Abstract: A method and system are provided to resolve Doppler ambiguity and multiple-input, multiple-output array phase compensation issues present in Time Division Multiplexing MIMO radars by estimating an unambiguous radial velocity measurement. Embodiments apply a disambiguation algorithm that dealiases the Doppler spectrum to resolve the Doppler ambiguity of a range-Doppler detection. Phase compensation is then applied for corrected reconstruction of the MIMO array measurements. The dealiasing processing first forms multiple hypotheses associated with the phase corrections for the radar transmitters based on a measured radial velocity of a range-Doppler cell being processed. A correct hypothesis, from the multiple hypotheses, is selected based on a least-spurious spectrum criterion. Using this approach, embodiments require only single-frame processing and can be applied to two or more transmitters in a TDM MIMO radar system.

Abstract: A linear chirp radar system, apparatus and method use a radar control processing unit to control an LFM radar front end which generates analog-to-digital (ADC) sample signals from one or more target return signals received in response to transmitted linear chirp radar signals, where the radar control processing unit is connected and configured to mitigate range migration by directly filtering the ADC samples using a modified Doppler filter that is tuned to fast-time scaled, slow-time frequencies to generate a focused ADC Doppler cube, and by applying a Fourier Transform on each Doppler cell in the focused ADC Doppler cube to generate a focused range-Doppler cube.

Abstract: A vehicle radar system, apparatus and method use a radar control processing unit to generate a target response signal in at least a first dimension from compressed radar data signals and to perform cell-averaging constant false alarm rate (CA-CFAR) target detection by convolving the target response signal with a weighted kernel window signal in a frequency domain using a Fast Fourier Transform hardware accelerator, an element-wise multiplier, and an Inverse Fast Fourier Transform hardware accelerator to generate an output signal having a sign that indicates a target detection decision.

Abstract: Embodiments of systems and methods for estimating direction of arrival are disclosed. A device includes a signal processing unit that includes processing circuitry and memory coupled to the processing circuitry, where the processing circuitry includes multiple vector processing units, each vector processing unit configured to receive an antenna input vector, receive an angular spectrum vector, retrieve a first and second weighting vectors from the memory, generate a processed antenna input vector by performing a circular convolution of the antenna input vector with the first weighting vector, generate a processed angular spectrum vector by performing a circular convolution of the angular spectrum vector with the second weighting vector, and generate a refined angular spectrum vector, which indicates angular position of one or more radar targets, by applying a non-linear activation function to a sum of the processed antenna input vector and the processed angular spectrum vector.

Abstract: A linear chirp radar system, apparatus and method use a radar control processing unit to control an LFM radar front end which includes a frequency-scanning transmit antenna and a frequency-scanning receive antenna which respectively sweep a transmit and receive energy focus across an angle space with each linear chirp signal, where the radar control processing unit processes digital output signals generated from target return signals received in response to transmitted linear chirp signals and extracts target range-angle information by applying time-frequency analysis processing to the digital output signals to generate a first range-angle map which includes range-biased angle information, and then applying a group delay compensation process to generate a second range-angle map which includes target range-angle information that is generated by selectively adjusting the range-biased angular information in the first range-angle map with an angular adjustment.

Abstract: A vehicle radar system, apparatus and method use a radar control processing unit to control an RF transmitter unit to generate a radiated beam by a long and medium range radar (LMRR) beam shaping antenna array which has a range coverage pattern with more power concentrated along a central direction axis for long range detection and less power spread off to sides of the central direction axis for medium range detection, wherein the LMRR beam shaping antenna array includes a plurality of transmit radiator elements stacked over a power dividing feeding network and separated by a conductive coupling aperture layer comprising a plurality of coupling apertures such that each transmit radiator element is aligned through a corresponding coupling aperture to a corresponding feeding line conductor from the power dividing feeding network.

