Real-time Environment Research Articles

Machine learning based data collection protocol for intelligent transport systems: a real-time implementation on Dublin M50, Ireland

The continuous global urbanization with rapid and dynamic transitioning in traffic situations among highly populated cities results in difficulty for data collection and communication. Data collection for millions of vehicles hinders by various problems, i.e., higher cost of energy, time, space, and storage resources. Moreover, higher data traffic results in higher delays, larger throughput, excessive bottlenecks, and frequent repetition of data. To better facilitate the aforementioned challenges and to provide a solution, we have proposed a lightweight Machine Learning based data collection protocol named ML-TDG to effectively deal with higher data volumes in a real-time traffic environment capable of bringing the least burden on the network while utilizing less space, time, and energy. ML-TDG is functional based on Apache Spark, an effective data processing engine that indexes the data based on two logs, i.e., old commuters or frequent/daily commuters and second new/occasional commuters. The proposed protocol’s main idea is to utilize real-time traffic, distinguish the indexes in parallel based on two assigned logs criteria to train the network, and collect data with the least sources. For energy and time optimization, dynamic segmentation switching is introduced which is an intelligent road segments division and switching for reducing bottlenecks and replication. ML-TDG is tested and verified on Dublin, Ireland’s busiest motorway M50. ML-TDG performs the data collection, data sorting, and network training to decide the next execution altogether for better optimization every time. The experimental results verify that our proposed protocol is attaining higher performance with lower resource requirements along with rich and time-efficient sustainable data collection clusters in comparison with baseline protocols.

Container-based Virtualization for Real-time Industrial Systems—A Systematic Review

Industrial Automation and Control systems have matured into a stable infrastructure model that has been kept fundamentally unchanged, using discrete embedded systems (such as Programmable Logic Controllers) to implement the first line of sensorization, actuation, and process control and stations and servers providing monitoring, supervision, logging/database and data-sharing capabilities, among others. More recently, with the emergence of the Industry 4.0 paradigm and the need for more flexibility, there has been a steady trend towards virtualizing some of the automation station/server components, first by using virtual machines and, more recently, by using container technology. This trend is pushing for better support for real-time requirements on enabling virtualization technologies such as virtual machines and containers. This article provides a systematic review on the use of container virtualization in real-time environments such as cyber-physical systems, assessing how existing and emerging technologies can fulfill the associated requirements. Starting by reviewing fundamental concepts related to container technology and real-time requirements, it goes on to present the methodology and results of a systematic study of 37 selected papers covering aspects related to the enforcement of real-time constrains within container hosts and the expected task latency on such environments, as well as an overview of container platforms and orchestration mechanisms for RT systems.

Conceptual Framework for Modelling of an Electric Tractor and Its Performance Analysis Using a Permanent Magnet Synchronous Motor

The purpose of this article is to model and validate the propulsion system of an electric tractor. Initially, the mathematical model of the ET and its load characteristics are developed based on the velocity profiles used in the agriculture ploughing operations. A PMSM is selected from the torque and power requirement of the ET. In order to replicate the ET load on the PMSM, the load is simulated in accordance with the specifications of the PMSM drive system. This article proposes a novel load-tracking control of the propulsion unit with the arrangement of a PMSM and dynamometer. The speed and torque of the propulsion motor are controlled separately using a decoupled control algorithm. The performance of the propulsion unit is verified under different load profiles of the ET. The load-tracking algorithm is used to improve the dynamic response of the propulsion motor. The propulsion system containing the mathematical model of the ET, PMSM and dynamometer is developed in MATLAB/SIMULINK and validated in a real-time environment using a hardware-in-the-loop (HIL) simulator (OPAL-RT OP5700). The result proves that the proposed load-tracking algorithm provided a better dynamic performance of the propulsion unit.

