Publications
-
Yousefi, Neda; Zamanifar, Azadeh; Farhadi, Amirfarhad & Taherkordi, Amirhosein
(2026).
Data Privacy and Security in AI-Enabled Healthcare,
AI Safety in Healthcare Systems Resilience, Transparency, and Accountability.
CRC Press.
ISSN 9781041249610.
p. 45–82.
doi:
10.1201/9781003743422-3.
Full text in Research Archive
-
Le, Thanh; La, Hoang Loc; Taherkordi, Amirhosein; Eliassen, Frank; Ha, Hoai Phuong & Guan, Peiyuan
(2026).
PM2Lat: Highly Accurate and Generalized Prediction of DNN Execution Latency on GPUs.
IEEE/ACM International Symposium on Cluster, Cloud, and Grid Computing.
ISSN 2376-4414.
p. 604–614.
doi:
10.1109/ccgrid68966.2026.00070.
Full text in Research Archive
Show summary
We present PM2Lat, a fast and generalized framework for accurately predicting the latency of deep neural network models on GPUs, with special focus on NVIDIA. Unlike prior methods that rely on deep learning models or handcrafted heuristics, PM2Lat leverages the Single-Instruction-Multiple-Thread architecture of GPUs to model execution time of DNN models. First, we conduct fine-grained GPU operation modeling by analyzing computational behavior and memory access patterns. Through this analysis, we observe that different GPU kernels can exhibit significant performance disparities, even when serving the same purpose. Hence, the core idea of PM2Lat is to differentiate kernels based on their configurations and analyze them accordingly. This kernel-aware modeling enables PM2Lat to achieve consistently low prediction error across diverse data types and hardware platforms. In addition, PM2Lat generalizes beyond standard matrix multiplication to support complex GPU kernels such as Triton, Flash Attention, and Cutlass Attention. Experimental results show that PM2Lat consistently achieves error rates below 10% across different data types and hardware platforms on Transformer models, outperforming the state-ofthe-art NeuSight by 10-20% for FP32 and by at least 50% for BF16. When applying to diverse kernels, the error rate is maintained at 3-8%.
-
-
Rafiee, Milad; Ocampo, Andres F.; Taherkordi, Amirhosein & Alay, Ozgu
(2025).
A Proactive Performance Prediction Framework for Virtual Network Functions in 5G Networks.
In Cerroni, Walter; Tortonesi, Mauro; Borsatti, Davide; Schaeffer-Filho, Alberto Egon; Tuncer, Daphné; François, Jérôme & Husák, Martin (Ed.),
2025 21st International Conference on Network and Service Management (CNSM).
IEEE (Institute of Electrical and Electronics Engineers).
ISSN 9783903176751.
doi:
10.23919/cnsm67658.2025.11297473.
Full text in Research Archive
-
Sharma, Jivitesh; Vallejo, Islen; Ødegård, Rune Åvar; Le, Truong Thanh; Taherkordi, Amirhosein & Eliassen, Frank
(2025).
Physics-Informed Deep Learning for Wind Downscaling over Oslo.
Proceedings - International Conference on Tools with Artificial Intelligence (ICTAI).
ISSN 1082-3409.
37,
p. 270–276.
doi:
10.1109/ICTAI66417.2025.00042.
Full text in Research Archive
Show summary
Running a numerical weather model such as WRF at kilometre or sub-kilometre grid spacing over a regional domain is computationally expensive. We present physics-informed deeplearning models that ingest a single 9km WRF wind field and simultaneously predict two finer-scale wind fields at 3 km and 1 km resolution via dual decoder heads. Four representative architectures are benchmarked-Deep Residual U-Net (DeepRU), DEVINE, a bespoke 3-D Transformer, and a Fourier Neural Operator (FNO)-each trained with divergence-free, vorticity, and Navier-Stokes residual constraints plus Charbonnier and gradient perceptual losses. We train and validate our models on the city of Oslo for the year 2018. DeepRU achieves R2=0.94 (RMSE =0.050) at 3km and R2=0.89(RMSE=0.065) at 1 km. DEVINE, Transformer 3-D, and FNO yield 3 km scores of 0.91−0.93, with 1km scores lower by 0.02−0.08, illustrating the increased difficulty of finer-scale reconstruction. Physicsinformed losses improve all models compared to MSE-only baselines, and the residual architecture (DeepRU) remains most effective for this dual-scale task.
-
Jamshidi, Omid; Abbasi, Mahdi; Ramazani, Abbas; Shahraki, Atefeh Salimi & Taherkordi, Amirhosein
(2025).
Modifier guided resilient CNN inference enables fault-tolerant edge collaboration for IoT.
Scientific Reports.
15(1).
doi:
10.1038/s41598-025-28454-z.
Full text in Research Archive
Show summary
In resource-constrained Internet of Things (IoT) scenarios, implementing robust and accurate deep learning inference is problematic due to device failures, limited computing power, and privacy concerns. We present a resilient, completely edge-based distributed convolutional neural network (CNN) architecture that eliminates cloud dependencies while enabling accurate and fault-tolerant inference. At its core is a lightweight Modifier Module deployed at the edge, which synthesizes predictions for failing devices by pooling peer CNN outputs and weights. This dynamic mechanism is trained via a novel fail-simulation technique, allowing it to mimic missing outputs in real-time without model duplication or cloud fallback. We assess our methodology using MNIST and CIFAR-10 datasets under both homogeneous and heterogeneous data partitions, with up to five simultaneous device failures. The system displays up to 1.5% absolute accuracy improvement, 30% error rate reduction, and stable operation even with over 80% device dropout, exceeding ensemble, dropout, and federated baselines. Our strategy combines significant statistical significance, low resource utilization (~ 15 KB per model), and real-time responsiveness, making it well-suited for safety-critical IoT installations where cloud access is infeasible.
-
Chen, Chen; Guan, Peiyuan; Chen, Ziru; Taherkordi, Amirhosein; Hou, Fen & Cai, Lin X.
(2025).
Cost Optimization for Serverless Edge Computing with Budget Constraints Using Deep Reinforcement Learning.
IEEE International Conference on Communications.
ISSN 1550-3607.
p. 5718–5723.
doi:
10.1109/ICC52391.2025.11162063.
Full text in Research Archive
-
Tavassolian, Fazeleh; Abbasi, Mahdi; Ramezani, Abbas; Salimi, Atefeh; Taherkordi, Amirhosein & Khosravi, M. Reza
(2025).
AdaptFedDS: adaptive federated learning edge devices selection by using a light-dual reinforcement learning approach.
Cluster Computing.
ISSN 1386-7857.
28(16).
doi:
10.1007/s10586-025-05765-4.
Full text in Research Archive
-
Xiong, Meng; Tian, Shengwei; Guan, Peiyuan; Pei, Xinjun; Li, Yushuai & Taherkordi, Amirhosein
(2025).
Client-Level Fault-Tolerant Federated Semi-Supervised Learning for Unlabeled Clients in Internet of Vehicles.
IEEE Transactions on Network Science and Engineering (IEEE TNSE).
12(6),
p. 4657–4670.
doi:
10.1109/TNSE.2025.3574118.
Full text in Research Archive
-
Zhang, Min; Eliassen, Frank; Taherkordi, Amirhosein; Jacobsen, Hans-Arno; Li, Yushuai & Zhang, Yan
(2025).
Self-Determination Theory and Deep Reinforcement Learning for Personalized Energy Trading in Smart Grid.
IEEE Transactions on Systems, Man & Cybernetics. Systems.
ISSN 2168-2216.
55(6),
p. 4216–4229.
doi:
10.1109/TSMC.2025.3551667.
Full text in Research Archive
-
Puka, Ergys; Herrmann, Peter & Taherkordi, Amirhosein
(2025).
An Efficient and Robust Protocol to Accelerate Message Delivery in Cellular Dead Spots.
IEEE Transactions on Vehicular Technology.
ISSN 0018-9545.
p. 1–16.
doi:
10.1109/TVT.2025.3631398.
Full text in Research Archive
-
-
Feng, Wendi; Dou, Xintan; Taherkordi, Amirhosein; Cheng, Bo & Zhang, Wei
(2025).
Compresso: Latency-Aware Transmission of Compressed IoT Measurement Data Over SDN.
IEEE Internet of Things Journal.
ISSN 2327-4662.
12(12),
p. 20462–20472.
doi:
10.1109/JIOT.2025.3543479.
Full text in Research Archive
-
Wu, Jia; Dai, Tingyi; Guan, Peiyuan; Chen, Ziru; Gou, Fangfang & Taherkordi, Amirhosein
(2025).
Opportunistic routing for mobile edge computing: A community detected and task priority aware approach.
Computer Networks.
ISSN 1389-1286.
258.
doi:
10.1016/j.comnet.2024.111000.
Full text in Research Archive
-
Zhang, Qixia; Taherkordi, Amirhosein & Ha, Hoai Phuong
(2025).
eU2U: Energy-efficient Wireless Charging and Trajectory Design for IoT Data Collection.
Consumer Communications and Networking Conference.
