Paal Engelstad

• Erdmann, Niklas; Bentsen, Lars Ødegaard; Stenbro, Roy; Riise, Heine Nygard; Warakagoda, Narada Dilp & Engelstad, Paal E. (2026). Multi-Horizon Time Series Forecasting of Non-Parametric CDFs with Deep Lattice Networks. Proceedings of the AAAI Conference on Artificial Intelligence. ISSN 2159-5399. 40(1), p. 255–264. Full text in Research Archive Show summary

  Probabilistic forecasting is not only a way to add more information to a prediction of the future, but it also builds on weaknesses in point prediction. Sudden changes in a time series can still be captured by a cumulative distribution function (CDF), while a point prediction is likely to miss it entirely. The modeling of CDFs within forecasts has historically been limited to parametric approaches, but due to recent advances, this no longer has to be the case. We aim to advance the fields of probabilistic forecasting and monotonic networks by connecting them and propose an approach that permits the forecasting of implicit, complete, and nonparametric CDFs. For this purpose, we propose an adaptation to deep lattice networks (DLN) for monotonically constrained simultaneous/implicit quantile regression in time series forecasting. Quantile regression usually produces quantile crossovers, which need to be prevented to achieve a legitimate CDF. By leveraging long short term memory units (LSTM) as the embedding layer, and spreading quantile inputs to all sub-lattices of a DLN with an extended output size, we can produce a multi-horizon forecast of an implicit CDF due to the monotonic constraintability of DLNs that prevent quantile crossovers. We compare and evaluate our approach's performance to relevant state of the art within the context of a highly relevant application of time series forecasting: Day-ahead, hourly forecasts of solar irradiance observations. Our experiments show that the adaptation of a DLN performs just as well or even better than an unconstrained approach. Further comparison of the adapted DLN against a scalable monotonic neural network shows that our approach performs better. With this adaptation of DLNs, we intend to create more interest and crossover investigations in techniques of monotonic neural networks and probabilistic forecasting.

• Zieglmeier, Sebastian Georg; Badyn, Mathias Hudoba de; Warakagoda, Narada Dilp; Krogstad, Thomas Røbekk & Engelstad, Paal E. (2025). Semi-Data-Driven Model Predictive Control: A Physics-Informed Data-Driven Control Approach. IEEE Conference on Decision and Control. Proceedings. ISSN 0743-1546. p. 2110–2117. doi: 10.1109/cdc57313.2025.11312410. Full text in Research Archive Show summary

  Data-enabled predictive control (DeePC) has emerged as a powerful technique to control complex systems without the need for extensive modeling efforts. However, relying solely on offline collected data trajectories to represent the system dynamics introduces certain drawbacks. Therefore, we present a novel semi-data-driven model predictive control (SD-MPC) framework that combines (limited) model information with DeePC to address a range of these drawbacks, including sensitivity to noisy data and a lack of robustness. In this work, we focus on the performance of DeePC in operating regimes not captured by the offline collected data trajectories and demonstrate how incorporating an underlying parametric model can counteract this issue. SD-MPC exhibits equivalent closed-loop performance as DeePC for deterministic linear time-invariant systems. Simulations demonstrate the general control performance of the proposed SD-MPC for both a linear time-invariant system and a nonlinear system modeled as a linear parameter-varying system. These results provide numerical evidence of the enhanced robustness of SD-MPC over classical DeePC.

• Aryan, Ramtin; Yazidi, Anis; Bouhoula, Adel & Engelstad, Paal E. (2025). A formal technique for automatic resolution of OpenFlow anomalies. International Journal of Information Security. ISSN 1615-5262. 24. doi: 10.1007/s10207-025-01035-x. Full text in Research Archive Show summary

  Conflicts between network policies are frequent in today’s computer networks due to the increasing complexity of network configuration. Resolving policy anomalies usually requires network administrator intervention, which is a time-intensive and error-prone process. This paper presents inference systems for the automatic resolution of OpenFlow anomalies. The approach uses high-level policies to detect conflict correction without policy violations. Our approach is fully automated and does not require interaction with the network administrator. Although there is a multitude of research papers on detecting anomalies in SDN, research to correct those anomalies in an automatic manner is very scarce, if not non-existent. We have formally proven the soundness and completeness of our inference systems. Furthermore, we provide experimental results based on real-life network configurations involving more than 1200 rules. The detection and correction processes exhibit very low computation overhead, in the order of milliseconds, when parallelization is used.

• Bentsen, Lars Ødegaard; Warakagoda, Narada Dilp; Stenbro, Roy & Engelstad, Paal (2024). Relative evaluation of probabilistic methods for spatio-temporal wind forecasting. Journal of Cleaner Production. ISSN 0959-6526. 434. doi: 10.1016/j.jclepro.2023.139944. Full text in Research Archive Show summary

  Short-term wind power forecasting has become a de facto tool to better facilitate the integration of such renewable energy resources into modern power grids. Instead of point predictors, which produce single-value predictions for the expected power, probabilistic forecasts predict probability distributions over the expected power output or associated confidence intervals. In this study, three different parametric and non-parametric methods for uncertainty modelling in wind power forecasting were studied, namely quantile regression (QR), variational inference and a maximum likelihood estimation (MLE) method. Johnson’s SU distribution was studied as a novel candidate for modelling wind power, which is a transformed normal distribution that exhibits both skew and heavy tails. This was one of the first studies to provide a thorough investigation of Johnson’s SU distribution for uncertainty modelling in a complex deep learning framework for wind forecasting. It was found that Johsnon’s SU likelihood and QR-based models significantly outperformed models using Gaussian likelihoods, based on a range of quantitative metrics to evaluate probability distributions and qualitative investigation of produced forecasts. Variational inference models using Johnson’s SU likelihoods performed remarkably well, with near-perfect calibration and higher precision than models using any of the other methods for uncertainty modelling, as evaluated through the pinball loss, Average Coverage Error (ACE) and Prediction Interval Coverage Percentage (PICP) metric. With the superior performance of Johnson’s SU likelihood models, the study mainly contributes to the literature by introducing another candidate distribution for probabilistic wind forecasting, which is analytical, unbounded and easy to integrate into modern deep learning frameworks.

• Khan, Mehak; Mello, Gustavo Borges Moreno e; Habib, Laurence; Engelstad, Paal & Yazidi, Anis (2024). HITS-based Propagation Paradigm for Graph Neural Networks. ACM Transactions on Knowledge Discovery from Data. ISSN 1556-4681. 18(4), p. 1–23. doi: 10.1145/3638779. Full text in Research Archive Show summary

  In this article, we present a new propagation paradigm based on the principle of Hyperlink-Induced Topic Search (HITS) algorithm. The HITS algorithm utilizes the concept of a “self-reinforcing” relationship of authority-hub. Using HITS, the centrality of nodes is determined via repeated updates of authority-hub scores that converge to a stationary distribution. Unlike PageRank-based propagation methods, which rely solely on the idea of authorities (in-links), HITS considers the relevance of both authorities (in-links) and hubs (out-links), thereby allowing for a more informative graph learning process. To segregate node prediction and propagation, we use a Multilayer Perceptron in combination with a HITS-based propagation approach and propose two models: HITS-GNN and HITS-GNN+. We provided additional validation of our models’ efficacy by performing an ablation study to assess the performance of authority-hub in independent models. Moreover, the effect of the main hyper-parameters and normalization is also analyzed to uncover how these techniques influence the performance of our models. Extensive experimental results indicate that the proposed approach significantly improves baseline methods on the graph (citation network) benchmark datasets by a decent margin for semi-supervised node classification, which can aid in predicting the categories (labels) of scientific articles not exclusively based on their content but also based on the type of articles they cite.

• Erdmann, Niklas; Bentsen, Lars Ødegaard; Stenbro, Roy; Riise, Heine Nygard; Warakagoda, Narada Dilp & Engelstad, Paal (2024). Deep and Probabilistic Solar Irradiance Forecast at the Arctic Circle. Conference record of the Photovoltaic Specialists Conference. ISSN 0160-8371. p. 1599–1606. doi: 10.1109/PVSC57443.2024.10748862. Full text in Research Archive
• Abolghasemi, Roza; Viedma, Enrique Herrera; Engelstad, Paal E.; Djenouri, Youcef & Yazidi, Anis (2024). A graph neural approach for group recommendation system based on pairwise preferences. Information Fusion. ISSN 1566-2535. 107. doi: 10.1016/j.inffus.2024.102343. Full text in Research Archive Show summary

  Pairwise preference information, which involves users expressing their preferences by comparing items, plays a crucial role in decision-making and has recently found application in recommendation systems. In this study, we introduce GcPp, a clustering algorithm that leverages pairwise preference data to generate recommendations for user groups. Initially, we construct individual graphs for each user based on their pairwise preferences and utilize a graph convolutional network to predict similarities between all pairs of graphs. These predicted similarity scores form the foundation of our research. We then construct a new graph where users are nodes and the edges are weighted according to the predicted similarities. Finally, we perform clustering on the graph’s nodes (users). By evaluating various metrics, we found that employing a similarity metric based on a convolutional neural network (SimGNN) with our proposed ground truth called Top-K yielded the highest accuracy. The proposed approach is specifically designed for group recommendation systems and holds significant potential for group decision-making problems. Code is available at https://github.com/RozaAbolghasemi/Group_Recommendation_Syatem_GcPp_clustering.

• Khan, Mehak; Mello, Gustavo Borges Moreno e; Engelstad, Paal E.; Habib, Laurence & Yazidi, Anis (2023). HITS-GNN: A Simplified Propagation Scheme for Graph Neural Networks. In Tsumoto, Shusaku; Ohsawa, Yukio; Chen, Lei; Poel, Dirk Van den; Hu, Xiaohua; Motomura, Yoichi; Takagi, Takuya; Wu, Lingfei; Xie, Ying; Abe, Akihiro & Raghavan, Vijay (Ed.), 2022 IEEE International Conference on Big Data. IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781665480451. doi: 10.1109/BigData55660.2022.10020815. Full text in Research Archive
• Rauniyar, Ashish; Østerbø, Olav Norvald; Håkegård, Jan Erik & Engelstad, Paal E. (2023). Exploiting Cooperative Downlink NOMA in D2D Communications. Sensors. 23(8). doi: 10.3390/s23083958. Full text in Research Archive Show summary

  We propose and investigate a bidirectional device-to-device (D2D) transmission scheme that exploits cooperative downlink non-orthogonal multiple access (NOMA) (termed as BCD-NOMA). In BCD-NOMA, two source nodes communicate with their corresponding destination nodes via a relaying node while exchanging bidirectional D2D messages simultaneously. BCD-NOMA is designed for improved outage probability (OP) performance, high ergodic capacity (EC) and high energy efficiency by allowing two sources to share the same relaying node for data transmission to their corresponding destination nodes while also facilitating bidirectional D2D communications exploiting downlink NOMA. Simulation and analytical expressions of the OP, EC and ergodic sum capacity (ESC) under both perfect and imperfect successive interference cancellation (SIC) are used to demonstrate the effectiveness of BCD-NOMA compared to conventional schemes.

• Bentsen, Lars Ødegaard; Warakagoda, Narada Dilp; Stenbro, Roy & Engelstad, Paal E. (2023). A Unified Graph Formulation for Spatio-Temporal Wind Forecasting. Energies. 16(20). doi: 10.3390/en16207179. Full text in Research Archive Show summary

  With the rapid adoption of wind energy globally, there is a need for accurate short-term forecasting systems to improve the reliability and integration of such energy resources on a large scale. While most spatio-temporal forecasting systems comprise distinct components to learn spatial and temporal dependencies separately, this paper argues for an approach to learning spatio-temporal information jointly. Many time series forecasting systems also require aligned input information and do not naturally facilitate irregular data. Research is therefore required to investigate methodologies for forecasting in the presence of missing or corrupt measurements. To help combat some of these challenges, this paper studied a unified graph formulation. With the unified formulation, a graph neural network (GNN) was used to extract spatial and temporal dependencies simultaneously, in a single update, while also naturally facilitating missing data. To evaluate the proposed unified approach, the study considered hour-ahead wind speed forecasting in the North Sea under different amounts of missing data. The framework was compared against traditional spatio-temporal architectures that used GNNs together with temporal long short-term memory (LSTM) and Transformer or Autoformer networks, along with the imputation of missing values. The proposed framework outperformed the traditional architectures, with absolute errors of around 0.73–0.90 m per second, when subject to 0–80% of missing input data. The unified graph approach was also better at predicting large changes in wind speed, with an additional 10-percentage-point improvement over the second-best model. Overall, this paper investigated a novel methodology for spatio-temporal wind speed forecasting and showed how the proposed unified graph formulation achieved competitive results compared to more traditional GNN-based architectures.

