US-12618940-B2 - Signal detection and denoising systems

US12618940B2US 12618940 B2US12618940 B2US 12618940B2US-12618940-B2

Abstract

Disclosed herein are systems and methods for estimating target ranges, angles of arrival, and speed using optimization procedures. Target ranges are estimated by performing an optimization procedure to obtain a denoised signal, performing a correlation of a transmitted waveform and the denoised signal, and using a result of the correlation to determine an estimate of a distance between the sensor and at least one target. Target angles of arrival are estimated by determining ranges at which targets are located, and, for each range, constructing an array signal from samples of received echo signals, and using the array signal, performing another optimization procedure to estimate a respective angle of arrival for each target of the at least one target. Doppler shifts may also be estimated using another optimization procedure. Certain of the optimization procedures use atomic norm techniques.

Inventors

Babak Hassibi
Behrooz Rezvani

Assignees

Neural Propulsion Systems, Inc.

Dates

Publication Date: 20260505
Application Date: 20240524

Claims (20)

1 . A method, comprising: a sensor receiving an echo signal off of at least one target; at least one analog-to-digital converter generating samples of the echo signal; using at least a portion of the samples of the echo signal, performing an optimization procedure to obtain a denoised signal, wherein the optimization procedure explicitly exploits a structure of the echo signal to select, from a set of estimates, a best estimate with regard to a criterion; performing a correlation of a transmitted waveform and the denoised signal; and using a result of the correlation of the transmitted waveform and the denoised signal, determining at least one range value, the at least one range value being an estimate of a distance between the sensor and the at least one target.
2 . The method of claim 1 , wherein the at least one range value comprises a first range value representing an estimate of a distance between the sensor and a first target, and a second range value representing an estimate of a distance between the sensor and a second target.
3 . The method of claim 1 , wherein performing the optimization procedure to obtain the denoised signal comprises: minimizing an atomic norm of the denoised signal subject to a constraint on a metric characterizing a closeness of the denoised signal to the samples of the echo signal.
4 . The method of claim 3 , wherein the atomic norm corresponds to a collection of atoms, and wherein: each atom is a time-shifted version of the transmitted waveform, and/or each atom is a planar wave with a respective arrival angle that differs from the respective arrival angle of every other atom.
5 . The method of claim 3 , wherein minimizing the atomic norm comprises performing a frequency-domain atomic norm optimization procedure, and/or performing a gradient descent.
6 . The method of claim 1 , wherein performing the optimization procedure to obtain the denoised signal comprises: minimizing a metric characterizing a closeness of the denoised signal to the samples of the echo signal subject to a constraint on an atomic norm of the denoised signal.
7 . The method of claim 1 , wherein performing the optimization procedure to obtain the denoised signal comprises: minimizing a weighted sum of (a) a metric characterizing a closeness of the denoised signal to the samples of the echo signal and (b) an atomic norm of the denoised signal.
8 . The method of claim 1 , wherein determining the at least one range value comprises: identifying at least one peak in the result of the correlation of the transmitted waveform and the denoised signal; and computing the estimate of the distance between the sensor and the at least one target based at least in part on a position of the at least one peak within the result of the correlation of the transmitted waveform and the denoised signal.
9 . The method of claim 1 , wherein determining the at least one range value comprises performing a Fourier transform.
10 . The method of claim 1 , wherein the echo signal is a first echo signal in a first frequency band, and the denoised signal is a first denoised signal, and the transmitted waveform is a first transmitted waveform, and the at least one range value is a first at least one range value, and the estimate of the distance between the sensor at the at least one target is a first estimate of the distance between the sensor and the at least one target, and wherein the method further comprises: the sensor receiving a second echo signal in a second frequency band; the at least one analog-to-digital converter generating samples of the second echo signal; using at least a portion of the samples of the second echo signal, performing the optimization procedure to obtain a second denoised signal; performing a correlation of a second transmitted waveform and the second denoised signal; and using a result of the correlation of the second transmitted waveform and the second denoised signal, determining a second at least one range value, the second at least one range value being a second estimate of the distance between the sensor and the at least one target.
11 . The method of claim 1 , wherein the sensor is a first sensor, the echo signal is a first echo signal, the samples of the echo signal are first samples of the first echo signal, the denoised signal is a first denoised signal, the at least one range value is a first at least one range value, and the optimization procedure is a first optimization procedure, and further comprising: a second sensor receiving a second echo signal off of the at least one target; the at least one analog-to-digital converter generating second samples of the second echo signal; using at least a portion of the second samples of the second echo signal, performing the first optimization procedure to obtain a second denoised signal; performing a correlation of the transmitted waveform and the second denoised signal; using a result of the correlation of the transmitted waveform and the second denoised signal, determining a second at least one range value, the second at least one range value being an estimate of a distance between the second sensor and the at least one target; using the first denoised signal, the second denoised signal, and the first at least one range value, constructing a first array signal; using the first array signal, performing a second optimization procedure to estimate a first respective angle of arrival for each target of the at least one target; using the first denoised signal, the second denoised signal, and the second at least one range value, constructing a second array signal; and using the second array signal, performing the second optimization procedure to estimate a second respective angle of arrival for each target of the at least one target.
12 . The method of claim 11 , wherein performing the second optimization procedure comprises: obtaining a denoised array signal from an array signal; and applying an angle-of-arrival estimation algorithm to the denoised array signal.
13 . The method of claim 12 , wherein obtaining the denoised array signal from the array signal comprises: minimizing an atomic norm of the denoised array signal subject to a constraint on a metric characterizing a closeness of the denoised array signal to the array signal.