ISSN 2331-9860.
doi:
10.1109/CCNC54725.2025.10976070.
Full text in Research Archive
Show summary
Thanks to their high maneuverability, high flexibility, and low cost, unmanned aerial vehicles (UAVs) have been widely used for data collection in the Internet of Things (IoT). To deal with UAV's onboard battery limitation, UAV-to-UAV (U2U) wireless charging mechanism emerges as a promising solution for extending flight distance and reducing mission completion time. However, U2U charging mechanisms encounter key challenges of limited wireless charging distance and energy loss. In this paper, we propose eU2U, a novel energy-efficient wireless charging and trajectory design approach for IoT data collection. We develop eU2U based on the distributed laser charging (DLC) system for its compact size and meter-level wireless power transmission. To minimize the total energy consumption, we build holistic power and energy consumption models for U2U-enabled data collection. With joint considerations on the wireless charging energy loss and delay constraint, we propose a heuristic algorithm to derive the most energy-efficient locations of the U2U charging points and the UAV trajectories, where the charging points are determined from the Fermat point and the battery drain point. Extensive evaluations show that eU2U can reduce the total energy consumption by 44.52% as compared to state-of-the-art schemes.
-
Abbasi, Mahmoud; Florez, SebastianLopez; Shahraki, Amin; Taherkordi, Amirhosein; Prieto, Javier & Corchado, Juan M.
(2025).
Class Imbalance in Network Traffic Classification: An Adaptive Weight Ensemble-of-Ensemble Learning Method.
IEEE Access.
13,
p. 26171–26192.
doi:
10.1109/ACCESS.2025.3538170.
Full text in Research Archive
Show summary
Network Traffic Classification (NTC) serves as a crucial element in network management, and the rapid progress in machine learning has inspired the utilization of learning methods to discern network traffic. The inherent characteristics of network traffics result in uneven class distributions when datasets are shaped, creating a phenomenon known as class imbalance. This phenomenon has garnered growing attention across various research fields. Despite encountering performance setbacks attributed to class imbalance, this challenge remains inadequately examined in the realm of network traffic classification. This paper introduces the Adaptive Weight Ensemble-of-Ensemble Learning (AWEE) method as an innovative solution to this challenge. The AWEE integrates multiple ensemble layers with a dynamic weight adjustment mechanism, showcasing the collaborative intelligence of diverse modeling strategies. Using a sliding window-based validation approach, the model enhances adaptability and robustness to address concept drift in dynamic data streams. Experimental studies on benchmark datasets demonstrate the superior performance of AWEE (achieving an outstanding accuracy rate of over 98%), highlighting its effectiveness in handling class imbalance challenges across diverse network traffic scenarios. AWEE, outperforms competitive methods, including algorithmic-level, cost-sensitive, and data-level techniques, showcasing its robustness and superior performance in addressing class imbalance challenges across a wide range of network traffic scenarios.
-
Le, Truong Thanh; Taherkordi, Amirhosein; Eliassen, Frank & Guan, Peiyuan
(2024).
Optimal Distribution of ML Models Over Edge for Applications with High Input Frequency.
IEEE International Conference on Cloud Computing Technology and Science (CloudCom).
ISSN 2330-2194.
p. 143–150.
doi:
10.1109/cloudcom62794.2024.00033.
Full text in Research Archive
Show summary
The rise of complex and sizeable Machine Learning (ML) models challenges traditional cloud computing models with respect to the high volume of incoming data which results in increased bandwidth usage and network congestion, as well as delays in inference. Such ML models are being rapidly developed thanks to advances in computing platforms and the real-time computing demands of ML-driven applications such as autonomous vehicles and video processing. To mitigate these challenges, ML model distribution and inference offloading to computing devices close to data sources have been explored, especially through partitioning the models across the IoT-Edge-Cloud continuum. Existing efforts in this area have not successfully mastered the fully automatic and efficient determination of optimal partition points. Additionally, they have not effectively integrated Early-Exit layers that allow for early termination of model inference at earlier stages when feasible. In this paper, we introduce a novel partitioning algorithm designed to distribute ML models across edge devices with the goal of reducing response time when facing high-rate input data streams. Our proposed approach leverages the principles of Dynamic Programming to determine optimal partition points and establish appropriate exit thresholds for Early-Exit layers. Our evaluation results reveal that, in the context of continuous, high-rate input data, our method consistently lowers the maximum round-trip time for processing inference requests compared to state-of-the-art methods such as NeuroSurgeon and Genetics.
-
Tavassolian, Fazeleh; Abbasi, Mahdi; Ramezani, Abbas; Taherkordi, Amirhosein & Khosravi, Mohammad R.
(2024).
ResFed: An Accurate and Light Federated Multi-Shot Pre-Trained Model on Edge Devices.
Tsinghua Science and Technology.
ISSN 1007-0214.
30(4),
p. 1539–1551.
doi:
10.26599/TST.2024.9010042.
Full text in Research Archive
Show summary
The rise of machine learning applications at the network edge demands real-time prediction with limited resources, requiring low computational complexity and accurate models. Also, these devices may have poor learning ability due to the lack of samples. Using federated learning enables devices to train machine learning models without sharing their private data. Meta-learning algorithms, particularly few-shot learning, are ideal for federated environments with highly personalized and decentralized training data due to their fast adaptation and good generalization to new tasks. Despite recent advances, the use of metric-based meta-learning methods remains ambiguous due to their simplicity, as well as their development to improve model learning ability and accuracy in a federated environment. This paper introduces ResFed, a federated meta-learning method specifically tailored for few-shot classification. The approach involves leveraging a pre-trained model and implementing data augmentation within the federated meta-learner, leading to favorable performance outcomes. The experimental results show that our approach, specifically designed for limited data scenarios in the federated environments, significantly improves convergence speed and accuracy. The values of accuracy obtained in CIFAR-100 and Omniglot datasets are 77.44% and 98.24%, respectively. Additionally, when compared to alternative methods, there is a notable reduction in resource costs ranging from 0.4 to 0.61 in diverse scenarios.
-
Motevallian, Mahsa; Alam, A M Esfar E; Taherkordi, Amirhosein & Abbasi, Golnoush
(2024).
Semantic Modeling of Waste Dataflow for Automating Circular Economy Systems,
2024 20th International Conference on Distributed Computing in Smart Systems and the Internet of Things (DCOSS-IoT 2024).
IEEE Computer Society Digital Library.
ISSN 9798350369441.
p. 677–684.
doi:
10.1109/DCOSS-IoT61029.2024.00105.
Full text in Research Archive
Show summary
Circular Economy (CE) is a model with a concrete action plan covering the whole life cycle of a product, from production and consumption to waste management (WM). Information technologies considerably contribute to the transition towards CE, e.g., waste tracking using Internet of Things (IoT). This will cause the businesses and organizations to confront a large diversity of data (i.e. waste amount, types, locations, etc.). The generated data is often stored and processed through manual or semi-manual methods by each business or organization. However, an automated method which can also interpret and integrate the diverse data in WM fields across different organizations is still in its infancy. Often, such data is not organized and falls short of reaching its full potential in facilitating coordinated management and enabling Circular Economy initiatives. In this paper, we aim to address this need through automated interpretation and integration of municipal waste data by applying semantic data modeling. Our approach proposes to capture the semantical description of entities in the WM process and their relations, which can appear between waste producers, authorities and consumers. Then, the obtained semantic model will facilitate and automate the required interpretation and integration of waste data, both for intra- and inter-organization scenarios. We realize intelligent semantic-based searching using natural language processing and large language models.
-
-
-
-
Rahmani, Maryam; Crumeyrolle, Suzanne; Allegri-Martiny, Nadége; Taherkordi, Amirhosein & Rouvoy, Romain
(2024).
PmForecast: leveraging temporal LSTM to deliver in situ air quality predictions.
Environmental Science and Pollution Research.
ISSN 0944-1344.
doi:
10.1007/s11356-024-34623-w.
Full text in Research Archive
-
Han, Song; Ma, Hui; Taherkordi, Amirhosein; Lan, Dapeng & Chen, Yange
(2024).
Privacy-preserving data integration scheme in industrial robot system based on fog computing and edge computing.
IET Communications.
ISSN 1751-8628.
18(7),
p. 461–476.
doi:
10.1049/cmu2.12749.
Full text in Research Archive
Show summary
To solve the security problems of the moving robot system in the fog network of the Industrial Internet of Things (IIoT), this paper presents a privacy-preserving data integration scheme in the moving robot system. First, a novel data collection enhancement algorithm is proposed to enhance the image effects, and a k-anonymous location and data privacy protection protocol based on Ad hoc network (Ad hoc-based KLDPP protocol) is designed in secure data collection phase to protect the privacy of location and network data. Second, the secure multiparty computation with verifiable key sharing is introduced to realize the valid computation against share cheating in the robot system. Third, the ciphertext classification method in a neural network is considered in the secure data storage process to realize the special application. Finally, experiments and simulations are conducted on the robot system of fog computing in the IIoT. The results demonstrate that the proposed scheme can improve the security and efficiency of the said robot system.