• Rauniyar, Ashish; Østerbø, Olav Norvald; Håkegård, Jan Erik & Engelstad, Paal E. (2023). Ergodic Performance Analysis of Reconfigurable Intelligent Surface Enabled Bidirectional NOMA. In Gregory, Mark A. & Li, Shuo (Ed.), Proceedings of the 2023 33rd International Telecommunication Networks and Applications Conference (ITNAC). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9798350317138. p. 128–133. doi: 10.1109/ITNAC59571.2023.10368523. Full text in Research Archive Show summary

  This paper proposes and investigates a reconfigurable intelligent surface (RIS) enabled bidirectional non-orthogonal multiple access (NOMA) network termed as NOMA-RIS. Here, RIS allows multiple NOMA users in one group to communicate with or share information with multiple NOMA users in another group. Specifically, the two NOMA user groups send the data intended for exchange to the RIS. RIS reflects the NOMA signals allowing bidirectional communication between two NOMA user groups. In particular, under the Rician fading environment, we pay close attention to how well RIS-enabled bidirectional NOMA networks operate. Analytical expressions for tight upper bounds for the ergodic capacity are mathematically derived and verified with the simulation results. Comprehensive performance comparisons are presented, showing that our proposed bidirectional NOMA-RIS system can achieve enhanced capacity gains than RIS-enabled traditional orthogonal multiple access schemes. These comparisons also offer practical insights into the impact of various system parameters on the overall network performance.

• Meng, Li; Goodwin, Morten; Yazidi, Anis & Engelstad, Paal E. (2023). Unsupervised State Representation Learning in Partially Observable Atari Games. In Tsapatsoulis, Nicolas (Eds.), Computer Analysis of Images and Patterns. CAIP 2023. Springer. ISSN 9783031442391. p. 212–222. doi: 10.1007/978-3-031-44240-7_21. Full text in Research Archive Show summary

  State representation learning aims to capture latent factors of an environment. Although some researchers realize the connections between masked image modeling and contrastive representation learning, the effort is focused on using masks as an augmentation technique to represent the latent generative factors better. Partially observable environments in reinforcement learning have not yet been carefully studied using unsupervised state representation learning methods. In this article, we create an unsupervised state representation learning scheme for partially observable states. We conducted our experiment on a previous Atari 2600 framework designed to evaluate representation learning models. A contrastive method called Spatiotemporal DeepInfomax (ST-DIM) has shown state-of-the-art performance on this benchmark but remains inferior to its supervised counterpart. Our approach improves ST-DIM when the environment is not fully observable and achieves higher F1 scores and accuracy scores than the supervised learning counterpart. The mean accuracy score averaged over categories of our approach is 66%, compared to 38% of supervised learning. The mean F1 score is 64% to 33%. The code can be found on https://github.com/mengli11235/MST_DIM.

• Abolghasemi, Roza; Engelstad, Paal E.; Herrera-Viedma, Enrique & Yazidi, Anis (2022). A personality-aware group recommendation system based on pairwise preferences. Information Sciences. ISSN 0020-0255. 595(595), p. 1–17. doi: 10.1016/j.ins.2022.02.033. Full text in Research Archive Show summary

  Human personality plays a crucial role in decision-making and it has paramount importance when individuals negotiate with each other to reach a common group decision. Such situations are conceivable, for instance, when a group of individuals want to watch a movie together. It is well known that people influence each other’s decisions, the more assertive a person is, the more influence they will have on the final decision. In order to obtain a more realistic group recommendation system (GRS), we need to accommodate the assertiveness of the different group members’ personalities. Although pairwise preferences are long-established in group decision-making (GDM), they have received very little attention in the recommendation systems community. Driven by the advantages of pairwise preferences on ratings in the recommendation systems domain, we have further pursued this approach in this paper, however we have done so for GRS. We have devised a three-stage approach to GRS in which we 1) resort to three binary matrix factorization methods, 2) develop an influence graph that includes assertiveness and cooperativeness as personality traits, and 3) apply an opinion dynamics model in order to reach consensus. We have shown that the final opinion is related to the stationary distribution of a Markov chain associated with the influence graph. Our experimental results demonstrate that our approach results in high precision and fairness.

• Bentsen, Lars Ødegaard; Warakagoda, Narada Dilp; Stenbro, Roy & Engelstad, Paal (2022). Wind Park Power Prediction: Attention-Based Graph Networks and Deep Learning to Capture Wake Losses. Journal of Physics: Conference Series (JPCS). ISSN 1742-6588. 2265. doi: 10.1088/1742-6596/2265/2/022035. Full text in Research Archive Show summary

  Abstract With the increased penetration of wind energy into the power grid, it has become increasingly important to be able to predict the expected power production for larger wind farms. Deep learning (DL) models can learn complex patterns in the data and have found wide success in predicting wake losses and expected power production. This paper proposes a modular framework for attention-based graph neural networks (GNN), where attention can be applied to any desired component of a graph block. The results show that the model significantly outperforms a multilayer perceptron (MLP) and a bidirectional LSTM (BLSTM) model, while delivering performance on-par with a vanilla GNN model. Moreover, we argue that the proposed graph attention architecture can easily adapt to different applications by offering flexibility into the desired attention operations to be used, which might depend on the specific application. Through analysis of the attention weights, it was showed that employing attention-based GNNs can provide insights into what the models learn. In particular, the attention networks seemed to realise turbine dependencies that aligned with some physical intuition about wake losses.

• Aryan, Ramtin; Yazidi, Anis; Brattensborg, Frode Henriksen; Kure, Øivind & Engelstad, Paal E. (2022). SDN Spotlight: A real-time OpenFlow troubleshooting framework. Future Generation Computer Systems. ISSN 0167-739X. 133, p. 364–377. doi: 10.1016/j.future.2022.03.014. Full text in Research Archive Show summary

  Troubleshooting in SDN-based networks is still a cumbersome task that can overwhelm human attention. Various anomalies, such as installation failure, disordered rules, and loops, remain unnoticed even when the most recent detection methods are used. In this paper, we address the issue of verifying SDN policies by actively probing the data plane. SDN Spotlight is presented as an anomaly detection framework that tries to detect installation failures, rule conflicts, and loops. In contrast to recent work, such as Monocle and Pronto, SDN Spotlight verifies a chain of rules using a single probing packet. This approach also reduces the number of monitoring rules, which has a direct effect on saving TCAM memory usage and minimizing the packet matching time. SDN Spotlight addresses two problems: verifying rule installation and forwarding behavior verification. Within the SDN Spotlight framework, we introduce two different approaches for forwarding anomaly detection: Hedge-SDN Spotlight and Open-SDN Spotlight. Furthermore, we devise an efficient and fast probe generation algorithm that generates one single probing packet per chain of rules. As opposed to other related work, Hedge-SDN Spotlight does not yield false positives and false negatives when detecting loops and forwarding failures. The results of the experiment demonstrate that SDN Spotlight is much faster than the SDNProbe and SDN traceroute method, in some cases by a factor of up to seven times as fast.

• Abolghasemi, Roza; Khadka, Rabindra; Lind, Pedro; Engelstad, Paal E.; Viedma, Enrique Herrera & Yazidi, Anis (2022). Predicting missing pairwise preferences from similarity features in group decision making. Knowledge-Based Systems. ISSN 0950-7051. 256(256). doi: 10.1016/j.knosys.2022.109860. Full text in Research Archive Show summary

  In group decision-making (GDM), fuzzy preference relations (FPRs) refer to pairwise preferences in the form of a matrix. Within the field of GDM, the problem of estimating missing values is of utmost importance, since many experts provide incomplete preferences. In this paper, we propose a new method called the entropy-based method for estimating the missing values in the FPR. We compared the accuracy of our algorithm for predicting the missing values with the best candidate algorithm from state of the art achievements. In the proposed entropy-based method, we took advantage of pairwise preferences to achieve good results by storing extra information compared to single rating scores, for example, a pairwise comparison of alternatives vs. the alternative’s score from one to five stars. The entropy-based method maps the prediction problem into a matrix factorization problem, and thus the solution for the matrix factorization can be expressed in the form of latent expert features and latent alternative features. Thus, the entropy-based method embeds alternatives and experts in the same latent feature space. By virtue of this embedding, another novelty of our approach is to use the similarity of experts, as well as the similarity between alternatives, to infer the missing values even when only minimal data are available for some alternatives from some experts. Note that current approaches may fail to provide any output in such cases. Apart from estimating missing values, another salient contribution of this paper is to use the proposed entropy-based method to rank the alternatives. It is worth mentioning that ranking alternatives have many possible applications in GDM, especially in group recommendation systems (GRS).

• Meng, Li; Yazidi, Anis; Goodwin, Morten & Engelstad, Paal (2022). Expert Q-learning: Deep Reinforcement Learning with Coarse State Values from Offline Expert Examples. Proceedings of the Northern Lights Deep Learning Workshop. 3. doi: 10.7557/18.6237. Full text in Research Archive Show summary

  In this article, we propose a novel algorithm for deep reinforcement learning named Expert Q-learning. Expert Q-learning is inspired by Dueling Q-learning and aims to incorporate semi-supervised learning into reinforcement learning through splitting Q-values into state values and action advantages. We require that an offline expert assesses the value of a state in a coarse manner using three discrete values. An expert network is designed in addition to the Q-network, which updates each time following the regular offline minibatch update whenever the expert example buffer is not empty. Using the board game Othello, we compare our algorithm with the baseline Q-learning algorithm, which is a combination of Double Q-learning and Dueling Q-learning. Our results show that Expert Q-learning is indeed useful and more resistant to the overestimation bias. The baseline Q-learning algorithm exhibits unstable and suboptimal behavior in non-deterministic settings, whereas Expert Q-learning demonstrates more robust performance with higher scores, illustrating that our algorithm is indeed suitable to integrate state values from expert examples into Q-learning.

• Meng, Li; Goodwin, Morten; Yazidi, Anis & Engelstad, Paal (2022). Improving the Diversity of Bootstrapped DQN by Replacing Priors With Noise. IEEE Transactions on Games (TG). ISSN 2475-1502. 15(4), p. 580–589. doi: 10.1109/TG.2022.3185330. Full text in Research Archive Show summary

  Q-learning is one of the most well-known Reinforcement Learning algorithms. There have been tremendous efforts to develop this algorithm using neural networks. Bootstrapped Deep Q-Learning Network is amongst them. It utilizes multiple neural network heads to introduce diversity into Q-learning. Diversity can sometimes be viewed as the amount of reasonable moves an agent can take at a given state, analogous to the definition of the exploration ratio in RL. Thus, the performance of Bootstrapped Deep Q-Learning Network is deeply connected with the level of diversity within the algorithm. In the original research, it was pointed out that a random prior could improve the performance of the model. In this article, we further explore the possibility of replacing priors with noise and sample the noise from a Gaussian distribution to introduce more diversity into this algorithm. We conduct our experiment on the Atari benchmark and compare our algorithm to both the original and other related algorithms. The results show that our modification of the Bootstrapped Deep Q-Learning algorithm achieves significantly higher evaluation scores across different types of Atari games. Thus, we conclude that replacing priors with noise can improve Bootstrapped Deep Q-Learning’s performance by ensuring the integrity of diversities.

• Bentsen, Lars Ødegaard; Warakagoda, Narada Dilp; Stenbro, Roy & Engelstad, Paal (2022). Probabilistic Wind Park Power Prediction using Bayesian Deep Learning and Generative Adversarial Networks. Journal of Physics: Conference Series (JPCS). ISSN 1742-6588. 2362. doi: 10.1088/1742-6596/2362/1/012005. Full text in Research Archive Show summary

  The rapid depletion of fossil-based energy supplies, along with the growing reliance on renewable resources, has placed supreme importance on the predictability of renewables. Research focusing on wind park power modelling has mainly been concerned with point estimators, while most probabilistic studies have been reserved for forecasting. In this paper, a few different approaches to estimate probability distributions for individual turbine powers in a real off-shore wind farm were studied. Two variational Bayesian inference models were used, one employing a multilayered perceptron and another a graph neural network (GNN) architecture. Furthermore, generative adversarial networks (GAN) have recently been proposed as Bayesian models and was here investigated as a novel area of research. The results showed that the two Bayesian models outperformed the GAN model with regards to mean absolute errors (MAE), with the GNN architecture yielding the best results. The GAN on the other hand, seemed potentially better at generating diverse distributions. Standard deviations of the predicted distributions were found to have a positive correlation with MAEs, indicating that the models could correctly provide estimates on the confidence associated with particular predictions.

