DA, Computer Systems (VT 2025)

Introduction to Design for Creative and Immersive Technology — HT2025

 

 

Research Interests:

• Distributed Computing Continuum Systems
• Learning Techniques in IoT
• AI/ML for Computing Systems
• Congnition, Causality in Computing Systems
• Cyber-physical Continuum

Editorial Activities:

• IEEE Internet of Things Journal
• Computing, Springer
• Scientific Reports, Nature
• Transactions on Emerging Telecommunications Technologies (ETT), Wiley

Call for Papers:

• TinyML for Edge-Driven 6G Networks
• Contemporary Mathematics for Changing Computational Demands

PhD Students:

• Alfreds Lapkovskis (Supervisor)
• Shubham Vaishnav (Co-Supervisor)

 

A selection from Stockholm University publication database

• Human-based Distributed Intelligence in Computing Continuum Systems

  2025. Praveen Kumar Donta (et al.). IEEE Internet Computing 29 (2), 61-68

  Article

  Distributed Computing Continuum Systems (DCCS) are integrated systems that combine cloud, edge, and IoT devices to deliver scalable and low-latency computing resources across diverse applications and environments. Composed of a heterogeneous mix of computational units, storage systems, and communication networks, DCCS facilitates real-time data processing and analysis by distributing tasks dynamically based on resource availability and demand. The complex structure of DCCS reflects the intricate organization of the human body, where different systems work together to maintain overall functionality. This paper draws parallels between the human body’s intelligence mechanisms and the operational strategies needed for DCCS. Especially exploring several human body analogies or principles that can be incorporated into DCCS to mitigate interpretable and non-interpretable challenges while enhancing overall performance.

  Read more about Human-based Distributed Intelligence in Computing Continuum Systems

• KDN-FLB: Knowledge-defined Networking through Federated Learning and Blockchain

  2025. Ying Li (et al.). Computer 58 (5), 16-26

  Article

  In this article, we explore the opportunities and benefits of integrating federated learning (FL) and blockchain technologies to build an adaptable and secure Knowledge-Defined Networking (KDN) system. Our aim is to enhance network performance by ensuring self-learning, self-adapting, and self-adjustment capabilities in dynamic and decentralized network environments. The proposed conceptual architecture, KDN-FLB, also strategically addresses critical challenges in knowledge sharing and privacy preservation within network environments. We discuss the constituents, architecture, processes, and use cases of KDN-FLB in contemporary networking applications. Additionally, we analyze the benefits, challenges, and future prospects associated with KDN-FLB, making it more intelligent for large-scale, dynamic, and decentralized network environments.

  Read more about KDN-FLB

• SatCooper: Enhancing Cooperative Inference Analytics for Satellite Service via Multi-exit DNNs

  2025. Qiyang Zhang (et al.). IEEE Transactions on Mobile Computing

  Article

  As a key technology of intelligent satellite-enabled services in B5G or 6G networks, deploying Deep Neural Networks (DNN) models on satellites has been a notable trend, catering to the daily demand for extensive computing-intensive and latency-sensitive tasks. The computing resources are strategically deployed on satellites where sensor data is generated or collected, facilitating the fine-grained computational inference of DNN-based tasks. However, no prior study has comprehensively explored the crucial inference challenges -- e.g., the trade-off between the number of tasks completed and accuracy and partitioning models in multi-exit models -- in the resource-constrained space environment. Effective scheduling frameworks cater to various streams of inference tasks are scarce because inference performance may deviate from the ideal situation due to changes in task system status, such as task profiles and network state. To this end, we first formulate a gain-aware in-orbit computing inference problem to strike a proper trade-off between inference latency and the number of tasks completed by dynamically selecting optimal early exit points and model partitioning points. We propose an offline dynamic programming-based algorithm that provides an effective solution when comprehensive system details are to be predicted. We have developed an online learning-based method to schedule inference tasks with uncertain and dynamic system statuses in real-world situations. Our evaluation shows that, compared to baseline methods, the online learning-based algorithm can improve task gain by an average of 87.3% across various tasks.

  Read more about SatCooper

• Follow-Me AI: Energy-Efficient User Interaction With Smart Environments

  2025. Alaa Saleh (et al.). IEEE pervasive computing

  Article

  This article introduces Follow-Me AI, a concept designed to enhance user interactions with smart environments, optimize energy use, and provide better control over data captured by these environments. Through AI agents that accompany users, Follow-Me AI negotiates data management based on user consent, aligns environmental controls as well as user communication and computes resources available in the environment with user preferences, and predicts user behavior to proactively adjust the smart environment. This article illustrates this concept with a detailed example of Follow-Me AI in a smart campus setting, detailing the interactions with the building's management system for optimal comfort and efficiency. Finally, this article looks into the challenges and opportunities related to Follow-Me AI.

