Virtual array for dual function MIMO radar communication systems using OTFS waveforms

Abstract

A radar-communication system using Orthogonal Time-Frequency Space (OTFS) waveforms is disclosed. The system includes a multiple-input multiple-output (MIMO) transceiver having transmit and receive chains coupled to multiple antennas, and processing circuitry configured to allocate, within a delay-Doppler grid and a corresponding time-frequency grid, a first set of bins as shared bins for concurrent communication and a second set of bins as private bins for sequential radar sensing. Signals received from the private bins are processed to generate a virtual array having an aperture exceeding that of the physical receive antennas, thereby enhancing sensing resolution with negligible loss of communication rate. The processing circuitry may employ time and frequency-shift operations including ISFFT / SFFT transforms. The disclosed approach enables efficient dual-function radar-communication (DFRC) operation by balancing sensing performance and communication throughput within a unified OTFS framework.

Description

VIRTUAL ARRAY FOR DUAL FUNCTION MIMO RADAR COMMUNICATION SYSTEMS USING OTFS WAVEFORMSGOVERNMENT SUPPORT

[0001] This invention was made with government support under Grants No. 2320568 and 2514270 awarded by the National Science Foundation (NSF ECCS) and under Grant No. W911NF2320103 awarded by the Army Research Office. The Government has certain rights in the invention.CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of U.S. Provisional Application No. 63 / 720,156, filed November 13, 2024. the entire contents of which are incorporated herein by reference.BACKGROUND

[0003] Dual-function radar-communication (DFRC) systems form a class of integrated sensing and communication (IS AC) architectures in which common hardware and waveforms support simultaneous radar and communication operations. In multiple-input multiple-output (MIMO) system DFRC implementations, different strategies exist for allocating waveform and spectrum resources among transmit antennas. An interleaved allocation can prevent overlapping between antennas but limits overall communication throughput. Conversely, fully shared resource allocation maximizes communication rate but constrains sensing performance and often requires computationally intensive target estimation techniques. Balancing the tradeoff between sensing accuracy and communication efficiency remains a key challenge for DFRC systems employing Orthogonal Time-Frequency Space (OTFS), a Doppler robust waveform for next-gen wireless systems, particularly when mapping resources within the delay-Doppler domain and coordinating antenna transmissions.BRIEF SUMMARY

[0004] Embodiments of the disclosure are directed to a dual-function radar-communication (DFRC) system that performs integrated sensing and communication tasks using multiple-input multiple-output (MIMO) antennas and Orthogonal Time-Frequency Space (OTFS) waveforms. The system operates efficiently in high-mobility and high-Doppler environments expected for next-generation wireless networks. Within an OTFS delay-Doppler and time-frequency grid.spectrum resources are allocated to enhance either communication or sensing functions. By default, all delay-Doppler bins and time-frequency bins are shared among multiple transmit antennas for maximized communication rate but coarse sensing performance, while when needed, a set of delay-Doppler bins is left empty and a set of time-frequency bins is privately assigned to selected antennas for enhanced sensing performance with a slightly reduction of communication rate. Signals obtained from the private bins can form a virtual array that provide enhanced sensing resolution without interrupting ongoing communication.

[0005] One aspect of the disclosure is directed to a radar-communication system configured to use OTFS waveforms. The system includes a MIMO transceiver having transmit and receive chains coupled to a plurality of antennas and processing circuitry configured to allocate shared and private resources, generate virtual array having an aperture exceeding that of the physical receive antennas, and sacrifice only a minimal portion of the communication rate.

[0006] Another aspect of the disclosure is directed to a method of operating a radarcommunication system configured to use OTFS waveforms. The method includes operating a MIMO transceiver, allocating shared and private resources within an OTFS frame, and generating virtual array from signals received over the private time-frequency bins, whose received signals are decoupled from those carried on the shared bins.

[0007] A further aspect of the disclosure is directed to a radar-communication processing circuit configured for use in a system employing OTFS waveforms. The circuit includes interface circuitry for coupling with a MIMO transceiver and processing circuitry configured to perform shared-private resources allocation, virtual array generation, and on demand activation of the above functions.