• Bentsen, Lars Ødegaard; Warakagoda, Narada Dilp; Stenbro, Roy & Engelstad, Paal (2022). Spatio-temporal wind speed forecasting using graph networks and novel Transformer architectures. Applied Energy. ISSN 0306-2619. 333. doi: 10.1016/j.apenergy.2022.120565. Full text in Research Archive Show summary

  To improve the security and reliability of wind energy production, short-term forecasting has become of utmost importance. This study focuses on multi-step spatio-temporal wind speed forecasting for the Norwegian continental shelf. In particular, the study considers 14 offshore measurement stations and aims to leverage spatial dependencies through the relative physical location of different stations to improve local wind forecasts and simultaneously output different forecasts for each of the 14 locations. Our multi-step forecasting models produce either 10-minute, 1- or 4-hour forecasts, with 10-minute resolution, meaning that the models produce more informative time series for predicted future trends. A graph neural network (GNN) architecture was used to extract spatial dependencies, with different update functions to learn temporal correlations. These update functions were implemented using different neural network architectures. One such architecture, the Transformer, has become increasingly popular for sequence modelling in recent years. Various alterations have been proposed to better facilitate time series forecasting, of which this study focused on the Informer, LogSparse Transformer and Autoformer. This is the first time the LogSparse Transformer and Autoformer have been applied to wind forecasting and the first time any of these or the Informer have been formulated in a spatio-temporal setting for wind forecasting. By comparing against spatio-temporal Long Short-Term Memory (LSTM) and Multi-Layer Perceptron (MLP) models, the study showed that the models using the altered Transformer architectures as update functions in GNNs were able to outperform these. Furthermore, we propose the Fast Fourier Transformer (FFTransformer), which is a novel Transformer architecture based on signal decomposition and consists of two separate streams that analyse the trend and periodic components separately. The FFTransformer and Autoformer were found to achieve superior results for the 10-minute and 1-hour ahead forecasts, with the FFTransformer significantly outperforming all other models for the 4-hour ahead forecasts. Our code to implement the different models are made publicly available at: https://github.com/LarsBentsen/FFTransformer.

• Rauniyar, Ashish; Østerbø, Olav Norvald; Håkegård, Jan Erik & Engelstad, Paal E. (2022). Secrecy Performance Analysis of Cooperative Nonorthogonal Multiple Access in IoT Networks. IEEE Sensors Journal. ISSN 1530-437X. 22(19), p. 19030–19045. doi: 10.1109/JSEN.2022.3199540. Full text in Research Archive Show summary

  Different system models utilizing Non-orthogonal multiple access (NOMA) have been successfully studied to meet the growing capacity demands of the Internet of Things (IoT) devices for the next-generation networks. However, analyzing the anti-eavesdropping for NOMA systems under different scenarios and settings still needs further exploration before it can be practically deployed. Therefore, in this paper, we study the secrecy performance of a cooperative NOMA system in IoT networks where two source nodes communicate with their respective destination nodes via a common relay in the presence of an eavesdropper. Specifically, two source node sends their data in parallel over the same frequency band to the common relay node using uplink NOMA. Then, the relay node forwards the decoded symbols to the respective destination nodes using downlink NOMA in the presence of an eavesdropper. To enhance the security performance of the considered system, we study and propose an artificial noise (AN)-aided scheme in which the two destination nodes emit a jamming signal to confuse the eavesdropper while receiving the signal from the common relay node. We also study the effect of NOMA power allocation, perfect successive interference cancellation (pSIC), and imperfect SIC (ipSIC) on the considered system. Analytical expressions for the Ergodic capacity, Ergodic secrecy sum rate (ESSR), and secrecy outage probability (SOP) are mathematically derived and verified with the simulation results. Our results demonstrate that a significantly higher ESSR and lower SOP of the system can be attained compared to a conventional NOMA system without a destination-assisted jamming signal scheme.

• Aryan, Ramtin; Yazidi, Anis; Bouhoula, Adel & Engelstad, Paal E. (2020). Net Auto-Solver: A formal approach for automatic resolution of OpenFlow anomalies. Conference on Local Computer Networks. ISSN 0742-1303. doi: 10.1109/LCN48667.2020.9314851. Full text in Research Archive Show summary

  Policy anomalies are frequent in nowadays’s computer networks due to their increasing configuration complexity. Resolving policy anomalies usually requires network administrator intervention, which is a time-intensive and error-prone process. In this paper, we present Net Auto-Solver, a formal approach for automatic resolution of OpenFlow anomalies. The approach resorts to the concept of high-level policies to not only detect policy violations but also correct them on-the-fly. Our approach is fully automated and does not require interaction with the network administrator. Although there is a multitude of research works on detecting anomalies in SDN, research to correct those anomalies in an automatic manner is extremely scarce. At the heart of our approach, we propose two inference systems to perform corrective actions to the policy. We provide some experimental results involving real-life network configurations to show the performance of our approach. The first results are very promising.

• Rauniyar, Ashish; Engelstad, Paal E. & Østerbø, Olav Norvald (2020). On the Performance of Bidirectional NOMA-SWIPT Enabled IoT Relay Networks. IEEE Sensors Journal. ISSN 1530-437X. 21(2), p. 2299–2315. doi: 10.1109/JSEN.2020.3018905. Full text in Research Archive Show summary

  In this paper, a bidirectional relaying (BR) non-orthogonal multiple access (NOMA) with simultaneous wireless information and power transfer (SWIPT) termed as BR NOMA-SWIPT is proposed and investigated for the Internet of Things (IoT) relay networks. Here, multiple NOMA users in one group can communicate or exchange information with multiple NOMA users in another group through a common energy harvesting (EH) based relay. The EH based relay exploits the radio frequency (RF) energy supplied by the two NOMA user groups to recharge itself, and then it exchanges the information between them. Specifically, the two groups of NOMA users transmit the information intended for the exchange to the relay node using the uplink NOMA protocol. The relay node first harvests the RF energy through the signals of the two group of NOMA users, and then it carries out the exchange of information between two NOMA user groups by using the downlink NOMA protocol. To the best of our knowledge, there is no existing study or research on BR NOMA with SWIPT. Therefore, in this paper, we examine in detail, the performance of BR NOMA-SWIPT enabled IoT relay networks. Specifically, we study a popular EH time switching (TS) relaying architecture with BR and NOMA. We also study the effect of both perfect successive interference cancellation (pSIC) and imperfect SIC (ipSIC) on the proposed BR NOMA-SWIPT system. Analytical expressions for the outage probability and ergodic capacity are mathematically derived. The analytical results of our proposed system model are validated by the simulation results, and representative performance comparisons are presented thoroughly, which not only provides practical insights into the effect of different system parameters on the overall network performance, but it also demonstrates that our proposed BR NOMA-SWIPT can attain significant throughput and capacity gains as compared to conventional BR multiple access schemes.

• Hagos, Desta Haileselassie; Løland, Martin V.; Yazidi, Anis; Kure, Øivind & Engelstad, Paal E. (2020). Advanced passive operating system fingerprinting using machine learning and deep learning. Proceedings - International Conference on Computer Communications and Networks (ICCCN). ISSN 1095-2055. p. 1–11. doi: 10.1109/ICCCN49398.2020.9209694. Full text in Research Archive Show summary

  Securing and managing large, complex enterprise network infrastructure requires capturing and analyzing network traffic traces in real-time. An accurate passive Operating System (OS) fingerprinting plays a critical role in effective network management and cybersecurity protection. Passive fingerprinting doesn't send probes that introduce extra load to the network and hence it has a clear advantage over active fingerprinting since it also reduces the risk of triggering false alarms. This paper proposes and evaluates an advanced classification approach to passive OS fingerprinting by leveraging state-of-the-art classical machine learning and deep learning techniques. Our controlled experiments on benchmark data, emulated and realistic traffic is performed using two approaches. Through an Oracle-based machine learning approach, we found that the underlying TCP variant is an important feature for predicting the remote OS. Based on this observation, we develop a sophisticated tool for OS fingerprinting that first predicts the TCP flavor using passive traffic traces and then uses this prediction as an input feature for another machine learning algorithm for predicting the remote OS from passive measurements. This paper takes the passive fingerprinting problem one step further by introducing the underlying predicted TCP variant as a distinguishing feature. In terms of accuracy, we empirically demonstrate that accurately predicting the TCP variant has the potential to boost the evaluation performance from 84% to 94% on average across all our validation scenarios and across different types of traffic sources. We also demonstrate a practical example of this potential, by increasing the performance to 91.3% on average using a tool for TCP variant prediction in an emulated setting. To the best of our knowledge, this is the first study that explores the potential for using the knowledge of the TCP variant to significantly boost the accuracy of passive OS fingerprinting.

• Rauniyar, Ashish; Engelstad, Paal E. & Østerbø, Olav Norvald (2020). Ergodic Sum Capacity Analysis of NOMA‐SWIPT Enabled IoT Relay Systems. Internet Technology Letters (ITL). 4(2). doi: 10.1002/itl2.218. Full text in Research Archive Show summary

  Instead of using the participating relay node only for relaying the data of the source node, in this letter, we have also considered transmitting the data of the relay node along with the source node data based on time switch-ing (TS), power splitting (PS) relaying and NOMA. Therefore, in this context, we investigate the ergodic sum capacity (ESC) analysis of simultaneous wireless information and power transfer (SWIPT) based non-orthogonal multiple access (NOMA) for the Internet of Things (IoT) relay systems. Furthermore, we also study the effect of both perfect successive interference cancellation (pSIC) and imperfect SIC (ipSIC) on the proposed system. Analytical expressions for the ESC under both pSIC and ipSIC scenarios are mathematically derived and corroborated through Monte-Carlo simulations. Our result demonstrates that the PS relaying with NOMA is more energy-efficient than TS relaying with NOMAfor the proposed system.

• Upreti, Ramesh; Rauniyar, Ashish; Kunwar, Jeevan; Haugerud, Hårek; Engelstad, Paal E. & Yazidi, Anis (2020). Adaptive Pursuit Learning for Energy‐efficient Target Coverage in Wireless Sensor Networks. Concurrency and Computation: Practice and Experience. ISSN 1532-0626. doi: 10.1002/cpe.5975. Full text in Research Archive Show summary

  With the proliferation of technologies such as wireless sensor networks (WSNs) and the Internet of things (IoT), we are moving towards the era of automation without any human intervention. Sensors are the principal components of the WSNs that bring the idea of IoT into reality. Over the last decade, WSNs are being used in many application fields such as target coverage, battlefield surveillance, home security, health care monitoring, and so on. However, the energy efficiency of the sensor nodes in WSN remains a challenging issue due to the use of a small battery. Moreover, replacing the batteries of the sensor nodes deployed in a hostile environment frequently is not a feasible option. Therefore, intelligent scheduling of the sensor nodes for optimizing its energy-efficient operation and thereby extending the lifetime of WSN has received a lot of research attention lately. In particular, this article investigates extending the lifetime of the WSN in the context of target coverage problems. To tackle this problem, we propose a scheduling technique for WSN based on a novel concept within the theory of learn-ing automata (LA) called pursuit LA. Each sensor node in the WSN is equipped with anLA so that it can autonomously select its proper state, that is, either sleep or active, with an aim to cover all targets with the lowest energy cost possible. Our comprehensive experimental testing of the proposed algorithm not only verifies the efficiency of our algorithm, but it also demonstrates its ability to yield a near-optimal solution. The results are promising, given the low computational footprint of the algorithm.

• Rauniyar, Ashish; Engelstad, Paal E. & Østerbø, Olav Norvald (2020). Ergodic Capacity Performance of D2D IoT Relay NOMA-SWIPT Systems with Direct Links. In Herencsar, Norbert & Benedetto, Francesco (Ed.), 2020 43rd International Conference on Telecommunications and Signal Processing (TSP). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781728163772. p. 117–123. doi: 10.1109/TSP49548.2020.9163552. Full text in Research Archive Show summary

  We investigate the Ergodic capacity (EC) performance of device-to-device (D2D) Internet of Things (IoT) relay non-orthogonal multiple access (NOMA)- simultaneous wireless information and power transfer (SWIPT) systems where the relayed communication is supported with direct link. A two-user case is considered in which a base station transmits symbols to two NOMA users, and the energy harvesting (EH) based relay node via a direct link. The EH based relay node harvests the energy from the BS's signal and again transmits a superimposed composite NOMA signal intended for the user with poor channel condition and for its D2D user to offload its data traffic. A D2D user offloading is considered to further enhance the spectral efficiency of the system. We derive the analytical expressions for the EC of each of the user and the ergodic sum capacity (ESC) of the system and validate them with simulation results. In such settings, our results demonstrate that the EC of a node and the ESC of the system can be improved through the maximal ratio combining (MRC) scheme compared to a system with single signal decoding scheme. Our results also indicate that the overall ESC of the NOMA-SWIPT system can be improved by having a direct link for a user in the system.

