601
Arşivlenen Tez
0
DOI Atanmış
0%
DOI Oranı
Arşivlenen Tez
FPGA'lerin hızlı, güvenli ve uzaktan çoklu yüklenmesi
The purpose of this thesis is to develop an efficient framework to implement secure FPGA-based (Field Programmable Gate Array) systems. An FPGA is a reconfigurable device that has the ability to adapt the hardware during runtime by loading a new circuit on the reconfigurable fabric. However, a circuit design formed as configuration data (bitstream) can be easily counterfeited and needs to be protected against the risks of cloning, overbuilding, and reverse-engineering. Although many applications could be implemented on low-cost FPGAs, protection schemes and dedicated hardware are mostly available on high-end FPGAs. In addition to this, only high-end FPGAs support dynamic partial self reconfiguration (DPSR), which is the ability tochange a part of a design at runtime. This thesis focuses on developing a security scheme leveraging hardware intrinsic features on low-cost FPGAs by using physical unclonable functions (PUFs). A PUF provides a way to extract security keys which are unique to each device. This thesis combines PUFs with another security scheme called obfuscation. Obfuscation is the act of intentionally modifying the description or structure of a circuit in order to conceal its functionality. Obfuscation is implemented in this thesis at RTL-level and is used to authenticate and control the device by using the keys by exploiting the PUF technique within a finite state machine (FSM). These methods are further used to implement ?secure MultiBoot?. The MultiBoot feature allows to reconfigure the FPGA fully at runtime as opposed to DPSR for devices which do not support partial reconfiguration. This thesis also establishes a framework that enables secure remote MultiBoot. A bitstream compression technique is applied to reduce the transmission time over the network. A proof-of-concept example is implemented using the proposed framework.
Yarı rezonans yapı ve yayılmış tayf metodu ile indüksiyon ocak tasarımı ve deneysel sonuçları
This thesis presents an induction cooker design and its detailed experimental results. The design is based on Quasi Resonant Topology and is unique in terms of the particular Jitter Method it employs to drive the semiconductor switches (IGBTs).Most induction cookers are based on Half Bridge Topology. Although Quasi Resonant Topology offers a more cost effective implementation, it introduces more EMI problems. However, we solve these problems with our Jitter Method. Although there are other works in the literature that solve the EMI issues by some sort of jitter methods (i.e., spread spectrum methods), these methods do not address the high voltage stress created across the IGBTs. However, the Jitter Method proposed in this thesis also reduces collector-emitter voltage stress besides improving EMI. Note that jitter methods also prevent mechanical vibration (i.e., resonance), which cause audible noise.Another thing that makes this thesis unique in the literature is the rich set of experimental results it contains ? based on an experimental prototype we built with a power rating of 2300W. While the existing literature mostly contains theoretical treatment, we have produced detailed experimental results for Quasi Resonant Topology and show that the current harmonics (drawn from the utility) stay within the published standards (TS EN 61000-3-2). We also include experimental results for Delivered Power Detection, Pan Detection, and Protection Circuits.
Çoklu-seçim dönmeli iş düzenleyici
In this thesis, we propose two multi(m)-pick Round Robin Arbiter (RRA) architectures. An m-pick RRA selects the m topmost requests out of n inputs with priority order indicated by an internally kept pointer (with an update policy that ensures fairness among requestors). The architectures that we propose are m-pick Thermo Coded-Parallel Prefix Arbiter (TC-PPA) and Three-Dimensional Programmable m-Selector RRA (3DPmS-RRA). Prior to this thesis, these two architectures existed in the literature as 2-pick and 1-pick arbiters, respectively. Our main contribution to the literature is the generalization of these architectures to m-pick. A logic building block that we call ?Saturated Adder? plays a key role in this generalization, which makes the 1-pick and 2-pick architectures simply special cases. We developed six differentvariants of 3DPmS-RRA and eight different variants of m-pick TC-PPA. We wrote automated HDL code generators for all variants as well as Cascade Architecture, which is a straight-forward way of implementing a multi-pick RRA using 1-pickProgrammable Priority Encoders. Then, all multi-pick architectures were verified and synthesized. Our experimental results show that 3DPmS-RRA architecture is the best choice for all pick sizes (except 2-pick) when timing is the primary design criterion. However, when area is more critical, TC-PPA architecture performs better. It is worthwhile to note that in terms of timing 3DPmS-RRA is better than TC-PPA by a mere 8% at the most based on our synthesis results. However, when we consider area, TC-PPA has significant improvements over 3DPmS-RRA, up to 53%.
