MOSAL: A Subspace-Based Forecasting Algorithm for Throughput Maximization in IoT Networks
IEEE Sensors Journal
Development of a deep wavelet pyramid scene parsing semantic segmentation network for scene perception in indoor environments
Journal of Ambient Intelligence and Humanized Computing
Wine Quality Assessment with Application Specific 2D Single Channel Convolutional Neural Networks
2021 13th International Conference on Electrical and Electronics Engineering (ELECO)
Electronic nose is becoming a popular tool for various application areas. The data of an electron... more Electronic nose is becoming a popular tool for various application areas. The data of an electronic nose is collected with various chemical sensor arrays and then odors are classified with suitable pattern recognition methods. This paper proposes a convolutional neural network for the the classification task of a wine quality electronic nose dataset. Method was tested on different portions of the dataset and compared with two previous studies. Proposed method managed to obtain high accuracy results within the relatively short time period. Additionally, method was tested by using portions of the sensor responses, hence allowing the user to assess wine quality earlier. Each training was repeated ten times in order to minimize the effects of random data selection.
Dynamic Automatic Forecaster Selection via Artificial Neural Network Based Emulation to Enable Massive Access for the Internet of Things
Journal of Network and Computer Applications, 2022
A 2-dimensional model of polynomial type for oscillatory ATM-Wipl dynamics in p53 network
2017 10th International Conference on Electrical and Electronics Engineering (ELECO), 2017
Under gamma irradiation, p53 gene regulatory network is able to exhibit three different modes, na... more Under gamma irradiation, p53 gene regulatory network is able to exhibit three different modes, namely low state, oscillations, and high state. There are experimental studies demonstrating that oscillatory behaviour of p53 is due to the interaction between upstream mediator of p53, i.e. ATM, and a negative feedback loop formed by Wipl with that upstream. By proposing a canonical model based on ordinary differential equations made up of polynomial type birth and death terms, we show mathematically that the simple interaction between ATM and Wipl is indeed able to exhibit three different behaviours relevant to DNA damage response of p53 network. We further carry out bifurcation analysis on the model with the aim of investigating the mutations such as Wip1 overexpression and ATM deficiency. Based on the proposed canonical model, we show that Wipl is an important target for curing these types of mutations.
Design and implementation of chaotic system based robust delta robot for blending graphene nanoplatelets
2016 21st International Conference on Methods and Models in Automation and Robotics (MMAR), 2016
In this paper, a new blending method for Graphene nanoplatelets was developed. A chaotic system b... more In this paper, a new blending method for Graphene nanoplatelets was developed. A chaotic system based robust delta robot was designed. Both the speed of the mixer motor and position of the propeller were chaotically changed. Performance of the systems was evaluated by a material analysis method. The results showed that the proposed method has a better performance than statically mounted mixers.
End-To-End Learning from Demonstation for Object Manipulation of Robotis-Op3 Humanoid Robot
2020 International Conference on INnovations in Intelligent SysTems and Applications (INISTA), 2020
Humanoid robots are deployed ranging from houses and hotels to healthcare and industry environmen... more Humanoid robots are deployed ranging from houses and hotels to healthcare and industry environments to help people. Robots can be easily programed by users to predefined tasks such as walking, grasping, stand-up, and shake-up. However, in these days, all robots are expected to learn itself from the obtained experience by watching the environment and people in there. In this study, it is aimed for Robotis-Op3 humanoid robot to grasp the objects by learning from demonstrations based on vision. A new algorithm is proposed for this purpose. Firstly, the robot is manipulated from user commands and the raw images from the camera of Robotis-Op3 are collected. Secondly, a semantic segmentation algorithm is applied to detect and recognize the objects. A new model using Convolutional Neural Networks (CNNs) and Long Short-Term Memory Networks (LSTMs) is then proposed to learn the user demonstrations. The results were compared in terms of training time, performance, and model complexity. Simulation results showed that new models produced a high performance for object manipulation.