Abstract: Aspects of the present disclosure are directed to radar and radar processing. As may be implemented in accordance with one or more embodiments involving multi-input multi-output (MIMO) co-prime radar signals transmitted by a plurality of transmitters and reflected from at least one target, the reflected radar signals are processed by resolving ambiguities associated with a range-Doppler detection based on unique pulse repetition frequencies (PRF)s associated with respective chirp groups of the reflected radar signals. Phase compensation is applied to compensate for motion-induced phased biases and, thereafter, Doppler estimates are reconstructed to provide a dealiased version of the reflected radar signals.

Abstract: A co-prime coded DDM MIMO radar system, apparatus, architecture, and method are provided with a reference signal generator (112) that produces a transmit reference signal; a plurality of DDM transmit modules (11) that produce, condition, and transmit a plurality of transmit signals over which each have a different co-prime encoded progressive phase offset from the transmit reference signal; a receiver module (12) that receives a target return signal reflected from the plurality of transmit signals by a target and generates a digital signal from the target return signal; and a radar control processing unit (20) configured to detect Doppler spectrum peaks in the digital signal, where the radar control processing unit comprises a Doppler disambiguation module (25) that is configured with a CPC decoder to associate each detected Doppler spectrum peak with a corresponding DDM transmit module, thereby generating a plurality of transmitter-associated Doppler spectrum peak detections.

Abstract: An early fusion network is provided that reduces network load and enables easier design of specialized ASIC edge processors through performing a portion of convolutional neural network layers at distributed edge and data-network processors prior to transmitting data to a centralized processor for fully-connected/deconvolutional neural networking processing. Embodiments can provide convolution and downsampling layer processing in association with the digital signal processors associated with edge sensors. Once the raw data is reduced to smaller feature maps through the convolution-downsampling process, this reduced data is transmitted to a central processor for further processing such as regression, classification, and segmentation, along with feature combination of the data from the sensors.

Abstract: A method and system are provided to resolve Doppler ambiguity and multiple-input, multiple-output array phase compensation issues present in Time Division Multiplexing MIMO radars by estimating an unambiguous radial velocity measurement. Embodiments apply a disambiguation algorithm that dealiases the Doppler spectrum to resolve the Doppler ambiguity of a range-Doppler detection. Phase compensation is then applied for corrected reconstruction of the MIMO array measurements. The dealiasing processing first forms multiple hypotheses associated with the phase corrections for the radar transmitters based on a measured radial velocity of a range-Doppler cell being processed. A correct hypothesis, from the multiple hypotheses, is selected based on a least-spurious spectrum criterion. Using this approach, embodiments require only single-frame processing and can be applied to two or more transmitters in a TDM MIMO radar system.

Abstract: A co-prime coded DDM MIMO radar system, apparatus, architecture, and method are provided with a reference signal generator (112) that produces a transmit reference signal; a plurality of DDM transmit modules (11) that produce, condition, and transmit a plurality of transmit signals over which each have a different co-prime encoded progressive phase offset from the transmit reference signal; a receiver module (12) that receives a target return signal reflected from the plurality of transmit signals by a target and generates a digital signal from the target return signal; and a radar control processing unit (20) configured to detect Doppler spectrum peaks in the digital signal, where the radar control processing unit comprises a Doppler disambiguation module (25) that is configured with a CPC decoder to associate each detected Doppler spectrum peak with a corresponding DDM transmit module, thereby generating a plurality of transmitter-associated Doppler spectrum peak detections.

Abstract: An early fusion network is provided that reduces network load and enables easier design of specialized ASIC edge processors through performing a portion of convolutional neural network layers at distributed edge and data-network processors prior to transmitting data to a centralized processor for fully-connected/deconvolutional neural networking processing. Embodiments can provide convolution and downsampling layer processing in association with the digital signal processors associated with edge sensors. Once the raw data is reduced to smaller feature maps through the convolution-downsampling process, this reduced data is transmitted to a central processor for further processing such as regression, classification, and segmentation, along with feature combination of the data from the sensors.