Time-constrained Scenarios Research Articles

A novel approach for classification of Tor and non-Tor traffic using efficient feature selection methods

In the dynamic realm of encrypted communications, traffic analysis and its classification are crucial for efficient resource utilization and network management. The prevalence of encryption technologies, The Onion Router (Tor) a globally recognized privacy-preserving network, poses a challenge for the task at hand by introducing complexity through its innovative onion routing mechanism. To overcome Tor’s limitations not only in terms of achieving better accuracy but also in performing classification in time-constrained scenarios, we propose a classification approach for Tor and non-Tor traffic classification, utilizing multiple models to enhance categorization and application identification. Leveraging the University of New Brunswick (UNB) Tor and non-Tor dataset, initially in a packet capture format, the preprocessing is done by transforming through CICFlowmeter. To expedite classification, we applied feature selection techniques like Principal Component Analysis (PCA) and t-Distributed Stochastic Neighbor Embedding (tSNE). Machine learning algorithms like support vector machine (SVM), Gradient Boosting, Random Forest, and Artificial Neural Network (ANN) are applied. Our approach achieves a remarkable recall score ratio of 1.00, demonstrating high accuracy in Tor traffic identification. Notably, efficient feature selection has significantly reduced classification time. This work also contributes to effective Tor and non-Tor network traffic analysis, offering an efficient model for enhanced security and management.

Assessing the Performance of Multipath Mitigation for Multi-GNSS Precise Point Positioning Ambiguity Resolution

Real-time GNSS PPP is commonly used for high-precision positioning, but its utility is constrained by factors that necessitate extended convergence periods for a dependable accuracy. Multipath, as an unmodeled error, significantly curtails PPP performance in time-constrained scenarios. Approximately 31 consecutive days of multi-GNSS data from the satellite positioning service of the German national survey (SAPOS) network were collected to evaluate the effectiveness of multipath correction for real-time PPP ambiguity resolution (AR). Using principal component analysis (PCA) to extract the common-mode error (CME) from observation residuals prior to multipath modeling, a multipath hemispherical map (MHM) and sidereal filtering (SF) approach were employed to alleviate the effects of multipath and assess the efficacy of multipath correction in real-time PPP-AR. The average RMS reductions of the carrier-phase and pseudorange residual of multi-GNSS were 25.5% and 20.1% with MHM 0.5, while being 24.4% and 18.3% using SF. With MHM 0.5 correction, the TTFF reductions were approximately 7.0%, 17.7%, 37.5%, and 23.7% for G/GE/GC/GEC kinematic PPP-AR, respectively; and the convergence times for G/GE/GC PPP-AR were reduced to 18.2, 11.7, and 8.6 min, while GEC achieved an average convergence time of 7.1 min; a remarkable improvement compared to the multipath-uncorrected result (18 min). Moreover, 80% of the stations achieved convergence within 10 min, while 40% achieved convergence within 5 min. The kinematic positioning accuracy for the GEC solution improved from 0.97, 0.88, and 2.07 cm, to 0.94, 0.70, and 1.72 cm. In the static results, the TTFF shortened by 30.1%, 19.1%, and 20.1% for G/GE/GC, and the GEC decreased from 10.5 to 9.7 min; the average convergence time for G/GE/GC shortened to 13.0, 10.0, and 11.3 min, and for GEC shortened from 12.5 to 8.3 min. For the GPS-only solution, 78.3% of stations achieved convergence within 15 min. Similarly, for the GE scheme, the convergence time was primarily concentrated within 10 min, and for GC and GEC, with the application of enhanced multipath error correction, some of the stations even achieved convergence of PPP-AR within 5 min. The static positioning accuracy for GEC PPP was 0.50, 0.30, and 0.71 cm for the east, north, and up components.

Heart Rate Estimation from Incomplete Electrocardiography Signals.

As one of the most remarkable indicators of physiological health, heart rate (HR) has become an unfailing investigation for researchers. Unlike many existing methods, this article proposes an approach to implement short-time HR estimation from electrocardiography in time series missing patterns. Benefiting from the rapid development of deep learning, we adopted a bidirectional long short-term memory model (Bi-LSTM) and temporal convolution network (TCN) to recover complete heartbeat signals from those with durations are less than one cardiac cycle, and the estimated HR from recovered segment combining the input and the predicted output. We also compared the performance of Bi-LSTM and TCN in PhysioNet dataset. Validating the method over a resting heart rate range of 60−120 bpm in the database without significant arrhythmias and a corresponding range of 30−150 bpm in the database with arrhythmias, we found that networks provide an estimated approach for incomplete signals in a fixed format. These results are consistent with real heartbeats in the normal heartbeat dataset (γ > 0.7, RMSE < 10) and in the arrhythmia database (γ > 0.6, RMSE < 30), verifying that HR could be estimated by models in advance. We also discussed the short-time limits for the predictive model. It could be used for physiological purposes such as mobile sensing in time-constrained scenarios, and providing useful insights for better time series analyses in missing data patterns.

