Elektrik Elektronik Mühendisliği Bölümü Yayın Koleksiyonu

Permanent URI for this collectionhttps://hdl.handle.net/20.500.12416/411

Browse

Now showing 1 - 8 of 8

Alternative Enhancement of Associativity Based Routing (Aeabr) for Mobile Networks
(Springer, 2010) Preveze, Barbaros; Safak, Aysel
This study proposes an alternative enhancement for the Enhanced Associativity Based Routing (EABR) method which is a derivation of ABR (Associativity Based Routing) by relative speed and relative distance estimation using the received power strength (RPS) of the nodes. In this study, it is shown that EABR outperforms some other well known protocols. The performance of EABR is improved in terms of number of route reconstructions (RRC) and connected status percentage (CSP). Message overhead and bandwidth utilization is also investigated.
Citation - WoS: 5
Citation - Scopus: 7
Comparison of Single Channel Indices for U-Net Based Segmentation of Vegetation in Satellite Images
(Spie-int Soc Optical Engineering, 2020) Ulku, Irem; Barmpoutis, Panagiotis; Stathaki, Tania; Akagunduz, Erdem
Hyper-spectral satellite imagery, consisting of multiple visible or infrared bands, is extremely dense and weighty for deep operations. Regarding problems related to vegetation as, more specifically, tree segmentation, it is difficult to train deep architectures due to lack of large-scale satellite imagery. In this paper, we compare the success of different single channel indices, which are constructed from multiple bands, for the purpose of tree segmentation in a deep convolutional neural network ( CNN) architecture. The utilized indices are either hand-crafted such as excess green index (ExG) and normalized difference vegetation index (NDVI) or reconstructed from the visible bands using feature space transformation methods such as principle component analysis (PCA). For comparison, these features are fed to an identical CNN architecture, which is a standard U-Net-based symmetric encoder-decoder design with hierarchical skip connections and the segmentation success for each single index is recorded. Experimental results show that single bands, which are constructed from the vegetation indices and space transformations, can achieve similar segmentation performances as compared to that of the original multi-channel case.
Comparison of Single Channel Indices for U-Net Based Segmentation of Vegetation in Satellite Images
(SPIE, 2020) Ülkü, İrem; Barmpoutis, P.; Stathaki, T.; Akagündüz, Erdem
Hyper-spectral satellite imagery, consisting of multiple visible or infrared bands, is extremely dense and weighty for deep operations. Regarding problems related to vegetation as, more specifically, tree segmentation, it is difficult to train deep architectures due to lack of large-scale satellite imagery. In this paper, we compare the success of different single channel indices, which are constructed from multiple bands, for the purpose of tree segmentation in a deep convolutional neural network (CNN) architecture. The utilized indices are either hand-crafted such as excess green index (ExG) and normalized difference vegetation index (NDVI) or reconstructed from the visible bands using feature space transformation methods such as principle component analysis (PCA). For comparison, these features are fed to an identical CNN architecture, which is a standard U-Net-based symmetric encoder-decoder design with hierarchical skip connections and the segmentation success for each single index is recorded. Experimental results show that single bands, which are constructed from the vegetation indices and space transformations, can achieve similar segmentation performances as compared to that of the original multi-channel case
Citation - WoS: 1
Citation - Scopus: 3
Hyperspectral Image Compression Using an Online Learning Method
(Spie-int Soc Optical Engineering, 2015) Ulku, Irem; Toreyin, B. Ugur
A hyperspectral image compression method is proposed using an online dictionary learning approach. The online learning mechanism is aimed at utilizing least number of dictionary elements for each hyperspectral image under consideration. In order to meet this "sparsity constraint", basis pursuit algorithm is used. Hyperspectral imagery from AVIRIS datasets are used for testing purposes. Effects of non-zero dictionary elements on the compression performance are analyzed. Results indicate that, the proposed online dictionary learning algorithm may be utilized for higher data rates, as it performs better in terms of PSNR values, as compared with the state-of-the-art predictive lossy compression schemes.