Two-stage stochastic unit commitment for renewable energy integrated power systems considering dynamic capacity-increase technologies of transmission lines

Dynamic capacity-increase technologies (DCITs), as a solution to reduce the stress of transmission networks, avoid long-term constructions and update the maximum transmitted capacities based on real-time environment factors and the maximum allowable operation temperatures (MAOTs), which provide flexibility in optimizing the operation modes for power systems. This paper proposes a detailed two-stage stochastic unit commitment model considering the physical processes of heat transfer as well as the uncertainties of renewable energy. The dynamic thermal rating (DTR) is calculated based on the heat balance of transmission lines in the steady state, including radiated and convective heat loss and gain from the solar heat and power flows. To simulate the uncertainties from renewable energy outputs, this paper applies the Latin hypercube sampling method to generate output scenarios in stochastic optimization. Finally, the proposed model is verified on a modified IEEE-24-bus test system. The effectiveness of DCITs is illustrated through the case study.

Fish-inspired robotic algorithm: mimicking behaviour and communication of schooling fish.

This study aims to present a novel flocking algorithm for robotic fish that will aid the study of fish in their natural environment. The algorithm, fish-inspired robotic algorithm (FIRA), amalgamates the standard flocking behaviors of attraction, alignment, and repulsion, together with predator avoidance, foraging, general obstacle avoidance, and wandering. The novelty of the FIRA algorithm is the combination of predictive elements to counteract processing delays from sensors and the addition of memory. Furthermore, FIRA is specifically designed to work with an indirect communication method that leads to superior performance in collision avoidance, exploration, foraging, and the emergence of realistic behaviors. By leveraging a high-latency, non-guaranteed communication methodology inspired by stigmergy methods inherent in nature, FIRA successfully addresses some of the obstacles associated with underwater communication. This breakthrough enables the realization of inexpensive, multi-agent swarms while concurrently harnessing the advantages of tetherless communication. FIRA provides a computational light control algorithm for further research with low-cost, low-computing agents. Eventually, FIRA will be used to assimilate robots into a school of biological fish, to study or influence the school. This study endeavors to demonstrate the effectiveness of FIRA by simulating it using a digital twin of a bio-inspired robotic fish. The simulation incorporates the robot's motion and sensors in a realistic, real-time environment with the algorithm used to direct the movements of individual agents. The performance of FIRA was tested against other collective flocking algorithms to determine its effectiveness. From the experiments, it was determined that FIRA outperformed the other algorithms in both collision avoidance and exploration. These experiments establish FIRA as a viable flocking algorithm to mimic fish behavior in robotics.

Texture feature analysis of MRI-ADC images to differentiate glioma grades using machine learning techniques

Apparent diffusion coefficient (ADC) of magnetic resonance imaging (MRI) is an indispensable imaging technique in clinical neuroimaging that quantitatively assesses the diffusivity of water molecules within tissues using diffusion-weighted imaging (DWI). This study focuses on developing a robust machine learning (ML) model to predict the aggressiveness of gliomas according to World Health Organization (WHO) grading by analyzing patients’ demographics, higher-order moments, and grey level co-occurrence matrix (GLCM) texture features of ADC. A population of 722 labeled MRI-ADC brain image slices from 88 human subjects was selected, where gliomas are labeled as glioblastoma multiforme (WHO-IV), high-grade glioma (WHO-III), and low-grade glioma (WHO I-II). Images were acquired using 3T-MR systems and a region of interest (ROI) was delineated manually over tumor areas. Skewness, kurtosis, and statistical texture features of GLCM (mean, variance, energy, entropy, contrast, homogeneity, correlation, prominence, and shade) were calculated using ADC values within ROI. The ANOVA f-test was utilized to select the best features to train an ML model. The data set was split into training (70%) and testing (30%) sets. The train set was fed into several ML algorithms and selected most promising ML algorithm using K-fold cross-validation. The hyper-parameters of the selected algorithm were optimized using random grid search technique. Finally, the performance of the developed model was assessed by calculating accuracy, precision, recall, and F1 values reported for the test set. According to the ANOVA f-test, three attributes; patient gender (1.48), GLCM energy (9.48), and correlation (13.86) that performed minimum scores were excluded from the dataset. Among the tested algorithms, the random forest classifier(0.8772 ± 0.0237) performed the highest mean-cross-validation score and selected to build the ML model which was able to predict tumor categories with an accuracy of 88.14% over the test set. The study concludes that the developed ML model using the above features except for patient gender, GLCM energy, and correlation, has high prediction accuracy in glioma grading. Therefore, the outcomes of this study enable to development of advanced tumor classification applications that assist in the decision-making process in a real-time clinical environment.

