Design of a wideband microstrip antenna and the use of artificial neural networks in parameter calculation

This paper deals with the design of a multi-slot hole-coupled microstrip antenna on a substrate of 2 mm thickness that gives multi-frequency (wideband) characteristics. The Method of Moments (MoM)-based IE3D software was used to simulate the results for return loss, VSWR, the Smith chart, and the radiation patterns. A tunnel-based artificial neural network (ANN) was also developed to calculate the radiation patterns of the antenna. The radiation patterns were measured experimentally at 10.5 GHz and 12 GHz. The experimental results were in good agreement with the simulated results from IE3D and those of the artificial neural network. A new method of using a genetic algorithm (GA) in an artificial neural network is also discussed. This new method was used to calculate the resonant frequency of a single-shorting-post microstrip antenna. The resonant frequency calculated using the genetic-algorithm-coupled artificial neural network was compared with the analytical and experimental results. The results obtained were in very good agreement with the experimental results.     

QOS aware channel and power allocation scheme for D2D enabled cellular networks

Telecommunication Systems - Device-to-device (D2D) communication enabled cellular system is capable of enhancing the spectrum utilization and throughput performance of the system. However, D2D...     

Influence of storage temperature on the quality of salted mackerel (Rastrelliger kangurta) and pink perch (Nemipterus japonicus)

Dry-salted mackerel and pink perch were stored at two temperatures: ambient (26·8 ± 3·3°C) and 2·5 ± 1°C. Changes in moisture content, salt content, water activity (a_w), peroxide value (PV), free fatty acid content (FFA), total volatile base nitrogen (TVBN) content, halophilic bacterial count and sensory scores for overall acceptability were studied. Loss of moisture and absorption of salt were considerably higher in the products stored at ambient temperature. The decrease in a_w was more pronounced at ambient temperature than at the lower temperature. Although the chemical indices of freshness (PV, FFA and TVBN) and the halophilic counts showed increasing trends, they were considerably lower in the products stored at the lower temperature. Sensory evaluation for overall acceptability indicated that storage at the lower temperature could considerably extend the shelf-life of salted fish.     

Fabrication of composites from different carbon content ferrous powders adopting thermomechanical processes

In this research, deep consolidation technique is utilized for processing of low and high carbon content ferrous powders to manufacture composite material. High carbon content ferrous powders were used as reinforcing material. The powder particles were allowed to mix with low carbon content ferrous powders in 50:50 weight percent for fabrication of composites. The composites were heat treated at three different temperatures (i.e., 800°C, 900°C, and 1000°C). The changes in microstructure, microhardness, the grain size, and bonding among powder particles were studied. Refinement of grains was observed and this led to improved hardness enabling the processed material to be used as a suitable composite. Abrasive wear tests were carried out using a laboratory tribometer in dry reciprocating sliding contact against Sic abrasive paper. The outcomes demonstrate that the abrasive wear resistance was notably affected by the treatment temperature and hardness of the composites.     

A multivariate framework to study spatio‐temporal dependency of electricity load and wind power

With massive wind power integration, the spatial distribution of electricity load centers and wind power plants make it plausible to study the inter‐spatial dependence and temporal correlation for the effective working of the power system. In this paper, a novel multivariate framework is developed to study the spatio‐temporal dependency using vine copula. Hourly resolution of load and wind power data obtained from a US regional transmission operator spanning 3 years and spatially distributed in 19 load and two wind power zones are considered in this study. Data collection, in terms of dimension, tends to increase in future, and to tackle this high‐dimensional data, a reproducible sampling algorithm using vine copula is developed. The sampling algorithm employs k ‐means clustering along with singular value decomposition technique to ease the computational burden. Selection of appropriate clustering technique and copula family is realized by the goodness of clustering and goodness of fit tests. The paper concludes with a discussion on the importance of spatio‐temporal modeling of load and wind power and the advantage of the proposed multivariate sampling algorithm using vine copula.     

An efficient actor‐critic reinforcement learning for device‐to‐device communication underlaying sectored cellular network

In this paper, a novel reinforcement learning (RL) approach with cell sectoring is proposed to solve the channel and power allocation issue for a device‐to‐device (D2D)‐enabled cellular network when the prior traffic information is not known to the base station (BS). Further, this paper explores an optimal policy for resource and power allocation between users intending to maximize the sum‐rate of the overall system. Since the behavior of wireless channel and traffic request of users in the system is stochastic in nature, the dynamic property of the environment allows us to employ an actor‐critic RL technique to learn the best policy through continuous interaction with the surrounding. The proposed work comprises of four phases: cell splitting, clustering, queuing model, and channel allocation and power allocation simultaneously using an actor‐critic RL. The implementation of cell splitting with novel clustering technique increases the network coverage, reduces co‐channel cell interference, and minimizes the transmission power of nodes, whereas the queuing model solves the issue of waiting time for users in a priority‐based data transmission. With the help of continuous state‐action space, the actor‐critic RL algorithm based on policy gradient improves the overall system sum‐rate as well as the D2D throughput. The actor adopts a parameter‐based stochastic policy for giving continuous action while the critic estimates the policy and criticizes the actor for the action. This reduces the high variance of the policy gradient. Through numerical simulations, the benefit of our resource sharing scheme over other existing traditional scheme is verified.     

Is a picture worth a thousand views? Measuring the effects of travel photos on user engagement using deep learning algorithms

Electronic Markets - Travel photos inform and inspire consumers by conveying a first-hand destination experience. Despite the proliferation of consumer-generated travel photos in online travel...