Bruno Lima, Rui Pinto
The widespread adoption of Internet of Things (IoT) systems in domains like healthcare, manufacturing, and smart cities has brought new challenges for software and systems testing. This survey systematically reviews studies published from 2012, identifying key testing types, methods, and tools across the IoT stack. Our analysis highlights sensor-specific challenges—including timing jitter, calibration drift, and environmental coupling—that remain underexplored in prior surveys. We propose a multidimensional conceptual framework that cross-correlates architectural layers, testing objectives, and technology targets, enabling structured comparison of testing strategies. The framework is empirically validated through an industrial case study, which reveals critical gaps such as a lack of cross-layer coverage, limited tool support for real-time constraints, and inadequate mechanisms for sensor fault resilience. We conclude by outlining future research directions, including digital twin-based validation, unified IoT testing pipelines, and lightweight security testing for constrained devices. This work provides researchers and practitioners with a sensor-aware, empirically grounded roadmap for advancing IoT testing.