Endodontic microsurgery virtual reality simulation and digital workflow process in a teaching environment.

Computer simulations are stimulating increased attention in dentistry. Augmented reality superimposes a virtual scenario over an existing reality and allows interaction with it. Virtual reality (VR) simulates a fully immersive situation permitting the user to experience the full environment in real time. Haptic technology provides tactile and realistic force feedback for the user to experience the immersive situation as if they were really there. Preclinical training is important to gain familiarity with difficult surgical techniques and to implement interpersonal skills. Developing a valid assessment of surgical simulation is challenging. This paper wants to present a newly realized VR simulation in endodontic microsurgery through the developmental digital workflow, the demonstration of a haptic VR scenario and student self-assessment and self-reflection feedback. The volumes were exported in a stereolithography format to prepare and optimize in terms of shape and shade for the VR simulation. The graphics and touchable haptic solid were created using Virteasy Editor, which allows the transformation of 3D surfaces into graphical and volumetric haptic solids depending on their material (enamel, dentine, pulp and bone). Users were asked to execute the osteotomy and root-resection preparation. The assessment criteria were determined, and the feedback statements were created by a questionnaire with fixed answers. Objective and qualitative criteria for assessing the preparation were obtained from the literature. This study provides proof that it is possible to provide reliable and clinically relevant qualitative feedback with a VR simulator. VR simulation offers an innovative approach with all the benefits of clinical experience. It permits you to save your own progress and review the assessment at any time.

An efficient power extraction technique for improved performance and reliability of solar PV arrays during partial shading

Partial shading in the arrays reduces the power output, creates hotspots and damages the modules affecting the performance of the system. The modules bypass diodes create non-convexity in the power curves causing additional power losses. Hence, this work proposes a black widow reconfiguration (BWR) with reduced switch counts for mitigating the effects of partial shading in the arrays. The methodology effectively reduces the current difference between rows of the PV array within a short period with a faster calculation rate and electrically reconnects the modules. The reconnection ensures reduced mismatch among modules, higher power output and smoother power curves during partial shading. To show the effectiveness of the BWR in arbitrary-sized arrays, three arrays of 2 × 4, 5 × 5, and 9 × 9 sizes are considered and tested under static and dynamic partial shadings. The results are then compared with three conventional, six static, and three dynamic techniques using power curves and various parameters. Besides validation in the MATLAB environment, the experimental setup and OPAL-RT platform are considered to show the application in real-time environment. From the investigation, an average power improvement of 25.49%, 15.47%, and 9.29% in BWR compared to the existing conventional, static, and dynamic techniques with 99.60% efficiency have been observed.

Leveraging Explainable Artificial Intelligence for Real-time Detection of Tidal Blade Damage

