M.S. Theses

Permanent URI for this collection

https://digitalarchive.library.bogazici.edu.tr/handle/123456789/11647

Browse

Now showing 1 - 20 of 24

A new generation embedded systems design for robocup SSL robots
(Thesis (M.A.) - Bogazici University. Institute for Graduate Studies in the Social Sciences, 2018., 2018.) Turan, Günay.; Kahya, Yasemin.; Akar, Mehmet.
RoboCup Small Size League (SSL) is one of the most competitive divisions in the RoboCup where international universities compete with their state of the art robotic hardware and software designs. In the SSL, teams build their own robots which move by four omniwheels; therefore, balance of the robot is trivial compared to humanoid competitions. This feature of the SSL enables teams to build faster robots and focus on more advanced AI development strategies. However, it is essential to have a robot pool that is robust enough to comply with AI directives. In this thesis, we have developed a new generation of robotic hardware and low level ﬁrmware as well as low level control algorithms for improvements in precise robot movement. Moreover, a fully modular system design is of importance due to highly evolving nature of the competition. This thesis focuses on the work and methods used to overcome the problems in previous robotic designs and improvements upon them. A robust, modular, and improvable robot pool is aimed to be developed. The robotic hardware is split into two main parts; motor driver circuit and main board. An integrated FET, Brushless DC (BLDC) motor driver IC is used separately from the main board to reduce noise and interference and to save space. In order to increase control performance and modularity of the design, two main logic units are used in the main board. The DSP chip used in the previous design is replaced with FPGA and MCU chips. Firmware development is done for both chips working simultaneously during the game. Tasks are divided among them according to their strengths. Finally, a PI control algorithm, developed on the FPGA in logic gate level, is discussed.
Adaptive optimization of EDCA parameters for QOS in multimedia applications
(Thesis (M.S.)-Bogazici University. Institute for Graduate Studies in Science and Engineering, 2009., 2009.) Çil, Alparslan Mustafa.; Anarım, Emin.; Akar, Mehmet.
The MAC layer of the IEEE 802.11 standard defines Distributed Channel Function (DCF), that provides simple and effective medium access mechanism. However with the increasing demand in multimedia applications and higher data rates, DCF can not provide QoS requirements since 802.11 standard does not meet the need of Quality of Service (QoS) in high data rate multimedia applications like VoIP and IPTV. For this reason, necessary extensions are made to improve data rates and MAC control schemes according to the dynamically changing needs. To provide QoS for the multimedia applications, IEEE de ned a new MAC mechanism, 802.11e Enhanced Distributed Channel Access (EDCA). With new access categories that enable prioritization between different traffic classes. Such prioritization provides QoS for multimedia applications by increasing the probability of higher priority class packets accessing the medium. The purpose of this thesis is to design an adaptive optimization system in EDCA mechanism in order to provide better QoS for multimedia applications from both delay and throughput perspectives. To achieve this goal, analytical model of the DCF mechanism, already de ned in the literature, is improved for the EDCA mechanism. Based on the back-off procedure in EDCA, that is represented using a Markov model, the estimates of throughput and delay are obtained by determining the probabilities in the Markov model. In order to validate the estimated throughput and delay, real time simulations in OPNET are carried out and the correlation between analytical results and simulation results are included in the thesis. Based on the validated analytical model, a new algorithm is proposed in order to maximize throughput while decreasing delay and minimizing collision probability for a higher access category.
Analysis of an averaging based synchronization algorithm in continuous-time
(Thesis (M.S.)-Bogazici University. Institute for Graduate Studies in Science and Engineering, 2010., 2010.) Ersöz, Didem.; Akar, Mehmet.
Synchronization in distributed networks plays a significant role in a broad range of applications that consist of mobile and/or wireless nodes which need to agree on a common notion of time to successfully fulfill the tasks that they are designed for. In this thesis, we introduce an averaging based distributed synchronization algorithm in continuous time and investigate its convergence properties for networks with time varying topology and time delay. The proposed distributed synchronization scheme is advantageous in the sense that it can be applied to large-scale networks due to its energy efficiency and its robustness to node failures. In the study of convergence properties, we utilize related results from graph theory, tools of Lyapunov functions, and Laplace transform. For fixed and time varying topology networks, it is shown that average consensus is achieved in the presence of free running frequencies, if the network is strongly connected and balanced. The result is also extended to networks with equal time delay, if the network topology is fixed, undirected, and connected. Extensive simulation results are carried out in Matlab to examine the related convergence scenarios.