[0008] This Summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1A illustrates an example radar-communication system in which a MIMO transceiver operates with processing circuitry that allocates shared-private resources for joint sensing and communication.

[0010] Figure IB is a schematic block diagram of representative processing-circuitry modules supporting modulation, shared-private resource allocation, waveform generation, and sensing coordination.

[0011] Figure 2A presents a mathematical model of a doubly-selective delay-Doppler channel as a superposition of per-target components.

[0012] Figure 2B shows an example magnitude response in the delay-Doppler domain where discrete peaks correspond to individual target reflections.

[0013] Figure 2C depicts a representative delay-Doppler grid and corresponding timefrequency grid used for OTFS symbol mapping and resource definition.

[0014] Figure 2D conceptually illustrates the transform chain (ISFFT / Heisenberg channel — > Wigner / SFFT) linking delay-Doppler. time-frequency, and time domains.

[0015] Figure 2E provides representative expressions for received signals associated with shared communication bins and private sensing bins.

[0016] Figure 2F illustrates formation of a virtual array by stacking private-bin measurements across antennas.

[0017] Figure 2G shows example communication data-rate performance versus SNR for different counts of private bins, indicating minimal throughput impact.

[0018] Figure 2H shows example BER performance versus SNR under various private-bin configurations and modulation formats, indicating robust link reliability.

[0019] Figure 3A is a flow diagram of an exemplary method for operating a MIMO DFRC system with shared and private resources and virtual array.

[0020] Figure 3B is a flow diagram of an illustrative refinement showing shared-private resource allocation, and virtual-array generation within the OTFS framework.DETAILED DESCRIPTION

[0021] The radio-frequency (RF) front-end architectures in modem radar and communication systems are highly similar. To reduce cost and improve spectral efficiency, these systems can be integrated on a common platform that shares antennas, converters, and baseband circuitry, forming an integrated sensing and communication (ISAC) system. Shared utilization of hardware reduces cost, weight, and power consumption.

[0022] Among ISAC designs, dual-function radar-communication (DFRC) systems represent a distinctive class in which radar and communication functionalities share not only hardware but also a common waveform. DFRC systems maximize scarce spectrum resourcesby performing simultaneous sensing and communication within the same frequency band. Radar sensing is then performed using the transmitted modulated communication symbols.

[0023] With the growth of autonomous vehicles, self-navigating drones, and other platforms that depend on both radar sensing and high-rate data links, efficient DFRC operation has become increasingly important for vehicular networks, indoor positioning, and unmanned aerial systems.

[0024] Conventional MIMO DFRC techniques typically employ either fully shared or fully partitioned spectrum resources among antennas. In fully shared approaches, all antennas use common waveform resources, maximizing communication throughput but limiting sensing resolution and often requiring complex maximum-likelihood estimation for target detection. Fully partitioned approaches, in contrast, transmit over non-overlapping resources, improving sensing diversity but reducing communication throughput.

[0025] These methods therefore struggle to achieve both high-resolution sensing and high-throughput communication, particularly in high-Doppler environments where orthogonal frequency-division multiplexing (OFDM) performance degrades.

[0026] Embodiments of the disclosure employ orthogonal time-frequency space (OTFS) waveforms, which modulate information in the delay-Doppler (DD) domain rather than the time-frequency domain of conventional OFDM. OTFS modulation transforms a time-vary ing channel into a quasi-time-invariant DD representation that is resilient to delay and Doppler spread, and supports full diversity.

[0027] Using OTFS, the disclosed DFRC system implements a shared-private resourceallocation strategy7in which processing circuitry7allocates bins within a delay-Doppler and time-frequency grid to balance communication and sensing. By default, all resources are shared by all antennas to maximize the communication rate. A low complexity7coarse sensing algorithm is proposed for this allocation. When high-resolution sensing is needed, some DD bins are left empty and some time-frequency bins are set private to certain antennas. That is, to support the creation of time-frequency bins each antenna leaves delay-Doppler bins empty7, such that there is an equal number of empty bins to the number of private bins on other antennas. (See Figure 2D) Signals received over the private bins are processed to construct a virtual array that effectively increase the aperture beyond the limits of physical receive antennas.

