In-depth Performance Analysis Research Articles

Performance comparison of machine learning algorithms for the estimation of blood pressure using photoplethysmography

This paper deals with an in-depth performance analysis on the estimation of Systolic Blood Pressure (SBP) and Diastolic Blood Pressure (DBP) by using features from the photoplethysmography (PPG) signal enhanced using the Maximal Overlap Discrete Wavelet Transform (MODWT), to train many machine learning (ML) regression models, including eXtreme Gradient Boost (XGBoost). The impact of different features selections methods, ML methods and training set sizes has been analyzed. One result is on the achievable improvements using features extracted from MODWT enhanced PPG signals. The most significant features have been selected using three different algorithms, namely RReliefF, Minimum Redundancy Maximum Relevance (MRMR) and Correlation-based Feature Selection (CFS). This comparison has been critical to underline that the exploitation of the new features allows to improve SBP and DBP estimation. Moreover, the authors have trained several ML algorithms to provide a comparison of their accuracy and training time, showing the Pareto frontier. RReliefF and MRMR selections algorithms, and several ML algorithms such as XGBoost, Gaussian Process Regression(GPR) and Ensemble stood out for their performance, with a different compromise between prediction error and training time. In addition, a further result has been obtained by varying the dimension of the dataset to understand the impact on Root Mean Square Error (RMSE) for models that have shown better performance, giving an empirical relationship on achievable RMSE as a function of training set size. From that relationship it has been extrapolated an upper boundary of the set size over which no further RMSE improvements are expected.

SENTINEY: Securing ENcrypted mulTI-party computatIoN for Enhanced data privacY and phishing detection

Phishing attacks have recently become a real danger that threatens the security of sensitive data. This research paper presents a new approach based on Secure Multi-Party Computation (SMPC) to identify phishing attacks on encrypted emails while ensuring data confidentiality and privacy. The proposed approach, SENTINEY (Securing ENcrypted mulTIparty computatIoN for Enhanced data privacY and phishing detection), combines unsupervised machine learning and string matching techniques to detect phishing links over encrypted data, making it adaptive. Subsequently, an adaptive system dynamically selects the appropriate phishing detection technique taking into account various factors (e.g, the volume of new attacks, the accuracy of the machine learning model, attack specificity and available system resources) is suggested. For efficiency reasons, the learning model uses network virtualization features to improve computational resources. This new approach has shown good performance in taking advantage of network virtualization to create a secure and collaborative environment for the SMPC. The dataset used to train and test the proposal is generated on the basis of real phishing emails, including real phishing URLs and keywords. An in-depth performance analysis evaluated the performance of the proposed approach in terms of efficiency (processing time) and robustness (accuracy, precision, recall and F1 score). Simulation results and comparison with relevant solutions show that the proposed approach achieves superior robustness at lower costs. Using a string-matching approach, the multilayer perceptron achieved the highest accuracy of 98% with a detection time of 0.89 s. On the other hand, Isolation Forest showed high efficiency in combating zero-day phishing attacks. The MLP model combined with other tools achieved an accuracy of 99.4%.

Design of a Fourier-transform spectrometer for detector characterization on CMB polarization telescope

We present the design of a Fourier-transform spectrometer (FTS) tailored for operation within the frequency range of 75 GHz to 300 GHz, offering a spectral resolution surpassing 2 GHz. This FTS can be used for characterizing detectors on a Cosmic Microwave Background (CMB) Polarization Telescope for the observation of CMB B-mode polarization. The FTS instrument is divided into the primary part, housing a Martin-Puplett interferometer (MPI) module, and the coupled part, which incorporates a beam expanding module and a beam steering module. An in-depth performance analysis focusing on the MPI module, involving considerations such as the orientation of the beam splitter, dihedral angle of the roof-top mirrors, and positioning uncertainty of the scanning mirror, has been conducted by using a simulation tool. Results indicate that the spectral characteristic remains unaffected by variations in the Reflection/Transmission ratio of the beam splitter, with any deviation from the 1:1 ratio uniformly decreasing intensity. The impact of dihedral angle errors of roof-top mirrors (below 0.02 degrees) and positioning uncertainty in the scanning mirror (lower than 3 × 10-3 mm) on the system performance is acceptable for the design of the FTS. The beam expanding module is capable of enlarging the output beam of the MPI module to roughly a 150 mm-radius. The beam steering module allows for directional adjustments within the range of +17 degrees to -17 degrees relative to the normal of the target plane. Preliminary testing on the MPI prototype reveals a spectral resolution of 1.5 GHz for the measured frequency.

