Abstract: Radar systems and methods are provided and include transmitting radar signals within a frame having N radar chirps with a stepped frequency waveform, receiving and sampling radar signals reflected from a target, performing range fast Fourier transform (FFT) processing, including a first domain FFT, and Doppler FFT processing, including a second domain FFT, on the receive values to generate range FFT values and Doppler FFT values. A range and a velocity of the target are estimated based on the range FFT values and Doppler FFT values, wherein the estimated range, r, and the estimated velocity, v, of the target are calculated based on at least one equation that includes (i) a first term corresponding to a target range, ro, measured by a first radar chirp of the N radar chirps, and (ii) a second term corresponding to a center wavelength, ?N/2, of the plurality of radar chirps of the frame.

Abstract: Distribution fitting Constant False Alarm Rate (CFAR) detection is described. Noise data in cells or bins around a target cell are fit to a noise distribution model, such as a Rayleigh distribution model. With a suitable noise distribution curve from the distribution model, a CFAR threshold for that cell along the curve can be determined. A quantile function of the noise distribution model for a bin or cell provides the CFAR threshold to use for that bin or cell. Distribution fitting CFAR enables a more-accurate CFAR threshold to be set for each bin or cell and may use far fewer computing resources than Ordered-Statistics CFAR. A radar detector can better prevent false alarm detections across multiple different driving scenarios by adapting to different environments and dynamically changing the noise distribution curve used depending on best-fit analysis by a noise distribution model of noise characteristics of the neighboring bins or cells.

Abstract: The techniques of this disclosure relate to object range and velocity detections from varying radar pulse repetition times. A device includes a transmitter, a receiver, and a processor. The transmitter is configured to transmit a series of electromagnetic pulses within a frame having a varying pulse repetition time between electromagnetic pulses. A center frequency of the electromagnetic pulses varies by a frequency step. The receiver is configured to receive a series of corresponding reflected electromagnetic pulses reflected from an object. The processor is configured to determine range and velocity information of the object based on the received series of corresponding reflected electromagnetic pulses. By varying the pulse repetition times between consecutive pulses in a radar frame, the radar system may improve a range resolution, particularly in the case of frequency-modulated continuous-wave (FMCW) radar systems.

Abstract: The techniques of this disclosure relate to object range and velocity detections from varying radar pulse repetition times. A device includes a transmitter, a receiver, and a processor. The transmitter is configured to transmit a series of electromagnetic pulses within a frame having a varying pulse repetition time between electromagnetic pulses. A center frequency of the electromagnetic pulses varies by a frequency step. The receiver is configured to receive a series of corresponding reflected electromagnetic pulses reflected from an object. The processor is configured to determine range and velocity information of the object based on the received series of corresponding reflected electromagnetic pulses. By varying the pulse repetition times between consecutive pulses in a radar frame, the radar system may improve a range resolution, particularly in the case of frequency-modulated continuous-wave (FMCW) radar systems.

Abstract: Distribution fitting Constant False Alarm Rate (CFAR) detection is described. Noise data in cells or bins around a target cell are fit to a noise distribution model, such as a Rayleigh distribution model. With a suitable noise distribution curve from the distribution model, a CFAR threshold for that cell along the curve can be determined. A quantile function of the noise distribution model for a bin or cell provides the CFAR threshold to use for that bin or cell. Distribution fitting CFAR enables a more-accurate CFAR threshold to be set for each bin or cell and may use far fewer computing resources than Ordered-Statistics CFAR. A radar detector can better prevent false alarm detections across multiple different driving scenarios by adapting to different environments and dynamically changing the noise distribution curve used depending on best-fit analysis by a noise distribution model of noise characteristics of the neighboring bins or cells.