[0028] Through this allocation, coarse target estimates can be obtained using low-complexity Fourier transform and cross-correlation operations in the DD grid, followed byrefinement through virtual-array synthesis and sparse signal recovery techniques. This approach enables high-resolution sensing with minimal loss of the communication rate.

[0029] Accordingly, embodiments described herein provide an OTFS-based DFRC framework capable of reliable, high-performance sensing and communication for intelligent-transportation, autonomous-navigation, and real-time environmental-monitoring applications.

[0030] As illustrated in Figure 1A, the framework may be embodied in a radarcommunication system 100 that integrates sensing and communication functionalities within a unified signaling architecture. The system 100 includes a multiple-input multiple-output (MIMO) transceiver 110 configured to both transmit and receive wireless signals through respective transmit (TX) chains 112 and receive (RX) chains 116, with signal propagation occurring through an environment 114 that represents both the wireless propagation path and radar targets such as reflectors or moving objects.

[0031] In contrast to conventional implementations that treat radar and communication as distinct subsystems, processing circuitry' 120 dynamically coordinates both functions through common waveform and resource management. The processing circuitry 120 receives and interprets sensing and communication data streams 102, which carry payload data and radarsensing information within a shared modulation framework, and also manages control signaling 104 that governs transmitter activation, timing, and adaptive mode selection in response to channel and sensing conditions.

[0032] At the core of this processing functionality, a shared-private resource allocator 128 performs j oint allocation of delay-Doppler bins 130 and time-frequency bins 132 under an orthogonal time-frequency space (OTFS) signaling scheme. As discussed in greater detail with respect to Figures 2A-2C, the delay-Doppler framework maps information symbols onto a two-dimensional grid of delay and Doppler coordinates, enabling joint time- and frequency-diverse signaling. The allocator 128 configures private bin generation for enhanced radar sensing, effectively forming a virtual array that coexists with data channels in the same spectral and temporal resources. Through this coordinated bin allocation and control interface, the system achieves unified radar-sensing and communication operation within a common delay-Doppler domain-enhancing spectral efficiency, environmental awareness, and operational robustness without requiring separate signaling channels.

[0033] In certain implementations, as few' as two transmit antennas are scheduled for private-bin sensing at a given time slice to bound overhead while preserving sensing aperture; other number of antennas may be used as conditions warrant.

[0034] As used herein, shared bins refer to time-frequency resource elements allocated for data communication within the unified radar-communication framework. Each shared bin corresponds to a region of the time-frequency grid used to convey modulated payload symbols associated with one or more communication users. As illustrated conceptually in Figure 2E, these bins may be dynamically assigned, reused, or jointly utilized across multiple transmit and receive antenna paths to optimize throughput, spectral efficiency, and interference management.

[0035] Conversely, private bins refer to time-frequency resources supported by delay-Doppler resource elements reserved for enhancing radar-sensing operations. The private-bin transmissions are coordinated in time and frequency with the shared-bins (see Figures 2E-2F for representative signal relationships). Signals received on private bins carry information only from certain antenna, which achieves orthogonality and enables formation of a virtual array that refines the radar-parameter estimation performance.

[0036] In various embodiments, the shared-private resource allocator 128 adaptively enables certain delay -Doppler domain and time-frequency resources to switch between shared and private bins based on real-time channel conditions, sensing requirements, or operational priorities. This adaptive allocation ensures that the sensing and communication processes remain spectrally coherent and temporally synchronized, while maintaining independent control of modulation, coding, and waveform generation for each bin type.

[0037] Radar-communication system 100 may operate under an OTFS signaling scheme or other time-frequency modulation format that supports mapping of information into delay-Doppler resources (see Figures 2A-2D). The architecture enables concurrent or alternating modes for sensing and communication, depending on application requirements and environmental conditions.

[0038] The MIMO transceiver 110 can include one or more transmit and receive antennas, radio-frequency (RF) front-end circuitry, and associated baseband interfaces that support spatial diversity', beamforming, and multiplexing for both radar and communication functions.

