Mobile edge computing (MEC) network provides near-users computing and communication functions and has become a potential 5G evolutionary architecture. In order to overcome the shortcomings of the ...existing MEC network in fixed base stations and limited computing resources, unmanned arial vehicle (UAV) is introduced as a relay edge computing node and UAV-enabled MEC networks are proposed. However, UAVs have limited energy. Thus, energy consumption would be an optimal target during the information interaction. Therefore, an energy efficiency optimization algorithm based on a three-layer computation offloading strategy is proposed in this paper by combining the UAV position optimization algorithm and the LSTM-based task prediction algorithm. The experiments show that the computation offloading strategy of the UAV-enabled MEC network can be dynamically programmed with the proposed algorithm and architecture, according to the required delay, UAV height, and data size in order to effectively reduce the energy consumption of the UAV.
Precision farming is made possible by rapid advances in deep learning (DL) and the internet of things (IoT) for agriculture, allowing farmers to upgrade their agriculture operations to sustainably ...fulfill the future food supply. This paper presents a comprehensive overview of recent research contributions in DL and IoT for precision agriculture. This paper surveys the diverse research on DL applications in agriculture, such as detecting pests, disease, yield, weeds, and soil, including fundamental DL techniques. Also, the work describes the IoT architecture and analyzes sensor categorization, agriculture sensors, and unmanned arial vehicles (UAVs) used in recent research. Besides that, data acquisition, annotation, and augmentation for agriculture datasets were covered, and a few widely used datasets were listed. This work also discusses some challenges and issues that DL and IoT face. Furthermore, the research proposed a bootstrapping approach of Transfer learning where fine-tuned VGG16 is fused with optimized and improved newly built fully connected layers for pest detection. The performance of the proposed model is evaluated and compared with other models, such as custom VGG16 as a classifier; fine-tuned VGG16 is optimized with other optimizers like SGD, RMSProp, and Adam. The results show that the proposed model for pest detection outperforms all other models with an accuracy of 96.58 % and a loss of 0.15%. The review and the proposed work presented in this paper will significantly direct researchers toward DL and IoT for intelligent farming.
•Precision agriculture is made possible by combining recent advances in Deep Learning (DL) and Internet of Things (IoT).•DL applications in precision agriculture were discussed such as detection of pest/disease, soil, yield, and more.•IoT architecture, smart devices like sensors and UAVs for smart agriculture were discussed.•Data set study covers data acquisition, augmentation, annotation, and also identifies benchmark dataset for smart agriculture.•The paper proposed a bootstrap approach, where fine-tuned VGG16 is fused with improved dense layers for plant pest detection.
Highway asset management requires capturing the highway's status. However, the onsite survey of the highway is very costly and time-consuming. This paper presents a novel approach for creating the ...digital twin of a highway using map data. The digital twin consists of primary highway components, including horizontal alignment, vertical alignment, cross-section, lanes and central reserves. It follows the engineering representation of a highway, which has excellent potential for further application in the field. The proposed approach was tested in a section of the A1(M) motorway in the UK. It requires minimum human input and has very high accuracy. Despite many outliers in the collected map data, the average vertical deviation per square metre between the surface of the generated digital twin and the actual data was at the centimetre level.
•A systematic approach is proposed for making the digital twin of a highway based on map data and engineering expertise.•Spatial data for creating the digital twin can be extracted from low-resolution online map data using the proposed method.•An algorithm is developed to fit the horizontal and vertical alignment of a highway.•Outliers in the spatial data of a highway can be identified and removed using the proposed method.•Cross-sections and the overall digital twin of a highway are generated considering various types of highway components.
•Remote sensing can provide relevant data on forest structure that influences species abundance.•Fecal counts linked to LiDAR or DAP derived forest metrics can provide predictive maps of density.•Our ...landscape-scale density model performed well on spatially-independent validation data.•Density is most affected by forest metrics related to stand age, ground cover, and canopy cover.•Our snowshoe hare model helped predict lynx habitat use, providing an important management tool.
Landscape-scale predictions of species abundance or density are of fundamental importance to conservation and management of ecosystems. Yet, developing these models remains challenging, as they require linking broad-scale population data with habitat characteristics that influence species abundance. Advances in remote sensing technology have resulted in increased availability of spatially continuous, high-resolution data that relate to ecologically important habitat characteristics. In forested systems, Light Detection and Ranging (LiDAR) and Digital Arial Photogrammetry (DAP) are of particular interest owing to their ability to estimate vegetative structure that drives variability in abundance or density of some forest-dependent species. We used an extensive dataset on the density of a keystone boreal forest species, the snowshoe hare (Lepus americanus) in northcentral Washington, USA, to examine which LiDAR- and DAP-derived habitat variables most strongly influence snowshoe hare density, and projected these relationships across the landscape to derive a hare density surface for our 53 km2 study area. We found snowshoe hare density is most influenced by habitat variables related to tree height (a proxy for stand age), horizontal cover, and vertical cover, and our model had high predictive performance on a spatially-independent validation dataset. Hare densities increased as horizontal cover and canopy cover increased, with our highest hare densities occurring in areas with >9% horizontal cover (% of LiDAR returns in 1–4 m height stratum), >65% canopy cover and tree height (a proxy for stand age) of ~5–10 m. To demonstrate the management implications of this work, we show that our landscape-scale model of predicted hare density helps understand habitat use by threatened Canada lynx (Lynx canadensis), a primary predator of hare. Our results show how coupling population data with remotely sensed forest structure metrics allows for continuous, large-scale population estimates. Such integration provides an important management tool for examining spatiotemporal changes in populations as boreal ecosystems come under increasing stress from climate and land use change.
