EP-4737934-A1 - SYSTEM AND METHOD FOR ESTIMATING FRACTIONAL BIN SHIFTS IN ISAC WITH MULTISTATIC RADAR

EP4737934A1EP 4737934 A1EP4737934 A1EP 4737934A1EP-4737934-A1

Abstract

Existing approaches for estimating delay-doppler values in an Integrated Sensing and Communication (ISAC) system with multistatic radar receivers have the disadvantage that they are estimating integer bin shifts and require separate estimation of delay and doppler values, requiring additional steps of computationally demanding data association and distinguishing the signal from different transmitters. Embodiments disclosed herein provide a method and system for estimating fractional bin shifts in ISAC with multistatic radar. In this approach, the system estimates paired fractional delay-doppler bin shifts by utilizing one or more phase information and a dictionary matrix, and computing the cross correlation of a at least one received signal and the associated transmitted signal. Cross correlation is utilized for identifying the one or more transmitters with respect to at least one transmitted signal.

Inventors

SURESH, Sweta
KUMAR, ACHANNA ANIL
Krishnamurthy, Giridhar

Assignees

Tata Consultancy Services Limited

Dates

Publication Date: 20260506
Application Date: 20250918

Claims (15)

A processor implemented method, comprising: receiving (302), via one or more hardware processors, at least one signal reflected from one or more targets in a delay-time domain frame at a radar receiver, wherein the received at least one signal comprises a M*N matrix, and wherein the M*N matrix comprises one or more pilot sequences, one or more guard symbols, and one or more communication data; demodulating (304), via the one or more hardware processors, the at least one signal to obtain a delay Doppler domain frame; extracting (306), via the one or more hardware processors, the one or more pilot sequences and the one or more guard symbols from the received at least one signal, wherein the one or more pilot sequences and the one or more guard symbols are known at the radar receiver and one or more transmitters from where the at least one signal is transmitted towards the one or more targets; determining (308), via the one or more hardware processors, a cross correlation of the received at least one signal and an associated transmitted signal by using the extracted one or more pilot sequences and a local version of a pilot sequence from each of the one or more transmitters; transforming (310), via the one or more hardware processors, the cross correlation of the received at least one signal and the associated transmitted signal to a frequency domain from the delay-doppler domain frame by performing a signal processing on the demodulated at least one signal, to obtain a transformed cross correlation of the received at least one signal and an associated transformed cross correlation of the transmitted signal; computing (312), via the one or more hardware processors, a plurality of phase information of the received at least one signal using the transformed cross correlation of the received at least one signal and the associated transformed cross correlation of the transmitted signal; constructing (314), via the one or more hardware processors, a dictionary matrix by selecting a plurality of pre-defined delay-doppler values computed using a 2D complex exponential signal technique; estimating (316), via the one or more hardware processors, one or more sparse vectors by solving a sparse recovery problem using the dictionary matrix and the plurality of phase information, wherein an estimation algorithm is used for solving the sparse recovery problem; and estimating (318), via the one or more hardware processors, one or more paired delay-doppler bin shifts with respect to at least one of the one or more targets by identifying one or more non-zero values of the one or more sparse vectors.
The processor implemented method as claimed in claim 1, wherein, the received at least one signal comprising the one or more pilot sequences and the one or more guard symbols is placed in the delay doppler domain frame as a plurality of row and column pilot structures, the one or more communication data are placed in the M*N matrix after the one or more pilot sequences and the one or more guard symbols, and wherein the at least one signal is transmitted from the one or more transmitters by converting the delay-Doppler domain frame to the delay-time domain frame by performing a signal modulation on the delay-Doppler domain frame and by adding a cyclic prefix to the at least one signal in the delay-time domain frame followed by a column-wise vectorization.
The processor implemented method as claimed in claim 2, wherein converting the received at least one signal from the delay-time domain frame to the delay-doppler frame by removing the cyclic prefix from the received at least one signal and performing the signal demodulation on the received at least one signal for converting to the delay-doppler frame is represented as: Y r = R r F N , where, Y ( r ) is the received at least one signal at r th receiver in the delay-Doppler domain frame, R ( r ) is the received at least one signal at the r th receiver in the delay-time domain frame, and F N is a N dimensional Discrete Fourier Transform matrix.
The processor implemented method as claimed in claim 1, wherein the at least one signal from each of the one or more transmitters is distinguished by determining the cross correlation of the received at least one signal and the associated transmitted signal with the local version of the pilot sequence from each of the one or more transmitters, and is represented as: Σ Y r , p t , where, Y ( r ) is the received at least one signal at r th receiver in the delay-Doppler domain frame and p ( t ) is a transmitted pilot signal.