-
Kamalian, Mahdieh; Taherkordi, Amirhosein; Payberah, Amir H. & Ferreira, Paulo
(2024).
FOGFLEET: Fog-Level Federated Transfer Learning for Adaptive Transport Mode Detection.
Proceedings of the IEEE.
ISSN 0018-9219.
p. 22–33.
doi:
10.1109/IC2E61754.2024.00010.
Full text in Research Archive
Show summary
Transport Mode Detection (TMD) systems play a pivotal role in facilitating applications in transport, urban planning, and more. Exploiting the advancements in smartphone sensing capabilities, TMD systems have evolved for mobile applications, with local classification on smartphones as a common approach. Yet, local approaches relying on centralized training raise privacy concerns due to the transmission of sensitive data (e.g., GPS logs) over the Internet. In this paper, we propose FOGFLEET, a novel Federated Transfer Learning (FTL) framework for TMD, addressing both privacy and performance concerns. Our approach relies on Federated Learning (FL) to train a global model on various datasets from different cities while employing transfer learning to adapt the global model to the specific characteristics of individual smartphones and cities. FOGFLEET relies on an architecture that integrates edge, fog, and cloud layers, with dedicated fog nodes for each city to simplify cross-silo federated learning. Experimental results demonstrate the effectiveness of FOGFLEET framework in higher TMD accuracy by up to 20% than its comparable centralized approach. Furthermore, it outperforms the FL solutions reported in the literature with at least an 8% increase in accuracy. In this work, we also highlight the importance of sufficient training data for distributed training and discuss the impact of smartphone sensor qualities on TMD system performance. Our work contributes to advancing TMD systems by providing an adaptive and privacypreserving solution suitable for deployment in diverse urban environments and across various geographical locations.
-
Kamalian, Mahdieh; Taherkordi, Amirhosein & Ferreira, Paulo
(2024).
TELEGAIT: Transfer Learning on Fog for Generalizable and Real-Time Transport Mode Detection.
Proceedings of the IEEE.
ISSN 0018-9219.
p. 196–203.
doi:
10.1109/FMEC62297.2024.10710203.
Full text in Research Archive
Show summary
The knowledge of users’ transport modes, e.g., cars or buses, enables innovative transportation services, such as automatic ticketing and navigation. A promising approach for transport mode detection (TMD) is analyzing smartphone sensor data to infer transport modes using machine learning (ML) techniques. However, in existing work, the accuracy of MLbased approaches for a given city is highly dependent on whether the target ML model is trained with the data collected in that specific city by various users, smartphones, etc. To address this limitation, we need a comprehensive ML-based TMD system whose accuracy is not significantly affected by the geographical location, user variation, smartphone type, or smartphone position (e.g., in a bag or pocket)—generalizable TMD. In this paper, we propose an approach, called TELeGAIT, for generalizable TMD which is based on the idea of fine-tuning a base TMD ML model for the target city. Specifically, built on a base ML model trained with the aggregated data of some different cities, we fine-tune the base model for a new city by employing Transfer Learning (TL) and using the local data of that city. Moreover, some transportation services require low-latency TMD such as automatic ticketing. To this end, TeLeGait relies on a fog-based architecture, where the fine-tuning process is offloaded to a fog node in each city, and the TMD results are also stored on the fog node. TeLeGait utilizes a Multilayer Perceptron (MLP) model trained using two approaches: segment-based and raw sensor data. Our analysis shows that the raw data-based approach is more effective for neural networks, achieving 94.42%
-
Lepioufle, Jean-Marie; Schneider, Philipp; Hamer, Paul David; Ødegård, Rune Åvar; Vallejo, Islen & Cao, Tuan-Vu
[Show all 8 contributors for this article]
(2024).
Data fusion of sparse, heterogeneous, and mobile sensor devices using adaptive distance attention.
Environmental Data Science (EDS).
3.
doi:
10.1017/eds.2024.18.
Full text in Research Archive
Show summary
In environmental science, where information from sensor devices are sparse, data fusion for mapping purposes is often based on geostatistical approaches. We propose a methodology called adaptive distance attention that enables us to fuse sparse, heterogeneous, and mobile sensor devices and predict values at locations with no previous measurement. The approach allows for automatically weighting the measurements according to a priori quality information about the sensor device without using complex and resource-demanding data assimilation techniques. Both ordinary kriging and the general regression neural network (GRNN) are integrated into this attention with their learnable parameters based on deep learning architectures. We evaluate this method using three static phenomena with different complexities: a case related to a simplistic phenomenon, topography over an area of 196 and to the annual hourly concentration in 2019 over the Oslo metropolitan region (1026 ). We simulate networks of 100 synthetic sensor devices with six characteristics related to measurement quality and measurement spatial resolution. Generally, outcomes are promising: we significantly improve the metrics from baseline geostatistical models. Besides, distance attention using the Nadaraya–Watson kernel provides as good metrics as the attention based on the kriging system enabling the possibility to alleviate the processing cost for fusion of sparse data. The encouraging results motivate us in keeping adapting distance attention to space-time phenomena evolving in complex and isolated areas.
-
Rahnemay, Mahdieh; Farzinvash, Leili; Zolfi, Mina & Taherkordi, Amirhosein
(2024).
ECMSH: An Energy-efficient and Cost-effective data harvesting protocol for Mobile Sink-based Heterogeneous WSNs using PSO-TVAC.
Ad hoc networks.
ISSN 1570-8705.
164.
doi:
10.1016/j.adhoc.2024.103629.
Full text in Research Archive
-
Liu, Liqing; Yuan, Xiaoming; Chen, Decheng; Zhang, Ning; Sun, Haifeng & Taherkordi, Amirhosein
(2023).
Multi-user dynamic computation offloading and resource allocation in 5g mec heterogeneous networks with static and dynamic subchannels.
IEEE Transactions on Vehicular Technology.
ISSN 0018-9545.
72(11),
p. 14924–14938.
doi:
10.1109/TVT.2023.3285069.
Full text in Research Archive
-
Liu, Liqing; Yuan, Xiaoming; Zhang, Ning; Chen, Decheng; Yu, Keping & Taherkordi, Amirhosein
(2023).
Joint computation offloading and data caching in multi-Access edge computing enabled internet of vehicles.
IEEE Transactions on Vehicular Technology.
ISSN 0018-9545.
72(11),
p. 14939–14954.
doi:
10.1109/TVT.2023.3285073.
Full text in Research Archive
-
Shahryari, Mohammad-Salar; Farzinvash, Leili; Feizi-Derakhshi, Mohammad-Reza & Taherkordi, Amirhosein
(2023).
High-throughput and energy-efficient data gathering in heterogeneous multi-channel wireless sensor networks using genetic algorithm.
Ad hoc networks.
ISSN 1570-8705.
139.
doi:
10.1016/j.adhoc.2022.103041.
Full text in Research Archive
-
-
Li, Chunhai; Chen, Qiyong; Chen, Mingfeng; Su, Zhaoyu; Ding, Yong & Lan, Dapeng
[Show all 7 contributors for this article]
(2023).
Blockchain enabled task offloading based on edge cooperation in the digital twin vehicular edge network.
Journal of Cloud Computing.
12(1).
doi:
10.1186/s13677-023-00496-6.
Full text in Research Archive
Show summary
The rapid development of the Internet of Vehicles (IoV) along with the emergence of intelligent applications have put forward higher requirements for massive task offloading. Even though Mobile Edge Computing (MEC) can diminish network transmission delay and ease network congestion, the constrained heterogeneous resources of a single edge server and the highly dynamic topology of vehicular edge networks may compromise the efficiency of task offloading, including latency and energy consumption. Vehicular edge networks are also vulnerable to malicious outside attacks. In this paper, we propose a new blockchain-enabled digital twin vehicular edge network (DTVEN) where digital twin (DT) is exploited to monitor network communication, computation, and caching (3C) resources management in real time to provide rich data for offloading decision-making, and blockchain is utilized to secure fair and decentralized offloading transactions among DTs. To ensure 3C resources sharing across edge servers, we design a DT-assisted edge cooperation scheme, which makes full use of edge resources in vehicular networks. Furthermore, a DT-based smart contract is built to achieve a quick and effective consensus process. Then, we apply a task offloading algorithm based on an improved cuckoo algorithm (ICA) and a resource allocation scheme based on greedy strategy to minimize network cost by comprehensively taking into account latency and energy consumption. Numerical results demonstrate that our proposed scheme outperforms the existing schemes in terms of network cost.
-
-
Liu, Lei; Zhao, Ming; Yu, Miao; Jan, Mian Ahmad; Lan, Dapeng & Taherkordi, Amirhosein
(2023).
Mobility-Aware Multi-Hop Task Offloading for Autonomous Driving in Vehicular Edge Computing and Networks.
IEEE Transactions on Intelligent Transportation Systems (T-ITS).
ISSN 1524-9050.
24(2),
p. 2169–2182.
doi:
10.1109/TITS.2022.3142566.
Full text in Research Archive
-
Sepahvand, Majid; Abdali-Mohammadi, Fardin & Taherkordi, Amirhosein
(2023).
