sensor fusion

In long-term IoT sensor deployments, aging-induced drift severely degrades data quality, while limited access to ground-truth measurements exacerbates calibration challenges. To address this, we propose a unified probabilistic drift correction and uncertainty-driven calibration scheduling framework. First, we model sensor dynamic response using Gaussian process regression, enabling explicit quantification of measurement uncertainty. Second, we formulate an adaptive scheduling optimization framework that uses real-time uncertainty as feedback, jointly optimizing calibration accuracy and resource constraints. Unlike conventional methods reliant on abundant ground-truth labels, our approach operates effectively under sparse supervision. Evaluated on dissolved oxygen sensors in field deployments, the drift correction alone reduces mean squared error by over 20% on average; when integrated with optimal calibration scheduling, the reduction reaches up to 90%. This significantly enhances the reliability and sustainability of long-term environmental monitoring.

This work addresses the challenge scientists face in efficiently transforming raw sensor data streams into actionable insights across edge-cloud infrastructures, hindered by the need for cross-domain expertise to manage heterogeneous systems and emerging platforms such as DPUs, which impedes rapid prototyping. To overcome this barrier, the authors propose a novel paradigm that integrates pattern-based workflow engineering with AI-assisted development. Implemented on the FABRIC testbed using the Pegasus workflow system and exemplified by the Orcasound hydrophone workflow, this approach enables swift construction of applications for air quality, seismic, and soil moisture monitoring. The framework supports modular extensibility and edge deployment, substantially lowering the barrier for non-expert users to iteratively develop distributed applications. Empirical validation across multiple use cases demonstrates its effectiveness in enhancing development efficiency, accelerating prototyping cycles, and accumulating practical deployment experience.

Large language models (LLMs) exhibit limited capability in jointly optimizing multiple parameters for composite tasks in code-based sensor data processing (e.g., IMU, ECG, audio), hindering their deployment in embedded perception systems. Method: We introduce SensorBench—the first dedicated benchmark for evaluating LLMs on sensor data processing—built upon diverse real-world sensing datasets and comprising two task categories: code generation and code interpretation. It systematically assesses LLM performance under four prompting paradigms: zero-shot, chain-of-thought, self-consistency, and our novel self-verification prompting. Contribution/Results: Experiments reveal robust LLM performance on simple tasks but significant degradation on compositionally complex, multi-parameter optimization tasks—underperforming domain experts. Self-verification prompting achieves state-of-the-art results on 48% of tasks. The benchmark, along with reproducible evaluation protocols and practical prompting guidelines tailored for sensor development, establishes a foundational methodology for integrating LLMs into embedded sensing applications.

This paper addresses the inherent trade-off between computation and communication latency in distributed real-time state estimation. Method: We formulate the first rigorous optimization framework jointly modeling computation latency, communication latency, and estimation performance; theoretically prove that transmitting raw sensor data is generally suboptimal in heterogeneous networks; and propose a joint convex optimization algorithm for sensor subset selection and adaptive linear preprocessing—explicitly respecting per-node computational constraints and network heterogeneity. Contributions/Results: Leveraging Kalman filtering theory and heuristic subset search, we validate the approach on multivariate discrete-time systems. Experiments demonstrate that our method significantly reduces estimation error compared to full-sensor transmission, and that judicious local preprocessing substantially improves overall estimation accuracy.

This work addresses the complexity of developing edge-to-cloud sensor applications, which typically requires cross-domain collaboration and hinders efficient transformation of raw data into actionable insights. To streamline this process, the authors propose an intent-driven, AI-assisted rapid development methodology that integrates reusable workflow patterns with intelligent configuration, enabling seamless edge adaptation and deployment without code rewriting. Built upon the Pegasus workflow system and deployed on the FABRIC testbed, the approach supports heterogeneous edge resources such as BlueField-3 DPUs and Raspberry Pi devices. Users can construct multi-stage sensing applications within 1–1.5 days, and the framework’s robustness and portability have been validated through real-world deployments in air quality, seismic activity, and soil moisture monitoring scenarios.

This study addresses the challenge of rapid change-point detection in high-dimensional multisensor systems under structural constraints and limited sensing resources. By integrating sparse modeling, heterogeneous data fusion, and a resource-adaptive sequential sampling strategy, the work extends classical change-point detection theory to large-scale, resource-constrained sensing scenarios and incorporates machine learning to handle cases with unknown system models. The proposed approach unifies sparse signal processing, multi-stream statistical decision-making, and resource-constrained optimization to enable simultaneous detection of multiple change points. This framework significantly enhances both applicability and scalability in high-dimensional, heterogeneous, and resource-limited environments while maintaining high detection efficiency.

This work addresses the challenges of classifying CBRNE threats using heterogeneous sensors, which suffer from limited sensing capabilities, significant reliability disparities, abundant indirect indicators, and high levels of clutter, compounded by the scarcity of high-quality labeled data. To overcome these issues, the authors propose an evidence hierarchy that integrates direct observations, indirect indicators, and contextual information. By incorporating open-source intelligence (OSINT) to enrich environmental context and leveraging domain knowledge to construct Bayesian priors, the framework enables robust fusion of multi-source heterogeneous sensor data for threat classification. Experimental results demonstrate that the proposed approach achieves 95% overall classification accuracy in simulated scenarios, significantly enhancing system robustness against clutter interference and prior mismatch.

This work addresses the challenges of high early-stage uncertainty in manufacturing monitoring system development—leading to redundant modeling and substantial training costs—and the limited transferability of filtering pipelines in cross-domain image segmentation tasks. To tackle these issues, the authors propose a problem-centric design paradigm that constructs an abstract system model to continuously accumulate and retrieve historical segmentation tasks along with their associated filtering pipelines, enabling solution reuse and incremental optimization. The approach integrates similarity-based problem retrieval, abstract modeling, pipeline reuse, and a retrieval-augmented evolutionary learning mechanism. Experimental results demonstrate that the method significantly reduces training costs and late-stage revision risks, provides the first systematic validation of filtering pipeline transferability across similar segmentation tasks, and achieves a favorable balance among complexity, technical requirements, and reliability under lightweight model constraints.