Toward Ultra-Low Latency SSDs: Analyzing the Impact on Data-Intensive Workloads

The recent trend in hyper-scale computing has shown that applications are getting hungrier for low, consistent latency for predictably fast service. In accordance with this change, storage vendors have innovated their SSDs to keep ultra-low latencies while delivering high IOPS. Even though these SSDs can provide significant value to a wide variety of workloads, their practical impact on performance has not been discussed much. In this context, our study aims to empirically examine the impact of emerging ultra-low latency SSDs on workload performance, particularly in comparison to the latest conventional SSDs. We conduct benchmarking using a diverse set of data-intensive workloads to comprehensively assess the performance of both ultra-low latency SSDs and conventional SSDs. Our research provides an in-depth performance analysis of the examined SSDs, shedding light on their comparative strengths and weaknesses. The results of our study contribute valuable insights into the practical implications of adopting ultra-low latency SSDs across various workloads, guiding future decisions in storage technology.

Investigating performance metrics for container-based HPC environments using x86 and OpenPOWER systems

Container-based High-Performance Computing (HPC) is changing the way computation is performed and reproduced without sacrificing the raw performance compared to hypervisor-assisted virtualization technologies. It primarily supports continuously evolving data-intensive applications such as computational fluid dynamics, seismic tomography, molecular biology, and Proteomics. OpenPOWER systems, unlike the x86 systems, use the POWER-compliant processor to exploit instruction-level and thread-level parallelism heavily. In our previous work, we designed and developed a Containerized HPC environment (cHPCe) from the scratch using Linux namespaces on OpenPOWER systems. This paper aims to provide an in-depth performance analysis of the Containerized HPC environment using x86 systems and Containerized HPC environment using the OpenPOWER system, on systems’ subcomponents, processor, memory, interconnect, and IO. This sub-component analysis provides an insight on several aspects of the system performance. To the best of our knowledge, no research has been reported yet for such a comparative analysis that designs cHPCe for both x86 and OpenPOWER systems. The performance of the developed cHPCe is compared with BareMetals, and VMs using the benchmarks HPCC, and IOZone. Our experimental results achieve 0.13% less compute performance penalty at its peak performance on cHPCe compared to the BareMetal-based solution for x86 systems. In contrast, a VM-based solution introduces an overhead of 20% and 4.83% in x86 and OpenPOWER cases, respectively. Moreover, the x86 and OpenPOWER systems observe inconsistent behavior for memory performance with a worst-case penalty of 9.68% and 6.64% compared to achieved peak performance, respectively. However, similar behavior is reported for cHPCe with an overhead of less than 3% and 2% in the worst case for the latency and bandwidth, respectively, compared to the BareMetal for network and disk performance. Our experimental results reveal that the containerized OpenPOWER environment represents a viable alternative to the counterpart x86 environment for the HPC solution.

Analysis of map-based models for reciprocating compressors and optimum selection of rating points

The present work includes an in-depth performance analysis in fixed-speed reciprocating compressors. The industry standard for compressor characterization is the AHRI-540, which uses a 10-term and third-degree polynomial to characterize mass flow rate and energy consumption. However, the suitability of such a high-degree polynomial is unclear, and the potential for overfitting and extrapolation errors cannot be ignored. This work analyzes the response surfaces of mass flow rate and energy consumption in reciprocating compressors to determine if more concise models with lower degrees are more suitable. For that purpose, a massive experimental dataset with multiple compressors using different refrigerant and suction conditions was analyzed to obtain overall conclusions in the compressor field. The results of the present work showed that mass flow rate modeling requires lower-degree polynomials. However, the energy consumption characterization is more complex, and the model reported in the standard may be justified. Additionally, it was found that, if the specific energy consumption is selected as the modeling variable, it is possible to use a compact polynomial expression, which can also be extended to scroll compressors and also has the advantage of reducing the experimental data necessary for the model fit. Finally, by selecting the mass flow rate and the specific energy consumption as response variables, this work also explores other critical issues related to the experimental points’ location and minimum sample sizes required in order to minimize the experimental costs and increase the model accuracy.

Constant or Variable? A Performance Analysis among Portfolio Insurance Strategies

In this paper, we propose a comparison among three portfolio insurance strategies, namely the constant proportion portfolio insurance, the time-invariant portfolio protection, and the exponential proportion portfolio insurance, via an in-depth performance analysis. We aim to ascertain whether strategies characterized by variable parameters can outperform those with constant parameters by measuring potential returns, investment riskiness, downside protection capability, and ability to capture market upside. The results, achieved in a model-free framework by exploiting bootstrapping techniques, advise that no winning strategy exists overall, even when considering different volatility regimes, rebalancing frequencies, and protection levels.