Analysis of cluster consensus in continuous - time networks with time delays
(Thesis (M.A.) - Bogazici University. Institute for Graduate Studies in the Social Sciences, 2018., 2018.) Develer, Ümit.; Akar, Mehmet.
In recent years, distributed coordination of multi-agent systems has the atten tion of many researchers due to its potential applications in various ﬁelds including group robots, distributed sensor networks, bird ﬂocks, the attitude alignment in multi ple spacecraft settings and control in communication systems. All of these applications indicate the importance of design and analysis of consensus protocols with which the agents agree on some particular variable of interest by exchanging information among themselves. Complete consensus where all agents in a system achieve a common ob jective is one of the most popular studies. A network using distributed consensus protocol may be divided into diﬀerent groups which are called clusters depending on the interaction topology of the network. In this thesis, we investigate the cluster con sensus problem for a multi-agent system with or without delay in continuous-time. Contrary to existing studies on cluster consensus in the literature, we do not assume that the clusters are pre-determined. The main contributions of the dissertation are to determine the number of clusters for a multi-agent system and to state delay con ditions that does not aﬀect the number of clusters and convergence properties of the system. The results are obtained by using primary and secondary layer subgraphs and sub-Laplacian matrices which are explained for the ﬁrst time in the literature. The upper bound of input delay that does not lead to instability is determined. It is also shown that bounded communication delay does not adversely aﬀect system stability properties, which is also supported by computer simulations.
Center of gravity estimation and rollover prevention using Kalman filtering techniques
(Thesis (M.S.)-Bogazici University. Institute for Graduate Studies in Science and Engineering, 2010., 2010.) Dere, Ali Dinçer.; Akar, Mehmet.
Due to its impacts on human safety and its economical cost, vehicle rollover is a very important safety issue that attracts the attention of major vehicle manufacturers and researchers. The objective of this thesis is to design a control system that only becomes active in emergency situations and prevents rollover by applying differential braking to the vehicle. More specifically, the main focus of this thesis is to estimate several unknown vehicle parameters, including the center of gravity height, that has major role in roll over, by using Kalman Filter algorithm. Subsequently, the estimated center of gravity height is used in determining the amount of differential braking force. Extensive simulations are carried out in MATLAB to demonstrate the superior performance of the proposed method.
Channel allocation system for wireless video communication in 802.11n MESH networks
(Thesis (M.S.)-Bogazici University. Institute for Graduate Studies in Science and Engineering, 2009., 2009.) Tekincan, Sinan.; Akar, Mehmet.; Anarım, Emin.
With the latest technological achievements, wireless communication is spreading all around the world and entering into the daily life of individuals more. Especially with the development of 802.11n standard, high data rate communication is becoming available, which is beneficial for multimedia applications like real-time video communi- cation. However, increased number of wireless communication users occupy the limited channel spectrum and prevent other users from accessing the channel as they desire. This issue requires proper channel allocation for access points (AP) essential. This thesis presents a complete solution for an effective channel allocation system in order to minimize the channel usage experienced by the APs. Our objective is to realize video communication for a mesh of access points operating on the same frequency. Therefore, minimizing channel switchings is very important in order not to interrupt the video communication frequently. The main steps of the proposed solution include measurement collection and processing techniques, channel switching decision and determining the optimal channel. Our proposed adaptive measurement processing method using Kalman and Median Filters can estimate channel usages accurately for both high-variance and low-variance data. By using multiple thresholding mechanisms, the proposed channel allocation algorithm not only minimizes the number of interfering APs, but it also strives to avoid unnecessary switching of the operating channel.
Combined formation control and path planning for mobile robot networks
(Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2011., 2011.) Dumlu, Semih Cemil.; Akar, Mehmet.
In this thesis, formation control and path planning algorithms in mobile robot networks are examined for both fixed and dynamically changing interaction topologies. The system is modelled using graph-based framework where the nodes denote the agents and the communication links are represented as edges of the graph. Based on this graph model, well-established algebraic graph theoretic results are employed to examine graph-based formation control for point mass holonomic and real omnidirectional robotic networks. In addition, path planning using potential fields that enables a leader as well as followers avoid obstacles in the environment while maintaining formation is analyzed. Subsequently, performance of the combined path planning and distributed formation control algorithms including saturated velocity control inputs is examined by the help of numerical simulations and it is verified that all followers reach and maintain the formation while avoiding obstacles.