Our current reliance on fossil fuels is the primary contributor to global warming and threatens our survival. Renewable energy is currently considered the leading solution to reduce greenhouse gas emissions. Energy extraction from the ocean tides (tidal turbines) can help fulfill the global renewable energy demand and combat world climate crises. Installations at this scale will have the associated benefit of reducing the levelised cost of tidal energy (LCOE) towards a target of EUR100/MWh, which will make ocean energy a viable option along with other renewable energy sources such as offshore wind. To achieve the target, increased performance and reliability of tidal energy devices are required. Tidal blades are a primary component of tidal turbines, possess a heterogeneous nature and can suffer from complex non-linear damage modes. For example, harsh marine environments can cause impact damage, delamination, matrix crack, fiber breakage or rupture, and others in tidal blades, which could lead to catastrophic failure of the system. Achieving reliable operational health and performance for the tidal blade is thus crucial for tidal energy companies. Fault diagnoses and maintenance operations are challenging in the sea; performance degradation, failure, or breakdown of the entire tidal energy system are more likely if unattended. Therefore, there is the need for real-time and reliable structure health monitoring (SHM) of tidal blades. The existing damage detection techniques have a limitation when dealing with the real-time environment, and do not take into account the along with uncertainty feature, we also addressed the issue of trusthworthiness in system decision employed with explanable artificial intelligence (XAI). This paper presents a real-time damage detection framework, information communication technologies (ICT) based infrastructure for real-time monitoring and proposes a novel model to classify/ detect the damages over blade structure. In addition a XAI based approach is proposed which based on supervised machine learning (ML) and uses an optimized convolutional neural network to classify from the heterogeneous data streams. Testing and evaluation of proposed approach in laboratory and operational settings is the future concern of this study.

Application of inertia emulation control strategy with energy storage system in multi-area hydro -thermal system using a novel metaheuristic optimized tilt controller

This article presents the application of a novel artificial hummingbirds’ algorithm (AHA) based tilt-integral second order of double-derivative controller (TIDD2) in a three-area hydro-thermal power system. Also, the inertia emulation control strategy has been incorporated using various energy storage systems (ESS) to established the frequency in interconnected power system. The supremacy of the proposed controller is established by comparing the system dynamic response with some existing controllers, such as proportional-integral-derivative (PID), tilt-integral-derivative (TID) and cascade of IDD - PID controllers. The fruitfulness of proposed AHA has also been evaluated with system dynamic and convergence characteristics comparison over some of existing metaheuristic optimization techniques such as harris hawk's optimization (HHO) and bird swarm algorithm (BSA). Further, effect of random load pattern (RLP), communication time delay (CTD) and variable CTD on the system dynamic is analyzed. The study of inertia emulation control strategy in a multi-area interconnected power system revealed that, when inertia is emulated with superconducting magnetic energy storage (SMES), the system dynamic exhibits better performance over battery energy storage (BES) and ultra-capacitor (UC). Furthermore, the proposed systems are validated in a real-time environment using OPAL-RT lab software. Also, the scalability of the proposed method is evaluated by using eigenvalue analysis.

Fast move: A prioritized vehicle rerouting strategy in smart city

Efficient and dynamic traffic management is an integral part of any smart city. Finding the shortest route considering congestion in real-time and managing traffic signals determine the efficiency of overall traffic flow. Route planning within a dynamic traffic environment remains an open challenge due to some essential unaddressed issues. Specially, considering priority vehicles to provide better traffic flow. For example, the routes of the vehicles need to be derived in such a way that higher priority vehicles like ambulances, police cars, etc. are provided efficient routes and green signals as per their given priority. To establish the solution we need to consider two factors. Vehicle rerouting to obtain shorter path and traffic signal optimization for lesser waiting time by a vehicle on a traffic signal. Existing works don't consider both factors in their work and tried to solve the issue reactively; upon request for clearance by a vehicle on a particular traffic signal. The proposed work seems to be a proactive approach that considers both factors. In this work, a dynamic real-time traffic environment with prioritized vehicles is modeled as a bi-level optimization problem. The solution ensures that the routes are offered to the vehicles in a way such that the unit travel time of a vehicle with higher priority must be less compared to vehicles with lesser priority. In addition, the traffic signal settings are optimized in a coordinated way to achieve the above-mentioned goal. Rigorous experimentation is done for validation of the proposed solution. Several traffic situations consisting of inherent dynamism are simulated through SUMO and the experimental findings substantiate our claims in terms of various traffic metrics.

Inside and outside the neighborhood: Short-term and long-term subjective well-being by geographical context.