Electricity Energy Forecasting for Turkey: A Review of the Years 2003–2020
Turkish Journal of Electrical Power and Energy Systems, 2021
A Multiscale Algorithm for Joint Forecasting–Scheduling to Solve the Massive Access Problem of IoT
IEEE Internet of Things Journal, 2020
The massive access problem of the Internet of Things (IoT) is the problem of enabling the wireles... more The massive access problem of the Internet of Things (IoT) is the problem of enabling the wireless access of a massive number of IoT devices to the wired infrastructure. In this article, we describe a multiscale algorithm (MSA) for joint forecasting–scheduling at a dedicated IoT gateway to solve the massive access problem at the medium access control (MAC) layer. Our algorithm operates at multiple time scales that are determined by the delay constraints of IoT applications as well as the minimum traffic generation periods of IoT devices. In contrast with the current approaches to the massive access problem that assume random arrivals for IoT data, our algorithm forecasts the upcoming traffic of IoT devices using a multilayer perceptron architecture and preallocates the uplink wireless channel based on these forecasts. The multiscale nature of our algorithm ensures scalable time and space complexity to support up to 6650 IoT devices in our simulations. We compare the throughput and energy consumption of MSA with those of reservation-based access barring (RAB), priority based on average load (PAL), and enhanced predictive version burst-oriented (E-PRV-BO) protocols, and show that MSA significantly outperforms these beyond 3000 devices. Furthermore, we show that the percentage control overhead of MSA remains less than 1.5%. Our results pave the way to building scalable joint forecasting–scheduling engines to handle a massive number of IoT devices at IoT gateways.
Double-strand break-induced (DSB) cells send signal that induces DSBs in neighbour cells, resulti... more Double-strand break-induced (DSB) cells send signal that induces DSBs in neighbour cells, resulting in the interaction among cells sharing the same medium. Since p53 network gives oscillatory response to DSBs, such interaction among cells could be modelled as an excitatory coupling of p53 network oscillators. This study proposes a plausible coupling model of three-mode two-dimensional oscillators, which models the p53-mediated cell fate selection in globally coupled DSBinduced cells. The coupled model consists of ATM and Wip1 proteins as variables. The coupling mechanism is realised through ATM variable via a mean-field modelling the bystander signal in the intercellular medium. Investigation of the model reveals that the coupling generates more sensitive DNA damage response by affecting cell fate selection. Additionally, the authors search for the cause-effect relationship between coupled p53 network oscillators and bystander effect (BE) endpoints. For this, they search for the possible values of uncertain parameters that may replicate BE experiments' results. At certain parametric regions, there is a correlation between the outcomes of cell fate and endpoints of BE, suggesting that the intercellular coupling of p53 network may manifest itself as the form of observed BEs.
IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 14, NO. 4, JULY 2003 891 A New Design Method for the Complex-Valued
A method to store each element of an integral memory set 1 2 ...
Engineering applications of a dynamical state feedback chaotification method
AIP Conference Proceedings, 2012
ABSTRACT This paper presents two engineering applications of a chaotification method which can be... more ABSTRACT This paper presents two engineering applications of a chaotification method which can be applied to any inputstate linearizable (nonlinear) system including linear controllable ones as special cases. In the used chaotification method, a reference chaotic and linear system can be combined into a special form by a dynamical state feedback increasing the order of the open loop system to have the same chaotic dynamics with the reference chaotic system. Promising dc motor applications of the method are implemented by the proposed dynamical state feedback which is based on matching the closed loop dynamics to the well known Chua and also Lorenz chaotic systems. The first application, which is the chaotified dc motor used for mixing a corn syrup added acid-base mixture, is implemented via a personal computer and a microcontroller based circuit. As a second application, a chaotified dc motor with a taco-generator used in the feedback is realized by using fully analog circuit elements.
Multilevel Data Classification and Function Approximation Using Hierarchical Neural Networks
Studies in Computational Intelligence, 2010
ABSTRACT Combining diverse features and multiple classifiers is an open research area in which no... more ABSTRACT Combining diverse features and multiple classifiers is an open research area in which no optimal strategy is found but successful experimental studies have been performed depending on a specific task at hand. In this chapter, a strategy for combining diverse features and multiple classifiers is presented as an exemplary new model in multilevel data classification using hierarchical neural networks. In the proposed strategy, each feature set and each classifier extracts its own representation from the raw data which results with measurements extracted from the original data (or a subset of original data) that are unique to each level of approximation/classification. Later on, the results of each level are linearly combined in function approximation or merged in classification. It is shown by advanced signal and image processing applications that proposed model of combining features/classifiers is especially important for applications that require integration of different types of features and classifiers.