Cooperative adaptive cruise control algorithms for vehicular platoons based on distributed model predictive control
(Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2019., 2019.) Taplı, Tuğba.; Akar, Mehmet.
Over the past decade, autonomous driving and driver assistance systems have become popular research topics with the aim of reducing traﬃc congestion, driver labor and rate of accidents. This has changed the requirements and priorities of today’s trans portation perception and reveals a need for an increased level of complexity. Changes in autonomous driving algorithms as a result of this diﬀerentiation have led the pla toon system applications turn into a diﬃcult control problem when evaluated together with bidirectional communication topologies and delays. Furthermore, the increased number of vehicles added to the platoon requires higher computational power. Using distributed controllers in such applications where response time is important helps us get more reliable results. The main objective of this thesis is to investigate Cooperative Adaptive Cruise Control Algorithms for Vehicular Platoons using Distributed Model Predictive Control (DMPC) under various communication links including unidirectional and bidirectional and also to propose a solution for a pre-known communication delay. Existing studies show that DMPC provides zero steady-state error with a fast response time under uni directional communication topologies, however bidirectional links and communication delay eﬀect are not adequately addressed. This thesis presents DMPC algorithms based on the information coming from the follower vehicles and delayed information received from the leader vehicle. It is shown that DMPC is the right tool to be utilized to both model and manage bidirectional communication. Simulation results demonstrate that the steady state error caused by the communication delay is successfully handled.
Delay based handover algorithm design for femtocell networks
(Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2013., 2013.) Kaymakçıoğlu, Piri.; Akar, Mehmet.
Mobile communication systems have limited frequency resources and macrocellular architecture utilizes smaller one tier cells to gain frequency reuse. One approach to enhance the overall system capacity is to use two tier femtocell-macrocell systems. The introduction of femtocells has caused a great excitement in that the femtocell system can be a solution to poor indoor coverage for mobile communications. Since femtocells are plug-and-play devices and not deployed with central planning, the handover systems used in one tier macrocell infrastructure should be updated according to the needs of two tier femtocell-macrocell systems. Femtocell base stations utilize the broadband Internet connection to connect to the operator core network. Since the Internet is a lossy network incapable of guaranteeing the quality service requirements of real time voice communications over Internet Protocol (IP) networks, the handover systems based only on the wireless received signal strength are unable to provide a solution to the quality requirements of wired systems. The quality of service requirements for wireless and wired medium are both taken into account to maintain a better call quality in the novel handover algorithms in this study. The handover rate and the signal degradation rate are the key performance indicators of the handover system optimization. The simulation studies demonstrate that better results are achieved with the proposed algorithms compared to existing ones in the literature.
Distributed consensus based adaptive ramp metering algorithms in freeway systems
(Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2019., 2019.) Dumen, Hafize Ceren.; Akar, Mehmet.
In recent years, traﬃc management has the attention of many researchers due to the public’s increasing demand for fast, eﬃcient, and convenient means of travel. In the literature, there has been a vast amount of research and improvement on traﬃc management control in freeways in various forms such as optimization, consensus pro tocols, and nonlinear control. The objective of this thesis is to develop a distributed consensus based ramp metering algorithm with the coordination of the traﬃc network. The applications of the distributed consensus algorithms in the literature indicate the importance of design and analysis of consensus protocols with which the agents in the systems achieve a common objective by exchanging information. Contrary to exist ing studies on coordinated ramp metering, centralized and decentralized on-ramp ﬂow control algorithms are proposed without assigning priorities to on-ramps with bot tlenecks at the upstream cells. Subsequently, the on-ramp ﬂow control algorithm is improved with a consensus based density control algorithm which provides a smooth traﬃc density in the traﬃc infrastructure. At the ﬁnal step, the mainstream inﬂow control is utilized in order to achieve the decided on-ramp ﬂow and the traﬃc density. The averaging based and minimum consensus protocols are used to control the main stream inﬂow. The performance and the convergence speed analysis of the proposed algorithms are evaluated with numerical simulations.
Distributed consensus in faulty networks
(Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2017., 2017.) Öksüz, Halil Yiğit.; Akar, Mehmet.
