Arama Sonuçları
Listeleniyor 1 - 10 / 94
Yayın On the sensitivity of desirability functions for multiresponse optimization(American Institute of Mathematical Sciences, 2008-11) Aksezer, Sezgin ÇağlarDesirability functions have been one of the most important multiresponse optimization technique since the early eighties. Main reasons for this popularity might be counted as the convenience of the implementation of the method and it's availability in many experimental design software packages. Technique itself involves somehow subjective parameters such as the importance coefficients between response characteristics that are used to calculate overall desirability, weights used in determining the shape of each individual response and the size of the specification band of the response. However, the impact of these sensitive parameters on the solution set is mostly uninvestigated. This paper proposes a procedure to analyze the sensitivity of the important characteristic parameters of desirability functions and their impact on pareto-optimal solution set. The proposed procedure uses the experimental design tools on the solution space and estimates a prediction equation on the overall desirability to identify the sensitive parameters. For illustration, a classical desirability example is selected from the literature and results are given along with the discussion.Yayın A survey of algorithms and architectures for H.264 sub-pixel motion estimation(World Scientific, 2012-05) Fatemi, Mohammad Reza Hosseiny; Ateş, Hasan Fehmi; Salleh, Rosli BinThis paper reviews recent state-of-the-art H. 264 sub-pixel motion estimation (SME) algorithms and architectures. First, H.264 SME is analyzed and the impact of its functionalities on coding performance is investigated. Then, design space of SME algorithms is explored representing design problems, approaches, and recent advanced algorithms. Besides, design challenges and strategies of SME hardware architectures are discussed and promising architectures are surveyed. Further perspectives and future prospects are also presented to highlight emerging trends and outlook of SME designs.Yayın Design of a global extremum seeking algorithm for an omni-directional robot model(Romanian Soc Control Tech Informatics, 2017-06) Dinçmen, ErkinA global extremum seeking algorithm is developed for a mobile robot model where the aim is to find the location of the most powerful signal source among the others. In other words, the control problem is to seek the global extremum point of a performance function when there are local extremas. The locations of the signal sources and signal distribution characteristics are unknown, i.e. the gradient of the performance function is unknown. The control algorithm also doesn't use any position measurement of the mobile robot itself. Henceforth, the controller is suitable for the missions where the robot moves in an unknown terrain with no GPS signal and no inertial measurements. Only the signal magnitude should be measured via a sensor mounted on the robot during the motion. A gradient estimator is designed to determine the motion direction towards the extremum point. When a local extremum is found, the robot will continue its search for another extremum points. Once each extremums have been visited, the robot will compare the signal levels on each source and identify the global extremum i.e. the most powerful signal source. In the absence of any position measurements, the robot can move towards the global extremum by repeating its motion history backwards. In the literature, this is the first global extremum seeking algorithm that has been developed for an omni-directional mobile robot model. Via the simulation studies it has been shown that the control algorithm can seek and find both stationary and non stationary signal sources and it can find the global extremum point when there are local extremas.Yayın Driver recognition using gaussian mixture models and decision fusion techniques(Springer-Verlag Berlin, 2008) Benli, Kristin Surpuhi; Düzağaç, Remzi; Eskil, Mustafa TanerIn this paper we present our research in driver recognition. The goal of this study is to investigate the performance of different classifier fusion techniques in a driver recognition scenario. We are using solely driving behavior signals such as break and accelerator pedal pressure, engine RPM, vehicle speed; steering wheel angle for identifying the driver identities. We modeled each driver using Gaussian Mixture Models, obtained posterior probabilities of identities and combined these scores using different fixed mid trainable (adaptive) fusion methods. We observed error rates is low as 0.35% in recognition of 100 drivers using trainable combiners. We conclude that the fusion of multi-modal classifier results is very successful in biometric recognition of a person in a car setting.Yayın Failure of an exchange-rate-based stabilization plan in Turkey(M E Sharpe, 2003-02) Gökkent, Giyas; Moslares, Carlos; Amiel-Saenz, RafaelThe Turkish exchange-rate-based stabilization plan adopted in 2000 has been a spectacular failure, lasting a mere fourteen months despite a relatively flexible peg regime and preannounced exit strategy. The final three months of the currency regime were marred by the eruption of a banking sector crisis that quickly developed into a currency crisis, quelled only by external loans and a blanket guarantee by the sovereign of all banking sector liabilities. This was ultimately to no avail as the lira was allowed to float following a full-fledged currency crisis in late February 2001. The usual indicators of crisis did not point to imminent turmoil in November 2000 despite widespread concern about eventual dire developments. To identify the source of the November crisis, one must weigh the factors that led economic agents, and banks in particular, to expect higher interest rates after the fall.Yayın Extremum seeking