  Read more about Follow-Me AI

• Communication-Efficient Federated Learning for Heterogeneous Clients

  2025. Ying Li (et al.). ACM Transactions on Internet Technology 25 (2)

  Article

  Federated learning stands out as a promising approach within the domain of edge computing, providing a framework for collaborative training on distributed datasets without necessitating data sharing. However, federated learning involves the frequent transmission of machine learning model updates between the server and clients, resulting in high communication costs. Additionally, heterogeneous clients can further complicate the Federated Learning process and deteriorate performance. To address these challenges, we propose adaptive self-knowledge distillation-based quality- and reputation-aware cross-device federated learning (ASDQR) - an efficient communication and inference framework designed for heterogeneous clients. ASDQR initiates the process by selecting high-reputation and high-quality clients to be involved in federated learning, significantly impacting communication efficiency and inference effectiveness. ASDQR also introduces a model of adaptive local self-knowledge distillation that incorporates multiple local personalized historical knowledge for more accurate inference, allowing the historical level to be dynamically adjusted across time. Finally, we present an inference-effective aggregation scheme that assigns higher weights to important and reliable local model updates based on clients’ contribution degrees when performing global model aggregation. ASDQR consistently outperforms baseline methods across all datasets and communication rounds, achieving 9.0% higher accuracy than FedAvg, 6.59% higher than MOON, 0.29% higher than FedProx, 0.2% higher than PFedSD, and 0.08% higher than FedMD on the MNIST dataset at 100 communication rounds. Similar improvements are observed on CIFAR, HAR, and WISDM datasets, demonstrating the robustness and efficiency of ASDQR in federated learning with non-IID data.

  Read more about Communication-Efficient Federated Learning for Heterogeneous Clients

• Spiking Neural Networks in Intelligent Edge Computing

  2024. Guanlei Zhang (et al.). IEEE Consumer Electronics Magazine, 1-9

  Article

  Deep neural networks (DNNs) have witnessed rapid advancements and remarkable success in recent years, leading to their increasingly widespread implementation on edge devices. However, the deployment, execution, and life cycle management of traditional artificial neural networks (ANNs) on resource-constrained edge devices present significant challenges. Spiking neural networks (SNNs) are a class of neuroscience-inspired neural networks that emulate the low-power operational mode of biological neurons. SNNs possess advantages such as low power consumption, low latency, event-driven processing, and reduced communication overhead, making them particularly well-suited for edge devices and intelligent edge computing. As a result, they have garnered significant attention in both research and practical applications. In this paper, we present a comprehensive survey of the fundamentals of SNNs and the advancements in SNN research for edge computing, exploring potential applications and future directions in this emerging field. We also present a case study highlighting that SNNs outperform ANNs in distributed learning, achieving a 6% improvement in accuracy and an 80% reduction in data transmission.

  Read more about Spiking Neural Networks in Intelligent Edge Computing

• On Distributed Computing Continuum Systems

  2023. Schahram Dustdar, Victor Casamayor Pujol, Praveen Kumar Donta. IEEE Transactions on Knowledge and Data Engineering 35 (4), 4092-4105

  Article

  This article presents our vision on the need of developing new managing technologies to harness distributed “computing continuum” systems. These systems are concurrently executed in multiple computing tiers: Cloud, Fog, Edge and IoT. This simple idea develops manifold challenges due to the inherent complexity inherited from the underlying infrastructures of these systems. This makes inappropriate the use of current methodologies for managing Internet distributed systems, which are based on the early systems that were based on client/server architectures and were completely specified by the application software. We present a new methodology to manage distributed “computing continuum” systems. This is based on a mathematical artifact called Markov Blanket, which sets these systems in a Markovian space, more suitable to cope with their complex characteristics. Furthermore, we develop the concept of equilibrium for these systems, providing a more flexible management framework compared with the one based on thresholds, currently in use for Internet-based distributed systems. Finally, we also link the equilibrium with the development of adaptive mechanisms. However, we are aware that developing the entire methodology requires a big effort and the use of learning techniques, therefore, we finish this article with an overview of the techniques required to develop this methodology.

  Read more about On Distributed Computing Continuum Systems

• Edge Intelligence—Research Opportunities for Distributed Computing Continuum Systems

  2023. Victor Casamayor Pujol (et al.). IEEE Internet Computing 27 (4), 53-74

  Article

  Edge intelligence and, by extension, any distributed computing continuum system will bring to our future society a plethora of new and useful applications, which will certainly revolutionize our way of living. Nevertheless, managing these systems challenges all previously developed technologies for Internet-distributed systems. In this regard, this article presents a set of techniques and concepts that can help manage these systems; these are framed in the main paradigm for autonomic computing, the well-known monitor–analyze–plan–execute over shared knowledge, or MAPE-K. All in all, this article aims at unveiling research opportunities for these new systems, encouraging the community to work together toward new technologies to make edge intelligence a reality.

  Read more about Edge Intelligence—Research Opportunities for Distributed Computing Continuum Systems

• A Graph-based Approach to Human Activity Recognition

  2024. Thomas Peroutka (et al.). 2024 IEEE 6th International Conference on Cognitive Machine Intelligence (CogMI), 117-126

  Conference

  Advanced wearable sensor devices have enabled the recording of vast amounts of movement data from individuals regarding their physical activities. This data offers valuable insights that enhance our understanding of how physical activities contribute to improved physical health and overall quality of life. Consequently, there is a growing need for efficient methods to extract significant insights from these rapidly expanding real-time datasets. This paper presents a methodology to efficiently extract substantial insights from these expanding datasets, focusing on professional sports but applicable to various human activities. By utilizing data from Inertial Measurement Units (IMU) and Global Navigation Satellite Systems (GNSS) receivers, athletic performance can be analyzed using directed graphs to encode knowledge of complex movements. Our approach is demonstrated on biathlon data and detects specific points of interest and complex movement sequences, facilitating the comparison and analysis of human physical performance.