• Rauniyar, Ashish; Engelstad, Paal E. & Østerbø, Olav Norvald (2020). Exploiting SWIPT for IoT NOMA-based Diamond Relay Networks. In Mühlhäuser, Max; Polyzos, George C.; Michahelles, Florian; Guinea, Alejandro Sánchez & Wang, Lin (Ed.), MobiQuitous '20: Computing, Networking and Services, Virtual Event / Darmstadt, Germany, December 7-9, 2020. ACM 2020. ACM Digital Library. ISSN 9781450388405. p. 137–146. doi: 10.1145/3448891.3448931. Full text in Research Archive Show summary

  Combining non-orthogonal multiple access (NOMA) with cooperative relaying networks has been considered as a viable approach for increasing the spectral efficiency and capacity enhancement of the system. With the exponential growth of the Internet of Things (IoT), integrating simultaneous wireless information and power transfer (SWIPT) with NOMA and diamond relaying networks can increase capacity and energy efficiency at the same time. Moreover, a diamond network topology has been established as a standard cooperative relaying model in the 3rd Generation Partnership Project (3GPP). Therefore, in this paper, we investigate SWIPT for IoT NOMA-based diamond relay networks wherein a source node transmits two symbols to a destination node through two energy harvesting (EH) based decode-and-forward relay nodes using the principle of downlink NOMA in the first time slot. The two EH based relay nodes harvest the energy from the source’s signal and then transmits the decoded symbol to the destination node using the uplink NOMA protocol in the next time slot. Analytical expressions for the achievable rate of the symbols and the achievable rate of the considered system are derived and verified with the simulation results. Moreover, an asymptotic closed-form solution for the achievable rate is also provided for mathematical tractability and simplified analysis. Our results demonstrate the effect of energy harvesting parameters and the effectiveness of the considered system over the similar system model using conventional orthogonal multiple access.

• Hagos, Desta Haileselassie; Yazidi, Anis; Kure, Øivind & Engelstad, Paal (2020). A Machine Learning-based Tool for Passive OS Fingerprinting with TCP Variant as a Novel Feature. IEEE Internet of Things Journal. ISSN 2327-4662. 8(5). doi: 10.1109/JIOT.2020.3024293. Full text in Research Archive Show summary

  With the emergence of Internet of Things (IoT), securing and managing large, complex enterprise network infrastructure requires capturing and analyzing network traffic traces in real-time. An accurate passive Operating System (OS) fingerprinting plays a critical role in effective network management and cybersecurity protection. Passive fingerprinting doesn’t send probes that introduce extra load to the network and hence it has a clear advantage over active fingerprinting since it also reduces the risk of triggering false alarms. This paper proposes and evaluates an advanced classification approach to passive OS fingerprinting by leveraging state-of-the-art classical machine learning and deep learning techniques. Our controlled experiments on benchmark data, emulated and realistic traffic is performed using two approaches. Through an Oracle-based machine learning approach, we found that the underlying TCP variant is an important feature for predicting the remote OS. Based on this observation, we develop a sophisticated tool for OS fingerprinting that first predicts the TCP flavor using passive traffic traces and then uses this prediction as an input feature for another machine learning algorithm for predicting the remote OS from passive measurements. This paper takes the passive fingerprinting problem one step further by introducing the underlying predicted TCP variant as a distinguishing feature. In terms of accuracy, we empirically demonstrate that accurately predicting the TCP variant has the potential to boost the evaluation performance from 84% to 94% on average across all our validation scenarios and across different types of traffic sources. We also demonstrate a practical example of this potential, by increasing the performance to 91.3% and 95.22% on average using a tool for loss-based and a combination of loss and delay-based TCP variant prediction in an emulated setting. To the best of our knowledge, this is the first study that explores the potential for using the knowledge of the TCP variant to significantly boost the accuracy of passive OS fingerprinting.

• Rauniyar, Ashish; Engelstad, Paal E. & Østerbø, Olav Norvald (2020). Ergodic Capacity Performance of NOMA-SWIPT Aided IoT Relay Systems with Direct Link. In Huang, Longbo; Koutsopoulos, Iordanis & Subramanian, Vijay (Ed.), 2020 18th International Symposium on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WiOpt). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9783903176294. p. 222–229. Full text in Research Archive Show summary

  For the delay-tolerant transmission mode, ergodic capacity (EC) is an appropriate measure for the performance analysis of the system. In this paper, we investigate the EC and ergodic sum capacity (ESC) performance of cooperative non-orthogonal multiple access (NOMA) simultaneous wireless information and power transfer (SWIPT) aided Internet of Things (IoT) relay systems with direct links over the Rayleigh fading channels. To the best of our knowledge, there is no published literature that investigates the EC and ESC of the NOMA-SWIPT aided IoT relay systems with the direct link, in which one source or base station (BS) transmits symbol to two destination nodes through the direct link and with the help of EH based relay node. Specifically, we study the time switching (TS), and power splitting (PS) relaying architecture for increasing the spectral and energy-efficiency of the considered system. Analytical expressions for the EC and the ESC are mathematically derived and validated by the simulation results. Our results not only provide a thorough comparison of the TS and PS relaying EH architecture for the considered system model, but it also demonstrates that the ESC performance could be significantly improved through the optimal choice of the power splitting ε factor for PS relaying with NOMA compared to TS relaying with NOMA.

• Rauniyar, Ashish; Yazidi, Anis; Engelstad, Paal E. & Østerbø, Olav Norvald (2020). A Reinforcement Learning based Game Theoretic Approach for Distributed Power Control in Downlink NOMA. In Avresky, Dimiter R. (Eds.), 2020 IEEE 19th International Symposium on Network Computing and Applications (NCA). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781728183275. doi: 10.1109/NCA51143.2020.9306737. Full text in Research Archive Show summary

  Optimal power allocation problem in wireless networks is known to be usually a complex optimization problem. In this paper, we present a simple and energy-efficient distributed power control in downlink Non-Orthogonal Multiple Access (NOMA) using a Reinforcement Learning (RL) based game theoretical approach. A scenario consisting of multiple Base Stations (BSs) serving their respective Near User(s) (NU) and Far User(s) (FU) is considered. The aim of the game is to optimize the achievable rate fairness of the BSs in a distributed manner by appropriately choosing the power levels of the BSs using trials and errors. By resorting to a subtle utility choice based on the concept of marginal price costing where a BS needs to pay a virtual tax offsetting the result of the interference its presence causes for the other BS, we design a potential game that meets the latter objective. As RL scheme, we adopt Learning Automata (LA) due to its simplicity and computational efficiency and derive analytical results showing the optimality and convergence of the game to a Nash Equilibrium (NE). Numerical results not only demonstrate the convergence of the proposed algorithm to a desirable equilibrium maximizing the fairness, but they also demonstrate the correctness of the proposal followed by thorough comparison with random and heuristic approaches.

• Rauniyar, Ashish; Engelstad, Paal E. & Østerbø, Olav Norvald (2020). An Adaptive User Pairing Strategy for Uplink Non-Orthogonal Multiple Access. In Darwazeh, Izzat & Aghvami, Hamid (Ed.), 2020 IEEE 31st Annual International Symposium on Personal, Indoor and Mobile Radio Communications. IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781728144917. doi: 10.1109/PIMRC48278.2020.9217383. Full text in Research Archive Show summary

  Non-orthogonal multiple access (NOMA) is consid-ered an important candidate for the next-generation cellular networks to address the issue of exponentially growing data traffic from technologies like the Internet of Things (IoT). Efficient user pairing between multiple users is needed to enhance the capacity of uplink NOMA systems, and this paper investigates a user pairing strategy that enhances the capacities of a cell for uplink NOMA. In general, NOMA separates nodes into two equally-sized groups where half of nodes with the best channel conditions are allocated to one group, while the other half of nodes are allocated to the other group. Then, NOMA pairs each node in one group with another node in the other group. First, we investigate an efficient pairing scheme that maximizes the sum capacity for the uplink NOMA system, with an even number of nodes, i.e. when there is an equal number of cellular users in each group. An additional problem arises when there is an odd number of nodes, and the number of users in one group is thus one larger than in the other, leaving one node potentially unpaired. Instead of using regular OMA for the unpaired node, we propose an efficient adaptive user pairing strategy where the unpaired node is allowed to pair with one of the other pairs, forming a cluster of three nodes. The objective of our proposed scheme is to find the best pair of nodes to serve the unpaired node in order to increase the overall sum capacity. Moreover, the impact of perfect and imperfect successive interference cancellation (SIC) is studied on the user pairing strategy for uplink NOMA systems. Numerical results verify the effectiveness of our proposed adaptive user pairing strategy over the user pairing scheme utilizing conventional NOMA and OMA schemes.

• Rauniyar, Ashish; Engelstad, Paal E. & Østerbø, Olav Norvald (2020). Capacity Enhancement of NOMA‐SWIPT IoT Relay System with Direct Links over Rayleigh Fading Channels. Transactions on Emerging Telecommunications Technologies. ISSN 1124-318X. doi: 10.1002/ett.3913. Full text in Research Archive Show summary

  It is known that when the direct links between the base station (BS) and the users exist and are nonnegligible, consolidating direct links could significantly enhance the performance of the cooperative relaying systems. Therefore, taking the impact of direct link into account, in this article, we investigate the capacity enhancement of the nonorthogonal multiple access (NOMA) with simultaneous wireless information and power transfer (SWIPT) for the Internet of Things (IoT) relay systems over the Rayleigh fading channels. Specifically, for the considered NOMA‐SWIPT system with direct links, we study a time‐switching energy harvesting (EH) architecture in which a BS transmits two symbols to the two users through the direct links and via an EH‐based relay node. On the receiving user node, we employ maximum ratio combining (MRC) to show the capacity enhancement of the considered system. Moreover, analytical expressions for the Ergodic capacity and Ergodic Sum Capacity are mathematically derived for the MRC and single signal decoding scheme, and the analytical expressions are corroborated with the Monte‐Carlo simulation results. This not only reveals the effect of different EH parameters on the system performance but it also demonstrates the capacity enhancement of the considered NOMA‐SWIPT system compared to a similar system without direct links and to conventional OMA schemes.

• Aryan, Ramtin; Yazidi, Anis; Kure, Øivind & Engelstad, Paal E. (2020). A parallel approach for detecting OpenFlow rule anomalies based on a general formalism. Concurrency and Computation: Practice and Experience. ISSN 1532-0626. doi: 10.1002/cpe.5907. Full text in Research Archive Show summary

  As the policies of a software-defined networking (SDN) network can be updateddynamically and often at a high pace, conflicts between policies can easily occur. Dueto the large number of switches and heterogeneous policies within a typical SDN net-work, detecting those conflicts is a laborious and challenging task. This article presentsthree main contributions. First, we devise an offline method for detecting unmatchedOpenFlow rules, that is, rules that are never fired. In our taxonomy such anomaliescan stem from either invalid or irrelevant unmatched rules. Second, we introduce anew set of definitions for the intraanomalies between rules in the same table, whichmight occur when using the multiaction feature of an OpenFlow rule. Third, our detec-tion method has been enhanced to support parallel execution, which makes it a viablesolution for troubleshooting large-scale networks. We provide some comprehensiveexperimental results based on both synthetic and real-life setup the synthetic set upis designed in such a way that the rule matching takes place in the last rules of theswitch and thus putting more stress on the rule detection process. The parallel methodis shown to outperform the single-threaded checking method by order of magnitudeup to 21.