[0039] The TX chain 112 represents the transmission path of the transceiver, including digital baseband generation, OTFS or modulation, digital-to-analog conversion, up-conversion, and power amplification. The TX chain 112 generates and emits the composite radar-communication yvaveform corresponding to the allocated shared-private resources.

[0040] The environment 114 models the wireless propagation medium between the transceiver antennas and surrounding objects. It encompasses both the communication channel linking the transmitter and intended receiver(s) and the radar channel formed by reflectionsfrom environmental targets such as vehicles, pedestrians, or infrastructure. As shown conceptually in Figure 2B, the environment introduces delay, Doppler, attenuation, and phase shifts that are jointly estimated and compensated through the processing circuitry.

[0041] The RX chain 116 represents the receiver path of the transceiver, including low-noise amplification, down-conversion, analog-to-digital conversion, synchronization, and demodulation. The RX chain 116 captures returning echoes and communication signals, providing baseband data streams for digital processing and sensing analysis.

[0042] The sensing and communication data streams 102 denote the digital interfaces that convey symbol data, processed waveforms, and sensing information between the transceiver and the processing circuitry. These streams may include data mapped to specific delay-Doppler bins as well as derived channel-state or target-state parameters produced by radar-sensing functions.

[0043] Control signaling 104 includes timing, activation, configuration, and feedback signals exchanged between the processing circuitry and transceiver elements. Such signaling supports adaptive mode selection, antenna coordination, and dynamic reconfiguration of delay-Doppler bins and time-frequency bins assignment.

[0044] As illustrated in Figure IB, processing circuitry 120 may include various functional modules configured to implement modulation, shared-private resource allocation, w aveform generation, and sensing coordination in support of the radar-communication framework introduced in Figure 1A. The illustrated modules collectively enable dynamic configuration of shared and private resources and the generation of virtual array that operate in parallel with data-communication channels. It should be understood that the particular organization shown in Figure IB is merely representative; equivalent functionality may be realized through hardware, firmware, or software in other configurations.

[0045] The processing chain may begin with a serial-to-parallel converter 122, which partitions incoming symbol or bit streams into multiple parallel paths suitable for domain mapping. A quadrature-amplitude mapper 124 then modulates the parallel data streams into complex-valued symbols for delay-Doppler placement.

[0046] The transformed data are provided to a shared-private resource allocator 128, which determines how delay-Doppler bins and time-frequency bins are used to balance sensing and communication functions. .

[0047] Within the allocator 128, a delay-Doppler bin controller 130 assign bins intended for payload communication while identify ing bins that will remain empty for later assignment of time-frequency private bins. An inverse symplectic finite Fourier transform (ISFFT) module126 converts the modulated symbols from delay-Doppler grid to a time-frequency representation. The time-frequency bin controller 132 designates bins reserved for certain antennas only to enhance radar-sensing operations. These controllers can coordinate in real time to adjust allocation ratios or reconfigure bin positions in response to environmental conditions, channel quality, or sensing priority. (See Figures 2A-2E for representative delay-Doppler mappings corresponding to shared and private bins.)

[0048] A digital / analog interface 134 manages exchange of waveform data between the digital processing domain and the analog front-end of the transceiver, ensuring proper formatting for transmission and reception.

[0049] To perform sensing on all bins (no matter whether it is shared or private), a low complexity approach is employed through the low complexity sensing operator 136. These approaches emulate the spatial and temporal structure of physical radar returns, enabling multitarget detection or environmental mapping while maintaining coexistence with data traffic (see Figure 2B). In embodiments, the transmitter side can end after the digital / analog interface 134 and the receiver side can start before operator 136.

[0050] A virtual array generator 138 synthesizes a virtual array with the received timefrequency private bins. The resulting virtual array achieves an effective aperture exceeding that of the physical receive antennas, thereby enhancing sensing resolution and target-parameter estimation. In some implementations, the operator 138 also controls activation codes and timing of private-bin functions to coordinate sequential antenna operation.

[0051] Building on the system architecture of Figures 1A and IB, Figure 2A presents a mathematical model of the delay-Doppler (DD) domain channel used to describe signal propagation within the radar-communication framework. In this representation, the wireless environment is modeled as a J-tap doubly selective DD channel, where each tap corresponds to an individual target or reflector contributing a distinct delay and Doppler shift.

