Keskin, Ali Ümit | Göker, İmran

Article | 2006 | Istanbul University Journal of Electrical and Electronics Engineering6 ( 1 ) , pp.113 - 115

One must first make sure that a given matrix subject to factorization is symmetric, and then use it in Cholesky algorithm, in MATLAB. This may cost machine time to check for symmetry, however, saves much more of it by preventing errors due to inherent structure of the built-in algorithm.

Çekiç, Yalçın | Şenay, Seda | Akan, Aydın | Chaparro, Luis F.

Article | 2005 | Istanbul University Journal of Electrical and Electronics Engineering5 ( 2 ) , pp.1411 - 1417

Multi-carrier spread spectrum (MCSS) communication systems are able to mitigate the effects of fading, interferences and Doppler frequency shifts as well as to support multiple access schemes. It is possible to get optimum performance from MCSS systems by carefully choosing the spreading sequences. However, due to inter-user interferences, performance degrades as the number of users increases unless the transmission channels are modeled and estimated. In this paper, we present a new spreading sequence obtained from a complex quadratic sequence multiplied by a pseudo random noise sequence, thus exploiting the properties of constant e . . .nvelope and statistical independence. Our approach is based on the time-frequency characterization of the wireless communication channel by means of discrete evolutionary transform. The properties of the new spreading sequence enables us to estimate the parameters of multi-user channels, both for uplink and downlink transmission cases, and design a receiver that is robust to channel noise, inter-user interference and intentional jammers. The performance of the proposed system is illustrated by simulations Daha fazlası Daha az

Utku, Feride Sermin | Basaran, Kubra | Sunar, Yagmur | Celebıoglu, Hulya | Kapıcı, Ibrahim

Article | 2017 | Istanbul University Journal of Electrical and Electronics Engineering17 ( 2 ) , pp.3439 - 3443

Mineralized natural protein based novel bone replacement materials are investigated for tissue engineering. Mineralized silk fibroin composite foams and films display excellent biocompatibility. In this study, the biomimetic and electrochemical mineralization of orderly oriented silk fibroin scaffolds was studied. Commercially obtained pure silk woven fabric was boiled in 0.02 M Na2CO3 for 20 min. Calcium phosphate was deposited at 37°C for twenty minutes in seven sequential immersion steps, using 250 mM CaCl2 2H2O and 120 mM K2HPO4, containing 0.15 M NaCl and 50mM TRIS-HCl, pH 7.4, followed by electrochemical treatment in modif . . .ied SBF solution at 40°C at a current density of -25mA/cm2 for 60 min. The amount of biomimetically deposited Ca-P increased with the number of immersion steps. SEM images and XRD analysis of the Ca-P deposit indicated the initial formation of brushite with its monoclinic crystal structure and characteristic peak at 11.76 2?, and electrochemical conversion of brushite to hydroxyapatite on silk after electrochemical cathodization as confirmed by XRD and SEM analysis. Thus, a silk-fibroin-hydroxyapatite composite material prepared as a xenograft consisting of biocompatible components, and easily prepared as an economical bone segment replacement material with highly oriented fibers Daha fazlası Daha az

Aygün, Birol

Other | 2004 | Turkish Journal of Electrical Engineering and Computer Sciences12 ( 2 ) , pp.107 - 116

The importance of projects in software engineering courses is well known. Both synthetic and real-life projects have various advantages and disadvantages. Our aim was to create a framework where students can develop projects which reflect some of the complexities of real-life, involving many concurrent, interacting, asynchronous processes, each in a different stage of development, with wide temporal differences among them - some occurring- within millisconds of each other and others executing sporadically over much longer periods. In this project, which was carried out in different arrangements in several software engineering course . . .s in three universities, the students developed both the sub- and super-structures required, with varying degrees of success. The projects were performed in parallel with and subsequent to one-semester courses in software engineering. The development was performed in accordance with the principles established in the lectures. The sub-structure consists of a discrete-time event simulator and a message passing mechanism. It can support many different super-structures. The super-structure we created was an e-business community simulator where a manufacturer, its trading partners and the transactions among them were simulated. In this paper we summarize the project and our experiences during the development. A literature search for similar projects for software engineering education did not yield any hits. However several reported industrial projects for virtual supply chain management were examined. Our project on the other hand was tailored for implementation by student groups in one semester with the primary purpose of getting experience in complex, multi-group software development rather than immediate industrial use of the software Daha fazlası Daha az

Küçük, Gürhan | Mesta, Mine

Article | 2012 | Turkish Journal of Electrical Engineering and Computer Sciences20 ( 1 ) , pp.125 - 139