The value of linking urban environment and subjective well-being (SWB) is now well recognized. But whether the geographical context inside and outside the neighborhood has differential influence on long- and short-term SWB remains unclear. Based on the activity perspective, we used survey data from Guangzhou, China, integrating GPS data, portable environmental sensors data to analyze time-weighted and real-time geographical context inside and outside the neighborhood on long- and short-term SWB. The results show that SWB is not only influenced by the neighborhood environment, but also the geographical context outside the neighborhood. Time-weighted geographical environment inside the neighborhood has a higher impact and explanatory ability on long-term SWB, while real-time geographical environment outside the neighborhood has a higher impact and explanatory ability on short-term SWB. This study provides a new understanding for geographies of SWB through the extension of time and space, and also provides reference for more refined urban planning and governance in the future.

TCMQMS System: A digital twin and blockchain-based platform for tracing the whole process of Chinese medicine quality information

Traditional Chinese medicine is an important medicine and health food in China. In the process of making Chinese medicines, the main active ingredients of the medicines are highly dependent on the production environment due to the special physical characteristics of the medicines themselves. Any slight change in the environment can lead to changes in the active ingredients as well as the final efficacy, even affecting the patient treatment cycle, and this characteristic is not conducive to quality control by national drug quality regulators. In order to enhance the quality management of TCM, a blockchain and digital twin based whole process monitoring system for TCM quality is proposed: Traditional Chinese Medicine Quality Monitoring System (TCMQMS), in order to achieve a whole process, full scope and highly transparent TCM quality monitoring model for TCM production from the source of cultivation to the final patient. And. A blockchain platform based on Fabric blockchain data development platform as well as Sia distributed data storage was designed.
 With the help of Sia distributed storage technology, the amount of data storage is significantly compressed, which achieves the purpose of unifying the huge information data flow into the TCMQMS system and quickly building the system platform using the Fabric platform. In addition, in order to ensure that data is read and written in real time during the manufacturing and transportation of Chinese medicine, environmental sensors, such as temperature and humidity sensors, are placed in the Chinese medicine processing plant and in the transport sector. The environmental sensors, such as temperature and humidity sensors, are placed inside the Chinese medicine processing plants and transport vehicles. The environmental data is combined with the production data recorded by the TCMQMS system and sent to the digital twin TCM manufacturing simulation environment in real time, and through the simulation of the TCM manufacturing environment, the TCM quality environment parameters are continuously iteratively updated to be more suitable for production. This is coupled with real-time linkage with environmental sensors to achieve real-time adjustment of environmental parameters. In addition, the TCMQMS system records the entire process data, giving each data block a unique and tamper-proof hash value and CA certificate, so that when there is a possible quality problem, the hash value and CA certificate will be used to quickly trace the person responsible and the details of the environment in which the TCM was produced in real time. Finally, we present a real-life example of the use of the TCMQMS system to validate our proposed approach.

Dental radiology: a convolutional neural network-based approach to detect dental disorders from dental images in a real-time environment

Dental radiology: a convolutional neural network-based approach to detect dental disorders from dental images in a real-time environment

Real-Time Fixed Priority Scheduling Synthesis using Affine DataFlow Graphs: from Theory to Practice

The major drawback of using static schedules to execute dataflow applications is their high inflexibility. In real-time systems, periodic schedules make it easier to assert safety guarantees and to decrease the schedule size, but their characteristics remain hard to compute. This article presents an approach to automatically generate fixed priority schedules from a dataflow specification. To do so, precedence dependencies between actors in the dataflow graphs are abstracted, as well as the task periods, by using affine relations . This abstraction allows us to synthesize schedules efficiently considering two main objectives: the maximization of throughput and the minimization of buffer sizes. Given a dataflow graph to execute in a real-time environment, we transform it into an Affine DataFlow Graph (ADFG) and compute the task priorities, their mapping, the number of delays in the buffers, and the buffer sizes. This article is the first to present an overview of both theoretical and practical aspects of ADFG. On the theoretical side, it presents corrections and improvements on the fixed priority case. On the practical side, benchmark evaluations demonstrate the robustness and maturity of the approach that our scheduling synthesizer implements. Synthesized schedules are evaluated by using scheduling simulation and real-time implementation. Last but not least, the synthesized periods reach the optimal throughput if enough processors are available, and most of the time the periods reach the maximal processor utilization factor in the uni-processor case. Moreover, execution time of the synthesis is about only one second for the main proposed algorithms.