An adaptive teacher–student learning algorithm with decomposed knowledge distillation for on-edge intelligence.
Engineering Applications of Artificial Intelligence.
ISSN 0952-1976.
117.
doi:
10.1016/j.engappai.2022.105560.
Full text in Research Archive
-
Wei, Zhaohui; Pei, Qingqi; Zhang, Ning; Liu, Xuefeng; Wu, Celimuge & Taherkordi, Amirhosein
(2023).
Lightweight Federated Learning for Large-Scale IoT Devices With Privacy Guarantee.
IEEE Internet of Things Journal.
ISSN 2327-4662.
10(4),
p. 3179–3191.
doi:
10.1109/JIOT.2021.3127886.
Full text in Research Archive
-
-
-
Shi, Yongpeng; Wu, Jinsong; Liu, Lei; Lan, Dapeng & Taherkordi, Amirhosein
(2022).
Energy-Aware Relay Optimization and Power Allocation in Multiple Unmanned Aerial Vehicles Aided Satellite-Aerial-Terrestrial Networks.
IEEE Systems Journal.
ISSN 1932-8184.
16(4),
p. 5293–5304.
doi:
10.1109/JSYST.2022.3147491.
Full text in Research Archive
-
-
-
Sepahvand, Majid; Abdali-Mohammadi, Fardin & Taherkordi, Amirhosein
(2022).
Teacher–student knowledge distillation based on decomposed deep feature representation for intelligent mobile applications.
Expert Systems With Applications.
ISSN 0957-4174.
202.
doi:
10.1016/j.eswa.2022.117474.
Full text in Research Archive
-
Shahraki, Amin; Abbasi, Mahmoud; Taherkordi, Amirhosein & Jurcut, Anca Delia
(2022).
Active Learning for Network Traffic Classification: A Technical Study.
IEEE Transactions on Cognitive Communications and Networking (TCCN).
ISSN 2372-2045.
doi:
10.1109/TCCN.2021.3119062.
Full text in Research Archive
-
Mao, Sun; Wu, Jinsong; Liu, Lei; Lan, Dapeng & Taherkordi, Amirhosein
(2022).
Energy-Efficient Cooperative Communication and Computation for Wireless Powered Mobile-Edge Computing.
IEEE Systems Journal.
ISSN 1932-8184.
16(1),
p. 287–298.
doi:
10.1109/JSYST.2020.3020474.
Full text in Research Archive
-
Oplenskedal, Magnus Karsten; Herrmann, Peter & Taherkordi, Amir
(2021).
DEEPMATCH2: A comprehensive deep learning-based approach for in-vehicle presence detection.
Information Systems.
ISSN 0306-4379.
doi:
10.1016/j.is.2021.101927.
Full text in Research Archive
Show summary
The accurate detection of the mobile context information of public transportation vehicles and their passengers is a key feature to realize intelligent transportation systems. A topical example is in-vehicle presence detection that can, e.g., be used to ticket passengers automatically. Unfortunately, most existing solutions in this field suffer from low spatiotemporal accuracy which impedes their use in practice. In previous work, we addressed this challenge through a deep learning-based framework, called DeepMatch, that allows us to detect in-vehicle presence with a high degree of accuracy. DeepMatch utilizes the smartphone of a passenger to analyse and match the event streams of its own sensors with the event streams of counterpart sensors provided by a reference unit that is installed inside the vehicle. This is achieved through a new learning model architecture using Stacked Convolutional Autoencoders to compress sensor input streams by feature extraction and dimensionality reduction as well as a deep convolutional neural network to match the streams of the user phone and the reference device. The sensor stream compression is offloaded to the smartphone, while the matching is performed in a server. In this paper, we introduce DeepMatch2. It is an amended version of DeepMatch that reduces the amount of data to be transferred from the user and reference devices to the server by the factor of four. Further, DeepMatch2 improves the already good accuracy of DeepMatch from 97.81% to 98.51%. Moreover, we propose a travel inference algorithm, based on DeepMatch2, to detect the duration of whole passenger trips in public transport vehicles with a high degree of precision. This is needed to create intelligent and highly reliable auto-ticketing systems. Thanks to the high accuracy of 98.51% by DeepMatch2, the inferences can be carried out with a negligible error rate.
-
Zhang, Min; Eliassen, Frank; Taherkordi, Amirhosein; Jacobsen, Hans-Arno; Chung, Hwei-Ming & Zhang, Yan
(2021).
Demand Response Games for Peer-to-Peer Energy Trading with the Hyperledger Blockchain.
IEEE Transactions on Systems, Man & Cybernetics. Systems.
ISSN 2168-2216.
doi:
10.1109/TSMC.2021.3111135.
Full text in Research Archive
-
Lohrasbinasab, Iraj; Shahraki, Amin; Taherkordi, Amirhosein & Jurcut, Anca Delia
(2021).
From statistical- to machine learning-based network traffic prediction.
Transactions on Emerging Telecommunications Technologies.
ISSN 1124-318X.
doi:
10.1002/ett.4394.
Full text in Research Archive
Show summary
Nowadays, due to the exponential and continuous expansion of new paradigms such as Internet of Things (IoT), Internet of Vehicles (IoV) and 6G, the world is witnessing a tremendous and sharp increase of network traffic. In such large-scale, heterogeneous, and complex networks, the volume of transferred data, as big data, is considered a challenge causing different networking inefficiencies. To overcome these challenges, various techniques are introduced to monitor the performance of networks, called Network Traffic Monitoring and Analysis (NTMA). Network Traffic Prediction (NTP) is a significant subfield of NTMA which is mainly focused on predicting the future of network load and its behavior. NTP techniques can generally be realized in two ways, that is, statistical- and Machine Learning (ML)-based. In this paper, we provide a study on existing NTP techniques through reviewing, investigating, and classifying the recent relevant works conducted in this field. Additionally, we discuss the challenges and future directions of NTP showing that how ML and statistical techniques can be used to solve challenges of NTP.
-
Abbasi, Mahmoud; Shahraki, Amin & Taherkordi, Amirhosein
(2021).
Deep Learning for Network Traffic Monitoring and Analysis (NTMA): A Survey.
Computer Communications.
ISSN 0140-3664.
170,
p. 19–41.
doi:
10.1016/j.comcom.2021.01.021.
Full text in Research Archive
Show summary
Modern communication systems and networks, e.g., Internet of Things (IoT) and cellular networks, generate a massive and heterogeneous amount of traffic data. In such networks, the traditional network management techniques for monitoring and data analytics face some challenges and issues, e.g., accuracy, and effective processing of big data in a real-time fashion. Moreover, the pattern of network traffic, especially in cellular networks, shows very complex behavior because of various factors, such as device mobility and network heterogeneity. Deep learning has been efficiently employed to facilitate analytics and knowledge discovery in big data systems to recognize hidden and complex patterns. Motivated by these successes, researchers in the field of networking apply deep learning models for Network Traffic Monitoring and Analysis (NTMA) applications, e.g., traffic classification and prediction. This paper provides a comprehensive review on applications of deep learning in NTMA. We first provide fundamental background relevant to our review. Then, we give an insight into the confluence of deep learning and NTMA, and review deep learning techniques proposed for NTMA applications. Finally, we discuss key challenges, open issues, and future research directions for using deep learning in NTMA applications.
-
Shahraki, Amin; Taherkordi, Amirhosein & Haugen, Øystein
(2021).
TONTA: Trend-based Online Network Traffic Analysis in ad-hoc IoT networks.
Computer Networks.
ISSN 1389-1286.
194.
doi:
10.1016/j.comnet.2021.108125.
Full text in Research Archive
Show summary
Internet of Things (IoT) refers to a system of interconnected heterogeneous smart devices communicatingwithout human intervention. A significant portion of existing IoT networks is under the umbrella of ad-hoc andquasi ad-hoc networks. Ad-hoc based IoT networks suffer from the lack of resource-rich network infrastructuresthat are able to perform heavyweight network management tasks using, e.g. machine learning-based NetworkTraffic Monitoring and Analysis (NTMA) techniques. Designing light-weight NTMA techniques that do notneed to be (re-) trained has received much attention due to the time complexity of the training phase. In thisstudy, a novel pattern recognition method, called Trend-based Online Network Traffic Analysis (TONTA), isproposed for ad-hoc IoT networks to monitor network performance. The proposed method uses a statisticallight-weight Trend Change Detection (TCD) method in an online manner. TONTA discovers predominant trendsand recognizes abrupt or gradual time-series dataset changes to analyze the IoT network traffic. TONTA isthen compared with RuLSIF as an offline benchmark TCD technique. The results show that TONTA detectsapproximately 60% less false positive alarms than RuLSIF.
-
Oplenskedal, Magnus Karsten; Taherkordi, Amirhosein & Herrmann, Peter
(2021).
Ataraxis: A Deep Learning Approach for Hardwareless In-Vehicle Presence Detection,
IEEE International Conference on Cognitive Machine Intelligence (CogMI).
Institute of Electrical and Electronics Engineers (IEEE).
ISSN 9781665416214.
p. 17–26.
doi:
10.1109/CogMI52975.2021.00012.