In-Depth Performance Analysis and Comparison of Monolithic and Particulate Zwitterionic Hydrophilic Interaction Liquid Chromatography Polymer Columns.

The kinetic performance of different zwitterionic hydrophilic interaction liquid chromatography polymer columns is evaluated and compared in-depth. For this purpose, two lab-made monolithic columns, synthesized with different crosslinkers, and a commercial particle packed column are considered. It is found that performance evaluation techniques, such as comparing plate height curves or fitted A-, B- and C-terms, obtained by fitting experimental plate height data to a plate height model, are complicated by the determination of a reliable characteristic length. This is due to the very different morphology of these column types, and the heterogeneity of the monolithic columns. The occurrence of a convective flow through the packed particle column further complicates the interpretation of the obtained fitting parameters, as part of the C-term is wrongfully attributed to the A-term. Therefore, the use of the kinetic plot method is suggested for the comparative evaluation of these columns, as kinetic plots do not require the determination of a characteristic length, nor rely on any fitting parameters. With the kinetic plot method, it is demonstrated that the lab-made monolithic columns outperform the packed particle column for plate counts between 10,000 and 800,000. This is attributed to the higher column efficiency of these columns, due to their small domain and skeleton sizes, and their high permeability, resulting from their high external porosity and the occasional occurrence of preferential flow paths.

A Maximum-Likelihood-Based Two-User Receiver for LoRa Chirp Spread-Spectrum Modulation

Long Range (LoRa) is an emerging low-power wide-area network technology offering long-range wireless connectivity to Internet of Things (IoT) devices. For energy efficiency reasons, LoRa end nodes implement a nonslotted ALOHA multiple access scheme to transmit packets to the gateway. Due to the lack of synchronization between end nodes, collisions between uplink packets have been identified as the main obstacle to the scaling of dense LoRa networks. To tackle this issue, we present in this article a LoRa receiver that is capable of decoding colliding packets from two interfering end nodes. The proposed two-user detector is derived from the maximum-likelihood principle using a detailed model of two colliding LoRa packets. As the complexity of the maximum-likelihood sequence estimation is prohibitive, complexity-reduction techniques are introduced to enable practical implementations of the receiver. An in-depth performance analysis highlights that the proposed two-user detector inherently leverages the differences in received power, time offsets, and frequency offsets between the users to separate and demodulate their respective signals. To demonstrate the practicality of the proposed detector, an interference-robust synchronization algorithm is then designed and evaluated. Simulation results indicate that a LoRa receiver combining the proposed synchronization algorithm and two-user detector is capable of detecting and demodulating two interfering users with satisfactorily error rates.

The listening chamber: experiencing scenography in Pan Pan’s All That Fall

ABSTRACT This article explores how scenography, specifically the visual-aural-spatial elements of a performance, shapes the response of an audience by acting as either a guiding element to a performance or its primary meaning. Through the case study of Pan Pan Theatre’s All That Fall (2011) in the Abbey Theatre, Dublin, this article investigates how the sound, lighting and spatial designs exploit the audience’s embodied responses. In conjunction with in-depth performance analysis, this article will utilise audience feedback and theatre practitioner interviews in order to investigate how scenography organises the relationship of the audience to the performance through the active foregrounding of perception, as well as demonstrating the evolving theatrical engagements/responses to Beckett’s work.

An in-depth investigation of the performance characteristics of Hyperledger Fabric

Private permissioned blockchains are deployed in ever greater numbers to facilitate cross-organizational processes in various industries, particularly in supply chain management. One popular example of this trend is Hyperledger Fabric. Compared to public permissionless blockchains, it promises improved performance and provides certain features that address key requirements of enterprises. However, also permissioned blockchains are still not as scalable as centralized systems, and due to the scarcity of theoretical results and empirical data, their real-world performance cannot be predicted with the necessary precision. We intend to address this issue by conducting an in-depth performance analysis of Hyperledger Fabric. The paper presents a detailed compilation of various performance characteristics using an enhanced version of the Distributed Ledger Performance Scan (DLPS). Researchers and practitioners alike can use the various performance properties identified and discussed as guidelines to better configure and implement their Hyperledger Fabric network. Likewise, they are encouraged to use the DLPS framework to conduct their measurements.

Fuzzy-Based Time Series Forecasting and Modelling: A Bibliometric Analysis

The purpose of this paper is to present the results of a systematic literature review regarding the development of fuzzy-based models for time series forecasting in the period 2017–2021. The study was conducted using a well-established review protocol and a couple of powerful tools for bibliometric analysis to know and analyse the main approaches adopted in the research field of interest. We analysed 118 articles published in peer-reviewed journals indexed in the 2020 Journal Citation Reports of the Web of Science. This allowed us to present an in-depth performance analysis and a science mapping regarding the current situation of fuzzy time series forecasting and modelling. The outputs of this study provide a practical base for further investigations that address this topic from both a methodological point of view and in terms of applicability.