  Read more about A Graph-based Approach to Human Activity Recognition

• A Privacy Enforcing Framework for Data Streams on the Edge

  2024. Boris Sedlak (et al.). IEEE Transactions on Emerging Topics in Computing, 1-12

  Article

  Recent developments in machine learning (ML) allow for efficient data stream processing and also help in meeting various privacy requirements. Traditionally, predefined privacy policies are enforced in resource-rich and homogeneous environments such as in the cloud to protect sensitive information from being exposed. However, large amounts of data streams generated from heterogeneous IoT devices often result in high computational costs, cause network latency, and increase the chance of data interruption as data travels away from the source. Therefore, this article proposes a novel privacy-enforcing framework for transforming data streams by executing various privacy policies close to the data source. To achieve our proposed framework, we enable domain experts to specify high-level privacy policies in a human-readable form. Then, the edge-based runtime system analyzes data streams (i.e., generated from nearby IoT devices), interprets privacy policies (i.e., deployed on edge devices), and transforms data streams if privacy violations occur. Our proposed runtime mechanism uses a Deep Neural Networks (DNN) technique to detect privacy violations within the streamed data. Furthermore, we discuss the framework, processes of the approach, and the experiments carried out on a real-world testbed to validate its feasibility and applicability.

  Read more about A Privacy Enforcing Framework for Data Streams on the Edge

• Equilibrium in the Computing Continuum through Active Inference

  2024. Boris Sedlak (et al.). Future Generation Computer Systems 160, 92-108

  Article

  Computing Continuum (CC) systems are challenged to ensure the intricate requirements of each computational tier. Given the system’s scale, the Service Level Objectives (SLOs), which are expressed as these requirements, must be disaggregated into smaller parts that can be decentralized. We present our framework for collaborative edge intelligence, enabling individual edge devices to (1) develop a causal understanding of how to enforce their SLOs and (2) transfer knowledge to speed up the onboarding of heterogeneous devices. Through collaboration, they (3) increase the scope of SLO fulfillment. We implemented the framework and evaluated a use case in which a CC system is responsible for ensuring Quality of Service (QoS) and Quality of Experience (QoE) during video streaming. Our results showed that edge devices required only ten training rounds to ensure four SLOs; furthermore, the underlying causal structures were also rationally explainable. The addition of new types of devices can be done a posteriori; the framework allowed them to reuse existing models, even though the device type had been unknown. Finally, rebalancing the load within a device cluster allowed individual edge devices to recover their SLO compliance after a network failure from 22% to 89%.

  Read more about Equilibrium in the Computing Continuum through Active Inference

• Markov Blanket Composition of SLOs

  2024. Boris Sedlak (et al.). 2024 IEEE International Conference on Edge Computing and Communications (EDGE), 128-138

  Conference

  Smart environments use composable microservices pipelines to process Internet of Things (IoT) data, where each service is dependent on the outcome of its predecessor. To ensure Quality of Service (QoS), individual services must fulfill Service Level Objectives (SLOs); however, SLO fulfillment is dependent on resources (e.g., processing or storage), which are scarcely available within the Edge. Hence, when distributing services over heterogeneous devices, this raises the question of where to deploy each service to best fulfill both its own SLOs as well as those imposed by dependent services. In this paper, we maximize SLO fulfillment of a pipeline-based application by analyzing these dependencies. To achieve this, services and hosting devices alike are extended with a Markov blanket (MB) - a probabilistic view into their internal processes - which are composed into one overarching model. Given a mutable set of services, hosts, and SLOs, the composed MB allows inferring the optimal assignment between services and edge devices. We evaluated our method for a smart city scenario, which assigned pipelined services (e.g., video processing) under constraints from subsequent services (e.g., consumer latency). The results showed how our method can support infrastructure providers by optimizing SLO fulfillment for arbitrary devices currently available.

  Read more about Markov Blanket Composition of SLOs

• Advanced Techniques for Anomaly Detection: Beyond the Basics

  2025. .

  Book (ed)

  This book is a comprehensive guide that explores the latest developments in anomaly detection techniques across a range of fields, including cybersecurity, finance, image processing, sensor networks, social network analysis, health systems, and IoT systems. With 6 chapters covering various topics such as deep learning-based anomaly detection, feature selection and extraction techniques, ensemble methods, and evaluation metrics, this book offers a comprehensive understanding of advanced anomaly detection techniques and their applications in different fields. This book will be an excellent resource for researchers, practitioners, and students interested in anomaly detection and its applications in various domains.

  Read more about Advanced Techniques for Anomaly Detection

Show all publications by Praveen Kumar Donta at Stockholm University