Full text in Research Archive
Show summary
Accurately detecting the mobile contexts, in which public transport vehicles and their passengers operate, is key
for future intelligent context-aware services in transportation systems. A prominent example is in-vehicle presence detection that, for instance, can be used to provide services such as automated ticketing, dynamic vehicle distribution, and live route optimization. To use such services in practice, the in-vehicle presence detection needs to be close to infallible. However, most existing solutions in this field suffer from low spatiotemporal accuracy. To address this challenge, in this paper we introduce ATARAXIS, an approach to hardwareless in-vehicle presence detection. In particular, we develop a deep convolutional neural network that can be trained to detect if a user is inside a public transportation vehicle such as a tram, subway, or bus from the raw sensor events generated by the sensors in a single ordinary smartphone. We show that this information can be used to infer the in-vehicle presence of users over time when combined with other sources such as the GPS trace of the user and that of the public transport vehicles. ATARAXIS has the capability to distinguish between the four user modes driving a car, riding a bike, walking, and using public transport with an accuracy of 98.69%. That is higher than existing techniques for transport mode detection. Since we are interested in applying ATARAXIS in practice, we also made experiments on the battery consumption and CPU overhead to be expected. The results show that the use of ATARAXIS in smartphones incurs a negligible computational overhead and power consumption, even though we base our approach on sensor data collection and a deep learning model.
-
Puka, Ergys; Herrmann, Peter & Taherkordi, Amirhosein
(2020).
Hybrid Context-aware Message Flooding for Dead Spot Mitigation in V2I Communication.
IEEE Vehicular Technology Conference (VTC).
ISSN 1090-3038.
doi:
10.1109/VTC2020-Fall49728.2020.9348721.
Full text in Research Archive
Show summary
An often overlooked practical problem in Intelligent Transportation Systems (ITS) is the presence of areas without
cellular network connectivity, so-called dead spots, which aggravates the communication between vehicles and the external infrastructure. In our previous work, we suggested to mitigate this problem by using a hybrid data dissemination protocol that combines cellular network communication with ad-hoc networks between vehicles. If the vehicles in such ad-hoc networks are in a dead spot but have a good estimation about the time they will leave it again, messages can be forwarded to the vehicle that is supposed to regain cellular network coverage first. Since this vehicle may transmit the stored messages immediately after having left the dead spot, the delivery time is improved. In this paper, we first analyze the behavior of the aforementioned data dissemination protocol in larger dead spots in which a message may be carried by several vehicles before being delivered via the cellular network. The analysis reveals that messages are not always delivered in the fastest possible time. To address this concern, a new protocol variant named context-aware message flooding protocol is introduced. This protocol, indeed, guarantees the fastest possible forwarding of messages to their recipients. This is achieved at the cost of delivering duplicates that, however, are only produced when the delivery of a message is sped up.
-
Azangoo, Mohammad; Taherkordi, Amirhosein & Blech, Jan Olaf
(2020).
Digital Twins for Manufacturing Using UML and Behavioral Specifications.
IEEE Conference on Emerging Technologies and Factory Automation.
ISSN 1946-0740.
p. 1035–1038.
doi:
10.1109/ETFA46521.2020.9212165.
Full text in Research Archive
-
Shahraki, Amin; Taherkordi, Amirhosein; Haugen, Øystein & Eliassen, Frank
(2020).
A Survey and Future Directions on Clustering: From WSNs to IoT and Modern Networking Paradigms.
IEEE Transactions on Network and Service Management.
18(2).
doi:
10.1109/TNSM.2020.3035315.
Full text in Research Archive
Show summary
Many Internet of Things (IoT) networks are created as an overlay over traditional ad-hoc networks such as Zigbee. Moreover, IoT networks can resemble ad-hoc networks over networks that support device-to-device (D2D) communication, e.g., D2D-enabled cellular networks and WiFi-Direct. In these ad-hoc types of IoT networks, efficient topology management is a crucial requirement, and in particular in massive scale deployments. Traditionally, clustering has been recognized as a common approach for topology management in ad-hoc networks, e.g., in Wireless Sensor Networks (WSNs). Topology management in WSNs and ad-hoc IoT networks has many design commonalities as both need to transfer data to the destination hop by hop. Thus, WSN clustering techniques can presumably be applied for topology management in ad-hoc IoT networks. This requires a comprehensive study on WSN clustering techniques and investigating their applicability to ad-hoc IoT networks. In this article, we conduct a survey of this field based on the objectives for clustering, such as reducing energy consumption and load balancing, as well as the network properties relevant for efficient clustering in IoT, such as network heterogeneity and mobility. Beyond that, we investigate the advantages and challenges of clustering when IoT is integrated with modern computing and communication technologies such as Blockchain, Fog/Edge computing, and 5G. This survey provides useful insights into research on IoT clustering, allows broader understanding of its design challenges for IoT networks, and sheds light on its future applications in modern technologies integrated with IoT.
-
Yuan, Xiaoming; Tian, Hansen; Wang, Haiyang; Su, Haoru; Liu, Jiemin & Taherkordi, Amirhosein
(2020).
Edge-Enabled WBANs for Efficient QoS Provisioning Healthcare Monitoring: A Two-Stage Potential Game-Based Computation Offloading Strategy.
IEEE Access.
8,
p. 92718–92730.
doi:
10.1109/ACCESS.2020.2992639.
Full text in Research Archive
-
Yuan, Xiaoming; Zhao, Zheyu; Wang, Haiyang; Zhang, Lin; Taherkordi, Amirhosein & Yu, Hua
(2020).
MDMH: An improved multi-attribute decision-making and highest response ratio next-based computation offloading approach for wireless body area networks.
Transactions on Emerging Telecommunications Technologies.
ISSN 1124-318X.
p. 1–20.
doi:
10.1002/ett.4070.
Full text in Research Archive
-
Lan, Dapeng; Taherkordi, Amir; Eliassen, Frank; Chen, Zhuang & Liu, Lei
(2020).
Deep Reinforcement Learning for Intelligent Migration of Fog Services in Smart Cities.
Lecture Notes in Computer Science (LNCS).
ISSN 0302-9743.
12453,
p. 230–244.
doi:
10.1007/978-3-030-60239-0_16.
Full text in Research Archive
-
Oplenskedal, Magnus Karsten; Taherkordi, Amirhosein & Herrmann, Peter
(2020).
DeepMatch: deep matching for in-vehicle presence detection in transportation,
DEBS '20: Proceedings of the 14th ACM International Conference on Distributed and Event-based Systems.
Association for Computing Machinery (ACM).
ISSN 9781450380287.
p. 97–108.
doi:
10.1145/3401025.3401741.
Full text in Research Archive
Show summary
A key feature of modern public transportation systems is the accurate detection of the mobile context of transport vehicles and their passengers. A prominent example is automatic in-vehicle presence detection which allows, e.g., intelligent auto-ticketing of passengers. Most existing solutions, in this field, are based on either using active RFID or Bluetooth Low Energy (BLE) technology, or mobile sensor data analysis. Such techniques suffer from low spatiotemporal accuracy in in-vehicle presence detection. In this paper, we address this issue by proposing a deep learning model and the design of an associated generic distributed framework. Our approach, called DeepMatch, utilizes the smartphone of a passenger to analyze and match the event streams of its own sensors and the event streams of the counterpart sensors in an in-vehicle reference unit. This is achieved through a new learning model architecture using Stacked Convolutional Autoencoders for feature extraction and dimensionality reduction, as well as a dense neural network for stream matching. In this distributed framework, feature extraction and dimensionality reduction is offloaded to the smartphone, while matching is performed in a server, e.g., in the Cloud. In this way, the number of sensor events to be transmitted for matching on the server side will be minimized. We evaluated DeepMatch based on a large dataset taken in real vehicles. The evaluation results show that the statistical accuracy of our approach is 0.978 for in-vehicle presence detection which, as we will argue, is sufficient to be used in, e.g., auto-ticketing systems.
-
Shahraki, Amin; Taherkordi, Amirhosein; Haugen, Øystein & Eliassen, Frank
(2020).
Clustering objectives in wireless sensor networks: A survey and research direction analysis.
Computer Networks.
ISSN 1389-1286.
180.
doi:
10.1016/j.comnet.2020.107376.
Full text in Research Archive
Show summary
Wireless Sensor Networks (WSNs) typically include thousands of resource-constrained sensors to monitor their surroundings, collect data, and transfer it to remote servers for further processing. Although WSNs are considered highly flexible ad-hoc networks, network management has been a fundamental challenge in these types of net- works given the deployment size and the associated quality concerns such as resource management, scalability, and reliability. Topology management is considered a viable technique to address these concerns. Clustering is the most well-known topology management method in WSNs, grouping nodes to manage them and/or executing various tasks in a distributed manner, such as resource management. Although clustering techniques are mainly known to improve energy consumption, there are various quality-driven objectives that can be realized through clustering. In this paper, we review comprehensively existing WSN clustering techniques, their objectives and the network properties supported by those techniques. After refining more than 500 clustering techniques, we extract about 215 of them as the most important ones, which we further review, catergorize and classify based on clustering objectives and also the network properties such as mobility and heterogeneity. In addition, statistics are provided based on the chosen metrics, providing highly useful insights into the design of clustering techniques in WSNs.