Two-layer coordinated reinforcement learning for traffic signal control in traffic network

Intersection traffic signal control considering vehicle emissions has become an important topic, however, the decision complexity of traffic signal control increases dramatically in a dynamic traffic environment with multi-intersections. It is a severe challenge to coordinate traffic signals at multi-intersections based on Internet of Things information to improve the traffic condition of the road network. This paper proposes a two-layer coordination algorithm based on multi-agent reinforcement learning—Multi-agent Coordinated Policy Optimization (MACoPO), for solving traffic signal control at multi-intersections. MACoPO consists of local cooperation, which adjusts the weights of individual rewards and neighborhood agents' rewards by using local cooperation factors (LCF), and global coordination, which updates the LCF to maximize global rewards. The state and reward functions are designed in terms of the current state of the signal, waiting queue length, vehicle density and emission concentration in the lane, vehicle delay, and vehicle emissions, thus making full use of the intersection state information. The proposed method is extensively assessed through simulation experiments using artificial and real road networks and the numerical results confirm its effectiveness in complex and dynamic real-time traffic environments with multi-intersections.

Performance evaluation and extraction of global power maxima under shading scenarios and load Variation: An experimental study

Due to environmental degradation, depletion of fossil resources and contaminants in the air, energy structure reform and development of sustainable environment friendly are approaching in the past few years, in which several renewable energy technologies such as solar, wind, bio and hydropower as efficient alternative energy have drawn considerable attention. Solar Photovoltaic (SPV) systems, in particular, have huge benefits for society and the economy as an essential means of extracting and utilization. However, these systems suffer from extracting maximum power from variable irradiance due to shading scenarios and variable load. As a result of PSCs, many LMPPs and GMPPs exist on their power voltage (P-V) characteristics. Thus, MPTT techniques are implemented in these systems to resolve these issues. Several researchers have conducted substantial research and discussion on these techniques to improve SPV system performances, but the practical applicability of MPPT approaches faces multiple obstacles. This study evaluates the performances of three such MPPT metaheuristic approaches, i.e., umbrella optimization technique (UOT), particle swarm optimization (PSO) and teaching learning-based optimization (TLBO) experimentally in real-time environment on small hardware under PSCs and load variation scenarios. Each technique is tested on a standalone SPV system incorporating an Inverse Single-Ended Primary Inductor Converter (SEPIC) and Luo DC-DC converter separately on various parameters of SPV systems such as tracking time, tracking efficiency, output power, output current and computational complexity, etc. With inverse SEPIC DC converter, UOT boosts GMPP up to 3.8% to 3.9% while the tracking time is reduced or faster by 20.11% to 27.7% compared to PSO & TLBO techniques. It also shows 98.6% average tracking efficiency under considered PSCs. On the other hand, UOT achieves 4.22% to 5.5% higher GMPP and tracking time is reduced by 38.38% to 57.87% with the Luo DC converter compared to PSO & TLBO techniques. UOT shows 98.19% average tracking efficiency under considered PSCs with this converter. Moreover, UOT can also maintain 3.80% to 6.48% high output current in the SPV system under varying load conditions, with stabilization time reduced to 34.06% to 35.80% compared to PSO & TLBO techniques. The further practical applicability of the study is tested on a 300 W standalone SPV system. On this system, UOT again shows its better tracking performance by boosting 3.5% to 3.64% GMPP as well as the system output current. It also takes 30.86% less tracking time with 98.56% average tracking efficiency compared to PSO and TLBO. Experimental validation of all three metaheuristic approaches on low and high-rated systems confirms the supremacy of UOT over PSO & TLBO techniques in terms of tracking time, tracking efficiency, output power and current under different working scenarios.