[0052] The composite channel response h(v, T) may be expressed as a superposition of pertarget components, each characterized by a reflection coefficient i. Doppler shift vj, and delay Tj. This abstraction captures how multipath propagation and target motion manifest in the DD domain and forms the analytical foundation for OTFS modulation and joint radarcommunication processing described in the following figures.

[0053] The DD-domain channel model provides a compact, quasi-time-invariant representation of range and velocity effects, enabling subsequent delay-Doppler mapping, transform operations, and virtual-array formation shown in Figures 2B through 2F.

[0054] Figure 2B illustrates an example magnitude response of the channel in the delay-Doppler domain. In this representation, discrete peaks correspond to individual target reflections, each characterized by a distinct combination of delay and Doppler shift as defined in the channel model of Figure 2A. The spatial positions and amplitudes of the peaks convey target range, velocity, and reflection strength, forming the observable basis for radar sensing within the unified OTFS framework. This depiction provides an intuitive view of how multipath and motion effects manifest as resolvable structures in the delay-Doppler plane, enabling the joint estimation techniques described in subsequent figures.

[0055] Figure 2C depicts a representative Doppler-delay grid and its corresponding timefrequency grid used for OTFS symbol mapping. Each grid defines the discrete coordinates over which delay-Doppler and time-frequency bins are used to place symbols and transformed symbols. The figure also illustrates the relationship between key waveform parameters — symbol duration, bandwidth, and orthogonality — that govern the mapping between delay-Doppler and time-frequency domains.

[0056] In some embodiments, shared and private resources are organized as subcarrier groups in the time-frequency domain that map bijectively to corresponding delay-Doppler bin groups;.

[0057] This structure establishes the two-dimensional resource space within which shared and private bins are defined for concurrent radar and communication functions (see Figures 2E-2F). The grid parameters Av and Ar correspond to the Doppler and delay resolutions achievable in the system and form the basis for the subsequent transform operations shown in Figure 2D.

[0058] In some embodiments, the private resources within the time-frequency (TF) bins that are intentionally “zero-forced” for specific transmit antennas to maintain orthogonality, while certain delay-Doppler (DD) bins remain empty to maintain invertibility of the modulation structure. During demodulation, a modified symplectic finite Fourier transform (SFFT) may be applied to account for these unused TF entries, thereby reconstructing the DD-domain signal without interference from the private-bin nulls.

[0059] Figure 2D conceptually illustrates the signal transformations that link the delay-Doppler, time-frequency, and time domains during joint radar-communication operation. Modulation symbols xL[k, Z]in the delay-Doppler grid are first converted to time-frequency representations X,[n, m] through an inverse symplectic finite Fourier transform (ISFFT). AHeisenberg transform then synthesizes the corresponding continuous -time transmit waveforms s,(t) for each transmit antenna.

[0060] The transmitted signals propagate through a channel containing one or more reflecting targets, modeled as a doubly selective delay-Doppler channel. At the receiver, the incoming waveform r„r(t)is analyzed using a Wigner transform to obtain the received timefrequency representation Kn?.[ , m], A subsequent symplectic finite Fourier transform (SFFT) maps the received data back into the delay-Doppler domain yrir[k, Z] , where information and sensing responses are jointly extracted.

[0061] This transform sequence demonstrates how the system processes both communication and sensing signals within a unified OTFS framework, maintaining domain coherence between the delay-Doppler resource allocation of Figures 1A-1B and the virtual-array formulations introduced in later figures.

[0062] In the described embodiments, all modulation and demodulation operations are implemented using an ISFFT and SFFT pair. These transforms provide the mathematical framework for mapping between the delay-Doppler domain and the time-frequency domain in the OTFS signaling process. References in earlier descriptions (e.g., in the provisional filing) to inverse Zak transform (IZT) or Zak transform (ZT) are to be understood as conceptually equivalent to ISFFT+Heisenberg Transform and Wigner Transform+SFFT, respectively.

[0063] Figure 2E presents representative expressions for received signals corresponding to shared and private delay-Doppler bins within the unified OTFS framework. The upper relation describes the received signal Ynr[n,m] associated with shared bins, in which modulated symbols X„t[n, m] convey communication data through the composite channel response H [n, m] .

