Rate-Limited Closed-Loop Distributed ISAC Systems: An Autoencoder Approach

In closed-loop distributed multi-sensor integrated sensing and communication (ISAC) systems, performance often hinges on transmitting high-dimensional sensor observations over rate-limited networks. In this paper, we first present a general framework for rate-limited closed-loop distributed ISAC systems, and then propose an autoencoder-based observation compression method to overcome the constraints imposed by limited transmission capacity. Building on this framework, we conduct a case study using a closed-loop linear quadratic regulator (LQR) system to analyze how the interplay among observation, compression, and state dimensions affects reconstruction accuracy, state estimation error, and control performance. In multi-sensor scenarios, our results further show that optimal resource allocation initially prioritizes low-noise sensors until the compression becomes lossless, after which resources are reallocated to high-noise sensors.

Bridging the Gap via Data-Aided Sensing: Can Bistatic ISAC Converge to Genie Performance?

We investigate data-aided iterative sensing in bistatic OFDM ISAC systems, focusing on scenarios with co-located sensing and communication receivers. To enhance target detection beyond pilot-only sensing methods, we propose a multi-stage bistatic OFDM receiver, performing iterative sensing and data demodulation to progressively refine ISAC channel and data estimates. Simulation results demonstrate that the proposed data-aided scheme significantly outperforms pilot-only benchmarks, particularly in multi-target scenarios, substantially narrowing the performance gap compared to a genie-aided system with perfect data knowledge. Moreover, the proposed approach considerably expands the bistatic ISAC trade-off region, closely approaching the probability of detection-achievable rate boundary established by its genie-aided counterpart.

Disjoint Delay-Doppler Estimation in OTFS ISAC with Deep Learning-aided Path Detection

In this work, the problem of communication and radar sensing in orthogonal time frequency space (OTFS) with reduced cyclic prefix (RCP) is addressed. A monostatic integrated sensing and communications (ISAC) system is developed and, it is demonstrated that by leveraging the cyclic shift property inherent in the RCP, a delay-Doppler (DD) channel matrix that encapsulates the effects of propagation delays and Doppler shifts through unitary matrices can be derived. Consequently, a novel low-complexity correlation-based algorithm performing disjoint delay-Doppler estimation is proposed for channel estimation. Subsequently, this estimation approach is adapted to perform radar sensing on backscattered data frames. Moreover, channel estimation is complemented by a deep learning (DL) architecture that improves path detection and accuracy under low signal-to-noise ratio (SNR) conditions, compared to stopping criterion (SC) based multipath detection. Simulation results indicate that the proposed estimation scheme achieves lower normalized mean squared error (NMSE) compared to conventional channel estimation algorithms and sensing performance close to the Cramer-Rao lower bound (CRLB). Furthermore, an iterative data detection algorithm based on matched filter (MF) and combining is developed by exploiting the unitary property of delay-Doppler parameterized matrices. Simulation results reveal that this iterative scheme achieves performance comparable to that of the linear minimum mean squared error (LMMSE) estimator while significantly reducing computational complexity.

UNILoc: Unified Localization Combining Model-Based Geometry and Unsupervised Learning

Accurate mobile device localization is critical for emerging 5G/6G applications such as autonomous vehicles and augmented reality. In this paper, we propose a unified localization method that integrates model-based and machine learning (ML)-based methods to reap their respective advantages by exploiting available map information. In order to avoid supervised learning, we generate training labels automatically via optimal transport (OT) by fusing geometric estimates with building layouts. Ray-tracing based simulations are carried out to demonstrate that the proposed method significantly improves positioning accuracy for both line-of-sight (LoS) users (compared to ML-based methods) and non-line-of-sight (NLoS) users (compared to model-based methods). Remarkably, the unified method is able to achieve competitive overall performance with the fully-supervised fingerprinting, while eliminating the need for cumbersome labeled data measurement and collection.

Distributed Intelligent Sensing and Communications for 6G: Architecture and Use Cases

The Distributed Intelligent Sensing and Communication (DISAC) framework redefines Integrated Sensing and Communication (ISAC) for 6G by leveraging distributed architectures to enhance scalability, adaptability, and resource efficiency. This paper presents key architectural enablers, including advanced data representation, seamless target handover, support for heterogeneous devices, and semantic integration. Two use cases illustrate the transformative potential of DISAC: smart factory shop floors and Vulnerable Road User (VRU) protection at smart intersections. These scenarios demonstrate significant improvements in precision, safety, and operational efficiency compared to traditional ISAC systems. The preliminary DISAC architecture incorporates intelligent data processing, distributed coordination, and emerging technologies such as Reconfigurable Intelligent Surfaces (RIS) to meet 6G's stringent requirements. By addressing critical challenges in sensing accuracy, latency, and real-time decision-making, DISAC positions itself as a cornerstone for next-generation wireless networks, advancing innovation in dynamic and complex environments.