Towards the Cognitive Factory in Industry 5.0: From Concept to Implementation

Industry 5.0 (I5.0) represents a shift towards a human-centered industry and emphasizes the integration of human and machine capabilities. A highly compatible concept for enabling the I5.0 implementation is intelligent spaces (ISs), i.e., physical spaces equipped with a network of sensors, which obtains information about the place it observes, and a network of actuators, which enables changes in the environment through computing services. These spaces can sense, interpret, recognize user behavior, adapt to preferences, and provide natural interactions between humans and intelligent systems, using the IoT, AI, computer vision, data analytics, etc., to create dynamic and adaptive environments in real time. The integration of ISs and I5.0 has paved the way for the development of cognitive factories, which transform industrial environments into ISs. In this context, this article explores the convergence of IS and I5.0 concepts and aims to provide insights into the technical implementation challenges of cognitive factories. It discusses the development and implementation of a laboratory replica of a cognitive cell as an example of a segment of a cognitive factory. By analyzing the key points and challenges associated with cognitive cell implementation, this article contributes to the knowledge base surrounding the advanced manufacturing paradigm of I5.0.

A Smart Control System for the Oil Industry Using Text-to-Speech Synthesis Based on IIoT

Oil refineries have high operating expenses and are often exposed to increased asset integrity risks and functional failure. Real-time monitoring of their operations has always been critical to ensuring safety and efficiency. We proposed a novel Industrial Internet of Things (IIoT) design that employs a text-to-speech synthesizer (TTS) based on neural networks to build an intelligent extension control system. We enhanced a TTS model to achieve high inference speed by employing HiFi-GAN V3 vocoder in the acoustic model FastSpeech 2. We experimented with our system on a low resources-embedded system in a real-time environment. Moreover, we customized the TTS model to generate two target speakers (female and male) using a small dataset. We performed an ablation analysis by conducting experiments to evaluate the performance of our design (IoT connectivity, memory usage, inference speed, and output speech quality). The results demonstrated that our system Real-Time Factor (RTF) is 6.4 (without deploying the cache mechanism, which is a technique to call the previously synthesized speech sentences in our system memory). Using the cache mechanism, our proposed model successfully runs on a low-resource computational device with real-time speed (RTF equals 0.16, 0.19, and 0.29 when the memory has 250, 500, and 1000 WAV files, respectively). Additionally, applying the cache mechanism has reduced memory usage percentage from 16.3% (for synthesizing a sentence of ten seconds) to 6.3%. Furthermore, according to the objective speech quality evaluation, our TTS model is superior to the baseline TTS model.

Fast retrieval of fMRI data for real-time applications: improving the transfer time through direct connection

One of the significant challenges in real-time fMRI environments is to ensure that the functional images are exported in real-time. The prerequired ability to reconstruct these images immediately after the acquisition has already been resolved in 2004. Nowadays, more sophisticated sequences allow for higher resolution and faster repetition times and thereby challenging the ability to export this data in real-time. In this article, we tackle the potentially arising problem of sending the reconstructed data from the MRI to an external PC to perform the real-time fMRI analysis. We show that depending on the implementation of the data transfer, long delays can occur that can differ drastically in time and how often they occur. In addition, we propose a solution for SIEMENS MRI devices which was tested and applied already on multiple MRI devices including 3T and 7T machines on different vendor software versions. This new technique can be used as a blueprint that can be directly applied to other manufacturers. We also provide the source code of the described solution and show that the delay in the data transfer can be significantly reduced to a tolerable level using our proposed procedure. Finally, we integrate measurement options for the data transfer times to improve quality measures in (clinical) real-time fMRI environments that can implement the proposed solution. Efforts should be taken by the real-time fMRI community and MRI manufacturers to employ a standardized real-time export e.g., similar to the lab streaming layer which is used as a standard export method in EEG environments.