• Rauniyar, Ashish; Kunwar, Jeevan; Haugerud, Hårek; Yazidi, Anis & Engelstad, Paal E. (2019). Energy Efficient Target Coverage in Wireless Sensor Networks Using Adaptive Learning. In Jemili, Imem & Mosbah, Mohamed (Ed.), Distributed Computing for Emerging Smart Networks. Springer Nature. ISSN 9783030401313. p. 133–147. doi: 10.1007/978-3-030-40131-3_9. Full text in Research Archive Show summary

  Over the past few years, innovation in the development of Wireless Sensor Networks (WSNs) has evolved rapidly. WSNs are being used in many application fields such as target coverage, battlefield surveillance, home security, health care supervision, and many more. However, power usage in WSNs remains a challenging issue due to the low capacity of batteries and the difficulty of replacing or charging them, especially in harsh environments. Therefore, this has led to the development of various architectures and algorithms to deal with optimizing the energy usage of WSNs. In particular, extending the lifetime of the WSNin the context of target coverage problems by resorting to intelligent scheduling has received a lot of research attention. In this paper, we propose a scheduling technique for WSN based on a novel concept within the theory of Learning Automata (LA) called pursuit LA. Each sensor node in the WSN is equipped with an LA so that it can autonomously select its proper state, i.e., either sleep or active with the aim to coverall targets with the lowest energy cost. Through comprehensive experimental testing, we verify the efficiency of our algorithm and its ability to yield a near-optimal solution. The results are promising, given the low computational footprint of the algorithm.

• Aryan, Ramtin; Brattensborg, Frode Henriksen; Yazidi, Anis & Engelstad, Paal (2019). Checking the OpenFlow Rule Installation and Operational Verification. In Andersson, Karl; Tan, Hwee-Pink & Oteafy, Sharief (Ed.), 44th Annual IEEE Conference on Local Computer Networks (LCN 2019). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781728110288. p. 250–253. doi: 10.1109/LCN44214.2019.8990808. Full text in Research Archive
• Hagos, Desta Haileselassie; Engelstad, Paal E. & Yazidi, Anis (2019). Classification of Delay-based TCP Algorithms From Passive Traffic Measurements. In Gkoulalas-Divanis, Aris; Marchetti, Mirco & Avresky, Dimiter R. (Ed.), 2019 IEEE 18th International Symposium on Network Computing and Applications (NCA). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781728125220. doi: 10.1109/NCA.2019.8935063. Full text in Research Archive Show summary

  Identifying the underlying TCP variant from passive measurements is important for several reasons, e.g., exploring security ramifications, traffic engineering in the Internet, etc. In this paper, we are interested in investigating the delay characteristics of widely used TCP algorithms that exploit queueing delay as a congestion signal. Hence, we present an effective TCP variant identification methodology from traffic measured passively by analyzing β, the multiplicative back-off factor to decrease the cwnd on a loss event, and the queueing delay values. We address how the β as a function of queueing delay varies and how the TCP variants of delay-based congestion control algorithms can be predicted both from passively measured traffic and real measurements over the Internet. We further employ a novel non-stationary time series approach from a stochastic nonparametric perspective using a two-sided Kolmogorov–Smirnov test to classify delay-based TCP algorithms based on the α, the rate at which a TCP sender’s side cwnd grows per window of acknowledged packets, parameter. Through extensive experiments on emulated and realistic scenarios, we demonstrate that the data-driven classification techniques based on probabilistic models and Bayesian inference for optimal identification of the underlying delay-based TCP congestion algorithms give promising results. We show that our method can also be applied equally well to loss-based TCP variants.

• Rauniyar, Ashish; Engelstad, Paal E. & Østerbø, Olav Norvald (2019). Performance Analysis of RF Energy Harvesting and Information Transmission based on NOMA with Interfering Signal for IoT Relay Systems. IEEE Sensors Journal. ISSN 1530-437X. 19(17), p. 7668–7682. doi: 10.1109/JSEN.2019.2914796. Full text in Research Archive Show summary

  Owing to the exponential proliferation of the Internet of Things (IoT), it is anticipated that the number of small IoT devices will grow expeditiously over the next few years. These billions of small IoT sensor and devices will consume a huge power for data transmission. In this fashion, radio frequency (RF) energy harvesting has been contemplated as an appealing solution to the architecture of long-term and self-sustainable next-generation wireless systems such as IoT network. However, in the practical environment, such as IoT networks or systems are subjected to external interference factors which often results in the loss of the system rate. In this paper, different from generic RF EH system, where only a source node data is relayed through intermediate EH relaying node, we have considered to transmit the data of IoT relay node along with source node data using non-orthogonal multiple access (NOMA) protocol in the presence of an interfering signal to their respective destinations. Specifically, in the presence on interfering signal, we study the combination of two popular energy harvesting relaying architectures-time switching (TS) relaying and power splitting (PS) relaying with NOMA protocol for IoT relay systems. Considering the interference from the external entity, we have mathematically derived the outage probability, throughput, and sum-throughput for our proposed system. The extensive simulations are carried out to find out the optimal TS and PS factor that maximizes the sum-throughput of the considered system in the presence of an interfering signal. The analytical results of our system model under consideration are validated by the simulation results, and representative performance comparisons are presented.

• Hagos, Desta Haileselassie; Engelstad, Paal E.; Yazidi, Anis & Griwodz, Carsten (2019). A Deep Learning Approach to Dynamic Passive RTT. In Wang, Yu & Wang, Mea (Ed.), 2019 IEEE 38th International Performance Computing and Communications Conference (IPCCC). IEEE Press. ISSN 9781728110257. p. 1–10. doi: 10.1109/IPCCC47392.2019.8958727. Full text in Research Archive Show summary

  The Round-Trip Time (RTT) is a property of the path between a sender and a receiver communicating with Transmission Control Protocol (TCP) over an IP network and over the public Internet. The end-to-end RTT value influences significantly the dynamics and performance of TCP, which is by far the most used communication protocol. Thus, in communication networks, RTT is an important network performancevariable.Bymeasuringthetrafficatanintermediate node, a network operator or service provider can estimate the RTT and use the estimation to study and troubleshoot the per-connection characteristics and performance. This paper aims at improving the accuracy and timeliness of the RTT estimation, to help network operators improving their analysis. We propose and evaluate a novel deep learning-based model capable of dynamically predicting at real-time the RTT between the sender and receiver with high accuracy based on passive measurements collected at an intermediate node, taking advantage of the commonly used TCP timestamps. We validate extensively our prediction methodology in a controlled experimental testbed and in a realistic scenario on the Google Cloud platform. We show that our model, which is based on classical deep learning algorithms, gives reasonably effective state-of-the-art performance results across multiple TCP congestion control variants. We also show that the model works well for transfer learning. Even though the RTT prediction model was trained on an emulated network, it performs well also when applied to a realistic scenario setting, as demonstrated in our experimental evaluation.

• Rauniyar, Ashish; Engelstad, Paal E. & Østerbø, Olav Norvald (2018). RF Energy Harvesting and Information Transmission in IoT Relay Systems based on Time Switching and NOMA. In Gregory, Mark A (Eds.), 2018 28th International Telecommunication Networks and Applications Conference (ITNAC). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781538671771. p. 168–174. doi: 10.1109/ATNAC.2018.8615403. Full text in Research Archive Show summary

  A huge expansion of billions of Internet of Things (IoT) sensor and devices is expected over the next few years which will consume more power. Therefore, energy efficiency is a major concern for the development of fifth generation (5G) wireless systems. In wireless communication systems, energy harvesting (EH) is an emerging paradigm that allows the sensor nodes to recharge themselves through radio frequency (RF) signals directed to them from the source node and then relaying or transmitting the information. Although a myriad of works have been carried out in the literature for EH, the absolute vast majority of those works only consider RF EH at relay node and transmission of source node data successfully to its destination node. Those approaches do not consider the data transmission of the relay node that may be an IoT node which needs to transmit its data along with the source node data to their respective destinations. Thus, such approaches are clearly ineffective for energy efficient IoT relay systems. In this paper, we rather focus on RF EH and information transmission based on time switching (TS) relaying and non-orthogonal multiple access (NOMA) for IoT relay systems. A source node information data is relayed through power constrained IoT relay node that first harvests the energy from source node RF signal using TS protocol and then transmits source node information along with its information using NOMA protocol. We have mathematically derived analytical expressions for outage probability, throughput and sum-throughput for our proposed system. We have also formulated an algorithm to find out optimal TS factor that maximizes the sum-throughput for our proposed system. Our proposed system analytical results are validated by the simulation results.

• Aryan, Ramtin; Yazidi, Anis & Engelstad, Paal E. (2018). An Incremental Approach for Swift OpenFlow Anomaly Detection. Conference on Local Computer Networks. ISSN 0742-1303. doi: 10.1109/LCN.2018.8638226. Full text in Research Archive Show summary

  Software Defined Networking (SDN) is designed for dynamic policy update where frequent changes are pushed to the forwarding devices. Different offline approaches for detecting misconfiguration anomalies in SDN by taking a snapshot of the state of the network have been developed in the literature. However, the detection process is time-consuming and unfeasible in the case of frequent changes to the OpenFlow tables as well in big size networks containing a large number of rules. This paper presents an incremental method for detecting potential anomalies in an online manner, i.e., after one or multiple simultaneous updates in the SDN policy. Whenever the OpenFlow tables are dynamically changed, a static approach that rechecks the whole policy is unnecessarily redundant in a sense that most of the policy remains intact. Hence the need for incremental verification method to reduce this overhead, and only the subset of the policy that is affected by the update is checked. Two different solutions are proposed based on whether the policy modifications take place in the ingress switches or in the middle switches. We provide some comprehensive experiments to demonstrate the detection performance for the case of single or multiple simultaneous changes in forwarding devices. The experiment results show that the incremental method is drastically faster than the static parallel approach, with a factor up to about 450 times in some cases.

• Rauniyar, Ashish; Engelstad, Paal E. & Østerbø, Olav Norvald (2018). RF energy harvesting and information transmission based on NOMA for wireless powered IoT relay systems. Sensors. 18:3254(10), p. 1–22. doi: 10.3390/s18103254. Full text in Research Archive Show summary

  Amidst the rapid development of the fifth generation (5G) networks, Internet of Things (IoT) is considered as one of the most important part of 5G next generation networks as it can support massive object communications. These massive object communications in the context of IoT is expected to consume a huge power. Furthermore, IoT sensors or devices are rather power constrained and are mostly battery operated. Therefore, energy efficiency of such network of IoT devices is a major concern. On the other hand, energy harvesting (EH) is an emerging paradigm that allows the wireless nodes to recharge themselves through radio frequency (RF) signals directed to them from the source node and then relaying or transmitting the information. Although a myriad of works have been carried out in the literature for EH, the vast majority of those works only consider RF EH at the relay node and successfully transmitting the source node data. Those approaches do not consider the data transmission of the relay node that may be an energy deprived IoT node which needs to transmit its own data along with the source node data to their respective destination nodes. Therefore, in this paper, we envisioned a RF EH and information transmission system based on time switching (TS) relaying, power splitting (PS) relaying and non-orthogonal multiple access (NOMA) which is suitable for wireless powered IoT relay systems. A source node information data is relayed through power constrained IoT relay node IoTR that first harvests the energy from source node RF signal using either TS and PS relaying protocol and then transmits the source node information along with its information using NOMA protocol to the respective destination nodes. Considering NOMA as a transmission protocol, we have mathematically derived analytical expressions for TS and PS relaying protocol for our proposed system. We have also formulated an algorithm to find out optimal TS and PS factor that maximizes the sum-throughput for our proposed system. Our proposed system analytical results for TS and PS protocol are validated by the simulation results.

• Hagos, Desta Haileselassie; Engelstad, Paal E.; Yazidi, Anis & Kure, Oivind (2018). A machine learning approach to TCP state monitoring from passive measurements. IFIP Wireless Days. ISSN 2156-9711. 2018-April, p. 164–171. doi: 10.1109/WD.2018.8361713. Full text in Research Archive
• Hagos, Desta Haileselassie; Engelstad, Paal E.; Yazidi, Anis & Kure, Øivind (2018). Towards a Robust and Scalable TCP Flavors Prediction Model from Passive Traffic. In Abdelzaher, Tarek F. (Eds.), Proceedings of 27th International Conference on Computer Communication Networks (ICCCN), 2018. Hangzhou, China. IEEE Sarnoff Symposium. ISSN 9781538651568. p. 1–11. doi: 10.1109/ICCCN.2018.8487396. Full text in Research Archive
• Hagos, Desta Haileselassie; Engelstad, Paal E.; Yazidi, Anis & Kure, Øivind (2018). Recurrent neural network-based prediction of TCP transmission states from passive measurements, 2018 IEEE 17th International Symposium on Network Computing and Applications (NCA). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781538676592. p. 1–10. doi: 10.1109/NCA.2018.8548064. Full text in Research Archive Show summary

  Long Short-Term Memory (LSTM) neural networks are a state-of-the-art techniques when it comes to sequence learning and time series prediction models. In this paper, we have used LSTM-based Recurrent Neural Networks (RNN) for building a generic prediction model for Transmission Control Protocol (TCP) connection characteristics from passive measurements. To the best of our knowledge, this is the first work that attempts to apply LSTM for demonstrating how a network operator can identify the most important system-wide TCP per-connection states of a TCP client that determine a network condition (e.g., cwnd) from passive traffic measured at an intermediate node of the network without having access to the sender. We found out that LSTM learners outperform the state-of-the-art classical machine learning prediction models. Through an extensive experimental evaluation on multiple scenarios, we demonstrate the scalability and robustness of our approach and its potential for monitoring TCP transmission states related to network congestion from passive measurements. Our results based on emulated and realistic settings suggest that Deep Learning is a promising tool for monitoring system-wide TCP states from passive measurements and we believe that the methodology presented in our paper may strengthen future research work in the computer networking community.