[0064] The lower relation depicts the received signal Ynr[np, mp] corresponding to private bins, where known or predetermined symbols Xp[np, mpare transmitted to facilitate radar sensing. These private-bin signals capture reflections from targets, enabling estimation of range, velocity, and angle parameters through subsequent radar algorithm processing.

[0065] In representative embodiments, the private resources may correspond to timefrequency (TF) bins that are intentionally zero-forced for certain transmit antennas, while certain delay-Doppler (DD) bins are left empty to preserve invertibility of the signal representation. During transmission, each antenna omits modulation at other antennas’ assigned private TF locations, creating known spectral gaps that are later exploited for sensing. At the receiver, the composite TF-domain signal is equalized and processed through a modifiedsymplectic finite Fourier transform (SFFT) that accounts for the omitted TF entries. This modified SFFT reconstructs the DD-domain symbols without interference from the private TF nulls, allowing radar-sensing reflections to be isolated from concurrently transmitted communication data.

[0066] Figure 2F illustrates formation of a virtual sensing array by stacking received signals from the private bins across multiple antennas. The combined response rprepresents a weighted sum of per-target array responses, each characterized by its reflection coefficient pj, Doppler shift Vj, delay Tj, and angular component 0j.

[0067] The resulting composite model, expressed asr ::::^ , captures the spatial, delay and Doppler structure of reflected signals within the unified framework. This formulation enables the system to synthesize virtual array whose number and resolution exceed those of the physical receive antennas, thereby improving target detection and parameter estimation while preserving coexistence with communication signaling.

[0068] In certain implementations, the delay-Doppler responses extracted from the private bins may undergo refinement using sparse-signal recovery (SSR) algorithms, such as orthogonal matching pursuit (OMP), basis pursuit (BP), or subspace pursuit (SP). These algorithms operate over a reduced search space centered around coarse angle-delay-Doppler estimates, improving angle, range and velocity accuracy with low computational complexity. The SSR-based refinement may also support estimation of fractional Doppler components and angle-of-arrival parameters when forming the virtual array response.

[0069] In high-mobility or high-Doppler environments, fractional Doppler shifts can cause inter-Doppler interference (IDI) that spreads symbol energy across multiple delay-Doppler indices. The virtual-array architecture described herein inherently provides Doppler diversity, allowing accurate recovery of fractional Doppler components and maintaining estimation performance near the Cramer-Rao lower bound (CRLB) across a wide range of signal-to-noise ratios.

[0070] In practical implementation, the system forms the virtual array by stacking privatebin responses corresponding to the zero-forced time-frequency locations across all transmit antennas. Each private-bin measurement contributes a distinct spatial signature determined by antenna position and waveform phase, enabling synthesis of a virtual aperture whose effective size equals the number of private bins multiplied by the number of transmit antennas. Coarse target estimates obtained from the delay-Doppler grid may be refined using sparse signal recovery (SSR) techniques such as orthogonal matching pursuit (OMP), basis pursuit (BP), orsubspace pursuit (SP). These algorithms operate over a reduced dictionary' centered around the coarse delay-Doppler-angle estimates, achieving high-resolution parameter recovery’ and fractional Doppler estimation with low computational cost.

[0071] Figure 2G shows an example relationship between communication data rate and signal-to-noise ratio (SNR) for OTFS-based operation under different allocations of private bins. The curves illustrate that introducing a small number of private bins for radar sensing produces only a minor reduction in communication throughput compared to using all bins for data transmission. As the number of private bins increases, sensing capability improves while maintaining nearly constant data-rate performance.

[0072] This demonstrates that the disclosed delay-Doppler resource-allocation framework achieves efficient coexistence of radar and communication functions, allowing dynamic tradeoffs between sensing resolution and communication throughput within a common spectral resource.

