Multi-Fidelity High-Throughput Optimization of Electrical Conductivity in P3HT-CNT Composites

Combining high-throughput experiments with machine learning accelerates materials and process optimization toward user-specified target properties. In this study, a rapid machine learning-driven automated flow mixing setup with a high-throughput drop-casting system is introduced for thin film preparation, followed by fast characterization of proxy optical and target electrical properties that completes one cycle of learning with 160 unique samples in a single day, a > 10 ×  improvement relative to quantified, manual-controlled baseline. Regio-regular poly-3-hexylthiophene is combined with various types of carbon nanotubes, to identify the optimum composition and synthesis conditions to realize electrical conductivities as high as state-of-the-art 1000 S cm⁻¹. The results are subsequently verified and explained using offline high-fidelity experiments. Graph-based model selection strategies with classical regression that optimize among multi-fidelity noisy input-output measurements are introduced. These strategies present a robust machine-learning driven high-throughput experimental scheme that can be effectively applied to understand, optimize, and design new materials and composites.     

A Robust Wearable Point-of-Care CNT-Based Strain Sensor for Wirelessly Monitoring Throat-Related Illnesses

Point-of-care testing (POC) has the ability to detect chronic and infectious diseases early or at the time of occurrence and provide a state-of-the-art personalized healthcare system. Recently, wearable and flexible sensors have been employed to analyze sweat, glucose, blood, and human skin conditions. However, a flexible sensing system that allows for the real-time monitoring of throat-related illnesses, such as salivary parotid gland swelling caused by flu and mumps, is necessary. Here, for the first time, a wearable, highly flexible, and stretchable piezoresistive sensing patch based on carbon nanotubes (CNTs) is reported, which can record muscle expansion or relaxation in real-time, and thus act as a next-generation POC sensor. The patch offers an excellent gauge factor for in-plane stretching and spatial expansion with low hysteresis. The actual extent of muscle expansion is calculated and the gauge factor for applications entailing volumetric deformations is redefined. Additionally, a bluetooth-low-energy system that tracks muscle activity in real-time and transmits the output signals wirelessly to a smartphone app is utilized. Numerical calculations verify that the low stress and strain lead to excellent mechanical reliability and repeatability. Finally, a dummy muscle is inflated using a pneumatic-based actuator to demonstrate the application of the affixed wearable next-generation POC sensor.     

A closed-loop healthcare processing approach based on deep reinforcement learning

Multimedia Tools and Applications - In healthcare, the human body is a controlled input-output system, which generates different observations with the variations of external interventions. The...     

Speech emotion recognition using optimized genetic algorithm-extreme learning machine

Multimedia Tools and Applications - Automatic Emotion Speech Recognition (ESR) is considered as an active research field in the Human-Computer Interface (HCI). Typically, the ESR system is...     

Emulating Software Defined Network using Mininet-ns3-WIFI Integration for Wireless Networks

SDN enables a new networking paradigm probable to improve system efficiency where complex networks are easily managed and controlled. SDN allows network virtualization and advance programmability for customizing the behaviour of networking devices with user defined features even at run time. SDN separates network control and data planes. Intelligently controlled network management and operation, such that routing is eliminated from forwarding elements (switches) while shifting the routing logic in a centralized module named SDN Controller. Mininet is Linux based network emulator which is cost effective for implementing SDN having in built support of OpenFlow switches. This paper presents practical implementation of Mininet with ns-3 using Wi-Fi. Previous results reported in literature were limited upto 512 nodes in Mininet. Tests are conducted in Mininet by varying number of nodes in two distinct scenarios based on scalability and resource capabilities of the host system. We presented a low cost and reliable method allowing scalability with authenticity of results in real time environment. Simulation results show a marked improvement in time required for creating a topology designed for 3 nodes with powerful resources i.e. only 0.077 sec and 4.512 sec with limited resources, however with 2047 nodes required time is 1623.547 sec for powerful resources and 4615.115 sec with less capable resources respectively.