Joint Localization and Synchronization in Downlink Distributed MIMO

We investigate joint localization and synchronization in the downlink of a distributed multiple-input-multiple-output (D-MIMO) system, aiming to estimate the position and phase offset of a single-antenna user equipment (UE) using downlink transmissions of multiple phase-synchronized, multi-antenna access points (APs). We propose two transmission protocols: sequential (P1) and simultaneous (P2) AP transmissions, together with the ML estimators that either leverage (coherent estimator) or disregard phase information (non-coherent estimator). Simulation results reveal that downlink D-MIMO holds significant potential for high-accuracy localization while showing that P2 provides superior localization performance and reduced transmission latency.

Optimized Vehicular Antenna Placement for Phase-Coherent Positioning

Distributed multi-antenna systems are an important enabling technology for future intelligent transportation systems (ITS), showing promising performance in vehicular communications and near-field (NF) localization applications. This work investigates optimal deployments of phase-coherent sub-arrays on a vehicle for NF localization in terms of a Cram\'er-Rao lower bound (CRLB)-based metric. Sub-array placements consider practical geometrical constraints on a three-dimensional vehicle model accounting for self-occlusions. Results show that, for coherent NF localization of the vehicle, the aperture spanned by the sub-arrays should be maximized and a larger number of sub-arrays results in more even coverage over the vehicle orientations under a fixed total number of antenna elements, contrasting with the outcomes of incoherent localization. Moreover, while coherent NF processing significantly enhances accuracy, it also leads to more intricate cost functions, necessitating computationally more complex algorithms than incoherent processing.

RIS-Enabled Self-Localization with FMCW Radar

In the upcoming vehicular networks, reconfigurable intelligent surfaces (RISs) are considered as a key enabler of user self-localization without the intervention of the access points (APs). In this paper, we investigate the feasibility of RIS-enabled self-localization with no APs. We first develop a digital signal processing (DSP) unit for estimating the geometric parameters such as the angle, distance, and velocity and for RIS-enabled self-localization. Second, we set up an experimental testbed consisting of a Texas Instrument frequency modulated continuous wave (FMCW) radar for the user and SilversIMA module for the RIS. Our results confirm the validity of the developed DSP unit and demonstrate the feasibility of RIS-enabled self-localization.

Joint Design of Radar Receive Filter and Unimodular ISAC Waveform with Sidelobe Level Control

Integrated sensing and communication (ISAC) has been considered a key feature of next-generation wireless networks. This paper investigates the joint design of the radar receive filter and dual-functional transmit waveform for the multiple-input multiple-output (MIMO) ISAC system. While optimizing the mean square error (MSE) of the radar receive spatial response and maximizing the achievable rate at the communication receiver, besides the constraints of full-power radar receiving filter and unimodular transmit sequence, we control the maximum range sidelobe level, which is often overlooked in existing ISAC waveform design literature, for better radar imaging performance. To solve the formulated optimization problem with convex and nonconvex constraints, we propose an inexact augmented Lagrangian method (ALM) algorithm. For each subproblem in the proposed inexact ALM algorithm, we custom-design a block successive upper-bound minimization (BSUM) scheme with closed-form solutions for all blocks of variable to enhance the computational efficiency. Convergence analysis shows that the proposed algorithm is guaranteed to provide a stationary and feasible solution. Extensive simulations are performed to investigate the impact of different system parameters on communication and radar imaging performance. Comparison with the existing works shows the superiority of the proposed algorithm.

Reduced-latency DL-based Fractional Channel Estimation in OTFS Receivers

In this work, we propose a deep learning (DL)-based approach that integrates a state-of-the-art algorithm with a time-frequency (TF) learning framework to minimize overall latency. Meeting the stringent latency requirements of 6G orthogonal time-frequency space (OTFS) systems necessitates low-latency designs. The performance of the proposed approach is evaluated under challenging conditions: low delay and Doppler resolutions caused by limited time and frequency resources, and significant interpath interference (IPI) due to poor separability of propagation paths in the delay-Doppler (DD) domain. Simulation results demonstrate that the proposed method achieves high estimation accuracy while reducing latency by approximately 55\% during the maximization process. However, a performance trade-off is observed, with a maximum loss of 3 dB at high pilot SNR values.