This paper explores data compression’s key role in cooperative sensing within integrated sensing and communication (ISAC) networks, where range-angle maps generated at each base station (BS) are shared with a fusion center (FC). Efficient compression schemes minimize network overhead while ensuring accurate target detection and localization. Motivated by this challenge, we propose three novel compression approaches tailored for a network of multiple-input multiple-output (MIMO)-orthogonal frequency division multiplexing (OFDM)-based mono-static sensors: i) excision filtering (EF) for map sifting, ii) principal component analysis (PCA) for dimensionality reduction, and iii) quantization for efficient encoding. To localize both point-like and extended targets from fused range-angle maps, we exploit the density-based spatial clustering of applications with noise (DBSCAN) clustering algorithm. Given that DBSCAN requires careful tuning of its clustering parameters, we introduce an artificial intelligence (AI)-driven method to optimize these settings dynamically. A comprehensive rate-distortion analysis evaluates the network’s localization performance under varying compression levels. Key metrics—including bit rate, generalized optimal sub-pattern assignment (GOSPA) error, missed detection rate, and false alarm rate—provide a holistic assessment that balances localization accuracy with network overhead.

Orthogonal time frequency space (OTFS) modulation is gaining recognition for its potential to facilitate integrated sensing and communication (ISAC) within future mobile networks. However, computing the sensing channel matrix in orthogonal time frequency space (OTFS), a crucial step for accurate target parameter estimation, presents significant challenges due to its high dimensionality. Therefore, this study introduces an innovative method to reduce such computational complexity by combining two ingredients. First, through algebraic operations, we decompose the sensing channel matrix into four lower-dimensional matrices whose elements can be associated with a Dirichlet kernel. Second, we formulate an analytical criterion, independent of system parameters, that leverages the properties of the Dirichlet kernel and identifies the most informative elements of these matrices that deserve computation. To demonstrate the effectiveness of our approach, we assess the computational complexity of this distilled channel matrix in terms of the number of elementary operations required. Numerical results indicate that our technique markedly decreases receiver complexity by up to three orders of magnitude without compromising sensing performance.

This work proposes a maximum likelihood-based parameter estimation framework for a multistatic millimeter wave integrated sensing and communication system using energy-efficient hybrid digital-analog arrays. Due to the typically large arrays used in the higher frequency bands to mitigate isotropic path loss, such arrays may operate in the near-field (NF) regime. To address this, we propose a two-step estimation process. Initially, we consider far-field (FF) propagation assumptions, followed by refined estimation based on NF assumptions, enhancing accuracy when the target is within the NF of the arrays. In particular, when operating in the NF of the transmitter (Tx), we select beamfocusing array weights designed to achieve constant gain over an extended spatial region. Subsequently, we re-estimate target parameters at the receivers (Rxs). The effectiveness of the proposed framework is evaluated over various scenarios through numerical simulations. In particular, the impact of customdesigned flat-gain beamfocusing codewords in improving both communication and sensing performance when the target is in the NF of the Tx is demonstrated. Additionally, the benefit of considering a correct NF channel model when the target is located near an Rx is shown.

This work proposes a maximum likelihood (ML)- based parameter estimation framework for a millimeter wave (mmWave) integrated sensing and communication (ISAC) system in a multistatic configuration using energy-efficient hybrid digital-analog (HDA) arrays. Due to the typically large arrays deployed in the higher frequency bands to mitigate isotropic path loss, such arrays may operate in the near-field (NF) regime. The proposed parameter estimation in this work consists of a two-stage estimation process, where the first stage is based on far-field (FF) assumptions, and is used to obtain a first estimate of the target parameters. In cases where the target is determined to be in the NF of the arrays, a second estimation based on NF assumptions is carried out to obtain more accurate estimates. In particular, when operating in the near-filed of the transmitter (Tx), we select beamfocusing array weights designed to achieve a constant gain over an extended spatial region and re-estimate the target parameters at the receivers (Rxs). We evaluate the effectiveness of the proposed framework in numerous scenarios through numerical simulations and demonstrate the impact of the custom-designed flat-gain beamfocusing codewords in increasing the communication performance of the system.

In this paper, we consider a joint sensing and communication (JSC) network in which multiple base stations (BSs) cooperate through a fusion center (FC) to detect and track the objects present in a supervised area. Every BS acts as a monostatic sensor capable of scanning the environment and sensing the targets while simultaneously communicating with user equipments (UEs). In particular, each BS generates range-angle maps, which are shared with a FC for data fusion and tracking via particle filter (PF) and multi-hypothesis tracker (MHT) algorithms. The performance of the proposed solutions is evaluated by varying the fraction of power and time devoted to sensing to manage the network overhead and offer a sensing/communication trade-off. Numerical results show that the proposed algorithms can successfully track multiple targets with different sizes and behavior in a vehicular scenario, ensuring, e.g., a root mean squared error (RMSE) of the estimated position of a pedestrian less than 50 cm when considering three BSs.

In this work, we consider a multistatic joint sensing and communication network composed of a transmitter and multiple receivers capable of estimating the position of a target in the monitored area. In particular, the system consists of a transmitter equipped with multiple antennas (typically a base station) and several receivers that can act as sensors with a single antenna. The transmitter adopts multiple beams to accommodate the communication towards the user equipment and the sensing functionality, with the ability to split the power between the two beams to adjust the sensing/communication trade-off. Processing of target echoes by the sensors produces a bistatic distance estimate or soft maps, which are combined by the fusion center. We then propose two sensor data fusion strategies, least square and soft maps fusion; the former has a negligible impact on the network overhead, while the latter always provides better performance in terms of root mean squared error of target position estimation when considering a small fraction of power devoted to sensing. Finally, we highlight the benefits of cooperation offered by the multistatic configuration, compared to the bistatic one, in terms of power saving for sensing.

This work proposes a low-complexity estimation approach for an orthogonal time frequency space (OTFS)-based integrated sensing and communication (ISAC) system. In particular, we first define four low-dimensional matrices used to compute the channel matrix through simple algebraic manipulations. Secondly, we establish an analytical criterion, independent of system parameters, to identify the most informative elements within these derived matrices, leveraging the properties of the Dirichlet kernel. This allows the distilling of such matrices, keeping only those entries that are essential for detection, resulting in an efficient, low-complexity implementation of the sensing receiver. Numerical results, which refer to a vehicular scenario, demonstrate that the proposed approximation technique effectively preserves the sensing performance, evaluated in terms of root mean square error (RMSE) of the range and velocity estimation, while concurrently reducing the computational effort enormously.

This work investigates a multibeam system for joint sensing and communication (JSC) based on multiple-input multiple-output (MIMO) 5G new radio (NR) waveforms. In particular, we consider a base station (BS) acting as a monostatic sensor that estimates the range, speed, and direction of arrival (DoA) of multiple targets via beam scanning using a fraction of the transmitted power. The target position is then obtained via range and DoA estimation. We derive the sensing performance in terms of probability of detection and root mean squared error (RMSE) of position and velocity estimation of a target under line-of-sight (LOS) conditions. Furthermore, we evaluate the system performance when multiple targets are present, using the optimal sub-pattern assignment (OSPA) metric. Finally, we provide an in-depth investigation of the dominant factors that affect performance, including the fraction of power reserved for sensing.