Recently, UAVs or Unnamed Aerial Vehicles have been proposed as flexible aerial support to assist ground vehicles for different applications such as rescue and traffic surveillance missions. UAVs can ...collect different data information about the road/traffic state usually as aerial photography and videos. The processing of this kind of data consists usually on pattern recognition and video processing which are complex tasks that necessitate powerful computing and energy resources. Unfortunately, the moderate UAV's computational and energy capabilities restrict local data processing. Fortunately, UAVs can leverage the computation resources of the surrounding edge network entities to enhance their computational capabilities. In this paper, we aim to achieve efficient data processing for the data collected by UAVs in the context of UAVs-aided vehicular networks for traffic monitoring missions. For this purpose, we propose a new system model where UAVs can offload and/or share intensive computation tasks with other nearby network nodes. Then, we use the computation response time, the energy consumed for the computation, the cost of cellular communication and the computation cost as the main system metrics to make any computation offloading/sharing decisions that optimize the system performance. We then modele the offloading/sharing decision-making problem as a sequential game, where we provide complete proof of the existence of the Nash equilibrium and propose an algorithm to reach such an equilibrium. The simulation results showed that the proposed game-based model outperforms other approaches by delivering better performance in terms of overall system utility with a data processing efficiency that varies between 43% and 97% depending on the computation approach, and provides a more efficient computation time and energy average.
MicroRNAs (miRNAs) are associated with cardiovascular disease and control gene expression and are detectable in the circulation.
The purpose of this study was to test the hypothesis that circulating ...miRNAs may be associated with atrial fibrillation (AF).
Using a prospective study design powered to detect subtle differences in miRNAs, we quantified plasma expression of 86 miRNAs by high-throughput quantitative reverse transcriptase-polymerase chain reaction in 112 participants with AF and 99 without AF. To examine parallels between cardiac and plasma miRNA profiles, we quantified atrial tissue and plasma miRNA expression using quantitative reverse transcriptase-polymerase chain reaction in 31 participants undergoing surgery. We also explored the hypothesis that lower AF burden after ablation would be reflected in the circulating blood pool by examining change in plasma miRNAs after AF ablation (n = 47).
Mean age of the cohort was 59 years; 58% of participants were men. Plasma miRs-21 and 150 were 2-fold lower in participants with AF than in those without AF after adjustment (P ≤.0006). Plasma levels of miRs-21 and 150 also were lower in participants with paroxysmal AF than in those with persistent AF (P <.05). Expression of miR-21, but not of miR-150, was lower in atrial tissue from patients with AF than in those without AF (P <.05). Plasma levels of miRs-21 and 150 increased 3-fold after AF ablation (P ≤.0006).
Cardiac miRs-21 and 150 are known to regulate genes implicated in atrial remodeling. Our findings show associations between plasma miRs-21 and 150 and AF, suggesting that circulating miRNAs can provide insights into cardiac gene regulation.
This paper deals with the motion of a hovering UAV with six degrees of freedom. The effects of the motion on the measured signal of a MIMO radar mounted underneath the UAV are analyzed. For each ...degree of freedom, namely the translation in x-, y- and z-direction as well as the rotation around x-, y- and z-axis, an algorithm is proposed to compensate the considered motion. The effectiveness of the proposed algorithms is demonstrated by measuring validated vital signs independent of the current UAV motion.
Applying Compressive Sensing (CS) to Received Signal Strength (RSS) based multi-emitter localization using Unmanned Arial Vehicles (UAVs) attracts much attention for its simplicity and efficiency. ...However, the RSS-based CS approach is vulnerable to the noise in a practical scenario. To mitigate this, we propose a robust localization framework for multiple emitters in UAV-based Wireless Sensor Network (WSN). We first approximate the lognormal noise's influence on the dictionary by a two-layer hierarchical prior model. Then, by exploiting multi-frequency measurements, the multi-emitter localization is transformed into the joint estimation for multiple sparse vectors and noise level. Finally, the joint estimation problem is solved by a Concurrent Variational Bayesian Inference (CVBI) algorithm, where an adaptive grid pruning mechanism is designed. The merits of the proposed framework are testified by numerical simulations.