Co-located with IEEE FiCloud conference –  Istanbul (Turkey) 11-13 August 2025

With the emergence of wireless communications, geolocation technologies, and cloud computing, innovative applications are designed for the IoT and CPS, targeting intensive data computations and virtualization technologies. Due to their constraints in terms of energy, computational power, memory, high mobility, sporadic connectivity, and sometimes security constraints, some smart devices need to outsource data storage, application hosting, and computation to the Cloud. Hence, the IoT and CPS require efficient and adaptive energy management solutions that optimize energy consumption through energy-aware communications protocols, scheduling methods, self-organization mechanisms, offloading techniques, and security solutions. We encourage submissions around developing energy-aware methodologies and technologies that improve the collection, transmission, storage, and processing.

The potential effects of these technologies on the machine learning processes executed in the cloud using the data collected by IoT and CPS should be considered and analyzed. In the presence of energy constraints, the machine learning processes may be hindered, and their performances may deteriorate. Many factors can be at the origin of these performance degradations. Energy-aware strategies, e.g., intermittent transmissions, decrease in sensing precision, reduction of the sampling frequency, etc., may compromise data availability on time and increase uncertainty in the data. Furthermore, in the case of distributed machine learning, energy-aware strategies may lead to heterogeneity in terms of the updates that the cloud receives from the distributed local IoT and CPS. For example, standard aggregation-based machine learning algorithms, like federated learning, are highly impacted by the heterogeneity in the IoT and CPS transmission rates. All these aspects impact the machine learning processes performed in the cloud; however, how it manifests and could be taken into account is not yet understood. This workshop provides a forum to discuss these aspects and devise new directions for future research.

Alternative energy sources available in the IoT and CPS surrounding environments could be used to achieve perpetual functioning without replacing or recharging batteries as often through energy harvesting. Machine learning is key to developing such strategies, e.g., analyzing patterns of energy consumption to adapt energy harvesting strategies, etc. Furthermore, energy management and optimization are also concerns for cloud data centers, which need efficient power and performance management for computing and air conditioning and environmental impact. These factors vastly improve cost versus energy optimization, which can be achieved through various approaches like optimization, computational intelligence, and machine learning.

This 10th edition of the International Workshop on Energy Management for Sustainable Internet-of-Things and Cloud Computing (EMSICC) 2025 seeks to bring together researchers, academicians, and industry experts to address the challenges and opportunities in designing energy-efficient and sustainable solutions employing distributed computing and AIML techniques for Cyber-Physical Systems (CPS), Internet of Things (IoT), and Cloud Systems.

Prospective authors are invited to submit original, previously unpublished work reporting on novel and significant research contributions, ongoing research projects, experimental results, and recent developments related to, but not limited to, the following topics:

  • Energy-efficient wireless communication protocols for IoT
  • Energy-aware network design for smart environments
  • Energy-efficient real-time systems and architectures for IoT and CPS
  • Energy-efficient Intelligent Transportation Systems
  • Dynamic energy optimization for real-time systems and applications
  • Adaptive middleware for energy-efficient computing
  • Enhanced performance and QoS in energy-efficient systems
  • Sensing, monitoring, control, and management of energy systems
  • Modelling, control, and architectures for renewable energy generation resources
  • Privacy and security in energy-aware platforms
  • Metrics, benchmarks, interfaces, and tools to consider energy dimension
  • Energy-aware hardware platforms
  • Specifications and validation of energy-aware platforms
  • Energy efficiency and virtualization in Cloud of Thing
  • Impact of energy-aware strategies in clouds using machine learning
  • Machine learning approaches for energy management
  • Learning of patterns of energy deployment in smart applications.