-
-
Lan, Dapeng; Pang, Zhibo; Fischione, Carlo; Liu, Yu; Taherkordi, Amirhosein & Eliassen, Frank
(2019).
Latency Analysis of Wireless Networks for Proximity Services in Smart Home and Building Automation: The Case of Thread.
IEEE Access.
7,
p. 4856–4867.
doi:
10.1109/ACCESS.2018.2888939.
Full text in Research Archive
Show summary
Proximity service (ProSe), using the geographic location and device information by considering the proximity of mobile devices, enriches the services we use to interact with people and things around us. ProSe has been used in mobile social networks in proximity and also in smart home and building automation (Google Home). To enable ProSe in smart home, reliable and stable network protocols and communication infrastructures are needed. Thread is a new wireless protocol aiming at smart home and building automation (BA), which supports mesh networks and native Internet protocol connectivity. The latency of Thread should be carefully studied when used in user-friendly and safety-critical ProSe in smart home and BA. In this paper, a system level model of latency in the Thread mesh network is presented. The accumulated latency consists of different kinds of delay from the application layer to the physical layer. A Markov chain model is used to derive the probability distribution of the medium access control service time. The system level model is experimentally validated in a multi-hop Thread mesh network. The outcomes show that the system model results match well with the experimental results. Finally, based on an analytical model, a software tool is developed to estimate the latency of the Thread mesh network, providing developers more network information to develop user-friendly and safety-critical ProSe in smart home and BA.
-
Dreibholz, Thomas; Mazumdar, Somnath; Zahid, Feroz; Taherkordi, Amirhosein & Gran, Ernst Gunnar
(2019).
Mobile Edge as Part of the Multi-Cloud Ecosystem: A Performance Study.
In Leporati, Francesco; Danese, Giovanni & Merelli, Ivan (Ed.),
Proceedings of the 27th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP).
IEEE (Institute of Electrical and Electronics Engineers).
ISSN 9781728116440.
p. 59–66.
doi:
10.1109/EMPDP.2019.8671599.
Full text in Research Archive
Show summary
Cloud computing has revolutionised the development and deployment of applications by running them cost-effectively in remote data centres. With the increasing need for mobility and micro-services, particularly with the emerging 5G mobile broadband networks, there is also a strong demand for mobile edge computing (MEC). It enables applications to run in small cloud systems in close proximity to the user in order to minimise latencies. Both cloud computing and MEC have their own advantages and disadvantages. Combining these two computing paradigms in a unified multi-cloud platform has the potential of obtaining the best of both worlds. However, a comprehensive study is needed to evaluate the performance gains and the overheads imposed by this combination to real-world cloud applications. In this paper, we introduce a baseline performance evaluation in order to identify the fallacies and pitfalls of combining multiple cloud systems and MEC into a unified MEC-multi-cloud platform. For this purpose, we analyze the basic, application-independent performance metrics of average round-trip time (RTT) and average application payload throughput in a setup consisting of two private and one public cloud systems. This baseline performance analysis confirms the feasibility of MEC-multi-cloud and provides guidelines for designing an autonomic resource provisioning solution in terms of an extension proposed to our existing Melodic middleware platform for multi-cloud applications.
-
Oplenskedal, Magnus Karsten; Taherkordi, Amirhosein & Herrmann, Peter
(2019).
Automated Product Localization through Mobile Data Analysis.
In Sarwat, Mohamed; Zheng, Baihua & Zheng, Kevin Kai (Ed.),
20th IEEE International Conference on Mobile Data Management (MDM).
IEEE (Institute of Electrical and Electronics Engineers).
ISSN 9781728133645.
p. 18–26.
doi:
10.1109/MDM.2019.00-78.
Full text in Research Archive
Show summary
Recent developments in the field of indoor RealTime Locating Systems (RTLS) using mobile devices stimulate decision support for users. For instance, smartphone-based navigation in shops can enable location-aware recommendations of certain products to customers. An impeding factor to realize such
systems is that they need the exact position of products. Existing product localization solutions, however, are based on tagging or manual location registering which tend to be quite costly and laborious. In this paper, we propose an automated product localization approach solving this problem. Our system infers the location of products based on the results of accumulating two sets of customer data, i.e., the locations at which the customers stop for picking up items as well as the list of the items, they purchase. These two data sets are accumulated for a large number of users, making it possible to build correct mappings between the products and their positions. We introduce a basic version of our localization algorithm and two extensions. One helps to improve calculating the position of relocated products while the other one fosters a faster localization using a smaller number of user data sets. We discuss the results of various simulation runs which give evidence that our system has a good potential to work in practice.
-
Shooshtarian, Leila; Lan, Dapeng & Taherkordi, Amirhosein
(2019).
A Clustering-Based Approach to Efficient Resource Allocation in Fog Computing.
Communications in Computer and Information Science (CCIS).
ISSN 1865-0929.
1080 CCIS,
p. 207–224.
doi:
10.1007/978-3-030-30143-9_17.
Full text in Research Archive
-
Oplenskedal, Magnus Karsten; Taherkordi, Amirhosein & Herrmann, Peter
(2019).
Automated Product Localization through Mobile Data Analysis.
Proceedings (IEEE International Conference on Mobile Data Management).
ISSN 1551-6245.
p. 18–26.
doi:
10.1109/MDM.2019.00-78.
Full text in Research Archive
Show summary
Recent developments in the field of indoor RealTime Locating Systems (RTLS) using mobile devices stimulate decision support for users. For instance, smartphone-based navigation in shops can enable location-aware recommendations of certain products to customers. An impeding factor to realize such
systems is that they need the exact position of products. Existing product localization solutions, however, are based on tagging or manual location registering which tend to be quite costly and laborious. In this paper, we propose an automated product localization approach solving this problem. Our system infers the location of products based on the results of accumulating two sets of customer data, i.e., the locations at which the customers stop for picking up items as well as the list of the items, they purchase. These two data sets are accumulated for a large number of users, making it possible to build correct mappings between the products and their positions. We introduce a basic version of our localization algorithm and two extensions. One helps to improve calculating the position of relocated products while the other one fosters a faster localization using a smaller number of user data sets. We discuss the results of various simulation runs which give evidence that our system has a good potential to work in practice.
-
Taherkordi, Amirhosein & Herrmann, Peter
(2018).
Pervasive Smart Contracts for Blockchains in IoT Systems.
In Parizi, Reza M. (Eds.),
Proceedings of the 2018 International Conference on Blockchain Technology and Application.
Association for Computing Machinery (ACM).
ISSN 9781450366465.
p. 1–6.
doi:
10.1145/3301403.3301405.
Full text in Research Archive
Show summary
The blockchain technology has gained tremendous attention thanks to its decentralized structure, immutability, and enhanced security and privacy guarantees. Blockchain has the potential to address security and privacy challenges of Internet of Things (IoT). By hosting and executing \textit{smart contracts}, blockchain allows secure and flexible message communication between IoT devices and traceability in IoT applications. The unique characteristics of IoT systems, such as heterogeneity and pervasiveness, pose challenges in designing smart contracts for IoT systems. In this paper, we study those challenges and propose a microservice-based approach to the design of IoT smart contracts. The proposed service model is aimed to encapsulate functionalities such as contract-level communication between IoT devices, access to data-sources within contracts, and supporting interoperability of heterogeneous IoT smart contracts. The flexibility, scalability and modularity of the microservice architecture model make it an efficient approach for developing IoT smart contracts.
-
Taherkordi, Amirhosein; Eliassen, Frank; McDonald, Michael & Horn, Geir
(2018).
Context-Driven and Real-Time Provisioning of Data-Centric IoT Services in the Cloud.
ACM Transactions on Internet Technology (TOIT).
ISSN 1533-5399.
19(1),
p. 7:1–7:24.
doi:
10.1145/3151006.
Full text in Research Archive
-
Zahid, Feroz; Taherkordi, Amirhosein; Gran, Ernst Gunnar; Skeie, Tor & Johnsen, Bjørn Dag
(2018).
A Self-Adaptive Network for HPC Clouds: Architecture, Framework, and Implementation.
IEEE Transactions on Parallel and Distributed Systems.
ISSN 1045-9219.
29(12),
p. 2658–2671.
doi:
10.1109/TPDS.2018.2842224.
Full text in Research Archive
Show summary
Clouds offer flexible and economically attractive compute and storage solutions for enterprises. However, the effectiveness of cloud computing for high-performance computing (HPC) systems still remains questionable. When clouds are deployed on lossless interconnection networks, like InfiniBand (IB), challenges related to load-balancing, low-overhead virtualization, and performance isolation hinder full potential utilization of the underlying interconnect. Moreover, cloud data centers incorporate a highly dynamic environment rendering static network reconfigurations, typically used in IB systems, infeasible. In this paper, we present a framework for a self-adaptive network architecture for HPC clouds based on lossless interconnection networks, demonstrated by means of our implemented IB prototype. Our solution, based on a feedback control and optimization loop, enables the lossless HPC network to dynamically adapt to the varying traffic patterns, current resource availability, workload distributions, and also in accordance with the service provider-defined policies. Furthermore, we present IBAdapt, a simplified ruled-based language for the service providers to specify adaptation strategies used by the framework. Our developed self-adaptive IB network prototype is demonstrated using state-of-the-art industry software. The results obtained on a test cluster demonstrate the feasibility and effectiveness of the framework when it comes to improving Quality-of-Service compliance in HPC clouds.