Robust spherical clustering as mixed integer optimization problem and its gradient network solution
Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, 2004.
Support vector based spherical clustering is described as an optimization problem posed in the in... more Support vector based spherical clustering is described as an optimization problem posed in the input space where the cluster indicators are also considered as variables. It is attempted to find the robust clustering by taking the objective function of the optimization problem as the energy function of the gradient network. The proposed method is an extension of the authors' work
Construction of energy landscape for discrete Hopfield associative memory with guaranteed error correction capability
First International IEEE EMBS Conference on Neural Engineering, 2003. Conference Proceedings.
An energy function-based auto-associative memory design method to store a given set of unipolar b... more An energy function-based auto-associative memory design method to store a given set of unipolar binary memory vectors as attractive fixed points of an asynchronous discrete Hopfield network is presented. The discrete quadratic energy function whose local minima correspond to the attractive fixed points of the network is constructed via solving a system of linear inequalities derived from the strict local
A 2D DPCM scheme using cellular neural networks
1998 Fifth IEEE International Workshop on Cellular Neural Networks and their Applications. Proceedings (Cat. No.98TH8359), 1998
We formulate differential pulse code modulation (DPCM) for image compression as the minimization ... more We formulate differential pulse code modulation (DPCM) for image compression as the minimization of a quadratic cost function. Non-causal interpolation error image in lieu of causal prediction error image can be coded in this fashion providing efficient compression. We implement the optimization process through the dynamics of cellular neural networks (CNNs). Two CNNs, one of them operated in binary mode and the other in gray level mode, are used in the coding stage. The first CNN creates an optimum differential image while the other tries to create a replica of the reconstructed image of the receiver. Decoding is realized by another gray level mode CNN fed by the differential image.
Real-Time Simulation Platform for Controller Design, Test, and Redesign
An Application of Support Vector Machine in Bioinformatics: Automated Recognition of Epileptiform Patterns in EEG Using SVM Classifier Designed by a Perturbation Method
Lecture Notes in Computer Science, 2004
We introduce an approach based on perturbation method for input dimension reduction in Support Ve... more We introduce an approach based on perturbation method for input dimension reduction in Support Vector Machine (SVM) classifiers. If there exists redundant data components in training data set, they can be discarded by analyzing the total disturbance of the SVM output corresponding to the perturbed inputs. Thus, input dimension size is reduced and network becomes smaller. Algorithm for input dimension reduction is first formulated and then applied to real electroencephalography (EEG) data for recognition of epileptiform patterns.
A canonical representation for piecewise-affine maps and its applications to circuit analysis
IEEE Transactions on Circuits and Systems, 1991
Abstract -A new canonical representation for a rather general class of piecewise-affine maps has ... more Abstract -A new canonical representation for a rather general class of piecewise-affine maps has been developed. The given canonical representation extends the canonical representation proposed by Chua and Kang into PWA partitions, which arise frequently in driving-point, ...
CNNs with radial basis input function
1996 Fourth IEEE International Workshop on Cellular Neural Networks and their Applications Proceedings (CNNA-96)
ABSTRACT This paper proposes a cellular neural network (CNN) model with radial basis input functi... more ABSTRACT This paper proposes a cellular neural network (CNN) model with radial basis input function (radial basis input CNN) for improving function approximation ability of CNNs. The model can be viewed as a cascade of two units: the first unit is a multi-input, multi-output radial basis function network (RBFN), the second unit is the original CNN model. The weights and centers of the RBFN unit are chosen identical for all RBFN outputs yielding a space-invariant connection weight pattern over the network. With such a weight sharing property, the proposed model becomes a special kind of nonlinear B-template CNN. The ability of the radial basis input CNN model in approximation to functions as its input-(steady state) output mapping is examined on an edge detection task for noisy images. A modified version of the recurrent perceptron learning algorithm (RPLA) is used for the training radial basis input CNN
Uploads
Papers by cuneyt guzelis