The processor implemented method as claimed in claim 1, wherein computing the plurality of phase information of the received at least one signal comprises of dividing the transformed cross correlation of the received at least one signal with the transformed cross correlation of an associated transmitted signal, and is represented as: ∑ u , v Y r ∼ , p t = ∑ l = 1 L ∑ t = 1 T a l e − j 2 π uτ l , bin M G + vμ l , bin N ∑ u , v X τ , μ t ∼ , p t + η ∼ , H u , v = ∑ l = 1 L ∑ t = 1 T a τl , μ l e − j 2 π uτ l , bin M G + vμ l , bin N + η ∼ , where, Y ( r ~ ) is the 2D-FFT of the extracted received signal in delay doppler (DD) domain, a τt,µ l is the path gain of l th path, µ l,bin is the Doppler bin shift of the l th path, τ l,bin is the delay bin shift of the l th path, T being the number of transmitters and L being the number of paths, H u,v is the plurality of phase information, Σ u , v X τ , μ t ∼ , p t is the corelated signal between an extracted transmitted signal ( X τ , μ t ∼ ) and ( p ( t )) the transmitted pilot signal, η ~ denotes a residual noise present in the received at least one signal, M G represents only those delay bins specific to pilot and guard symbols and N denotes the number of doppler bins.
The processor implemented method as claimed in claim 1, wherein estimating the one or more zero values of the one or more sparse vectors by comparing the plurality of columns of a dictionary matrix with the plurality of phase information using the estimation algorithm and identifying the one or more non-zero value of the one or more sparse vectors as paired delay-doppler shifts associated to the one or more targets is represented as: u s ∼ = min u s u s l 1 such that h − ϕu s l 2 < ϵ , where, h is the vectorised version of H , u s is the sparse vector with L significant values, l 1 is the l 1 norm, l 2 is the l 2 norm and ε is the threshold used in the estimation algorithm.
A system (100), comprising: one or more hardware processors (104); a communication interface (106); and a memory (102) storing a plurality of instructions, wherein the plurality of instructions cause the one or more hardware processors to: receive at least one signal reflected from one or more targets in a delay-time domain frame at a radar receiver, wherein the received at least one signal comprises a M*N matrix, and wherein the M*N matrix comprises one or more pilot sequences, one or more guard symbols, and one or more communication data; demodulate the at least one signal to obtain a delay Doppler domain frame; extract the one or more pilot sequences and the one or more guard symbols from the received at least one signal, wherein the one or more pilot sequences and the one or more guard symbols are known at the radar receiver and one or more transmitters from where the at least one signal is transmitted towards the one or more targets; determine a cross correlation of the received at least one signal and associated transmitted signal by using the extracted one or more pilot sequences and a local version of a pilot sequence from each of the one or more transmitters; transform the cross correlation of the received at least one signal and the associated transmitted signal to a frequency domain from the delay-doppler domain frame by performing a signal processing on the demodulated at least one signal, to obtain a transformed cross correlation of the received at least one signal and an associated transformed cross correlation of the transmitted signal; compute a plurality of phase information of the received at least one signal using the transformed cross correlation of the received at least one signal and the associated transformed cross correlation of the transmitted signal; construct a dictionary matrix by selecting a plurality of predefined delay-doppler values computed using a 2D complex exponential signal technique; estimate one or more sparse vectors by solving a sparse recovery problem using the dictionary matrix and the plurality of phase information, wherein an estimation algorithm is used for solving the sparse recovery problem; and estimate one or more paired delay-doppler bin shifts with respect to at least one of the one or more targets by identifying one or more non-zero values of the one or more sparse vectors.
The system as claimed in claim 7, wherein, the received at least one signal comprising the one or more pilot sequences and the one or more guard symbols is placed in the delay doppler domain frame as a plurality of row and column pilot structures, the one or more communication data are placed in the M*N matrix after the one or more pilot sequences and the one or more guard symbols, and wherein the at least one signal is transmitted from the one or more transmitters by converting the delay-Doppler domain frame to the delay-time domain frame by performing a signal modulation on the delay-Doppler domain frame and by adding a cyclic prefix to the at least one signal in the delay-time domain frame followed by column-wise vectorization.
The system as claimed in claim 8, wherein the one or more hardware processors are configured for converting the received at least one signal from the delay-time domain frame to the delay-doppler frame by removing the cyclic prefix from the received at least one signal and performing the signal demodulation on the received at least one signal for converting to the delay-doppler frame is represented as: Y r = R r F N , where, Y ( r ) is the received at least one signal at r th receiver in the delay-Doppler domain frame, R ( r ) is the received at least one signal at the r th receiver in the delay-time domain frame and F N is a N dimensional Discrete Fourier Transform matrix.
