set of query entities and a set of target entities, each query entity having one or more target entities matched thereto by the ML model, processing the set of inference results to generate a set of matched sub-sets of target entities by executing a search over target entities in the set of target entities based on constraints, for each problem in a set of problems, providing the problem as a tuple including an index value representative of a target entity in the set of target entities and a value associated with the query entity, the value including a constraint relative to the query entity, and executing at least one task in response to one or more matched sub-sets in the set of matched sub-sets.

Abstract: A radar device having a plurality of transmit antennas and a plurality of receive antennas. The radar device further comprises at least one high-frequency (HF) radar chip which is configured to generate a high-frequency signal and transmit it via the transmit antennas. The transmit antennas have an offset with respect to the receive antennas in a first direction and at least one transmit antenna has an overlap with at least one receive antenna in a second direction. The first direction is a vertical direction and the second direction a horizontal direction. Alternatively, the first direction is a horizontal direction and the second direction a vertical direction.

Abstract: Methods, systems, and computer-readable storage media for receiving IRF data sets, the IRF data sets including a set of records including inference results determined by the ML model during production use of the ML model and at least one correction to an inference result, executing incremental training of the ML model to provide an updated ML model by selectively filtering one or more records of the set of records to adjust a negative sample to positive sample proportion of a sub-set of records based on a negative sample to positive sample proportion of initial training of the ML model, for each record in the sub-set of records, determining a weight, and during incremental training, applying the weight of a respective record being in a loss function in determining an accuracy of the ML model based on the respective record, and deploying the updated ML model for production use.

Abstract: Pairwise entity matching systems and methods are disclosed herein. A deep learning model may be used to match entities from separate data tables. Entities may be preprocessed to fuse textual and numeric data early in the neural network architecture. Numeric data may be represented as a vector of a geometrically progressing function. By fusing textual and numeric data, including dates, early in the neural network architecture the neural network may better learn the relationships between the numeric and textual data. Once preprocessed, the paired entities may be scored and matched using a neural network.

Abstract: A radar sensor system having a defined number of HF components, with each of the HF components having at least one antenna for transmitting and/or receiving of radar waves in each case, and at least one antenna control for operating the at least one antenna; and a synchronization network, which is connected to all HF components and via which an operating frequency of all HF components is able to be synchronized; with a synchronization master according to at least one defined criterion being able to be provided by all HF components.

Abstract: A device for emitting and receiving electromagnetic radiation, in which different antennas are used for the emitting and receiving, a first antenna or first group being used for the transmission in a first polarization form, a second antenna or second group being used for the transmission in a second polarization form, and a third antenna or third group being used for receiving the reflected electromagnetic radiation that was emitted by the first antenna or first group and by the second antenna or second group. The device may be fixed in place on a motor vehicle and used for object detection within the framework of a distance and speed control or a collision avoidance, and the polarimetric information obtained from the different receiving levels during the propagation of the two differently polarized electromagnetic waves via different propagation paths is able to be used for ascertaining a weather-related road condition.

Abstract: A radar sensor having at least one high-frequency component and at least one waveguide structure in the form of a plastic body provided with an electrically conductive surface layer. The radar sensor has at least one further plastic body provided with an electrically conductive surface layer, and the plastic bodies with their conductive surface layers are thermally bonded to one another.

Abstract: An assortment of radar sensors in different variant embodiments. Each radar sensor has: a housing terminated by a radome, a circuit board that is equipped on the side facing away from the radome with at least one radio-frequency module, and an antenna structure on the side of the circuit board facing the radome. The housing is realized identically in all variant embodiments. The antenna structure has a planar antenna structure in at least one variant embodiment, and has a hollow conductor structure in at least one variant embodiment.

Abstract: A radar system for surroundings detection of a motor vehicle. The radar system includes an antenna for sending and/or receiving radar signals, a circuit board including at least one high-frequency component and an element for sending and/or receiving high-frequency signals. At least one fastening element is situated on/at the circuit board. The antenna includes at least one connecting element. The connecting element interacts with the fastening element in such a way that the antenna is fastenable and at the same time positionable on the circuit board by being pushed onto the at least one fastening element.

Abstract: Pairwise entity matching systems and methods are disclosed herein. A deep learning model may be used to match entities from separate data tables. Entities may be preprocessed to fuse textual and numeric data early in the neural network architecture. Numeric data may be represented as a vector of a geometrically progressing function. By fusing textual and numeric data, including dates, early in the neural network architecture the neural network may better learn the relationships between the numeric and textual data. Once preprocessed, the paired entities may be scored and matched using a neural network.

Abstract: A method for calibrating two receiving units of a radar sensor that includes an array of receiving antennas formed by two sub-arrays and an evaluation unit, which is designed to carry out an angle estimation for located radar targets based on phase differences between the signals received by the receiving antennas, each receiving unit including parallel reception paths for the signals of the receiving antennas of one of the sub-arrays. The method includes: analyzing the received signals and deciding whether a multi-target scenario or a single-target scenario is present, in the case of a single-target scenario, measuring phases of the signals received in the sub-arrays and calculating a phase offset between the two sub-arrays, and calibrating the phases in the two receiving units based on the calculated offset.

Abstract: A radar sensor for motor vehicles. The radar sensor includes an antenna arrangement having multiple antenna elements linearly arranged at uniform distances along a connecting line, the distance between each two adjacent antenna elements being equal to half of the mean wavelength ? of the emitted radar signal, and the arrangement including at least one triple of adjacent antenna elements, in which the two antenna elements located on the outside in the triple, on the one hand, and the interposed antenna element, on the other hand, originate in opposite directions from the connecting line. The antenna arrangement includes at least one pair of antenna elements which originate in opposite directions from the connecting line and whose distance from one another is an integer multiple of the wavelength ?, so that one of these antenna elements has a negative amplitude taper in relation to the antenna elements of the triple.

Abstract: A radar sensor system for transmitting and receiving radar waves. The system includes a first sub-sensor having a first antenna and a first antenna control for operating the first antenna, and a second sub-sensor including a second antenna and a second antenna control for operating the second antenna. The system further includes a frequency-generating device having a clock pulse generator for generating a usable frequency and having a control unit for actuating and controlling the first antenna control, the second antenna control and the frequency-generating device, the frequency-generating device having a first clock pulse generator and a second clock pulse generator, the first clock pulse generator and the second clock pulse generator being able to be connected via at least two multiplexers to the first antenna control and the second antenna control for the supply of a usable frequency in each case.