• Kure, Endre Hegland; Engelstad, Paal E.; Maharjan, Sabita; Gjessing, Stein & Zhang, Yan (2018). Distributed Uplink Offloading for IoT in 5G Heterogeneous Networks under Private Information Constraints. IEEE Internet of Things Journal. ISSN 2327-4662. 6(4), p. 6151–6164. doi: 10.1109/JIOT.2018.2886703. Full text in Research Archive
• Kure, Endre Hegland; Engelstad, Paal E.; Maharjan, Sabita; Gjessing, Stein; Zhang, Xing & Zhang, Yan (2018). Energy Usage Forecasting for LTE: A Network-Wide Traffic Measurements Study. In NN, N (Eds.), 2018 IEEE GLOBAL COMMUNICATIONS CONFERENCE (GLOBECOM). IEEE Press. ISSN 9781538647271. doi: 10.1109/GLOCOMW.2018.8644423. Full text in Research Archive Show summary

  Energy usage in LTE base stations are driven by spectral efficiency and traffic. To predict the energy usage these parameters must be forecasted. In this work we analyse hourly measurements collected from more than 12000 base station cells spread across more than 3700 base stations over the course of one month. We show that the two parameters are very weakly correlated, and therefore we investigated them separately. Further, we evaluated the possible gains for advanced prediction methods using a large scale search for individually fitted time series (SARIMA models) for each base station. In total, we examined and evaluated approximately 31000 time series models from an identified group of 4 million potential models. We found that the spectral efficiency measurements can be represented fairly well with time series models, with only an average 6.5% relative error. The time series models have to be individually adapted for each base station as the unsupervised clustering showed that each cluster's members have a wide variety of best fitted models. However, for traffic the time series models have relative high prediction errors, and we believe there is a potential for new methods to improve the forecasts.

• Hagos, Desta Haileselassie; Engelstad, Paal E.; Yazidi, Anis & Kure, Øivind (2018). General TCP State Inference Model From Passive Measurements Using Machine Learning Techniques. IEEE Access. 6, p. 28372–28387. doi: 10.1109/ACCESS.2018.2833107. Full text in Research Archive Show summary

  Many applications in the Internet use the reliable end-to-end Transmission Control Protocol (TCP) as a transport protocol due to practical considerations. There are many different TCP variants widely in use, and each variant uses a specific congestion control algorithm to avoid congestion, while also attempting to share the underlying network capacity equally among the competing users. This paper shows how an intermediate node (e.g., a network operator) can identify the transmission state of the TCP client associated with a TCP flow by passively monitoring the TCP traffic. Here, we present a robust, scalable and generic machine learning-based method which may be of interest for network operators that experimentally infers Congestion Window (cwnd) and the underlying variant of loss-based TCP algorithms within a flow from passive traffic measurements collected at an intermediate node. The method can also be extended to predict other TCP transmission states of the client. We believe that our study also has a potential benefit and opportunity for researchers and scientists in the networking community from both academia and industry who want to assess the characteristics of TCP transmission states related to network congestion. We validate the robustness and scalability approach of our prediction model through a large number of controlled experiments. It turns out, surprisingly enough, that the learned prediction model performs reasonably well by leveraging knowledge from the emulated network when it is applied on a real-life scenario setting. Thus, our prediction model is general bearing similarity to the concept of transfer learning in the machine learning community. The accuracy of our experimental results both in an emulated network, realistic and combined scenario settings and across multiple TCP congestion control variants demonstrate that our model is reasonably effective and has considerable potential.

• Rauniyar, Ashish; Engelstad, Paal E. & Østerbø, Olav Norvald (2018). RF Energy Harvesting and Information Transmission based on Power Splitting and NOMA for IoT Relay Systems, 2018 IEEE 17th International Symposium on Network Computing and Applications (NCA). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781538676592. doi: 10.1109/NCA.2018.8548068. Full text in Research Archive Show summary

  Due to the proliferation of the Internet of Things (loT), an unprecedented expansion of wireless devices and communication among these devices for data transmission is expected over the next few years. IoT devices are rather power constrained and are mostly battery operated devices. Therefore, energy efficiency of such IoT devices or systems is a major concern. In this paper, radio frequency (RF) energy harvesting and information transmission based on Non-orthogonal multiple access (NOMA) protocol is considered for wireless powered IoT relay systems. A source node needs the help of power constrained IoT relay node IoTR to transmit its data. IoTR node first harvests the energy from the RF signal of the source node and process the information in the first stage and then transmits the source node information data along with its data using NOMA protocol in the next stage. Unlike most of the previous works in this domain, this paper also considers the data transmission of the IoTR node along with the source node to its intended destination nodes. Specifically, considering the energy constrained nature of IoT nodes and devices, we have used the combination of power splitting (PS) and NOMA protocol and have mathematically derived the outage probability and sum-throughput for the proposed system. We have also formulated an algorithm to find out the optimal power splitting factor that maximizes the sum-throughput of the proposed system. Our proposed system analytical results are validated by the simulation results.

• Rauniyar, Ashish; Engelstad, Paal E. & Moen, Hans Jonas Fossum (2018). A New Distributed Localization Algorithm Using Social Learning Based Particle Swarm Optimization for Internet of Things. IEEE Vehicular Technology Conference (VTC) Proceedings. ISSN 1550-2252. doi: 10.1109/VTCSpring.2018.8417665. Full text in Research Archive Show summary

  Emerging applications in the Internet of Things (IoT) will depend on the accurate location of thousands of deployed sensors. However, accurate localization of deployed sensors nodes is a classical optimization problem which falls under NP-hard class of problems. Therefore in this work, we propose a new distributed localization algorithm using social learning based particle swarm optimization (SL-PSO) for IoT. With SL-PSO algorithm, we aim to do precise localization of deployed sensor nodes and reduce the computational complexity which will further enhance the lifetime of these resource-constrained IoT sensor nodes. Extensive simulations are carried out to show the effective performance of the SL-PSO algorithm in accurate localization. Experimental results depict that SL-PSO algorithm can not only increase convergence rate but also significantly reduce average localization error compared to traditional particle swarm optimization (PSO) and its other variants.

• Hammer, Hugo Lewi; Yazidi, Anis; Bai, Aleksander & Engelstad, Paal E. (2017). Improving Classification of Tweets Using Linguistic Information from a Large External Corpus. In Maglaras, Leandros (Eds.), Industrial Networks and Intelligent Systems. Springer. ISSN 9783319525686. p. 122–134. doi: 10.1007/978-3-319-52569-3_11. Full text in Research Archive Show summary

  The bag of words representation of documents is often unsat- isfactory as it ignores relationships between important terms that do not co-occur literally. Improvements might be achieved by expanding the vocabulary with other relevant word, like synonyms. In this paper we use word-word co-occurence information from a large corpus to expand the vocabulary of another corpus consisting of tweets. Several different methods on how to include the co-occurence information are constructed and tested out on the classification of real twitter data. Our results show that we are able to reduce the number of erroneous classifications by 14% using co-occurence information.

• Aryan, Ramtin; Yazidi, Anis; Engelstad, Paal E. & Kure, Øivind (2017). A General Formalism for Defining and Detecting OpenFlow Rule Anomalies. In Jens, Tölle (Eds.), 42nd IEEE Conference on Local Computer Networks. IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781509065233. p. 426–434. doi: 10.1109/LCN.2017.94. Full text in Research Archive Show summary

  SDN network's policies are updated dynamically at a high pace. As a result, conflicts between policies are prone to occur. Due to the large number of switches and heterogeneous policies within a typical SDN network, detecting those conflicts is a laborious and challenging task. This paper presents two-fold contributions. First, we devise an offline method for detecting unmatched OpenFlow rules, i.e., those rules that are never fired. At the heart of our scheme is a formal approach for predicting the packet's path inside a SDN network. In this perspective, we proffer the taxonomy: invalid and irrelevant anomalies for the unmatched rules. Second, we introduce a new set of definitions for the intra-anomalies, which might occur when using the OpenFlow rule's multi-action feature. We provide some comprehensive experimental results that show the feasibility of our approach and its ability to scale within large SDN network.

• Do, Van Thuan; Engelstad, Paal E.; Feng, Boning & Do, van Thanh (2017). Detection of DNS tunneling in mobile networks using machine learning. Lecture Notes in Electrical Engineering. ISSN 1876-1100. 424, p. 221–230. doi: 10.1007/978-981-10-4154-9_26. Full text in Research Archive Show summary

  Lately, costly and threatening DNS tunnels on the mobile networks bypassing the mobile operator’s Policy and Charging Enforcement Function (PCEF), has shown the vulnerability of the mobile networks caused by the Domain Name System (DNS) which calls for protection solutions. Unfortunately there is currently no really adequate solution. This paper proposes to use machine learning techniques in the detection and mitigation of a DNS tunneling in mobile networks. Two machine learning techniques, namely One Class Support Vector Machine (OCSVM) and K-Means are experimented and the results prove that machine learning techniques could yield quite efficient detection solutions. The paper starts with a comprehensive introduction to DNS tunneling in mobile networks. Next the challenges in DNS tunneling detections are reviewed. The main part of the paper is the description of proposed DNS tunneling detection using machine learning.

• Do, van Thanh; Engelstad, Paal E.; Feng, Boning & Do, Van Thuan (2017). A near real time SMS Grey Traffic detection. In Mezhuyev, Vitaliy & Benedicenti, Luigi (Ed.), Proceedings of 2017 6th International Conference on Software and Computer Applications (ICSCA 2017) February 26-28, 2017 Bangkok, Thailand. Association for Computing Machinery (ACM). ISSN 9781450348577. p. 244–249. doi: 10.1145/3056662.3056687. Full text in Research Archive Show summary

  Lately, mobile operators experience threats from SMS grey routes which are used by fraudsters to evade SMS fees and to deny them millions in revenues. But more serious are the threats to the user’s security and privacy and consequently the operator’s reputation. Therefore, it is crucial for operators to have adequate solutions to protect both their network and their customers against this kind of fraud. Unfortunately, so far there is no sufficiently efficient countermeasure against grey routes. This paper proposes a near real time SMS grey traffic detection which makes use of Counting Bloom Filters combined with blacklist and whitelist to detect SMS grey traffic on the fly and to block them. The proposed detection has been implemented and proved to be quite efficient. The paper provides also comprehensive explanation of SMS grey routes and the challenges in their detection. The implementation and verification are also described thoroughly.