The system as claimed in claim 7, wherein the at least one signal from each of the one or more transmitters is distinguished by determining the cross correlation of the received at least one signal and the associated transmitted signal with the local version of the pilot sequence from each of the one or more transmitters, and is represented as: Σ Y r , p t , where, Y ( r ) is the received at least one signal at r th receiver in the delay-Doppler domain frame and p ( t ) is a transmitted pilot signal.
The system as claimed in claim 7, wherein the one or more hardware processors are configured for computing the plurality of phase information of the received at least one signal comprises of dividing the transformed cross correlation of the received at least one signal with the transformed cross correlation of an associated transmitted signal, and is represented as: ∑ u , v Y r ∼ , p t = ∑ l = 1 L ∑ t = 1 T a l e − j 2 π uτ l , bin M G + vμ l , bin N ∑ u , v X τ , μ t ∼ , p t + η ∼ , H u , v = ∑ l = 1 L ∑ t = 1 T a τl , μ l e − j 2 π uτ l , bin M G + vμ l , bin N + η ∼ , where, Y ( r ~ ) is the 2D-FFT of the extracted received signal in delay doppler (DD) domain, a τt,µ l is the path gain of l th path, µ l,bin is the Doppler bin shift of the l th path, τ l,bin is the delay bin shift of the l th path, T being the number of transmitters and L being the number of paths, H u,v is the plurality of phase information, ∑ u , v X τ , μ t ∼ , p t is the corelated signal between an extracted transmitted signal ( X τ , μ t ∼ ) and ( p ( t )) is the transmitted pilot signal, η ~ denotes a residual noise present in the received at least one signal, M G represents only those delay bins specific to pilot and guard symbols and N denotes the number of doppler bins.
The system as claimed in claim 7, wherein the one or more hardware processors are configured for estimating the one or more zero values of the one or more sparse vectors by comparing the plurality of columns of a dictionary matrix with the plurality of phase information using the estimation algorithm and identifying the one or more non-zero value of the one or more sparse vectors as paired delay-doppler shifts associated to the one or more targets is represented as: u s ∼ = min u s u s l 1 such that h − ϕu s l 2 < ϵ , where, h is the vectorised version of H , u s is the sparse vector with L significant values, l 1 is the l 1 norm, l 2 is the l 2 norm and ε is the threshold used in the estimation algorithm.
One or more non-transitory machine-readable information storage mediums comprising one or more instructions which when executed by one or more hardware processors cause: receiving at least one signal reflected from one or more targets in a delay-time domain frame at a radar receiver, wherein the received at least one signal comprises a M*N matrix, and wherein the M*N matrix comprises one or more pilot sequences, one or more guard symbols, and one or more communication data; demodulating the at least one signal to obtain a delay Doppler domain frame; extracting the one or more pilot sequences and the one or more guard symbols from the received at least one signal, wherein the one or more pilot sequences and the one or more guard symbols are known at the radar receiver and one or more transmitters from where the at least one signal is transmitted towards the one or more targets; determining a cross correlation of the received at least one signal and an associated transmitted signal by using the extracted one or more pilot sequences and a local version of a pilot sequence from each of the one or more transmitters; transforming the cross correlation of the received at least one signal and the associated transmitted signal to a frequency domain from the delay-doppler domain frame by performing a signal processing on the demodulated at least one signal, to obtain a transformed cross correlation of the received at least one signal and an associated transformed cross correlation of the transmitted signal; computing a plurality of phase information of the received at least one signal using the transformed cross correlation of the received at least one signal and the associated transformed cross correlation of the transmitted signal; constructing a dictionary matrix by selecting a plurality of pre-defined delay-doppler values computed using a 2D complex exponential signal technique; estimating one or more sparse vectors by solving a sparse recovery problem using the dictionary matrix and the plurality of phase information, wherein an estimation algorithm is used for solving the sparse recovery problem; and estimating one or more paired delay-doppler bin shifts with respect to at least one of the one or more targets by identifying one or more non-zero values of the one or more sparse vectors.
The one or more non-transitory machine readable information storage mediums in claim 13, wherein, the received at least one signal comprising the one or more pilot sequences and the one or more guard symbols is placed in the delay doppler domain frame as a plurality of row and column pilot structures, the one or more communication data are placed in the M*N matrix after the one or more pilot sequences and the one or more guard symbols, and wherein the at least one signal is transmitted from the one or more transmitters by converting the delay-Doppler domain frame to the delay-time domain frame by performing a signal modulation on the delay-Doppler domain frame and by adding a cyclic prefix to the at least one signal in the delay-time domain frame followed by a column-wise vectorization.