The objective of this thesis is to develop fault tolerant consensus algorithms in the presence of two types of fault models: (i) Structured Byzantine (StrBYZ) faults that send diﬀerent structured information to its in-neighbors, (ii) Unstructured Byzan tine (uStrBYZ) faults which send diﬀerent random erroneous information to its in neighbors. For the StrBYZ fault model, it is shown that the network of non-faulty nodes can achieve approximate Byzantine consensus for synchronous and asynchronous networks without using a fault tolerant algorithm. Furthermore, the analysis is extended to the study of approximate Byzantine group consensus in which the non-faulty nodes reach more than one equilibrium. As opposed to the StrBYZ fault model, it is impossible to guarantee approximate Byzantine consensus under the uStrBYZ fault model without employing a fault tolerant algorithm. To remedy this situation, two fault tolerant algorithms, so called Layered Mean-Select-Reduced (L-MSR) and Rooted Mean-Select-Reduced (R-MSR) are pro posed. Necessary and suﬃcient conditions for the success of these algorithms for single equilibrium are presented. Moreover, it is shown that L-MSR and R-MSR algorithms can also be used to solve the approximate Byzantine group consensus problem. Another contribution of this thesis is to introduce a parameter independent fault tolerant algorithm that can be used to guarantee approximate Byzantine consensus. Using the proposed distributed fault detection scheme, the requirement on the know ledge of the number of faulty agents is relaxed. Convergence analysis for the proposed algorithm is carried out.
Distributed synchronization in delayed and topology varying networks
(Thesis (M.S.)-Bogazici University. Institute for Graduate Studies in Science and Engineering, 2009., 2009.) Cihan, Onur.; Akar, Mehmet.
There are numerous network applications where nodes require a common notion of time, and as such distributed synchronization is an important task in topology varying networks. In this thesis, we introduce a well known averaging based distributed synchronization algorithm and investigate its convergence conditions under varying topologies and delay. When data are transmitted through the network, communication delays are un- avoidable and it might degrade system performance. Furthermore, delay can cause a stable system go unstable unless certain conditions are met. In this thesis, the conver- gence of the consensus algorithm for delay varying networks is studied using properties of scrambling matrices. It is shown that delay does not affect the convergence of the algorithm so long as it is bounded. The effect of delay on convergence speed for some well known topologies is also discussed. In order to reduce convergence time, fastest converging system matrices are found for networks with symmetric connections by us- ing Linear Matrix Inequalities. It is also shown that the fastest converging matrices for fixed topologies will not provide the fastest convergence for varying topology networks. Consensus algorithms are investigated for networks not only with non–faulty nodes, but also with the faulty ones (or Byzantine nodes) that try to obstruct syn- chronization by sending wrong clock information to other nodes. It is shown that a network with Byzantine nodes will not be synchronized if the network topology and synchronization algorithm do not meet some conditions. Theoretical results are also illustrated by numerical examples.
Hardware and software development for robocup SSL robots
(Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2013., 2013.) Esen, Huzeyfe.; Akar, Mehmet.
In Small Size League (SSL) which is one of the most dynamic and fastest competitions of RoboCup, a soccer game is played between two autonomous teams of 6 robots on a field of size 6050 mm 4050 mm. Because of this challenging feature of SSL, the robots should be controlled precisely so that they can react in real-time with intelligent soccer-like moves. In this thesis, we not only focus on the electronic board design of the robots, but also introduce new low and high level control algorithms that enable the robots to compete in SSL successfully. In the new electronic board that controls the main functions of the robot, the microcontroller MC9S08DZ128 and encoders with higher resolution have been utilized for precise control and the features of the motor driver MC33035 have been adapted accordingly. Morover, a new kicker circuit is designed so that the capacitors can be charged in a shorter amount of time. The algorithmic contributions of this thesis include not only a new low level control algorithm to drive brushless DC motors, but also implementation of the extended rapidly exploring random trees (ERRT) algorithm in our system for path planning and obstacle avoidance. Morover, new formation control algorithms have been developed for coordination of the robots. In this context, a Kalman filter is implemented to compansate for vision based system delay and least squares techniques are used to predict the ball position.
Joint frequency/power update algorithms for self - organizing femtocell networks
(Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2019., 2019.) Sevim, Oğuzhan.; Akar, Mehmet.