A Novel Cascade Model for End-to-End Aspect-Based Social Comment Sentiment Analysis

The end-to-end aspect-based social comment sentiment analysis (E2E-ABSA) task aims to discover human’s fine-grained sentimental polarity, which can be refined to determine the attitude in response to an object revealed in a social user’s textual description. The E2E-ABSA problem includes two sub-tasks, i.e., opinion target extraction and target sentiment identification. However, most previous methods always tend to model these two tasks independently, which inevitably hinders the overall practical performance. This paper investigates the critical collaborative signals between these two sub-tasks and thus proposes a novel cascade social comment sentiment analysis model for jointly tackling the E2E-ABSA problem, namely CasNSA. Instead of treating the opinion target extraction and target sentiment identification as discrete procedures in previous works, our new framework takes the contextualized target semantic encoding into consideration to yield better sentimental polarity judgment. Additionally, extensive empirical results show that the proposed approach effectively achieves a 68.13% F1-score on SemEval-2014, 62.34% F1-Score on SemEval-2015, 56.40% F1-Score on SemEval-2016, and 50.05% F1-score on a Twitter dataset, which is higher than the existing approaches. Ablated experiments demonstrate that the CasNSA model substantially outperforms state-of-the-art methods, even when using fixed words embedding rather than pre-trained BERT fine tuning. Moreover, in-depth performance analysis on the social comment datasets further validates that our work gains superior performance and reliability effectively and efficiently in realistic scenarios.

Multiplay Multiarmed Bandit Algorithm Based Sensing of Noncontiguous Wideband Spectrum for AIoT Networks

To bring large-scale AIoT to reality, wireless networks need intelligence to identify resources in a non-contiguous spectrum. We address this challenge via a sub-Nyquist sampling-based wideband spectrum analyzer deployed in AIoT gateway. The non-contiguous nature demands learning the channel occupancy. However, the identification of channel status can fail when the number of busy channels in a selected subset is higher than the number of ADCs, K. We model this subset selection problem as Multi-Play Multi-Armed Bandit. First, we demonstrate the learnability of such a problem via a learning algorithm with a subset size of K (no sensing failure). For wideband sparse spectrum, we extend this algorithm using a novel subset size estimation approach to identify the optimal subset that gives the best possible throughput and could have a size potentially larger than K. These algorithms are mapped on the system-on-chip, and their in-depth performance analysis demonstrates their superiority over state-of-the-art.

Heat Dissipation by Streams of Bifurcated Tubes

There is a need for solutions to provide sufficient cooling from power devices, which produce large amounts of heat. This paper focuses on the influence of design of bifurcated fluid streams to dissipate heat. In this study, a single Y-tubes, a double Y-tubes, and an X-tubes designs are studied numerically under space constraints. For a comprehensive and in-depth performance analysis, both heat dissipation and hydraulic performances are analyzed. The distributions of velocity and temperature in the fluid streams is simulated, also the flow resistances and dissipated heat are calculated. Based on the results obtained, a thermo-hydraulic performance factor is introduced for the designs under study. In addition, the accumulation of undesired substances on the wall surface (fouling) that may influence the heat exchanging capability is studied.

Self-Learning Pipeline for Low-Energy Resource-Constrained Devices

The trend of bringing machine learning (ML) to the Internet of Things (IoT) field devices is becoming ever more relevant, also reducing the overall energy need of the applications. ML models are usually trained in the cloud and then deployed on edge devices. Most IoT devices generate large amounts of unlabeled data, which are expensive and challenging to annotate. This paper introduces the self-learning autonomous edge learning and inferencing pipeline (AEP), deployable in a resource-constrained embedded system, which can be used for unsupervised local training and classification. AEP uses two complementary approaches: pseudo-label generation with a confidence measure using k-means clustering and periodic training of one of the supported classifiers, namely decision tree (DT) and k-nearest neighbor (k-NN), exploiting the pseudo-labels. We tested the proposed system on two IoT datasets. The AEP, running on the STM NUCLEO-H743ZI2 microcontroller, achieves comparable accuracy levels as same-type models trained on actual labels. The paper makes an in-depth performance analysis of the system, particularly addressing the limited memory footprint of embedded devices and the need to support remote training robustness.