-
-
Oplenskedal, Magnus Karsten; Herrmann, Peter; Blech, Jan Olaf & Taherkordi, Amirhosein
(2018).
Simulation-driven Development of Self-adaptive Transportation Systems,
Proceedings of the 2018 IEEE Conference on Network Softwarization.
IEEE (Institute of Electrical and Electronics Engineers).
ISSN 9781538646335.
p. 372–377.
doi:
10.1109/NETSOFT.2018.8460073.
Full text in Research Archive
Show summary
Modern Intelligent Transportation Systems (ITS) operate highly automatically. Therefore, they have to be able to
handle a large variety of situations each demanding a particular system behavior. That aggravates the development of control software that has to guarantee safe and expedient operation in all possible situations. To support a suitable reconfiguration of the controllers to changing environments, the use of self-adaptation seems to be a highly promising approach. In this paper, we propose to combine model-based engineering of control software with simulation. That allows us to create and test controller software in parallel with the physical systems, it shall operate. Moreover, this approach makes it possible to safely confront a transport system with situations that, otherwise, could only be reproduced taking a significant risk. In particular, we introduce a framework for the creation of control software using simulators together with a development structure. The suggested design process is illustrated with a mobile robot example.
-
Taherkordi, Amirhosein; Herrmann, Peter; Blech, Jan Olaf & Fernandez, Alvaro Fernandez
(2017).
Service virtualization for self-adaptation in mobile cyber-physical systems.
Lecture Notes in Computer Science (LNCS).
ISSN 0302-9743.
10380 LNCS,
p. 56–68.
doi:
10.1007/978-3-319-68136-8_5.
Full text in Research Archive
-
Svae, Alexander; Taherkordi, Amirhosein; Herrmann, Peter & Blech, Jan Olaf
(2017).
Self-adaptive control in cyber-physical systems: The autonomous train experiment.
In Shin, Sung Y.; Shin, Dongwan & Lancastre, Maria (Ed.),
Proceedings of the Symposium on Applied Computing.
Association for Computing Machinery (ACM).
ISSN 9781450344869.
p. 1436–1443.
doi:
10.1145/3019612.3019651.
Full text in Research Archive
Show summary
Autonomous systems become more and more important in today’s
transport sector. They often operate in dynamic environments in which unpredictable events may occur at any time. These events may affect the safe operation of vehicles, calling for highly efficient control software technologies to reason about and react on their appearance. A crucial efficiency parameter is timeliness as vehicles often operate under high speed. The contribution of this paper is the presentation and analysis of design aspects of dynamic control software in the context of an autonomous train experiment. This is achieved through a self-adaptation software framework intended for autonomous trains and built on a demonstrator using Lego Mindstorms. The main mission of the framework is to collect context information, reason about it, and adapt the train behavior accordingly. The adaptation framework is implemented using the development tool Reactive Blocks and tested on the demonstrator. The evaluation results provide useful insights into the performance of the framework, particularly about the time needed to reason about the context and to carry out reconfigurations.
-
-
-
-
-
-
-
-
Taherkordi, Amirhosein & Eliassen, Frank
(2015).
Models@run.time for creating in-cloud dynamic cyber-physical ecosystems.
IEEE International Conference on Cloud Computing Technology and Science (CloudCom).
ISSN 2330-2194.
2015-February(February),
p. 976–982.
doi:
10.1109/CloudCom.2014.158.
Full text in Research Archive
-
Dar, Kashif Sana; Taherkordi, Amirhosein; Baraki, Harun; Eliassen, Frank & Geihs, Kurt
(2014).
A Resource Oriented Integration Architecture for the Internet of Things: A Business Processes Perspective.
Pervasive and Mobile Computing.
ISSN 1574-1192.
20,
p. 145–159.
doi:
10.1016/j.pmcj.2014.11.005.
Full text in Research Archive
Show summary
The vision of the Internet of Things (IoT) foresees a future Internet incorporating smart physical objects that offer hosted functionality as IoT services. These services when integrated with the traditional enterprise level services form the creation of ambient intelligence for a wide range of applications. To facilitate seamless access and service life cycle management of large, distributed and heterogeneous IoT resources, service oriented computing and resource oriented approaches have been widely used as promising technologies. However, a reference architecture integrating IoT services into either of these two technologies is still an open research challenge. In this article, we adopt the resource oriented approach to provide an end-to-end integration architecture of front-end IoT devices with the back-end business process applications. The proposed architecture promises a programmer friendly access to IoT services, an event management mechanism to propagate context information of IoT devices, a service replacement facility upon service failure, and a decentralized execution of the IoT aware business processes.
-
-
-
-
Taherkordi, Amirhosein; Loiret, Frederic; Rouvoy, Romain & Eliassen, Frank
(2013).
Optimizing Sensor Network Reprogramming via In Situ Reconfigurable Components.
ACM transactions on sensor networks.
ISSN 1550-4867.
9(2).
doi:
10.1145/2422966.2422971.
Full text in Research Archive
-
Romero, Daniel; Hermosillo, Gabriel; Taherkordi, Amirhosein; Nzekwa, Russel; Rouvoy, Romain & Eliassen, Frank
(2013).
The DigiHome Service-Oriented Platform.
Software, Practice & Experience.
ISSN 0038-0644.
43(10),
p. 1205–1218.
doi:
10.1002/spe.1125.
Full text in Research Archive
-
Taherkordi, Amirhosein; Loiret, Frederic; Rouvoy, Romain & Eliassen, Frank
(2011).
A Generic Component-Based Approach for Programming, Composing and Tuning Sensor Software.
Computer journal.
ISSN 0010-4620.
54(8),
p. 1248–1266.
doi:
10.1093/comjnl/bxq102.
Full text in Research Archive
-
Dar, Kashif Sana; Taherkordi, Amirhosein; Rouvoy, Romain & Eliassen, Frank
(2011).
Adaptable Service Composition for Very-Large-Scale Internet of Things Systems.
In Eyers, David (Eds.),
Proceedings of the 8th Middleware Doctoral Symposium of the 12th ACM/IFIP/USENIX International Middleware Conference.
ACM Publications.
ISSN 9781450310727.
doi:
10.1145/2093190.2093192.
Full text in Research Archive
View all works in NVA
-
Le, Thanh; La, Hoang Loc; Taherkordi, Amirhosein; Eliassen, Frank; Ha, Hoai Phuong & Guan, Peiyuan
(2026).
Empirical Analysis of GPU Frequency Behavior Under ML Workloads.
doi:
10.1109/ccgridw69005.2026.00018.
Full text in Research Archive
Show summary
This work presents ongoing research on the frequency-scaling behavior of NVIDIA GPUs when executing ML/AI workloads. Our preliminary findings show that, on lower-performance GPUs, the operating frequency is strongly affected by the recent workload history-typically within an 80ms window. This behavior challenges a common assumption under-lying several state-of-the-art ML latency-prediction techniques, which treat individual GPU kernel latencies as independent and therefore estimate total execution time by summing isolated per-kernel measurements. Our results indicate that such an assumption does not always hold, as the GPU's dynamic frequency scaling introduces inter-kernel dependencies. We also outline several promising directions for leveraging this observation in future work, including improved latency-prediction models, GPU kernel-reordering strategies, and NAS-driven guidelines for frequency/latency/energy-aware model design.
-
Huang, He; Chen, Shigang; Basat, Ran Ben; Dai, Haipeng; Taherkordi, Amirhosein & Xu, Jun
(2024).
Guest Editorial Introduction to the Special Section on Next-Generation Traffic Measurement With Network-Wide Perspective and Artificial Intelligence.
IEEE Transactions on Network Science and Engineering (IEEE TNSE).
11(3),
p. 2332–2337.
doi:
10.1109/TNSE.2024.3389428.
Full text in Research Archive
-
Oplenskedal, Magnus Karsten; Taherkordi, Amirhosein & Herrmann, Peter
(2021).
Ataraxis: A Deep Learning Approach for Hardwareless In-Vehicle Presence Detection.