Influence of Atmospheric Turbulence on Quality of Multichannel Laser Radiation and Correction for Distortion
(Spie-int Soc Optical Engineering, 2016) Kanev, F. Yu.; Lukin, V. P.; Makenova, N. A.; Antipov, O. L.; Eyyuboglu, Halil Tanyer
In this paper the results of simulation are presented of multichannel radiation propagation in the atmosphere, and correction for turbulent distortion on the base of the beam phase control is considered. The results demonstrate dependence of correction effectiveness on number of channels and on precision of a reference beam phase reconstruction. Additionally increase of effectiveness is possible with adjustment of amplification in the channels of the optical system, i.e., with the use of amplitude-phase control of radiation.
Citation - Scopus: 1
Intensity Fluctuations of Incoherently Superposed Gaussian Beams in Atmospheric Turbulence
(Spie-int Soc Optical Engineering, 2010) Baykal, Yahya
Intensity fluctuations of incoherently superposed Gaussian beams are formulated in weak turbulence by employing the extended Huygens-Fresnel principle. Each individual beam superposed is taken to be fully incoherent. The scintillation index evaluated for different number of beams indicates that as the number of beams increase, scintillations decrease. Incoherent superposition of smaller sized Gaussian sources exhibits smaller fluctuations. Comparing the scintillation index arising from incoherently superposed Gaussian beams to the scintillation index of coherently superposed Gaussian beams of the same structure shows that incoherent superposition yields lower intensity fluctuations, thus can be advantageous in atmospheric optical communication links.
Citation - WoS: 2
Citation - Scopus: 3
Method of Evaluation of the Mutual Coherence Function of Laser Beams and Its Application for Symmetric Dark Hollow Beams
(Spie-int Soc Optical Engineering, 2011) Banakh, V. A.; Marakasov, D. A.; Rytchkov, D. S.; Baykal, Y. K.; Eyyuboglu, T.; Eyyuboǧlu, H.T.
In our report, a method of evaluation of the mutual coherence function (MCF) of optical wave propagating in turbulent atmosphere is proposed. The method is based on using inverse Fresnel transform and 2-D Fast Fourier Transformation procedure (FFT) and provides a high accuracy in various propagation conditions from weak up to strong optical turbulence regime. This technique allows significantly reducing the evaluation time of MCF. The proposed method is suitable for optical waves with arbitrary initial distribution of amplitude and phase. Results of investigations on the degradation of coherence of symmetric dark hollow beams (DHB) propagating in turbulent atmosphere are presented. Analysis of evolution of MCF is brought up, and some characteristics of DHB, such as mean intensity distribution, and moments of Wigner distribution are calculated for various kinds of profiles of structural characteristic of refractive index. All of the evaluations for DHB are fulfilled for Kolmogorov spectrum of correlation function of refractive index fluctuations. The comparison between mean intensity calculations within the proposed method and method based on semi-analytical approach using of quadratic approximation of spherical wave structure function, is presented.
Citation - WoS: 6
Citation - Scopus: 8
Performance Analysis of Mimo Fso Systems With Radial Array Beams and Finite Sized Detectors
(Spie-int Soc Optical Engineering, 2014) Kamacioglu, Canan; Uysal, Murat; Baykal, Yahya; Gokce, Muhsin C.; Kamacıoğlu, Canan
Multiple-input multiple-output (MIMO) systems are employed in free space optical (FSO) links to mitigate the degrading effects of atmospheric turbulence. In this paper, we consider a MIMO FSO system with practical transmitter and receiver configurations that consists of a radial laser array with Gaussian beams and finite sized detectors. We formulate the average received intensity and the power scinitillation as a function of the receiver coordinates in the presence of weak atmospheric turbulence by using the extended Huygens-Fresnel principle. Then, integrations over the finite sized multiple detectors are performed and the effect of the receiver aperture averaging is quantified. We further derive an outage probability expression of this MIMO system in the presence of turbulence-induced fading channels. Using the derived expressions, we demonstrate the effect of several practical system parameters such as the ring radius, the number of array beamlets, the source size, the link length, structure constant and the receiver aperture radius on the system performance.

Browse

Browsing Elektrik Elektronik Mühendisliği Bölümü Yayın Koleksiyonu by Scopus Q "Q4"