Application of Deep Learning to Internet of Things Systems

Dear Colleagues,

In recent years, the evolution of Internet of Things (IoT) technologies has catalyzed an explosion of data from myriad IoT sensors, now installed globally, propelling the advancement of sophisticated services and applications across diverse sectors. Deep learning methodologies have emerged as key tools, elevating IoT functionalities from conventional desktop and mobile applications to resource-constrained IoT ecosystems. These strides have yielded measurable outcomes in domains ranging from image recognition and medical data analysis to language processing and smart forest, smart agriculture, smart city solutions.

This Special Issue delves deep into the continuing exploration and implementation of IoT technologies, with a keen focus on multimodal signal processing, sensor data extraction, data visualization, and allied subjects. Central objectives encompass probing the efficacy of deep neural network architectures in processing and amalgamating multimodal sensor inputs for diverse IoT applications, refining existing designs, and pioneering novel approaches to curtail resource consumption in deploying deep learning models on IoT devices. Additionally, it seeks to accurately compute reliability metrics for deep learning predictions within constrained computational frameworks and streamline the utilization of labeled IoT data for signal learning amid operational constraints.

Special topics of interest we invite to contribute, but are not confined to, the following:

• Edge computing and its pivotal role in optimizing deep learning for IoT systems;
• Federated learning approaches for collaborative model training in distributed IoT environments;
• Explainable AI techniques aimed at augmenting the transparency and interpretability of deep learning models in IoT applications;
• Security and privacy considerations inherent in deploying deep learning models on IoT devices;
• Integration of reinforcement learning algorithms for fostering autonomous decision-making in IoT systems;
• Energy-efficient deep learning implementations tailored for sustainable IoT applications;
• Meta-learning strategies for enhancing adaptability and generalization of deep learning models in dynamic IoT environments;
• Hybrid AI architectures combining symbolic reasoning with deep learning for robust IoT decision-making;
• Bio-inspired computing paradigms for optimizing resource utilization and fault tolerance in IoT networks;
• Human-in-the-loop approaches for leveraging human intelligence to improve IoT data annotation and model training;
• Ethical considerations and responsible AI practices in the development and deployment of deep learning-based IoT solutions;
• Innovative IoT applications in forestry management and agriculture, including precision farming, crop monitoring, and environmental conservation;
• Real-time asset tracking and management systems for logistics and supply chain optimization;
• Smart building solutions for energy management, occupancy monitoring, and predictive maintenance
• Healthcare IoT applications for remote patient monitoring, personalized medicine, and medical device integration;
• IoT-enabled transportation systems for traffic management, vehicle tracking, and autonomous vehicle control;
• Retail IoT applications for inventory management, customer analytics, and personalized shopping experiences;
• IoT-enabled utility infrastructure for smart grid management, water quality monitoring, and waste management optimization.

We invite researchers and practitioners to contribute their insights and research findings to this Special Issue, shedding light on critical aspects of deep learning applications for IoT systems. Submissions addressing sensor data integration, model efficiency, and reliability assessment in IoT contexts are highly encouraged. Join us in advancing the frontiers of deep learning in IoT ecosystems and shaping the future of intelligent and interconnected technologies.

Dr. Rytis Maskeliunas
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Computers is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• Internet of Things
• deep learning
• data fusion
• multimodal signal processing
• data processing and visualization