The one or more non-transitory machine readable information storage mediums in claim 14, wherein converting the received at least one signal from the delay-time domain frame to the delay-doppler frame by removing the cyclic prefix from the received at least one signal and performing the signal demodulation on the received at least one signal for converting to the delay-doppler frame is represented as: Y r = R r F N , where, Y ( r ) is the received at least one signal at r th receiver in the delay-Doppler domain frame, R ( r ) is the received at least one signal at the r th receiver in the delay-time domain frame, and F N is a N dimensional Discrete Fourier Transform matrix.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS AND PRIORITY The present application claims priority from Indian application no. 202421083424, filed on October 30, 2024. TECHNICAL FIELD The disclosure herein generally relates to, telecommunication, and, more particularly, to a system and method for estimating fractional bin shifts in Information Sharing and Analysis Center (ISAC) with a multistatic radar. BACKGROUND In recent times, there has been a huge surge in research related to advanced communication systems due to its performance in using higher frequency and efficiency in communication systems such as intelligent transportation systems for smart traffic control, autonomous vehicles, and so on. The next generation communication system uses Integrated Sensing and Communication system (ISAC), that aims to develop a single hardware system which processes both radar and communications and is beneficiary of both. ISAC system uses the same communication waveform for communicating with the receiver as well as for detecting the target. Earlier the Orthogonal Frequency Division multiplexing (OFDM) waveforms were used in ISAC, however, it was unable to retain its orthogonality and suffered a loss in performance due to high-speed vehicle communication. The Orthogonal Time Frequency Space (OTFS) was later introduced to deal with such problems. The OTFS channel in delay-Doppler domain (DD) gives the ability to work well in high doppler channels. The existing system fails in accurate estimation of fraction DD bin shifts. Most of the existing systems are used for computing integer values of DD bin shifts which results inaccurate target detection. The existing system computes the delay and doppler parameters in different iterations, which may cause an increase in computation time and add inaccuracy in target detection. Further and most of the systems are dependent on the shape/design of the pilots which makes the system to be dependent on the respective shape/design of the pilots. The existing system would not be able to distinguish between the different signals transmitted from the number of transmitters in high doppler channels. SUMMARY Embodiments of the present disclosure present technological improvements as solutions to one or more of the above-mentioned technical problems recognized by the inventors in conventional systems. For example, in one embodiment, a method for estimating fractional bin shifts in ISAC with multistatic radar is provided. The method includes: receiving, via one or more hardware processors, at least one signal reflected from one or more targets in a delay-time domain frame at a radar receiver, wherein the received at least one signal comprises a M*N matrix, and wherein the M*N matrix comprises one or more pilot sequences, one or more guard symbols, and one or more communication data; demodulating, via the one or more hardware processors, the at least one signal to obtain a delay Doppler domain frame; extracting, via the one or more hardware processors, the one or more pilot sequences and the one or more guard symbols from the received at least one signal, wherein the one or more pilot sequences and the one or more guard symbols are known at the radar receiver and one or more transmitters from where the at least one signal is transmitted towards the one or more targets; determining, via the one or more hardware processors, a cross correlation of the received at least one signal and an associated transmitted signal by using the extracted one or more pilot sequences and a local version of a pilot sequence from each of the one or more transmitters; transforming, via the one or more hardware processors, the cross correlation of the received at least one signal and the associated transmitted signal to a frequency domain from the delay-doppler domain frame by performing a signal processing on the demodulated at least one signal, to obtain a transformed cross correlation of the received at least one signal and an associated transformed cross correlation of the transmitted signal; computing, via the one or more hardware processors, a plurality of phase information of the received at least one signal using the transformed cross correlation of the received at least one signal and the associated transformed cross correlation of the transmitted signal; constructing, via the one or more hardware processors, a dictionary matrix by selecting a plurality of predefined delay-doppler values computed using a 2D complex exponential signal technique; estimating, via the one or more hardware processors, one or more sparse vectors by solving a sparse recovery problem using the dictionary matrix and the plurality of phase information, wherein an estimation algorithm is used for solving the sparse recovery problem; and estimating, via the one or more hardware processors, one or more paired delay-doppler bin shifts with respect to at least one of the one or more targets by identifying one or more non-zero values of the one