Low Complexity Super-Resolution Technique for Object Detection in Frequency Modulation Continuous Wave Radar

Patent metadata

Field	Value
Publication number	US-2016334502-A1
Application number	US-201514951014-A
Country	US
Kind code	A1
Filing date	Nov 24, 2015
Priority date	May 15, 2015
Publication date	Nov 17, 2016
Grant date	—

Abstract

In the proposed low complexity technique a hierarchical approach is created. An initial FFT based detection and range estimation gives a coarse range estimate of a group of objects within the Rayleigh limit or with varying sizes resulting from widely varying reflection strengths. For each group of detected peaks, demodulate the input to near DC, filter out other peaks (or other object groups) and decimate the signal to reduce the data size. Then perform super-resolution methods on this limited data size. The resulting distance estimations provide distance relative to the coarse estimation from the FFT processing.

First claim

What is claimed is:

1 . A method of object detection comprising the steps of:
generating a signal with a linearly changing frequency; transmitting said signal in the direction of the objects to be detected; receiving the reflected signal from said objects; mixing the received signal with the transmitted signal to form a heterodyne or beat frequency proportional to the distance to said objects; performing an Fourier Transform on the beat frequency where the peaks of the Fourier Transform output correspond to the objects detected, and the frequencies of the peaks correspond to a coarse estimate of the distance to said objects; demodulating said input signal to near DC for each group of detected peaks; filtering out other peaks; decimating the resulting signal to reduce data size; performing super resolution calculation on the reduced data set, giving more precise distance estimations relative to the coarser estimates given by the Fourier Transform calculation.

2 . The method of claim 1 , wherein:
the step of performing the super resolution calculation employs an eigen analysis of the reduced data set.

3 . The method of claim 1 , wherein:
the step of performing the super resolution calculation employs the Multiple Signal Classification (MUSIC) algorithm.

4 . The method of claim 3 , further comprising the steps of:
dividing the signal auto-correlation matrix into signal and noise subspaces; performing singular value decomposition on the subspaces; extracting the noise subspace by extracting the eigenvectors with the lowest eigenvalues; creating the MUSIC pseudo spectrum orthogonal to the noise subspace; and searching for peaks in the above spectrum.

5 . The method of claim 1 , wherein:
the step of performing the super resolution calculation employs the Matrix Pencil Method (MPM) algorithm.

6 . The method of claim 5 , further comprising the steps of:
creating a Hankel matrix with a delayed signal vector; computing the generalized eigenvalues of the matrix; performing singular value decomposition; selecting the highest eigenvalues; extracting two eigenvector matrices; performing a second singular value decomposition; and searching for peaks within the resulting eigenvalues.

7 . An apparatus for object detection comprising of:
a linear ramp generator; a voltage controlled oscillator controlled by the output of the linear ramp generator; an antenna operable to transmit the output of the voltage controlled oscillator; an antenna operable to receive the signal reflected from a plurality of objects; a mixer operable to mix the output of the voltage controlled oscillator and the received reflected signal forming a beat frequency proportional to the distance to the objects reflecting the signal; a processor operable to perform a Fourier transform on the beat frequency wherein the peaks of the Fourier transform output correspond to the objects detected, and the frequencies of the peaks correspond to an estimate of the distance to the said objects; said processor is further operable to demodulate to near DC the output of said Fourier transform for each group of detected peaks; said processor is further operable to sub-sample the demodulated data, and perform super resolution calculation on the sub-sampled data set.

8 . The apparatus of claim 7 , wherein:
said processor is further operable to perform said super resolution calculation by employing an eigen analysis of the sub-sampled data set.

9 . The apparatus of claim 7 , wherein:
said processor is further operable to perform said super resolution calculation by employing the Multiple Signal Classification algorithm (MUSIC).

10 . The apparatus of claim 9 , wherein said processor is further operable to:
divide the signal auto-correlation matrix into signal and noise subspaces; perform singular value decomposition on the subspaces; extract the noise subspace by extracting the eigenvectors with the lowest eigenvalues; create a MUSIC pseudo spectrum orthogonal to the noise subspace; and search for peaks in the above spectrum.

11 . The apparatus of claim 7 , wherein:
said processor is further operable to perform said super resolution calculation by employing the matrix pencil method algorithm.

12 . The apparatus of claim 11 , wherein said processor is further operable to:
create a Hankel matrix with a delayed signal vector; compute the generalized eigenvalues of the matrix; perform singular value decomposition; select the highest eigenvalues; extract two eigenvector matrices; perform a second singular value decomposition; and search for peaks within the resulting eigenvalues.

Assignees

• Texas Instruments Inc

Inventors

• Ali Murtaza
• Wang Dan
• Ikram Muhammad Zubair

Classifications

• G01S13/343 Primary

  using sawtooth modulation · CPC title

• G01S13/34 Primary

  using transmission of continuous, frequency-modulated waves while heterodyning the received signal, or a signal derived therefrom, with a locally-generated signal related to the contemporaneously transmitted signal · CPC title

• G01S7/354 Primary

  Extracting wanted echo-signals (Doppler systems G01S13/50) · CPC title

• G01S2007/356

  Physics · mapped topic

• G01S7/356

  involving particularities of FFT processing · CPC title