Evaluating Distributed Computing Infrastructures: An Empirical Study Comparing Hadoop Deployments on Cloud and Local Systems

The popularity of distributed computing platforms (e.g., Hadoop) is largely due to their ability to address scalability issues that arise from data storage and processing limitations of standard computing systems. However, the decision to dedicate organizational resources and capital for such systems needs a careful consideration of several factors including evaluation of cloud-based distributed computing options. We propose a framework of metrics which we used to conduct an in-depth performance and cost benefit analysis of two standard Hadoop infrastructural choices, i.e., a Platform as a Service ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">PaaS</i> ) on-demand cloud setup and a local organizational setup. We evaluated the framework by means of an exploratory data analysis use-case for a large-scale graph processing research problem. Our analysis considered highly granular aspects of distributed computing performance and studied how utilization rates and infrastructure amortization times affect break-even times. We identified that virtual memory management adversely affects the performance of a cloud cluster during the reduce phase with the magnitude of degradation dependent on the type of MapReduce operation. Our study is intended not only as an evaluation of infrastructural choices but also a development of a metric framework that can serve as a baseline for researchers examining distributed infrastructures.

Valorization of Waste Apple Pomace for Production of Platform Biochemicals: A Multi-Objective Optimization Study

In line with the prevailing global interest in value extraction from biomass waste streams, the current study explored the technical feasibility of valorizing waste apple pomace (WAP) to produce high-value biochemicals of 5-hydroxymethylfurfural (HMF), lactic acid, and xylitol. Technical feasibility was demonstrated via a process simulation study that employed experimental data and incorporated previously reported approaches in the literature. Economic and environmental performances of the WAP based biorefinery were assessed using the internal rate of return (IRR) and the mass of greenhouse gas emission per unit mass of feedstock (GF) as sufficient performance indicators, respectively. The study was able to show that as the IRR value increased (better economic performance), the GF increased (poorer environmental performance). This suggested that the determination of the optimal condition of the environmental and economic performances would require the imposition of trade-offs. The preferred trade-off condition for enhanced economic and environmental performances was subsequently determined via multi-objective optimization, with a Pareto front containing non-dominated equally optimal solutions subsequently developed. The present work, therefore, provides an in-depth performance analysis of WAP based biorefinery as a waste management strategy. Notably, the proposed strategy of multiple product generation from biomass may be extended to other organic waste based biorefineries.

A Comprehensive Worst Case Bounds Analysis of IEEE 802.15.7

Visible Light Communication (VLC) has been emerging as a promising technology to address the increasingly high data-rate and time-critical demands that the Internet of Things (IoT) and 5G paradigms impose on the underlying Wireless Sensor Actuator Networking (WSAN) technologies. In this line, the IEEE 802.15.7 standard proposes several physical layers and Medium Access Control (MAC) sub-layer mechanisms that support a variety of VLC applications. Particularly, at the MAC sub-layer, it can support contention-free communications using Guaranteed Timeslots (GTS), introducing support for time-critical applications. However, to effectively guarantee accurate usage of such functionalities, it is vital to derive the worst-case bounds of the network. In this paper, we use network calculus to carry out the worst-case bounds analysis for GTS utilization of IEEE 802.15.7 and complement our model with an in-depth performance analysis. We also propose the inclusion of an additional mechanism to improve the overall scalability and effective bandwidth utilization of the network.

Dual-composite drag-reduction surface based on the multilayered structure and mechanical properties of tuna skin.

Energy efficiency and friction reduction have attracted considerable research attention. To design low drag surfaces, researchers derived inspiration from nature on various types of drag reduction methods with exceptional functional surfaces, such as fish skin that possesses low friction. Fishes with high-performance swimming possess a range of physiological and mechanical adaptations that are of considerable interest to physiologists, ecologists, and engineers. Although tuna is a fast-swimming ocean-based predator, most people focus their attention on its nutritional value. In this study, the multilayered structures and mechanical properties of tuna skin are first analyzed, and then the drag-reduction effect of the bionic fish-scale and dual-composite surfaces are studied based on the computational fluid dynamics method. The results indicate that tuna skin is composed of five layers, with the fish scale covered by a flexible epidermis layer. According to the uniaxial tension results, the modulus and tensile strength of the epidermis are obtained as 1.17 and 20 MPa, respectively. The nanoindentation results show that the modulus and hardness of the outer surface of the fish scale are larger than that of the inner surface, while those of the dry state are larger than those of the hydrated state. The simulation results show that both the bionic fish-scale and dual-composite surfaces display drag reduction, with the maximum drag-reduction rate of 25.7% achieved by the bionic dual-composite surface. These findings can offer a reference for in-depth performance analysis of the hydrodynamics of tuna and provide new sources of inspiration for drag reduction and antifouling.