Search

With the advent of millimeter wave (mmWave) communications, the combination of a detailed 5G network simulator with an accurate antenna radiation model is required to analyze the realistic performance of complex cellular scenarios. However, due to the complexity of both electromagnetic and network models, the design and optimization of antenna arrays is generally infeasible due to the required computational resources and simulation time. In this paper, we propose a Machine Learning framework that enables a simulation-based optimization of the antenna design. We show how learning methods are able to emulate a complex simulator with a modest dataset obtained from it, enabling a global numerical optimization over a vast multi-dimensional parameter space in a reasonable amount of time. Overall, our results show that the proposed methodology can be successfully applied to the optimization of thinned antenna arrays., Comment: 5 pages, 7 figures. This paper has been presented at EuCAP 2020. Copyright IEEE 2020. Please cite it as: M. Lecci, P. Testolina, M. Rebato, A. Testolin, and M. Zorzi, "Machine Learning-aided Design of Thinned Antenna Arrays for Optimized Network Level Performance," 14th European Conference on Antennas and Propagation (EuCAP 2020), Copenhagen, Mar. 2020

In this paper, a novel metaheuristic algorithm called Social Network Search (SNS) is developed for solving optimization problems. The SNS algorithm simulates the attempts of users in social networks to gain more popularity by modeling the moods of users in expressing their opinions. These moods are named Imitation, Conversation, Disputation, and Innovation, which are real-world behaviors of users in social networks. These moods are used as optimization operators and model how users are affected and motivated to share their new views. To evaluate the performance of the SNS algorithm, two comparative studies with different properties were conducted. In the first step, 210 mathematical functions have been chosen, which include 120 fixed-dimension, 60 N-dimension, and 30 CEC 2014 problems. Seven metaheuristics are selected from the literature, and the statistical results of these methods are calculated and analyzed. Also, to provide a valid judgment about the performance of the new algorithm, four nonparametric statistical tests have been used. In the next step, the performance of the proposed algorithm is compared to some state-of-the-art algorithms in dealing with CEC 2017 problems. According to the performance of algorithms, the SNS method is capable of achieving better results compared to the other metaheuristics in 101 cases (48%) and performed the same or comparatively in dealing with the other problems.

Globally, wind energy is growing rapidly and has received huge consideration to fulfill global energy requirements. An accurate wind power forecasting is crucial to achieve a stable and reliable operation of the power grid. However, the unpredictability and stochastic characteristics of wind power affect the grid planning and operation adversely. To address these concerns, a substantial amount of research has been carried out to introduce an efficient wind power forecasting approach. Artificial Intelligence (AI) approaches have demonstrated high precision, better generalization performance and improved learning capability, thus can be ideal to handle unstable, inflexible and intermittent wind power. Recently, AI-based hybrid approaches have become popular due to their high precision, strong adaptability and improved performance. Thus, the goal of this review paper is to present the recent progress of AI-enabled hybrid approaches for wind power forecasting emphasizing classification, structure, strength, weakness and performance analysis. Moreover, this review explores the various influential factors toward the implementations of AI-based hybrid wind power forecasting including data preprocessing, feature selection, hyperparameters adjustment, training algorithm, activation functions and evaluation process. Besides, various key issues, challenges and difficulties are discussed to identify the existing limitations and research gaps. Finally, the review delivers a few selective future proposals that would be valuable to the industrialists and researchers to develop an advanced AI-based hybrid approach for accurate wind power forecasting toward sustainable grid operation.

Teaching-Learning-based Optimization is an optimization technique which does not require any algorithm-specific parameters and is popular for its less computational cost and high consistency. Therefore, it has achieved great success application by the researchers in various disciplines of engineering. It works on the philosophy of teaching and learning which is used to solve multi-dimensional, linear and nonlinear problems with appreciable efficiency. Recently the basic TLBO algorithm is improved to enhance its exploration and exploitation capacities and the performance. However, there is less surveys on TLBO algorithm recent advances and its application. In this paper, the successful researches of TLBO algorithm of the past decade are surveyed. Firstly, the available intelligent optimization algorithms were reviewed. Then the application fields of TLBO and the improved TLBO were discussed and analyzed. Furthermore, some representative TLBO methods were classified into three main groups: 1) Improvement of teaching process; and 2) Fusion with Other Optimization Methods; and 3) Weight Methods and Others. Finally, our viewpoints were shared on the open issues and challenges in TLBO as well as research trends in the future.

The productions quality has become one of the essential issues in the modern manufacturing industry and several techniques have introduced for control and monitoring the production process. Control charts are the most practical and popular tools for continuously monitoring and, if required, make adjustments to the product or process. A new automatic method based on deep learning and optimization algorithms for nine control chart patterns (CCPs) recognition are proposed in this paper. This method has two principal parts: the classification part and the tuning part. In the last few years, a convolutional neural network (ConvNet) has led to an excellent performance on various tasks, like image processing, speech recognition, and signal processing. Therefore, in the classification part, ConvNet is used as the intelligent classifier for CCPs recognition. One significant difficulty of ConvNet is that it requires considerable proficiency to select suitable parameters like a number of kernels and their spatial sizes, learning rate, etc. The ConvNet parameters have domestic dependencies which make the tuning of these parameters a challenging task. According to these issues, in the tuning part of the proposed method, the Harris hawks optimization (HHO) algorithm is used for optimal tuning of ConvNet parameters. Contrasting the common CCPs recognition methods, the proposed method takes unprocessed data and passes to more than one hidden layer for extracting the optimal feature representation instead of relying on any feature engineering mechanisms. The quantitative and simulation results show the superiority of the proposed method over the previous techniques in terms of its performance.