14 . The method of claim 12 , wherein obtaining the denoised array signal from the array signal comprises: minimizing a metric characterizing a closeness of the denoised array signal to the array signal subject to a constraint on an atomic norm of the denoised array signal.
15 . The method of claim 12 , wherein obtaining the denoised array signal from the array signal comprises: minimizing a weighted sum of (a) a metric characterizing a closeness of the denoised array signal to the array signal and (b) an atomic norm of the denoised array signal.
16 . The method of claim 11 , further comprising: using the first denoised signal, the second denoised signal, the first at least one range value, and the second at least one range value, performing a third optimization procedure to determine a velocity of the at least one target.
17 . The method of claim 11 , further comprising: transmitting, for a time period, a first instance of the transmitted waveform modulated onto a first carrier signal having a carrier frequency and a first phase; and transmitting, for the time period and substantially synchronously with the first instance of the transmitted waveform, a second instance of the transmitted waveform modulated onto a second carrier signal having the carrier frequency and a second phase, the second phase being different from the first phase, wherein the first instance and the second instance of the transmitted waveform are substantially identical.
18 . The method of claim 1 , further comprising: modifying an aspect of the optimization procedure in response to additional information other than the samples of the echo signal.
19 . The method of claim 1 , further comprising downconverting the echo signal to baseband, and wherein the at least one analog-to-digital converter is configured to generate a set of in-phase samples and a set of quadrature samples for the sensor, and wherein the samples of the echo signal comprise the set of in-phase samples and the set of quadrature samples.
20 . The method of claim 1 , further comprising downconverting the echo signal to an intermediate frequency, and wherein sampling the echo signal is performed while the echo signal resides at the intermediate frequency.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS This application is a continuation of U.S. application Ser. No. 18/174,612, filed Feb. 25, 2023 and entitled “SIGNAL DETECTION AND DENOISING SYSTEMS,” which is a division of U.S. Application Ser. No. 17/120,127, filed Dec. 12, 2020 and entitled “SIGNAL DETECTION AND DENOISING SYSTEMS,” which is a continuation of U.S. application Ser. No. 16/569,011, filed Sep. 12, 2019 and entitled “SIGNAL DETECTION AND DENOISING SYSTEMS,” which claims the benefit of U.S. Provisional Application No. 62/730,405, filed Sep. 12, 2018 and entitled “Low Power Radar.” All of the above-referenced applications are hereby incorporated by reference in their entireties for all purposes. BACKGROUND Radio detection and ranging (“radar”) systems use radio signals to determine, among other things, the position and/or speed of an object, which may be referred to as a target. The radar system transmits a radio signal, the target reflects the transmitted signal, and the radar system receives the reflected signal and uses it to determine the speed, location, or velocity (speed and direction) of the target. Radar is used in many applications, such as weather monitoring, air traffic control, speed enforcement, autonomous driving, medical imaging, and military applications. Continuous-wave (CW) radar systems transmit radio signals on a continuous basis. Pulsed radar transmitters are switched on and off (pulsed) to provide range timing information with each pulse. The maximum target detection range of either type of radar system is proportional to the transmitter output power. Other systems, such as, for example, such as light detection and ranging (LIDAR) and sonar systems, have similar limitations. It would be desirable to increase the range and angular resolution of these systems. And it would also desirable to reduce the output power of these systems' transmitters without a commensurate degradation to the range or accuracy of the target detection ability. Conventional systems also tend to be expensive, and it would be desirable to reduce their cost. There is, therefore, an ongoing need for improvements in target detection systems. SUMMARY This summary represents non-limiting embodiments of the disclosure. Disclosed herein are systems and methods for detecting targets and estimating their positions/distances, angles of arrival, and/or movement (e.g., velocity or speed), and/or identifying characteristics of such targets (e.g., radar cross section, or composition material, etc.). In some embodiments, a system comprises a sensor (e.g., an antenna) for receiving an echo signal off of at least one target, at least one analog-to-digital converter (ADC) coupled to the sensor and configured to generate samples of the echo signal, and at least one processor. The at least one processor is configured to execute at least one machine-executable instruction that, when executed, causes the at least one processor to (a) perform an optimization procedure using at least a portion of the samples of the echo signal to obtain a denoised signal, (b) perform a correlation of a transmitted waveform and the denoised signal, and (c) using a result of the correlation of the transmitted waveform and the denoised signal, determine at least one range value, the at least one range value being an estimate of a distance between the sensor and the at least one target. In some such embodiments, the at least one range value comprises a first range value representing an estimate of a distance between the sensor and a first target, and a second range value representing an estimate of a distance between the sensor and a second target. In some embodiments, the optimization procedure comprises a projected gradient descent procedure. In some embodiments, when executed by the at least one processor, the at least one machine-executable instruction further causes the at least one processor to modify an aspect of the optimization procedure in response to additional information other than the samples of the echo signal. In some embodiments, performing the optimization procedure to obtain the denoised signal comprises minimizing an atomic norm of the denoised signal subject to a constraint on a metric characterizing a closeness of the denoised signal to the samples of the echo signal. In some such embodiments, the atomic norm corresponds to a collection of atoms, wherein each atom is a time-shifted version of the transmitted waveform. In some embodiments, minimizing the atomic norm comprises performing a frequency-domain atomic norm optimization procedure, or performing a projected gradient descent. In some embodiments, the metric characterizing the closeness of the denoised signal to the samples of the echo signal is a squared Euclidean distance. In some embodiments, the metric characterizing the closeness of the denoised signal to the samples of the echo signal is an information-theoretic metric. In some embodiments, the metric characterizing the c