Full text in Research Archive
Show summary
Accurately detecting the mobile contexts, in which public transport vehicles and their passengers operate, is key
for future intelligent context-aware services in transportation systems. A prominent example is in-vehicle presence detection that, for instance, can be used to provide services such as automated ticketing, dynamic vehicle distribution, and live route optimization. To use such services in practice, the in-vehicle presence detection needs to be close to infallible. However, most existing solutions in this field suffer from low spatiotemporal accuracy. To address this challenge, in this paper we introduce ATARAXIS, an approach to hardwareless in-vehicle presence detection. In particular, we develop a deep convolutional neural network that can be trained to detect if a user is inside a public transportation vehicle such as a tram, subway, or bus from the raw sensor events generated by the sensors in a single ordinary smartphone. We show that this information can be used to infer the in-vehicle presence of users over time when combined with other sources such as the GPS trace of the user and that of the public transport vehicles. ATARAXIS has the capability to distinguish between the four user modes driving a car, riding a bike, walking, and using public transport with an accuracy of 98.69%. That is higher than existing techniques for transport mode detection. Since we are interested in applying ATARAXIS in practice, we also made experiments on the battery consumption and CPU overhead to be expected. The results show that the use of ATARAXIS in smartphones incurs a negligible computational overhead and power consumption, even though we base our approach on sensor data collection and a deep learning model.
-
Ramezanifarkhani, Toktam & Taherkordi, Amirhosein
(2021).
IKTPLUSS Grant Proposal- How to write a successful proposal?
Full text in Research Archive
-
Puka, Ergys; Herrmann, Peter & Taherkordi, Amirhosein
(2020).
Hybrid Context-aware Message Flooding for Dead Spot Mitigation in V2I Communication.
Full text in Research Archive
Show summary
An often overlooked practical problem in Intelligent Transportation Systems (ITS) is the presence of areas without
cellular network connectivity, so-called dead spots, which aggravates the communication between vehicles and the external infrastructure. In our previous work, we suggested to mitigate this problem by using a hybrid data dissemination protocol that combines cellular network communication with ad-hoc networks between vehicles. If the vehicles in such ad-hoc networks are in a dead spot but have a good estimation about the time they will leave it again, messages can be forwarded to the vehicle that is supposed to regain cellular network coverage first. Since this vehicle may transmit the stored messages immediately after having left the dead spot, the delivery time is improved. In this paper, we first analyze the behavior of the aforementioned data dissemination protocol in larger dead spots in which a message may be carried by several vehicles before being delivered via the cellular network. The analysis reveals that messages are not always delivered in the fastest possible time. To address this concern, a new protocol variant named context-aware message flooding protocol is introduced. This protocol, indeed, guarantees the fastest possible forwarding of messages to their recipients. This is achieved at the cost of delivering duplicates that, however, are only produced when the delivery of a message is sped up.
-
Oplenskedal, Magnus Karsten; Taherkordi, Amirhosein & Herrmann, Peter
(2020).
Deep Matching for In-Vehicle Presence Detection in Transportation.
Full text in Research Archive
Show summary
A key feature of modern public transportation systems is the accurate detection of the mobile context of transport vehicles and their passengers. A prominent example is automatic in-vehicle presence detection which allows, e.g., intelligent auto-ticketing of passengers. Most existing solutions, in this field, are based on either using active RFID or Bluetooth Low Energy (BLE) technology, or mobile sensor data analysis. Such techniques suffer from low spatiotemporal accuracy in in-vehicle presence detection. In this paper, we address this issue by proposing a deep learning model and the design of an associated generic distributed framework. Our approach, called DeepMatch, utilizes the smartphone of a passenger to analyze and match the event streams of its own sensors and the event streams of the counterpart sensors in an in-vehicle reference unit. This is achieved through a new learning model architecture using Stacked Convolutional Autoencoders for feature extraction and dimensionality reduction, as well as a dense neural network for stream matching. In this distributed framework, feature extraction and dimensionality reduction is offloaded to the smartphone, while matching is performed in a server, e.g., in the Cloud. In this way, the number of sensor events to be transmitted for matching on the server side will be minimized. We evaluated DeepMatch based on a large dataset taken in real vehicles. The evaluation results show that the statistical accuracy of our approach is 0.978 for in-vehicle presence detection which, as we will argue, is sufficient to be used in, e.g., auto-ticketing systems.
-
Oplenskedal, Magnus Karsten; Taherkordi, Amirhosein & Herrmann, Peter
(2019).
Automated Product Localization through Mobile Data Analysis.
Full text in Research Archive
Show summary
Recent developments in the field of indoor RealTime Locating Systems (RTLS) using mobile devices stimulate decision support for users. For instance, smartphone-based navigation in shops can enable location-aware recommendations of certain products to customers. An impeding factor to realize such
systems is that they need the exact position of products. Existing product localization solutions, however, are based on tagging or manual location registering which tend to be quite costly and laborious. In this paper, we propose an automated product localization approach solving this problem. Our system infers the location of products based on the results of accumulating two sets of customer data, i.e., the locations at which the customers stop for picking up items as well as the list of the items, they purchase. These two data sets are accumulated for a large number of users, making it possible to build correct mappings between the products and their positions. We introduce a basic version of our localization algorithm and two extensions. One helps to improve calculating the position of relocated products while the other one fosters a faster localization using a smaller number of user data sets. We discuss the results of various simulation runs which give evidence that our system has a good potential to work in practice.
-
Taherkordi, Amirhosein & Herrmann, Peter
(2018).
Pervasive Smart Contracts for Blockchains in IoT Systems.
Full text in Research Archive
Show summary
The blockchain technology has gained tremendous attention thanks to its decentralized structure, immutability, and enhanced security and privacy guarantees. Blockchain has the potential to address security and privacy challenges of Internet of Things (IoT). By hosting and executing \textit{smart contracts}, blockchain allows secure and flexible message communication between IoT devices and traceability in IoT applications. The unique characteristics of IoT systems, such as heterogeneity and pervasiveness, pose challenges in designing smart contracts for IoT systems. In this paper, we study those challenges and propose a microservice-based approach to the design of IoT smart contracts. The proposed service model is aimed to encapsulate functionalities such as contract-level communication between IoT devices, access to data-sources within contracts, and supporting interoperability of heterogeneous IoT smart contracts. The flexibility, scalability and modularity of the microservice architecture model make it an efficient approach for developing IoT smart contracts.
-
Oplenskedal, Magnus Karsten; Herrmann, Peter; Blech, Jan Olaf & Taherkordi, Amirhosein
(2018).
Simulation-driven Development of Self-adaptive Transportation Systems.
Full text in Research Archive
Show summary
Modern Intelligent Transportation Systems (ITS) operate highly automatically. Therefore, they have to be able to
handle a large variety of situations each demanding a particular system behavior. That aggravates the development of control software that has to guarantee safe and expedient operation in all possible situations. To support a suitable reconfiguration of the controllers to changing environments, the use of self-adaptation seems to be a highly promising approach. In this paper, we propose to combine model-based engineering of control software with simulation. That allows us to create and test controller software in parallel with the physical systems, it shall operate. Moreover, this approach makes it possible to safely confront a transport system with situations that, otherwise, could only be reproduced taking a significant risk. In particular, we introduce a framework for the creation of control software using simulators together with a development structure. The suggested design process is illustrated with a mobile robot example.
-
Svae, Alexander; Taherkordi, Amirhosein; Herrmann, Peter & Blech, Jan Olaf
(2017).
Self-Adaptive Control in Cyber-Physical Systems: The Autonomous Train Experiment.
Full text in Research Archive
Show summary
Autonomous systems become more and more important in today’s
transport sector. They often operate in dynamic environments in which unpredictable events may occur at any time. These events may affect the safe operation of vehicles, calling for highly efficient control software technologies to reason about and react on their appearance. A crucial efficiency parameter is timeliness as vehicles often operate under high speed. The contribution of this paper is the presentation and analysis of design aspects of dynamic control software in the context of an autonomous train experiment. This is achieved through a self-adaptation software framework intended for autonomous trains and built on a demonstrator using Lego Mindstorms. The main mission of the framework is to collect context information, reason about it, and adapt the train behavior accordingly. The adaptation framework is implemented using the development tool Reactive Blocks and tested on the demonstrator. The evaluation results provide useful insights into the performance of the framework, particularly about the time needed to reason about the context and to carry out reconfigurations.
-
Taherkordi, Amirhosein; Herrmann, Peter; Blech, Jan Olaf & Fernandez, Alvaro
(2016).
Service Virtualization for Self-Adaptation in Mobile Cyber-Physical Systems.
Full text in Research Archive
Show summary
Mobile Cyber-Physical Systems (mCPS) consist of cooperating units that often operate in an unpredictably changing environment. Thus, they need to adapt quickly to varying spatial and temporal conditions during operation, e.g., to avoid collisions. The control software of the mobile units has to reflect this complex dynamics, and traditional device-level adaptation models are usually not efficient enough to engineer them smoothly. We address this challenge by proposing a Virtual Adaptation Services Framework (VASF). It provides a virtualized application-level view to adaptation requirements, enabling adaptation coordination between cooperative mCPS devices. In particular, VASF allows us to describe the contextual conditions of mCPS by abstract rules that are analyzed at runtime by the tool-set BeSpaceD. Based on this analysis, the control systems of the involved mCPS units are automatically reconfigured using the OSGi framework.
The approach is demonstrated with DiddyBorg robots that are operated by Raspberry Pi boards.
View all works in NVA
Published
Oct. 16, 2013 11:04 AM
- Last modified
Jan. 2, 2020 12:10 PM