Nowadays, all the designers of systems from high-performance servers to battery-operated handheld devices aim for reliability, high-performance and longevity. Central within these aims the issue of processor power consumption is becoming increasingly important. In this study, we aim to adapt our already-provenmethod for single-threaded superscalar processors to simultaneous multi-threaded (SMT) processors for energy savings. The original method focused on resizing datapath resources according to the demands of running applications. To achieve this, the targeted resources are physically divided into multiple partitions, and turned on . . . and off according to the needs of the applications. Since, the energy consumption of the turned-off datapath resources is quite low, as a result, it becomes possible to have great amount of energy savings within a processor. However, special care must be taken when there are multiple threads racing against each other to gain access to shared datapath resources. As a result, our proposed microarchitectural technique achieves 0.5% Instructions Per Cycle (IPC) and 3.2% Total number of instructions Per Cycle (TPC) improvement, while it turns off 45% of the Reorder Buffer (ROB), 59% of the Load-Store Queue (LSQ), 43% of the Issue Queue (IQ), 30% of the integer Physical Register Files (PRF) and, finally, 48% of the floating PRF, on the average across all simulated benchmarks. According to our estimates, the total processor power is reduced by 12%, on the average Daha fazlası Daha az

Kalay, Saban | Blanchet, Clement | Culha, Mustafa

Article | 2014 | Turkish Journal of Chemistry38 ( 5 ) , pp.686 - 700

The charge and size of molecules chemically attached to nanoparticles (NPs) play an important role in their interaction behavior in suspensions. Gold nanoparticles (AuNPs) were modified systematically with peptides and the modification was verified with surface-enhanced Raman scattering (SERS). The behavior of the modified AuNPs in suspension and at the liquidsolid interface was monitored using small angle X-ray scattering (SAXS), UV/Vis spectroscopy and dynamic light scattering (DLS) in suspension, and atomic force microscopy (AFM) at the solid liquid interface. It was found that while negatively charged peptide modified AuNPs beha . . .ve very similar to citrate reduced AuNPs due to their negatively charged surface, positively charged peptide modified AuNPs showed significantly different assembly/aggregation properties in suspension. The formation of linear assemblies of positively charged peptide (CKRHSKRHRSKRHSKRHSKRHSKR) modified AuNPs was clearly observed from the AFM analysis of the droplet areas of its colloidal suspension. The combined analyses of data obtained from the employed techniques suggest that the positively charged large peptide modified AuNPs can form linear and 3D-like networks in the suspension. This study reveals important information regarding the surface property dependent behavior of NPs that may help in efforts to build higher structures using NPs as building blocks Daha fazlası Daha az

Ertaş, Gökhan

Article | 2018 | Electrica18 ( 2 ) , pp.242 - 248

Melanocytic lesions are the main cause of death from skin cancer, and early diagnosis is the key to decreasing the mortality rate. This studyassesses the role of input-vector encoding in neural network-based classification of melanocytic lesions in dermoscopy. Twelve dermoscopicmeasures from 200 melanocytic lesions are encoded by compact encoding, ACD encoding, 1-of-N encoding, normalized encoding, and rawencoding, resulting in five different input-vector sets. Feed-forward neural networks with one hidden layer and one output layer are designedwith several neurons in the hidden layer, ranging from two to twenty-two for each type of . . .input-vector set, to classify a melanocytic lesion intocommon nevus, atypical nevus, and melanoma. Accordingly, 105 networks are designed and trained using supervised learning and then testedby performing a 10-fold cross validation. All the neural networks achieve high sensitivities, specificities, and accuracies in classification. However,the network with seven neurons in the hidden layer and raw encoded dermoscopic measures as the input vector realizes the highest sensitivity(97.0%), specificity (98.1%), and accuracy (98.0%). The practical use of the network can facilitate lesion classification by retaining the neededexpertise and minimizing diagnostic variability among dermatologists Daha fazlası Daha az

İnce, Ahmet T. | Snook, Richard D. | Dawson, John B.

Other | 2000 | Turkish Journal of Chemistry24 ( 1 ) , pp.43 - 58

A study of inherent noise sources present in a purpose built flame atomic magneto-optic rotation (AMOR) spectrometer system was carried out with a view to identifying their sources. Two different optical configurations were employed, a crossed polariser and a 45° offset polariser configuration. The offset polariser configuration was operated in both an AMOR and atomic absorption (AA) mode for the noise measurements. Two main interference frequencies were found to occur in all the optical measurement modes. An interference flame feature frequency (15.5 Hz -39.0 Hz) which arose from an "organ-pipe" effect in the gas flow . . . through the flame cooling sheets. The second interference frequency (at 100 Hz) was thought, to arise due to modulation of the analytes magneto-optic properties due to ripple on the electromagnet d.c. power supply. Photon noise and fluctuation noise were found to be present as background noise. The fluctuation noise was dominant at. higher analyte concentrations. The characteristics and origins of fluctuation noise are discussed and investigated for each measurement mode. The noise spectra of the conventional AA measurements were compared with those of the contemporary AMOR measurements, to ascertain any differences in noise power spectra which may affect, the signal-to-noise ratio Daha fazlası Daha az