Predicting Secondary Task Performance: A Directly Actionable Metric for Cognitive Overload Detection

In this article, we address cognitive overload detection from unobtrusive physiological signals for users in dual-tasking scenarios. Anticipating cognitive overload is a pivotal challenge in interactive cognitive systems and could lead to safer shared-control between users and assistance systems. Our framework builds on the assumption that decision mistakes on the cognitive secondary task of dual-tasking users correspond to cognitive overload events, wherein the cognitive resources required to perform the task exceed the ones available to the users. We propose DecNet, an end-to-end sequence-to-sequence deep learning model that infers in real time the likelihood of user mistakes on the secondary task, i.e., the practical impact of cognitive overload, from eye-gaze and head-pose data. We train and test DecNet on a data set collected in a simulated driving setup from a cohort of 20 users on two dual-tasking decision-making scenarios, with either visual or auditory decision stimuli. DecNet anticipates cognitive overload events in both scenarios and can perform in time-constrained scenarios, anticipating cognitive overload events up to 2 s before they occur. We show that DecNet’s performance gap between audio and visual scenarios is consistent with user-perceived difficulty. This suggests that single modality stimulation induces higher cognitive load on users, hindering their decision-making abilities.

Why Should We Add Early Exits to Neural Networks?

Deep neural networks are generally designed as a stack of differentiable layers, in which a prediction is obtained only after running the full stack. Recently, some contributions have proposed techniques to endow the networks with early exits, allowing to obtain predictions at intermediate points of the stack. These multi-output networks have a number of advantages, including: (i) significant reductions of the inference time, (ii) reduced tendency to overfitting and vanishing gradients, and (iii) capability of being distributed over multi-tier computation platforms. In addition, they connect to the wider themes of biological plausibility and layered cognitive reasoning. In this paper, we provide a comprehensive introduction to this family of neural networks, by describing in a unified fashion the way these architectures can be designed, trained, and actually deployed in time-constrained scenarios. We also describe in-depth their application scenarios in 5G and Fog computing environments, as long as some of the open research questions connected to them.

A QUIC Look at Web Tracking

Abstract QUIC has been developed by Google to improve the transport performance of HTTPS traffic. It currently accounts for approx. 7% of the global Internet traffic. In this work, we investigate the feasibility of user tracking via QUIC from the perspective of an online service. Our analysis reveals that the protocol design contains violations of privacy best practices through which a tracker can passively and uniquely identify clients across several connections. This tracking mechanisms can achieve reduced delays and bandwidth requirements compared to conventional browser fingerprinting or HTTP cookies. This allows them to be applied in resource- or time-constrained scenarios such as real-time biddings in online advertising. To validate this finding, we investigated browsers which enable QUIC by default, e.g., Google Chrome. Our results suggest that the analyzed browsers do not provide protective measures against tracking via QUIC. However, the introduced mechanisms reset during a browser restart, which clears the cached connection data and thus limits achievable tracking periods. To mitigate the identified privacy issues, we propose changes to QUIC’s protocol design, the operation of QUIC-enabled web servers, and browser implementations.

Enhancing Quality for HEVC Compressed Videos

The latest High Efficiency Video Coding (HEVC) standard has been increasingly applied to generate video streams over the Internet. However, HEVC compressed videos may incur severe quality degradation, particularly at low bit-rates. Thus, it is necessary to enhance the visual quality of HEVC videos at the decoder side. To this end, this paper proposes a Quality Enhancement Convolutional Neural Network (QE-CNN) method that does not require any modification of the encoder to achieve quality enhancement for HEVC. In particular, our QE-CNN method learns QE-CNN-I and QE-CNN-P models to reduce the distortion of HEVC I and P frames, respectively. The proposed method differs from the existing CNN-based quality enhancement approaches, which only handle intra-coding distortion and are thus not suitable for P frames. Our experimental results validate that our QE-CNN method is effective in enhancing quality for both I and P frames of HEVC videos. To apply our QE-CNN method in time-constrained scenarios, we further propose a Time-constrained Quality Enhancement Optimization (TQEO) scheme. Our TQEO scheme controls the computational time of QE-CNN to meet a target, meanwhile maximizing the quality enhancement. Next, the experimental results demonstrate the effectiveness of our TQEO scheme from the aspects of time control accuracy and quality enhancement under different time constraints. Finally, we design a prototype to implement our TQEO scheme in a real-time scenario.