• Kongsgård, Kyrre Wahl; Nordbotten, Nils Agne; Mancini, Federico; Haakseth, Raymond & Engelstad, Paal E. (2017). Data Leakage Prevention for Secure Cross-Domain Information Exchange. IEEE Communications Magazine. ISSN 0163-6804. 55(10), p. 37–43. doi: 10.1109/MCOM.2017.1700235. Full text in Research Archive Show summary

  Cross-domain information exchange is an increasingly important capability for conducting efficient and secure operations, both within coalitions and within single nations. A data guard is a common cross-domain sharing solution that inspects the security labels of exported data objects and validates that they are such that they can be released according to policy. While we see that guard solutions can be implemented with high assurance, we find that obtaining an equivalent level of assurance in the correctness of the security labels easily becomes a hard problem in practical scenarios. Thus, a weakness of the guard-based solution is that there is often limited assurance in the correctness of the security labels. To mitigate this, guards make use of content checkers such as dirty word lists as a means of detecting mislabeled data. To improve the overall security of such cross-domain solutions, we investigate more advanced content checkers based on the use of machine learning. Instead of relying on manually specified dirty word lists, we can build data-driven methods that automatically infer the words associated with classified content. However, care must be taken when constructing and deploying these methods as naive implementations are vulnerable to manipulation attacks. In order to provide a better context for performing classification, we monitor the incoming information flow and use the audit trail to construct controlled environments. The usefulness of this deployment scheme is demonstrated using a real collection of classified and unclassified documents. © 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

• Hagos, Desta Haileselassie; Yazidi, Anis; Kure, Øivind & Engelstad, Paal E. (2017). Enhancing security attacks analysis using regularized machine learning techniques. Advanced Information Networking and Applications. ISSN 1550-445X. p. 909–918. doi: 10.1109/AINA.2017.19. Full text in Research Archive Show summary

  With the increasing threats of security attacks, Machine learning (ML) has become a popular technique to detect those attacks. However, most of the ML approaches are black-box methods and their inner-workings are difficult to understand by human beings. In the case of network security, understanding the dynamics behind the classification model is a crucial element towards creating safe and human-friendly systems. In this article, we investigate the most important features in identifying well-known security attacks by using Support Vector Machines (SVMs) and l1-regularized method with Least Absolute Shrinkage and Selection Operator (LASSO) for robust regression both to binary and multiclass attack classification. SVMs are one of the standards of ML classification techniques that give a reasonably good performance but with some drawbacks in terms of interpretability. On the other hand, LASSO is a regularized regression method often performing comparably well and it has extra compelling advantages of being very easily interpretable. LASSO provides coefficients that contribute how individual features affect the probability of specific security attack classes to occur. Hence, we finally use LASSO in particular for multiclass classification to help us better understand which actual features shared by attacks in a network are the most important ones. To perform our analysis, we use the recent NSL-KDD intrusion detection public dataset where the data are labeled into either anomalous (denial-of-service (DoS), remote-to-local (R2L), user-to-root (U2R) and probe attack classes) or normal. Empirical results of the analysis and computational performance comparison over the competing methods used are also presented and discussed. We believe that the methodology presented in this paper may strengthen a future research in network intrusion detection settings.

• Yazidi, Anis; Hammer, Hugo Lewi; Bai, Aleksander & Engelstad, Paal E. (2016). On the Feasibility of Machine Learning as a Tool for Automatic Security Classification: A Position Paper. In NN, NN (Eds.), 2016 International Conference on Computing, Networking and Communications, ICNC 2016. IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781467385794. p. 6–6. doi: 10.1109/ICCNC.2016.7440588. Full text in Research Archive
• Hammer, Hugo Lewi; Yazidi, Anis; Bai, Aleksander & Engelstad, Paal E. (2016). Improving classification of tweets using word-word co-occurrence information from a large external corpus. In Ossowski, Sascha (Eds.), Proceedings of the 31st Annual ACM Symposium on Applied Computing (SAC '16). Association for Computing Machinery (ACM). ISSN 9781450337397. p. 1174–1177. doi: 10.1145/2851613.2851986. Full text in Research Archive Show summary

  Classifying tweets is an intrinsically hard task as tweets are short messages which makes traditional bags of words based approach ine cient. In fact, bags of words approaches ig- nores relationships between important terms that do not co-occur literally. In this paper we resort to word-word co-occurence informa- tion from a large corpus to expand the vocabulary of another corpus consisting of tweets. Our results show that we are able to reduce the number of erroneous classi cations by 14% using co-occurence information.

• Do, Van Thuan; Engelstad, Paal E.; Feng, Boning & Do, van Thanh (2016). Strengthening mobile network security using machine learning. In Younas, Muhammad; Awan, Irfan; Kryvinska, Natalia; Strauss, Christine & Do, van Thanh (Ed.), Mobile Web and Intelligent Information Systems 13th International Conference, MobiWIS 2016 Vienna, Austria, August 22-24, 2016 Proceedings. Springer. ISSN 9783319442143. p. 173–183. doi: 10.1007/978-3-319-44215-0_14. Full text in Research Archive Show summary

  Lately, several episodes of tapping and tracking of mobile phones in Europe including Norway have been revealed, showing the vulnerabilities of both the mobile network and mobile phones. A better protection of the user’s confidentiality and privacy is urgently required. This paper will present an in-novative mobile network security system using machine learning. The paper will start with a vulnerability and threat analysis of the evolving mobile network, which is a fusion of mobile wireless technologies and Internet technologies, complemented with the Internet of Things. The main part of the paper will concentrate on clarifying how machine learning can help improving mobile network security. The focus will be on elucidating what makes machine learning superior to other techniques. A special case study on the detection of IMSI Catcher, the fake base station that is used in mobile phone tracking and tapping, will be explained.

• Rauniyar, Ashish; Engelstad, Paal E.; Feng, Boning & Do, van Thanh (2016). Crowdsourcing-based Disaster Management using Fog Computing in Internet of Things Paradigm. In Bilof, Randall (Eds.), 2016 IEEE 2nd International Conference on Collaboration and Internet Computing (CIC). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781509046072. p. 490–494. doi: 10.1109/CIC.2016.074. Full text in Research Archive Show summary

  In internet of things (IoT) paradigm, crowdsourcing is the process of obtaining and analyzing information or input to a particular task or project generated by a number of sources such as sensors, mobile devices, vehicles and human. Cloud computing is widely used for the services such as analyzing crowdsourced data and application implementation over the IoT. Nowadays, every country and human are prone to natural and artificial disasters. Early detection about disasters such as earthquakes, fire, storms, and floods can save thousands of people’s life and effective preventive measure can be taken for the public safety. All the crowdsourced data which are providing the information of a certain geographic region are analyzed in a cloud platform. But, by the time the crowdsourced data makes its way to the cloud for analysis, the opportunity to act on it might be gone. Moreover, thousands of people’s life will be lost. Therefore, fog computing is the new and efficient way to analyze such critical crowdsourced IoT data of disasters. In this paper, in order to detect and take necessary steps for public safety during a disaster, we propose a crowdsourcing-based disaster management using fog computing (CDMFC) model in IoT. Further, we also proposed a data offloading mechanism for our CDMFC model to send disaster-related IoT data to the fog even if a direct link to the fog is not available. Our proposed CDMFC model and its data offloading mechanism can detect real-time disasters and disseminate early information for public safety as compared to the conventional cloud computing based disaster management models.

• Engelstad, Paal E.; Hammer, Hugo Lewi; Kongsgård, Kyrre Wahl; Yazidi, Anis; Nordbotten, Nils Agne & Bai, Aleksander (2016). Automatic security classification with Lasso. Lecture Notes in Computer Science (LNCS). ISSN 0302-9743. 9503, p. 399–410. doi: 10.1007/978-3-319-31875-2_33. Full text in Research Archive
• Pham, Hai Ngoc; Zhang, Yan; Skeie, Tor; Engelstad, Paal E. & Eliassen, Frank (2016). Joint energy-efficient cooperative spectrum sensing and power allocation in Cognitive Machine-to-Machine Communications, Proceedings of the 2016 International Wireless Communications and Mobile Computing Conference (IWCMC 2016). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781509003044. doi: 10.1109/IWCMC.2016.7577208. Full text in Research Archive Show summary

  In battery-powered Cognitive Machine-to-Machine Communications (CM2M), the energy consumption, opportunis- tic data access capacity and interference to the licensed system need to be optimized simultaneously. We consider this as joint cooperative spectrum sensing and power allocation, and model this as a constraint multiobjective optimization problem of three objectives. Our model helps to find a Pareto optimal variable set of sensing duration, detection threshold and transmission power for each individual sensor in cooperative spectrum sensing. The evaluation of our model shows that energy consumption, opportunistic data capacity and interference are optimized simultaneously while keeping the total cooperative spectrum sensing error lower than a predefined threshold. Pareto optimal results show that better energy efficiency [bits/joule] makes lower harmful interference to the primary system.

• Kongsgård, Kyrre Wahl; Nordbotten, Nils Agne; Mancini, Federico & Engelstad, Paal E. (2016). Data loss prevention based on text classification in controlled environments. Lecture Notes in Computer Science (LNCS). ISSN 0302-9743. doi: 10.1007/978-3-319-49806-5_7. Full text in Research Archive Show summary

  Loss of sensitive data is a common problem with potentially severe consequences. By categorizing documents according to their sensitivity, security controls can be performed based on this classification. However, errors in the classification process may effectively result in information leakage. While automated classification techniques can be used to mitigate this risk, little work has been done to evaluate the effectiveness of such techniques when sensitive content has been transformed (e.g., a document can be summarized, rewritten, or have paragraphs copy-pasted into a new one). To better handle these more difficult data leaks, this paper proposes the use of controlled environments to detect misclassification. By monitoring the incoming information flow, the documents imported into a controlled environment can be used to better determine the sensitivity of the document(s) created within the same environment. Our evaluation results show that this approach, using techniques from machine learning and information retrieval, provides improved detection of incorrectly classified documents that have been subject to more complex data transformations.

• Hammer, Hugo Lewi; Yazidi, Anis; Bai, Aleksander & Engelstad, Paal E. (2015). Building domain specific sentiment lexicons combining information from many sentiment lexicons and a domain specific corpus. In Abdelmalek, Amine; Ladjel, Bellatreche; Zakaria, Elberrichi; J., Neuhold Erich & Robert, Wrembel (Ed.), 5th IFIP TC 5 International Conference, CIIA 2015. Springer. ISSN 9783319195773. p. 205–216. doi: 10.1007/978-3-319-19578-0_17. Full text in Research Archive Show summary

  Most approaches to sentiment analysis requires a sentiment lexicon in order to automatically predict sentiment or opinion in a text. The lexicon is generated by selecting words and assigning scores to the words, and the performance the sentiment analysis depends on the quality of the assigned scores. This paper addresses an aspect of sentiment lexicon generation that has been overlooked so far; namely that the most appropriate score assigned to a word in the lexicon is dependent on the domain. The common practice, on the contrary, is that the same lexicon is used without adjustments across different domains ignoring the fact that the scores are normally highly sensitive to the domain. Consequently, the same lexicon might perform well on a single domain while performing poorly on another domain, unless some score adjustment is performed. In this paper, we advocate that a sentiment lexicon needs some further adjustments in order to perform well in a specific domain. In order to cope with these domain specific adjustments, we adopt a stochastic formulation of the sentiment score assignment problem instead of the classical deterministic formulation. Thus, viewing a sentiment score as a stochastic variable permits us to accommodate to the domain specific adjustments. Experimental results demonstrate the feasibility of our approach and its superiority to generic lexicons without domain adjustments.

• Engelstad, Paal E. (2015). Security Challenges with Cross-Domain Information Exchange: Integrity and Guessing Attacks. MILCOM IEEE Military Communications Conference. ISSN 2155-7578. p. 1596–1605. doi: 10.1109/MILCOM.2015.7357673. Full text in Research Archive Show summary

  Current research on cross-domain information ex- change is advocating to move away from the inflexible Bell-La Padula (BLP) model, into a more complex policy-driven security model where information objects and end-users are characterized in terms of complex meta-data. It will lead to higher flexibility but will also rely not only on guards, but also on automatic or semi-automatic tools for forming and processing the meta- data. In this paper, we point out some potential pitfalls with this approach. The paper focuses specifically on the relaxation of the BLP security model for confidentiality and discusses security concerns that arise from the use of such tools in combination with guards.