dead-zone pre-compensator for an industrial control system(Walter De Gruyter GMBH, 2018-06-26) Dinçmen, ErkinPID type industrial controllers such as PI, PD, PID are mature control algorithms and they are intensively used in industry due to their simplicity and easily implementability. However, they start to fail when there is an unknown or unpredictable nonlinear behavior in the plant or actuator. In this paper, a novel compensation algorithm is proposed for PD type industrial control systems, which possess an unknown dead-zone nonlinearity. An extremum-seeking technique is utilized in the compensation algorithm. The aim is to propose a new, effective and robust compensator which can be added easily to an existing industrial controller without any need to change/retune the controller settings/parameters. It is shown that by adding the compensator to an existing PD control system, the sensitivity of the controller to the dead-zone nonlinearity is removed.Yayın Small molecule sorption and desorption in and out of iota-carrageenan gels(Taylor & Francis Group, 2007-08) Ataman, Evren; Pekcan, Mehmet ÖnderSmall molecule sorption and desorption in and out of Iota-Carrageenan was studied by using steady-state fluorescence (SSF) technique. Pyranine dissolved in water used as fluorescence probe. Fluorescence emission intensity, I-p from pyranine was monitored for studying sorption and desorption processes at various temperatures. The Fickian model was applied to produce sorption, D-s, early desorption, D-ed, and desorption, D-d, coefficients. Corresponding activation energies were obtained and found to be 20.5 kJ mol(-1), 7.0 kJ mol(-1) and 34.9 kJ mol(-1), respectively. The observed D-ed value is an order of magnitude smaller than the D-s and D-d coefficients. On the other hand, sorption processes were shown to be twice as fast as desorption processes.Yayın On numerical design technique of wideband microwave amplifiers based on GaN small-signal device model(Springer, 2014-10) Köprü, Ramazan; Kuntman, Hulusi Hakan; Yarman, Bekir Sıddık BinboğaThis work presents an application of Normalized Gain Function (NGF) method to the design of linear wideband microwave amplifiers based on small-signal model of a device. NGF has been originally developed to be used together with an S-parameter (*.s2p) file, whereas this work enables the NGF to be able to work with explicit S-parameter formulae derived from the small-signal model of the device. This approach provides the designer to be able to use simple set of S-parameter equations instead of S-parameter file of the device. Representation of the device simply by several model equations not only eliminates the need of carrying large number of data but also provides the capability of equation-based easy, realistic and equispaced S-parameter data generation in any desired resolution in frequency axis without requiring interpolation. NGF is defined as the ratio of T and |S-21|(2), i.e. T-N = T/|S-21|(2), gain function of the amplifier to be designed and transistor forward gain function, respectively. Synthesis of output/input matching networks (OMN/IMN) of the amplifier requires two target gain functions in terms of T-N, to be used in two sequential non-linear optimization procedures, respectively. An amplifier with a flat gain of similar to 10 dB operating in 0.8-2.35 GHz is designed using a small-signal model of an experimental GaN-HEMT. Theoretical amplifier performance obtained in Matlab is shown to be in excellent agreement with the simulated performance in MWO (Microwave Office, AWR Inc.). A prototype low-power amplifier having a similar to 10 to 12 dB gain, operating in (0.9-1.5 GHz) is also produced and measured which yielded good performance results.Yayın Classification-based macroblock layer rate control for low delay transmission of H.263 video(IS & T - Soc Imaging Science Technology, 2003-07) Bayazıt, UluğPuri and Aravind's method of macroblock bit count estimation for video rate control is based on the classification of the macroblock data into discrete classes and assigning a unique non-linear estimate for each class and quantization parameter pair. This method stands apart from other methods in the literature, since the model of the bit count versus the quantization parameter relation, parameterized by macroblock variance, is a discrete model generated solely from measurements, We extend their technique for low-delay video rate control (tight buffer regulation) in two ways. We propose a strategy of near-uniform quantization parameter assignments to the macroblocks of a frame that can come close to maximizing an objective spatial quality function, such as PSNR, over the entire frame. We also adaptively update the quantization parameter assignments for the yet to be coded macroblocks, after the encoding of each macroblock, to compensate for any errors in the bit count estimation of the encoded macroblock. Our experiments demonstrate that the proposed rate control method can more accurately control the number of bits expended for a frame, as well as yield a higher objective spatial quality than the method adopted by TMN8.Yayın Construction of a Turkish proposition bank(Tubitak Scientific & Technical Research Council Turkey, 2018) Ak, Koray; Toprak, Cansu; Esgel, Volkan; Yıldız, Olcay TanerThis paper describes our approach to developing the Turkish PropBank by adopting the semantic role-labeling guidelines of the original PropBank and using the translation of the English Penn-TreeBank as a resource. We discuss the semantic annotation process of the PropBank and language-specific cases for Turkish, the tools we have developed for annotation, and quality control for multiuser annotation. In the current phase of the project, more than 9500 sentences are semantically analyzed and predicate-argument information is extracted for 1330 verbs and 1914 verb senses. Our plan is to annotate 17,000 sentences by the end of 2017.