Heterogenous networks are resorted as one of the most promising ways formeetingthe rapidly increasing data demand. As the smallest members of heterogenous networks, femto base stations also carry a huge potential for increasing the service quality in indoor areas. However, their unplanned deployment by the end users increases the possibility of having dense femtocell networks with unknown topologies. Therefore, self–organizing methods have great importance in resource allocation of femtocell networks. In this thesis, power and frequency allocation problems are studied for OFDMA femtocell networks. First, we present a power update algorithm for the general wireless networks as an alternative for a well known power control algorithm from the literature. Then, this algorithm and another power control algorithm from the literature are extended for the OFDMAfemtocellnetworks. Asopposedtothepreviousversions,whichareapplicableonly in networks where each base station can have at most one user, the extended algorithms can be used by base stations that have more than one user. Additionally, a frequency allocation scheme is developed in order to increase the maximum achievable SINR in femtocell networks. Furthermore,bymergingthisschemewiththeproposedpowerupdatealgorithms,we present two joint frequency/power update algorithms with increased performance. Convergence and optimality analyses of all the proposed algorithms are carried out and illustrated with numerical results.
Model predictive control of diesel engine air path with actuator delays
(Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2019., 2019.) Kekik, Betul.; Akar, Mehmet.
In recent years, emission regulations of the countries have been tightened with the increase in the number of cars. Nevertheless, reducing the emissions of diesel engines is a diﬃcult technical issue. Since the Diesel Engine Air Path (DEAP) is a MIMO system with 2 actuators, EGR valve and VGT valve, and 2 outputs, manifold absolute pressure (MAP) and air mass ﬂow (MAF), control of this system with SISO PID controllers requires an iterative ﬁne-tuning process for controller parameters due to coupled eﬀect of VGT and EGR on MAP and MAF. The main objective of this thesis is the Model Predictive Control (MPC) of MIMO Diesel Engine Air Path system. MPC is a well-known technique in the literature with its many applications on MIMO systems. Existing results show that MPC can satify the desired settling time, zero steady-state error, and overshoot criteria for both outputs MAP and MAF. However, the eﬀect of actuator delay that considerably aﬀects system performance, is not addressed suﬃciently. In this thesis, MPC of DEAP is extended with a delay term added to actuators EGR and VGT on the plant model. A linear state-space model of the plant is obtained by using the System Identiﬁcation techniques and the states of the identiﬁed model are extended due to delay term. It is shown that MPC performs better when the delay is taken into account in the algorithm. Another contribution of this thesis is that SISO PID controllers are optimized by Particle Swarm Optimization (PSO) method. The PID gains found by Ziegler-Nichols (ZN) method are taken as the initial points of the PSO and it is shown that PSO improves the PID controller performance for the MIMO system.
Model-based and model-free control algorithms for textile processes
(Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2022., 2022) Çom, Mustafa.; Akar, Mehmet.
Textile processes consist of several control loops that require accurate reference tracking. One of the most crucial control loops is the temperature control where the temperature of the corresponding medium must track the reference value with sufficient accuracy to obtain a high-quality textile product. Even though several studies can be found on designing control algorithms for industrial processes in the literature, none of them focus particularly on the aforementioned textile processes. In this study, in order to achieve successful control, several adaptive control algorithms are developed. In addition, corresponding processes are modelled, and a simulation environment is built to increase the speed and safety of development works. Modelling is realized by dividing the corresponding process into several regions of operation, preparing sub-models for each region and building a composite model by combining these sub-models. A simulation environment is created by examining currently used control algorithms and process dynamics. The simulations of designed models result in significant accuracies. A model-based control algorithm, based on the Model Predictive Control (MPC) approach that utilises previously designed pro cess models, is developed and verified in the simulation environment. Two model-free control algorithms, referred to as Adaptive PI Control and Error Predictive Control (EPC), are developed and verified not only in the simulation environment but also in the field.
Multi agent intersection management for autonomous vehicles
(Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2019., 2019.) Ugranlı, Burak.; Akar, Mehmet.
Traditional transportation systems cause traﬃc congestion especially at the in tersections as the number of vehicles keeps increasing. This is also the main reason of air pollution and time wasted. Most of the people lose their time and money because of traﬃc congestion. Thanks to recent research on autonomous vehicles, intelligent transportation and wireless communication systems, eﬃcient traﬃc management at the intersections with multi-agent scheduling methods will be possible. The main objective of this thesis is intersection coordination for multi-agent sys tems by using time-based optimization and Model Predictive Control (MPC) methods while considering fuel economy at the intersections. Existing results show that these methods are eﬃcient in comparison to the traditional methods when all the vehicles are autonomous. However, better trajectory planning can improve the total delay of the system. Besides, including fuel economy in the optimization function can also decrease fuel consumption which would be good for both humanity and nature. In this thesis, the eﬀect of trajectory planning and diﬀerent communication ranges on time-based op timization method is studied. It is shown that a wider communication range and better trajectory planning provide less time delay. Another contribution of this thesis is to propose centralized and decentralized MPC algorithms by including fuel consumption related costs in the objective function. As a result, fuel consumption is decreased at the expense of an increase in the time delay. In simulations, it is also observed that centralized MPC performs better than decentralized MPC.