• Engelstad, Paal E. (2015). On the usability of clustering for topic-oriented multi-level security models. In Al-Dabass, David; Romero, Gregorio; Orsoni, Alessandra & Pantelous, Athanasios (Ed.), 9th IEEE European modelling symposium on Mathematical Modelling and Computer Simulation (EMS), Proceedings. IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781509002061. p. 14–20. doi: 10.1109/EMS.2015.13. Full text in Research Archive
• Engelstad, Paal E. (2015). On the Trade-Off Between Multi-Level Security Classification Accuracy and Training Time. In Al-Dabass, David; Saad, Ismail; Bolong, Nurmin & Hijazi, Mohd Hanafi Ahmad (Ed.), Third International Conference on Artificial Intelligence, Modelling and Simulation. IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781467386753. p. 349–355. doi: 10.1109/AIMS.2015.62. Full text in Research Archive
• Engelstad, Paal E.; Hammer, Hugo Lewi; Yazidi, Anis & Bai, Aleksander (2015). Advanced Classification Lists (Dirty Word Lists) for Automatic Security Classification. In Bin, Xie; Hee, Yong Youn & Min, Chen (Ed.), Proceedings of the 2015 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781467391993. p. 44–53. doi: 10.1109/CyberC.2015.103. Full text in Research Archive
• Yazidi, Anis; Bai, Aleksander; Hammer, Hugo Lewi & Engelstad, Paal E. (2015). A Simple and Efficient Algorithm for Lexicon Generation Inspired by Structural Balance Theory. In Ali, Moonis; Kwon, Sig Young; Chang-Hwan, Lee; Kim, Juntae & Kim, Yongdai (Ed.), 28th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems. Springer. ISSN 9783319190662. p. 336–347. doi: 10.1007/978-3-319-19066-2_33. Full text in Research Archive
• Yazidi, Anis; Hammer, Hugo Lewi; Bai, Aleksander & Engelstad, Paal E. (2015). On Enhancing the Label Propagation Algorithm for Sentiment Analysis Using Active Learning with an Artificial Oracle. In Leszek, Rutkowski; Marcin, Korytkowski; Rafal, Scherer; Ryszard, Tadeusiewicz; Lotfi, A. Zadeh & Jacek, M. Zurada (Ed.), Artificial Intelligence and Soft Computing: 14th International Conference, ICAISC 2015, Zakopane, Poland, June 14-18, 2015, Proceedings, Part II. Springer. ISSN 9783319193694. p. 799–810. doi: 10.1007/978-3-319-19369-4_71. Full text in Research Archive
• Engelstad, Paal E.; Hammer, Hugo Lewi; Yazidi, Anis & Bai, Aleksander (2015). Analysis of Time-Dependencies in Automatic Security Classification. In Bin, Xie; Hee, Yong Youn & Min, Chen (Ed.), Proceedings of the 2015 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781467391993. p. 54–61. doi: 10.1109/CyberC.2015.104. Full text in Research Archive
• Bratterud, Alfred; Walla, Alf-André; Haugerud, Hårek; Engelstad, Paal E. & Begnum, Kyrre (2015). IncludeOS: A minimal, resource efficient unikernel for cloud services. IEEE International Conference on Cloud Computing Technology and Science (CloudCom). ISSN 2330-2194. p. 250–257. doi: 10.1109/CloudCom.2015.89. Full text in Research Archive Show summary

  The emergence of cloud computing as a ubiquitous platform for elastically scaling services has generated need and opportunity for new types of operating systems. A service that needs to be both elastic and resource efficient needs A) highly specialized components, and B) to run with minimal resource overhead. Classical general purpose operating systems designed for extensive hardware support are by design far from meeting these requirements. In this paper we present IncludeOS, a single tasking library operating system for cloud services, written from scratch in C++. Key features include: extremely small disk-and memory footprint, efficient asynchronous I/O, OS-library where only what your service needs gets included, and only one device driver by default (virtio). As a test case a bootable disk image consisting of a simple DNS server with OS included is shown to require only 158 kb of disk space and to require 5-20% less CPU-time, depending on hardware, compared to the same binary running on Linux.

• Hammer, Hugo Lewi; Kongsgård, Kyrre Wahl; Bai, Aleksander; Yazidi, Anis; Nordbotten, Nils Agne & Engelstad, Paal E. (2015). Automatic Security Classification by Machine Learning for Cross-Domain Information Exchange. MILCOM IEEE Military Communications Conference. ISSN 2155-7578. doi: 10.1109/MILCOM.2015.7357672. Full text in Research Archive Show summary

  Cross-domain information exchange is necessary to obtain information superiority in the military domain, and should be based on assigning appropriate security labels to the information objects. Most of the data found in a defense network is unlabeled, and usually new unlabeled information is produced every day. Humans find that doing the security labeling of such information is labor-intensive and time consuming. At the same time there is an information explosion observed where more and more unlabeled information is generated year by year. This calls for tools that can do advanced content inspection, and automatically determine the security label of an information object correspondingly. This paper presents a machine learning approach to this problem. To the best of our knowledge, machine learning has hardly been analyzed for this problem, and the analysis on topical classification presented here provides new knowledge and a basis for further work within this area. Presented results are promising and demonstrates that machine learning can become a useful tool to assist humans in determining the appropriate security label of an information object.

• Bai, Aleksander; Hammer, Hugo Lewi; Yazidi, Anis & Engelstad, Paal E. (2014). Constructing sentiment lexicons in Norwegian from a large text corpus. In Liu, Xingang; Hsu, Ching-Hsien; Kang, Kai & Chen, Weifeng (Ed.), The 17th IEEE International conference on Computational science and Engineering (CSE 2014). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781479979806. p. 231–237. doi: 10.1109/CSE.2014.73. Full text in Research Archive
• Hammer, Hugo Lewi; Bai, Aleksander; Yazidi, Anis & Engelstad, Paal E. (2014). Building sentiment Lexicons applying graph theory on information from three Norwegian thesauruses. NIKT: Norsk IKT-konferanse for forskning og utdanning. ISSN 1892-0713. doi: 10.1109/cse.2014.73. Full text in Research Archive
• Grønsund, Pål R.; Engelstad, Paal E.; Pawelczak, Przemyslaw; Grøndalen, Ole; Lehne, Per Hjalmar & Cabric, Danijela (2013). Spectrum sensing aided long-term spectrum management in cognitive radio networks. Conference on Local Computer Networks. ISSN 0742-1303. 38. doi: 10.1109/lcn.2013.6761241. Full text in Research Archive
• Borgenholt, Gaute; Begnum, Kyrre & Engelstad, Paal (2013). Audition: a DevOps-oriented service optimization and testing framework for cloud environments. NIKT: Norsk IKT-konferanse for forskning og utdanning. ISSN 1892-0713. Full text in Research Archive
• Flathagen, Joakim; Larsen, Erlend; Engelstad, Paal Einar & Kure, Øivind (2012). O-CTP: Hybrid Opportunistic Collection Tree Protocol for Wireless Sensor Networks, Proceedings of the 37th Conference on Local Computer Networks Workshops (LCN Workshops 2012 IEEE). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781467321303. doi: 10.1109/lcnw.2012.6424086. Full text in Research Archive
• Grønsund, P.; MacKenzie, R.; Lehne, Per Hjalmar; Briggs, K.; Grøndalen, Ole & Engelstad, Paal E. [Show all 7 contributors for this article] (2012). Towards spectrum micro-trading, Future Network and Mobile Summit (FUNEMS 2012), Berlin, Germany, July 2012. IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781905824298. Full text in Research Archive
• Shams, S M Saif; Engelstad, Paal E. & Kvalbein, Amund (2012). PreeN: Improving steady-state performance of ISP-friendly P2P applications, Proceedings of the 13th International Conference, ICDCN 2012,Lecture Notes in Computer Science, Theoretical Computer Science and General Issues, Vol. 7129. Springer. ISSN 9783642259593. doi: 10.1007/978-3-642-25959-3_34. Full text in Research Archive
• Shams, S M Saif; Engelstad, Paal Einar & Kvalbein, Amund (2012). On Randomness in ISP-friendly P2P Applications. Advanced Information Networking and Applications. ISSN 1550-445X. p. 431–437. doi: 10.1109/AINA.2012.101. Full text in Research Archive

View all works in NVA

• Berlijn, Sonja Monica; Engelstad, Paal E.; Noll, Josef; Chockalingam, Sabarathinam & Pirbhulal, Sandeep (2026). WP4. Annual Progress Report INTPART-ResCri (2026). Full text in Research Archive
• Holen, Martin; Ruud, Else-Line Malene; Warakagoda, Narada Dilp; Granmo, Ole-Christoffer; Engelstad, Paal E. & Knausgård, Kristian Muri (2022). Towards Using Reinforcement Learning for Autonomous Docking of Unmanned Surface Vehicles. doi: 10.1007/978-3-031-08223-8_38. Full text in Research Archive Show summary

  Providing full autonomy to Unmanned Surface Vehicles (USV) is a challenging goal to achieve. Autonomous docking is a subtask that is particularly difficult. The vessel has to distinguish between obstacles and the dock, and the obstacles can be either static or moving. This paper developed a simulator using Reinforcement Learning (RL) to approach the problem. We studied several scenarios for the task of docking a USV in a simulator environment. The scenarios were defined with different sensor inputs and start-stop procedures but a simple shared reward function. The results show that the system solved the task when the IMU (Inertial Measurement Unit) and GNSS (Global Navigation Satellite System) sensors were used to estimate the state, despite the simplicity of the reward function.

• Aryan, Ramtin; Brattensborg, Frode; Yazidi, Anis & Engelstad, Paal E. (2019). Checking the OpenFlow Rule Installation and Operational Verification. Full text in Research Archive
• HaileselassieHagos, Desta; Engelstad, Paal E. & Yazidi, Anis (2019). Classification of Delay-based TCP Algorithms From Passive Traffic Measurements. Full text in Research Archive
• Aryan, Ramtin; Yazidi, Anis & Engelstad, Paal E. (2018). An Incremental Approach for Swift OpenFlow Anomaly Detection. Full text in Research Archive
• Kongsgård, Kyrre Wahl; Nordbotten, Nils Agne; Mancini, Federico & Engelstad, Paal E. (2017). An Internal/Insider Threat Score for Data Loss Prevention and Detection. Full text in Research Archive
• Do, van Thanh; Engelstad, Paal E.; Feng, Boning & Do, Van Thuan (2017). A near real time SMS Grey Traffic detection. Full text in Research Archive
• Thuan, Do Van; Engelstad, Paal E.; Feng, Boning & Do, van Thanh (2017). Detection of DNS tunneling in mobile networks using machine learning. Full text in Research Archive
• Aryan, Ramtin; Yazidi, Anis; Engelstad, Paal E. & Kure, Øivind (2017). A General Formalism for Defining and Detecting OpenFlow Rule Anomalies. Full text in Research Archive
• Kongsgård, Kyrre Wahl; Nordbotten, Nils Agne; Engelstad, Paal E. & Mancini, Federico (2017). Data Loss Prevention for Cross-Domain Instant Messaging. Full text in Research Archive Show summary

  This paper proposes a cascading classifier for inspecting and validating the payload of chat messages in (military) instant messaging. The first step in the cascading classifier pipeline is an anomaly detection-based method whose purpose is to ensure that the message channel is not used to exfiltrate non-message data such as images, documents, binary files or encrypted content. Messages that pass the filtering phase then proceed to have their content analyzed for the presence of known sensitive information. This data loss prevention step is enhanced by incorporating an author profile signal that assesses the validity of the claimed authorship by capturing the stylometric signature embedded in each user's past message stream. The hypothesis being that the inference and subsequent inclusion of latent author traits such as gender, age and ethnicity will aid the data loss prevention solution by reducing the number of incorrect classifications. Experiments were conducted using message traffic that was generated during a field-training exercise conducted by members of the armed forces, as well as an internal repository of classified documents and a myriad of non-message based data sources. The results demonstrated that our proposed traffic-filtering classifier is successful in distinguishing between legitimate and illegitimate traffic. Further, the experiments showed that constructing authorship verification models, using sparse messages as a training set, is feasible and that including this signal in the data loss prevention solution leads to a significant increase in the predictive performance for the cross-domain messaging setting.

• Do, Van Thuan; Engelstad, Paal E.; Boning, Feng & Do, van Thanh (2016). Strengthening mobile network security using machine learning. Full text in Research Archive
• Engelstad, Paal E.; Hammer, Hugo Lewi; Yazidi, Anis & Bai, Aleksander (2015). Analysis of Time-Dependencies in Automatic Security Classification. Full text in Research Archive
• Engelstad, Paal E.; Hammer, Hugo Lewi; Yazidi, Anis & Bai, Aleksander (2015). Advanced Classification Lists (Dirty Word Lists) for Automatic Security Classification. Full text in Research Archive
• Hammer, Hugo Lewi; Yazidi, Anis; Bai, Aleksander & Engelstad, Paal E. (2015). Building domain specific sentiment lexicons combining information from many sentiment lexicons and a domain specific corpus. Full text in Research Archive
• Yazidi, Anis; Hammer, Hugo Lewi; Bai, Aleksander & Engelstad, Paal E. (2015). On Enhancing the Label Propagation Algorithm for Sentiment Analysis Using Active Learning with an Artificial Oracle. Full text in Research Archive
• Yazidi, Anis; Bai, Aleksander; Hammer, Hugo Lewi & Engelstad, Paal E. (2015). A Simple and Efficient Algorithm for Lexicon Generation Inspired by Structural Balance Theory. Full text in Research Archive
• Bai, Aleksander; Hammer, Hugo Lewi; Yazidi, Anis & Engelstad, Paal E. (2014). Constructing sentiment lexicons in Norwegian from a large text corpus. Full text in Research Archive
• Hauge, Mariann; Landmark, Lars; Larsen, Erlend; Engelstad, Paal E. & Kure, Øivind (2014). Resilient internetwork routing with QoS support over heterogeneous mobile military networks. Full text in Research Archive
• Begnum, Kyrre & Engelstad, Paal E. (2013). HIOA får sin egen sky. [Internet]. https://www.hioa.no/Aktuelle-saker. Full text in Research Archive

View all works in NVA