[0073] Figure 2H illustrates example bit-error-rate (BER) performance as a function of signal-to-noise ratio (SNR) for various private-bin configurations and modulation formats. The results show that OTFS signaling maintains robust communication performance even as private bins are allocated for radar sensing. Systems employing a small number of private bins (e.g., Np= 2or Np= 4) exhibit only marginal BER degradation relative to configurations using all shared bins, demonstrating that high-order modulations such as 16QAM and 16PSK remain viable under integrated operation.

[0074] These results further confirm that the delay-Doppler resource-allocation framework achieves joint sensing and communication without significant compromise in link reliability', supporting the adaptive, coexistence-oriented design described in Figures 1A through 2G.

[0075] Figure 3A presents an exemplary' method for performing integrated radar and communication functions within an OTFS-based delay-Doppler framework.

[0076] At step 310, the system operates a multiple-input multiple-output (MIMO) transceiver having transmit and receive chains that include digital-to-analog and analog-to-digital conversion circuitry coupled to a plurality of antennas. The transceiver transmits and receives signals that carry both sensing and communication information.

[0077] At step 312, processing circuitry coupled to the transceiver allocates, within a delay-Doppler grid and time-frequency grid of an OTFS frame. By default, All delay-Doppler grid and time-frequency grid are shared by all transmit antenna to maximize the communication throughput. When enhanced sensing is required, some time-frequency bins are reassigned asprivate bins for specific antennas by enforcing zeros on the corresponding bins of other antennas. To preserve the invertibility, each antenna leaves an equal number of delay-Doppler bins empty that corresponds to private-bin locations of other antennas.

[0078] At step 314, the system generates, from signals received over the private bins, a virtual array having an aperture exceeding the number of physical receive antennas. The virtual array is formed independently of communication signals carried on the shared bins, thereby enhanced sensing resolution without interrupting data transmission.

[0079] Figure 3B expands on the resource-allocation process introduced in Figure 3A, illustrating how the system dynamically transitions between shared- and private-bin operation within the OTFS delay-Doppler and time-frequency framework.

[0080] At step 320. the MIMO transceiver operates by default with all delay-Doppler and time-frequency bins shared among transmit antennas to maximize communication throughput. When enhanced sensing is required, control signaling activates a transmit-private-bin mode for selected antennas. This signaling is coordinated by the processing circuitry shown in Figures 1A-1B, which governs antenna scheduling and timing of private-bin activation.

[0081] At step 322, an inverse symplectic finite Fourier transform (ISFFT) converts the modulated symbols from the delay-Doppler grid to their time-frequency representation. Private time-frequency bins are then established for designated transmit antennas by enforcing zeros on the corresponding bins of other antennas, thereby maintaining orthogonality between sensing and communication signals.

[0082] At step 324, the system maps information-bearing symbols back onto the delay-Doppler grid while leaving a number of bins empty7equal to the number of private bins assigned to other antennas.

[0083] Although the subject matter has been described in language specific to structural features and / or acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as examples of implementing the claims and other equivalent features and acts that would be recognized by one skilled in the art are intended to be within the scope of the claims.

[0084] For instance, elements presented herein may be realized using discrete logic, programmable circuitry, or software-defined processing architectures, and are not limited to any particular hardware topology7. In various embodiments, the same or equivalent functional blocks can be distributed, combined, or reconfigured depending on system bandwidth, antennaconfiguration, and implementation technology, thereby supporting the full range of operational modes described in the present disclosure.

Claims

CLAIMSWhat is claimed is:

1. A radar-communication system configured to use Orthogonal Time-Frequency Space (OTFS) waveforms, comprising:a multiple-input multiple-output (MIMO) transceiver having a transmit chain and a receive chain, each including digital-to-analog and analog-to-digital conversion circuitry coupled to a plurality of transmit and receive antennas;processing circuitry coupled to the MIMO transceiver and configured to:allocate, within a delay -Doppler grid and a corresponding time-frequency grid of an OTFS frame,a first set of time-frequency bins as shared bins accessible to multiple transmit antennas for concurrent communication, anda second set of time-frequency bins as private bins each associated with a respective transmit antenna for sequential sensing transmission; and generate, from signals received over the private bins, a virtual array having an aperture exceeding a number of physical receive antennas of the system, the virtual array being formed independently of communication signals carried on the shared bins.