Köksal, Meriç Akkoc | Yarım, Mine Yüksel | Durmaz, İrem | Çetin, Rengül Atalay

Article | 2012 | Turkish Journal of Chemistry36 ( 4 ) , pp.503 - 514

A series of 3-[(4-substitutedpiperazin-1-yl)methyl]-1H -indole derivatives were synthesized, and their structures were confirmed by spectral analysis. All the compounds were tested for their cytotoxic activity in vitro against 3 human tumor cell lines: human liver (HUH7), breast (MCF7), and colon (HCT116). Among the designed derivatives, most of the compounds showed significant cytotoxicity against liver and colon cancer cell lines with lower IC50 concentrations than the standard drug 5-fluorouracil. Compound 3s, with 3,4-dichlorophenyl substituent on the piperazine ring, was the most active in suppressing the growth of all screened . . . cancer cells Daha fazlası Daha az

Yeğin, Korkut

Article | 2014 | Turkish Journal of Electrical Engineering and Computer Sciences22 ( 2 ) , pp.371 - 379

Wire antennas can be made wideband if the antenna is loaded with passive elements and connected to a lossless matching network. However, realization of the load component values and matching network can easily become impractical. In this study, using only a surface mount and standard component values, antenna loads and a matching network are optimized using genetic algorithms. The optimized design achieves a 25:1 bandwidth ratio, from 20 MHz to 500 MHz, with a maximum voltage standing wave ratio (VSWR) of 3.5 and minimum system gain of --5 dBi. The antenna system gain at azimuth is taken as the objective function and an exact penalt . . .y function is formulated to take into account the VSWR over the design frequency band. A loaded antenna is built and measured to corroborate the simulations results. The realized antenna is only 0.14 lambda long at 20 MHz Daha fazlası Daha az

Gökhan, Nesrin | Köksal, Meriç | Küpeli, Esra | Yeşilada, Erdem | Erdoğan, Hakkı

Article | 2005 | Turkish Journal of Chemistry29 ( 4 ) , pp.445 - 454

The synthesis of a novel series of Mannich bases of 5-methyl-3-substituted piperazinomethyl-2-benzoxazolinones is described. The structures attributed to compounds 3c, 3d and 3f-3n were elucidated using IR and 1H NMR spectroscopic techniques as well as elemental analysis. The compounds were examined for their in vivo anti-inflammatory and analgesic activities in 2 different bioassays, namely, carrageenan-induced hind paw edema and p-benzoquinone-induced abdominal constriction tests in mice, respectively. In addition, the ulcerogenic effects of the compounds were determined. Among the derivatives tested the most promising results wer . . .e obtained for the compounds bearing electron-withdrawing substituents (F, Cl, COCH_3) in the para position of the phenyl nucleus on the piperazine ring at the 3 position of benzoxazolinone moiety (3a, 3c, 3i). The analgesic activities of all compounds are higher than their anti-inflammatory activities and therefore these high analgesic activities indicated that the compounds could a show central effect Daha fazlası Daha az

Şafak, Koray Kadir

Article | 2014 | Turkish Journal of Electrical Engineering and Computer Sciences22 ( 6 ) , pp.1611 - 1624

In this paper, methods to achieve actively powered walking on level ground using a simple 2-dimensional walking model (compass-gait walker) are explored. The walker consists of 2 massless legs connected at the hip joint, a point mass at the hip, and an infinitesimal point mass at the feet. The walker is actuated either by applying equal joint torques at the hip and ankle, by an impulse applied at the toe off, immediately before the heel strike, or by the combination of both. It is shown that actuating the walker by equal joint torques at the hip and ankle on level ground is equivalent to the dynamics of the passive walker on a downh . . .ill slope. The gait cycle for the simplified walker model is determined analytically for a given initial stance angle. Stability of the gait cycle by an analytical approximation to the Jacobian of the walking map is calculated. The results indicate that the short-period cycle always has an unstable eigenvalue, whereas stability of the long-period cycle depends on selection of the initial stance angle. The effect of the torso mass by adding a third link attached at the hip joint is investigated. The torso link is kept in the vertical position by controlling the torque applied to it. The proportional-derivative control law is utilized to regulate the angular position error of the torso link. Using linearized dynamics for this walker, active control is applied to the ankle, which reduces the dynamics of the walker to the passive walker without the torso. The proposed walker is capable of producing stable walking while keeping the torso in an upright positio Daha fazlası Daha az