Novel fault-tolerant distributed algorithms for approximate Byzantine consensus
(Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2014., 2014.) Haseltalab, Ali.; Akar, Mehmet.
Reaching consensus in a network which contains faulty nodes is a critical problem in the eld of distributed and multi-agent systems. A distributed system which intends to do a certain task needs to display robustness against adverse behavior of some of its faulty nodes, known also as Byzantine nodes. In this thesis, a novel Mean- Select-Reduced (MSR) fault tolerant algorithm is proposed for achieving Approximate Byzantine Consensus. It is shown that the topological condition required for the success of the algorithm is more relaxed compared to the previous results. In contrary to results that appeared in the literature, it is proved that synchronicity of networks and presence of delay on communication paths do not change this condition. Subsequently, the convergence rate and time analysis for the proposed fault-tolerant algorithm is carried out and the results are extended to time-varying networks. In most of the fault-tolerant algorithms that have been introduced for Byzantine networks, it is assumed that each node has knowledge of the maximum number of faulty nodes, ft, in the network. In this thesis, we also propose a new family of algorithms which do not require this a priori information and evaluate their performance facing Byzantine failures
Predictive maintenance on flexible impeller pumps based on acoustic data
(Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2021., 2021.) Çoker Turan, Ceren.; Akar, Mehmet.
The idea of smart factories has aroused with the emergence of Industry 4.0. A way to contribute to the construction of smart factories is to develop intelligent main tenance strategies which enable factories to reduce maintenance costs and keep them away from catastrophic damages. Predictive maintenance warns users regarding the need for maintenance of assets at a specific moment. In this thesis, predictive mainte nance strategies for Flexible Impeller Pumps (FIP) are developed by using supervised and unsupervised learning methods. After observation of real textile dye houses, an emulation setup has been designed to collect data from real use cases of FIPs. Data consisting of three conditions, healthy, looseness, and cavitation, have been collected via an acoustic sensor involving 3 mi crophones. Obtained time-series signals are converted to informative features which are utilized for training a supervised model, convolutional neural network (CNN). The trained model is able to generate a warning for a specific failure state. In addition to the supervised model, an unsupervised model has also been trained for anomaly de tection of FIPs via extracted features of healthy data due to the applicability issues of unhealthy state data acquisition. The model can produce a warning when an anomaly is detected but it cannot specify the failure mode. Both models give satisfactory re sults for classifying failures and detecting anomaly situations of FIPs. Mel Frequency Cepstrum Coefficients (MFCC) have shown superiority among the features extracted to confirm human hearing based modeling can be applied successfully in maintenance methods.
Radio direction finding & smart antennas
(Thesis (M.S.) - Bogazici University. Institute for Graduate Studies in Science and Engineering, 2016., 2016.) Kebeli, Murat.; Akar, Mehmet.
In this thesis, Radio Direction Finding (DF) and Smart Antennas are studied. DF refers to the science of determining the location of radio transmission sources. A DF system uses the characteristics of the radio signal such as amplitude and phase at each antenna element to estimate the direction of arrival (DOA). In practice, three basic methods are preferred in DF systems; Watson Watt, Interferometer and High Resolution methods. This work focuses on the implementation of a novel DF system based on the interferometer method. However, an ambiguity problem occurs in estimating DOA when antenna aperture becomes larger than the half of the incident signal's wavelength. The dominant work in the rst part of the thesis is resolving this ambiguity in interferometric DF systems. It is veri ed that the FFT based DF system performance is improved with the increasing antenna aperture and that the working frequency bandwidth of the system is also increased by using a single antenna array. The second part of the thesis focuses on Smart Antennas that can dynamically adjust the beam-pointing direction of the antenna array by applying phase delays to the antenna elements. There is a tremendous need and increase in use of smart antennas with the growth of mobile networks. Three basic beamforming approaches are used in smart antennas: xed, switched and adaptive. We propose a novel multi-model beamforming model that integrates the switched and adaptive beamforming systems that combines the advantages of a switched and an adaptive beamforming systems. It gives faster response than an adaptive only system and provides perfect tracking to all possible combinations of desired signal and unwanted signals compared to a switched beamforming only system.

Browse

Browsing M.S. Theses by Author "Akar, Mehmet."

Results Per Page

Sort Options