2. The radar-communication system of claim 1, wherein the private bins are defined by enforcing zeros on corresponding bins of other transmit antennas and leaving complementary bins empty to preserve data invertibility.

3. The radar-communication system of claim 1, wherein all delay-Doppler bins are operated as shared communication bins during normal operation, andwherein, when a sensing function is required, the processing circuitry temporarily reassigns a subset of the shared bins as private bins for radar sensing.

4. The radar-communication system of claim 1, wherein the processing circuitry is configured to generate the delay-Doppler bins by superimposing time-shift and frequency-shift operations applied to modulation symbols.

5. The radar-communication system of claim 1, wherein the processing circuitry is configured to process the signals in the private bins to enhance target angle, range and velocity estimation based on angle, delay and Doppler separation of the received signals.

6. The radar-communication system of claim 1, wherein the processing circuitry performs an inverse Zak transform (ISFFT+Heisenberg Transform) for modulation in the transmit chain and a Zak transform (Wigner Transform+SFFT) for demodulation in the receive chain.

7. The radar-communication system of claim 1, wherein as few as two transmit antennas are assigned to the private bins.

8. The radar-communication system of claim 1, wherein the processing circuitry activates the private bins sequentially according to an antenna-specific activation demand.

9. A method of operating a radar-communication system configured to use Orthogonal Time-Frequency Space (OTFS) waveforms, the method comprising:operating a multiple-input multiple-output (MIMO) transceiver having a transmit chain and a receive chain, each including digital -to-analog and analog-to-digital conversion ci rcui try coupled to a plurality of transmit and receive antennas;allocating, by processing circuitry coupled to the MIMO transceiver and within a delay -Doppler grid and a corresponding time-frequency grid of an OTFS frame,a first set of delay-Doppler bins as shared bins accessible to multiple transmit antennas for concurrent communication, anda second set of delay-Doppler bins as private bins each associated with a respective transmit antenna for sequential sensing transmission; andgenerating, from signals received over the private bins, a virtual array having an aperture exceeding a number of physical receive antennas of the system, the virtual array being formed independently of communication signals carried on the shared bins.

10. A radar-communication processing circuit configured for use in a radar-communication system employing Orthogonal Time-Frequency Space (OTFS) waveforms, the circuit comprising:interface circuitry configured to couple with a multiple-input multiple-output (MIMO) transceiver having a transmit chain and a receive chain; andprocessing circuitry configured to:allocate, within a delay-Doppler grid and a corresponding time-frequency grid of an OTFS frame,a first set of time-frequency bins as shared bins accessible to multiple transmit antennas for concurrent communication, anda second set of time-frequency bins as private bins each associated with a respective transmit antenna for sequential sensing transmission; andgenerate, from signals received over the private bins, a set of virtual array having an aperture exceeding a number of physical receive antennas of the system, the virtual array being formed independently of communication signals carried on the shared bins.

11. The radar-communication processing circuit of claim 10, wherein the interface circuitry includes a serial-to-parallel converter configured to distribute modulation symbols to multiple transmit-antenna paths corresponding to the shared bins of the delay-Doppler grid.

12. The radar-communication processing circuit of claim 10, wherein the processing circuitry includes a quadrature-amplitude-modulation (QAM) mapper configured to map data symbols to complex modulation symbols for transmission through the shared bins while preserving sensing allocation for the private bins.

13. The radar-communication processing circuit of claim 10, wherein the processing circuitry includes an inverse symplectic finite Fourier transform (ISFFT) module for modulation in a transmit path and a symplectic finite Fourier transform (SFFT) module for demodulation in a receive path, each operating on symbols arranged within the delay-Doppler grid allocated to the shared and private bins.

14. The radar-communication processing circuit of claim 10, wherein the interface circuitry further includes digital-to-analog converters (D / A) coupled to transmit antennas and analog-to-digital converters (A / D) coupled to receive antennas, the converters being synchronized with sequential activation of the private bins according to an antenna-specific activation code.

15. The radar-communication processing circuit of claim 10, wherein the processing circuitry is configured to apply an activation on-demand to control sequential activation of transmitantennas associated with the private bins such that as few as two transmit antennas are active for sensing at a given time.

Images