     

DLSA: Delay and Link Stability Aware Routing Protocol for Flying Ad-hoc Networks (FANETs)

Flying Ad-hoc Network (FANET) is a new class of Mobile Ad-hoc Network in which the nodes move in three-dimensional (3-D) ways in the air simultaneously. These nodes are known as Unmanned Aerial Vehicles (UAVs) that are operated live remotely or by the pre-defined mechanism which involves no human personnel. Due to the high mobility of nodes and dynamic topology, link stability is a research challenge in FANET. From this viewpoint, recent research has focused on link stability with the highest threshold value by maximizing Packet Delivery Ratio and minimizing End-to-End Delay. In this paper, a hybrid scheme named Delay and Link Stability Aware (DLSA) routing scheme has been proposed with the contrast of Distributed Priority Tree-based Routing and Link Stability Estimation-based Routing FANET’s existing routing schemes. Unlike existing schemes, the proposed scheme possesses the features of collaborative data forwarding and link stability. The simulation results have shown the improved performance of the proposed DLSA routing protocol in contrast to the selected existing ones DPTR and LEPR in terms of E2ED, PDR, Network Lifetime, and Transmission Loss. The Average E2ED in milliseconds of DLSA was measured 0.457 while DPTR was 1.492 and LEPR was 1.006. Similarly, the Average PDR in %age of DLSA measured 3.106 while DPTR was 2.303 and LEPR was 0.682. The average Network Lifetime of DLSA measured 62.141 while DPTR was 23.026 and LEPR was 27.298. At finally, the Average Transmission Loss in dBm of DLSA measured 0.975 while DPTR was 1.053 and LEPR was 1.227.

     

Optimal path planning in cluttered environment using RRT*-AB

Rapidly exploring Random Tree Star (RRT*) has gained popularity due to its support for complex and high-dimensional problems. Its numerous applications in path planning have made it an active area of research. Although it ensures probabilistic completeness and asymptotic optimality, its slow convergence rate and large dense sampling space are proven problems. In this paper, an off-line planning algorithm based on RRT* named RRT*-adjustable bounds (RRT*-AB) is proposed to resolve these issues. The proposed approach rapidly targets the goal region with improved computational efficiency. Desired objectives are achieved through three novel strategies, i.e., connectivity region, goal-biased bounded sampling, and path optimization. Goal-biased bounded sampling is performed within boundary of connectivity region to find the initial path. Connectivity region is flexible enough to grow for complex environment. Once path is found, it is optimized gradually using node rejection and concentrated bounded sampling. Final path is further improved using global pruning to erode extra nodes. Robustness and efficiency of proposed algorithm is tested through experiments in different structured and unstructured environments cluttered with obstacles including narrow and complex maze cases. The proposed approach converges to shorter path with reduced time and memory requirements than conventional RRT* methods.     

TIE algorithm: a layer over clustering-based taxonomy generation for handling evolving data

Taxonomy is generated to effectively organize and access large volume of data. A taxonomy is a way of representing concepts that exist in data. It needs to continuously evolve to reflect changes in data. Existing automatic taxonomy generation techniques do not handle the evolution of data; therefore, the generated taxonomies do not truly represent the data. The evolution of data can be handled by either regenerating taxonomy from scratch, or allowing taxonomy to incrementally evolve whenever changes occur in the data. The former approach is not economical in terms of time and resources. A taxonomy incremental evolution (TIE) algorithm, as proposed, is a novel attempt to handle the data that evolve in time. It serves as a layer over an existing clustering-based taxonomy generation technique and allows an existing taxonomy to incrementally evolve. The algorithm was evaluated in research articles selected from the computing domain. It was found that the taxonomy using the algorithm that evolved with data needed considerably shorter time, and had better quality per unit time as compared to the taxonomy regenerated from scratch.     

Forecasting wheat yield from weather data and MODIS NDVI using Random Forests for Punjab province,Pakistan

Wheat is the staple food of Punjab province of Pakistan, which contributes more than 75% of the total national production. Accurate and timely forecasting of wheat yield is a cornerstone for monitoring food security and planning for agricultural markets, but the efficiency of the current system for near real-time forecasting should be improved. In this research paper, we developed a model to forecast wheat yield before harvest for each of eight individual districts and for Punjab province as a whole. The model uses weather and Moderate Resolution Imaging Spectroradiometer (MODIS)-derived normalized difference vegetation index (NDVI) data for 2001–2014 (14 years) to calculate Random Forest (RF) statistical models using 15 independent variables. Temperature, rainfall, sunshine hours, growing degree days, and MODIS-derived NDVI for each of the eight districts of Punjab province were used to forecast yield for the year 2014. The same independent variables were used to forecast wheat yield of the whole Punjab from 2001 to 2014 by excluding the respective year from training set. Sunshine hour data were not available for all districts and therefore we tested using temperature data and average latitude-based solar radiation as surrogates. The root mean square errors (RMSEs) of the forecast results of the whole of Punjab province were 147.7 kg ha^?1 and 148.7 kg ha^?1 with a mean error of less than 5% using average and generic RFs, respectively. Forecasts for individual districts showed R² of 0.95 with RMSE of 175.6 kg ha^?1 and 5.86% mean error.