Ileana Popescu

Applications of generalized RBF-NN for path loss prediction

This paper presents the results of the generalized radial basis function neural networks applications for the prediction of propagation path loss in urban and suburban environment. We have studied two types of neural network based models; the first one is used for path loss prediction while the second one is a hybrid prediction model based on error control. The performances of the neural models are compared to the path loss values measured in the city of Kavala and in Oia village on Santorini Island, Greece, based on the absolute mean error, standard deviation and root mean square error between predicted and measured values.

Generalized regression neural network prediction model for indoor environment

This paper presents the results of our studies regarding the applications of the neural networks to the propagation path loss prediction in indoor environment. The proposed model consists of a generalized regression neural network trained with measurements. The result of the prediction made by the proposed model has shown a good agreement with the measurements.

ANN Prediction Models for Outdoor Environment

This paper presents the results of our studies concerning the applications of feedforward artificial neural networks to the propagation path loss prediction in outdoor environment. An error correction model is proposed, based on the combination between a theoretical model and a neural network. The performances of the proposed artificial neural network models are compared to the measured path loss values, based on the absolute mean error, standard deviation and root mean square error. Also, the proposed neural network models are compared to each other and to the COST 231-Walfisch-Ikegami

Comparison of neural network models for path loss prediction

This work presents the results of the studies concerning the applications of the feedforward neural networks to the prediction of propagation path loss in urban and suburban environment. First, neural network models are designed in order to predict the path loss. Further investigations are made on an error correction model, based on the combination between a theoretical model and a neural network. The performances of the neural models are compared to the measured path loss values from the measurements conducted in the city of Kavala and in Oia village on Santorini Island, Greece, based on the absolute mean square error, standard deviation and root mean square error between predicted and measured values. Also, the neural networks models are compared to each other and to the COST 231-Walfisch-Ikegami.

ANN Prediction Models for Indoor Environment

This work presents the results of the studies concerning the application of the feedforward neural networks to the prediction of propagation path loss in indoor environment. The proposed models consist of a multilayer perceptron and a generalized regression neural network trained with measurements. The results of the prediction made by the proposed neural models show a good agreement with the measurements

Field strength prediction in indoor environment with a neural model

This paper presents the results of our studies concerning the application of the neural networks to the field strength prediction in an indoor environment. The proposed model consists of a multilayer perceptron trained with measurements. The results of the prediction show good agreement with the measurements.

Real-Time WLAN Monitoring in a 4G Multiplatform Environment

Real-time monitoring in 4G networks is of prime importance for network operators of present and future generation. Nowadays operators utilize stored historical data from their data warehouses to improve networks and services, at the same time the number of mobile subscribers rapidly increases and additionally, the new data-technologies for wireless access add extra traffic to the already overloaded networks, often causing serious problems to their performance. A real-time monitoring tool is necessary to operators in order to guarantee high quality of service. In this paper a novel 4G multiplatform real-time monitoring system is presented and analyzed, giving emphasis to WLAN part. The main idea of this system is to collect reports from numerous network elements in such way that the system is compatible and operational in any kind of network of any manufacturer and operator. Additionally the system architecture discussed in this paper is capable of accommodating and supporting 4G networks effectively

A new approach for the radar mixed signal separation

The retrieval of several electromagnetic signal sources received by an antenna array is presented. Spatial diversity information is employed to separate instantaneous mixtures. The approach relies on an adaptive algorithm for the minimization of the mutual information between receiver channels. The examples are included to illustrate the procedure feasibility. The number of sources was equal to the number of receivers.

Mobile Terminal Subarea Localisation Method in GSM Networks

In this paper, a novel method for mobile terminal (MT) subarea localisation in GSM networks is presented. No base stations position and additional hardware are required. It is a variation of the fingerprinting localisation method aiming to determine not the MT geographic position but the MT subarea and to reduce necessary initialisation measurements. Several GSM-based methods have been developed for MT localisation. The localisation method, proposed in this paper, is based on the fingerprinting method but is differentiated regarding the produced result. In this method the location server does not compare the MT collected radio network measurements with the stored radio map but with the subareas proprietary characteristics indicating the subarea where the MT is located and not the exact MT geographic position. Finally, if an LBS requires as input the subscriber subarea and not the geographic position then the novel method presented here reduces significantly the localisation initialisation requirements and computational cost in order to provide the service.

Comparison of ANN Based Models for Path Loss Prediction in Indoor Environment

This work presents the results of the studies concerning the application of the neural networks to the propagation path loss prediction in indoor environment. The proposed models consist of a multilayer perceptron and a generalized regression neural network trained with measurements. The results of the prediction made by the proposed neural models show a good agreement with the measurements.