4th International Symposium on Multidisciplinary Studies and Innovative Technologies, ISMSIT 2020 -- 22 October 2020 through 24 October 2020 -- -- 165025 2-s2.0-85097674127 Particle swarm optimization (PSO) is a stochastic optimization algorithm belonging to the swarm intelligence algorithms. The main characteristic of these algorithms is the communication that exists between the members of the swarm (particles), with the aim of solving a common problem, in this case an optimization problem. This is achieved by organizing the swarm in a topology that tells each particle with whom it will have direct communication during the evolution of the algorithm, in this paper we implement a routing scheme in wireless sensor network using topology optimization in particle swarm algorithm. © 2020 IEEE.

Accurate load and response forecasts are a criticalenabler for high demand response penetrations and optimizationof responses and market actions. Project RESPONSE studies anddevelops methods to improve the forecasts. Its objectives are toimprove 1) load and response forecast and optimization modelsbased on both data- driven and physical modelling, and theirhybrid models, 2) utilization of various data sources such assmart metering data, weather data, measurements fromsubstations etc., and 3) performance criteria of load forecasting.The project applies, develops, compares, and integrates variousmodelling approaches including partly physical models, machinelearning, modern load profiling, autoregressive models, andKalman-filtering. It also applies non-linear constrainedoptimization to load responses. This paper gives an overview ofthe project and the results achieved so far.

This paper presents a preliminary application of the support vector machine regression to the modeling of the eye diagram heights in high speed links for design and optimization purposes. The support vector machine regression is applied to generate a compact surrogate model of the eye diagram heights at a specific node of the link from a set of randomly selected training samples. The surrogate can be suitably adopted both for design optimization and for stochastic analysis. The feasibility and accuracy of the surrogate model calculated via the support vector machine regression are investigated on a realistic high-speed communication channel.

Cloud platforms are an essential part of modern world. Used in all kind of fields, from education to science, they became inseparable from information technologies and computing sphere. However the problem of performance optimization still exists and is not entirely solved even today. With introduction of machine learning and artificial intelligence this task became easier to solve. This paper presents a way to utilize decision trees to control performance of a computational system and to balance load on different nodes in attempt to increase quality of service.

Kidney exchange programs enable willing, but incompatible, donor-patient pairs to swap donors, thus allowing persons suffering from organ failure to access transplantation. Choosing which pairs to match requires solving a stochastic online optimization problem where patients and donors arrive over time. Despite this, most of the related scientific literature has focused on deterministic offline models. In this paper, we present a simple approach to employ a model for the offline Kidney Exchange Problem (KEP) as the basis of an on-line anticipatory algorithm. Our approach grounds on existing techniques for the on-line KEP, but it generalizes them and provides a more accurate estimate of the expected impact of current decisions. In an experimentation based on a state-of-the-art donor pool generation method, the approach provides improvements in terms of quality and is able to deal with realistic instance size in reasonable time.

In this paper, a novel distributed model-based prediction method is proposed using sensor networks. Each sensor communicates with the neighboring nodes for state estimation based on a consensus protocol without centralized coordination. The proposed distributed estimator consists of a consensus-filtering scheme, which uses a weighted combination of sensors information, and a model-based predictor. Both the consensus-filtering weights and the model-based prediction parameter for all the state components are jointly optimized to minimize the variance and bias of the prediction error in a Pareto framework. It is assumed that the weights of the consensus-filtering phase are unequal for the different state components, unlike consensus-based approaches from literature. The state, the measurements, and the noise components are assumed to be individually correlated, but no probability distribution knowledge is assumed for the noise variables. The optimal weights are derived and it is established that the consensus-filtering weights and the model-based prediction parameters cannot be designed separately in an optimal way. The asymptotic convergence of the mean of the prediction error is demonstrated. Simulation results show the performance of the proposed method, obtaining better results than distributed Kalman filtering.

Smart environments are physical places that are richly and invisibly populated with sensors, actuators and computational elements. The objective of such environments is to adapt themselves to its users in order to increase their comfort and usefulness. This paper proposes a platform, named APOLLO, capable of inferring behaviour rules from a smart environment and apply them to provide an intelligent space. The APOLLO platform is built upon a Service Oriented Architecture (SOA), in which collected context information is used to infer behaviour rules though statistical and machine learning techniques. The proposed platform is to be deployed in a home automation scenario.

International audience; In this paper, we propose a new approach to sensor localization problems, based on recent developments in machine leaning. The main idea behind it is to consider a matrix regression method between the ranging matrix and the matrix of inner products between positions of sensors, in order to complete the latter. Once we have learnt this regression from information between sensors of known positions (beacons), we apply it to sensors of unknown positions. Retrieving the estimated positions of the latter can be done by solving a linear system. We propose a distributed algorithm, where each sensor positions itself with information available from its nearby beacons. The proposed method is validated by experimentations.