ALTO servisli eş görevli ağlarda gerçek zamanlı ölçeklenebilir video akıtımı
It has been measured that, as high as 80% of the average Internet traffic is now from peer-to-peer (P2P) applications, and upwards of 90% during peak hours. These results tell us that a significant portion of the traffic that ISPs route is P2P traffic. However, since the typical P2P application overlay network model is network-oblivious, due to its nature, it causes inefficient network utilization, and low qualityof experience, from the perspective of the network providers and the end users, respectively. This inefficient resource utilization increases inter-ISP traffic, and leads to serious disruptions on ISP economics as well as lower QoE by the users.A significant portion of the P2P traffic on the Internet is for video applications. Furthermore, real networks are heterogeneous in link rates. Thus a live video streamingservice should support streams with different qualities for users using different data rates. A possible solution to this problem is achieved by creating a different video file for each quality level. However, this solution is inefficient due to data duplication.To improve ISP economics, improve network efficiency and achieve acceptable QoE performance, we implement a P2P Live Scalable Video Streaming Mechanism and Application Layer Traffic Optimization (ALTO) Server, by using different peer ranking algorithms in this thesis. The main objective of this thesis, to show that the ALTO protocol reduces the inter-ISP traffic significantly while maintaining the application performance. Results reveal that, placing an ALTO Server decreases the inter-ISP traffic dramatically while maintaining a satisfactory QoE performance for the streaming video application.
Performance analysis and optimization of relay-assisted free space optical communication systems
Free-space optical (FSO) communication has received an increasing attention in recent years with their ability to achieve ultra-high data rates (at the order of multiple gigabits per second) over unlicensed optical spectrum. A major degrading factor, particularly in long links, is the atmospheric turbulence induced fading. Smartly exploiting the fact that fading variance is distance-dependent in FSO channel, relay assisted transmission takes advantage of the resulting shorter hops and yields significant performance improvements. In this thesis, we make several contributions to the performance analysis and design of relay-assisted FSO systems. In the first part, we investigate how to determine optimal relay locations in serial and parallel FSO relaying as to minimize the outage probability and quantify performance improvements obtained through optimal relay placement. In the second part, we investigate the combined use of serial and parallel relaying for FSO mesh networks. We derive the outage probability expressions of FSO mesh networks and demonstrate performance improvements with respect to both stand-alone serial and parallel relaying schemes. We also present a diversity gain analysis and quantify the achievable diversity orders in terms of the number of relays and turbulence channel parameters. Finally, we investigate the use of all-optical relaying removing the need for optical-electrical and electrical-optical conversions. Building on all-optical relay assumption, we investigate the outage performance of a dual-hop FSO link taking into account practical limitations such as amplifier noise and filtering effects.
Femto-hücre ağlarında en düşük inen bağlantı girişimi için ortak frekans tahsisi ve güç kontrolü
Due to the fact that the most of data services and more than half of calls are requested indoors, it has been dramatically important and even a must for cellular operators to provide more than adequate indoor coverage. Instead of expensive traditional approach which is to use outdoor macrocell base stations to provide good indoor coverage, indoor systems for the cases of small offices and homes are found more convenient and efficient.Femtocell is one of the developing technologies as a recent solution to aforementioned problems. They offer to improve the indoor coverage and reduce the cellular network load. However, in case of the most feasable solution, co-channel deployment of the femtocell network over the existing macrocell network, a closed access approach could result in serious co-channel interference problems.In this thesis, improvements to the performances of both femtocell and indoor macrocellular users in terms of average achieved data rates are considered. A joint spectrum and power control algorithm is proposed to mitigate the co-tier and cross-tier interferences. The algorithm aims to save the indoor macrocell subscribers from the violation of femtocells whilst providing throughput to femtocell subscribers as higher as possible. Simulations including physical layer and system level are conducted to determine the performance of the proposed method. Results reveal that an algorithm that combines the management of two important resources, spectrum and power, performs really well in two-tier networks, even better than the techniques managing only one resource, in terms of overall system performance.
Femto hücre ağları için uygulama bazlı çizelgeleyici
This thesis is concerned with designing a QoS-enabled, application-based schedulerfor indoor femtocell networks. For this purpose, a novel scheduler is proposed whichtakes into account the quality of service (QoS) requirements of each active applicationas well as whether their sustainability is at risk at a given time. The scheduler aimsto maintain an efficient use of the wireless radio resources and strives to achieve auser and application specific notion of fairness of service reception. Unlike current so-lutions, the scheduler differentiates between not only different users but also betweenapplications of each user, and schedules data transmission amongst these applicationsso that QoS requirements for all applications are satisfied.
Morfolojik olarak zengin diller için melez istatistiksel/birim seçmeli metinden konuşma sentezleme sistemi
Two most prominent examples of Text-to-Speech (TTS) systems are Unit Selection based TTS (UTTS) and the Hidden Markov Model (HMM) based TTS (HTTS). UTTS has been the dominant approach of the last decade while HTTS has been increasingly getting more attention from the TTS research community. Both systems have distinct pros and cons. Despite its success, UTTS has some disadvantages such as the sudden discontinuities in speech which cause distraction whereas HTTS lacks of those artifacts. However, UTTS systems offer high quality speech given a huge unit database where the storage is not a problem. On the other hand, the small memory footprint requirement of HTTS systems makes them attractive for embedded devices. Here, a novel hybrid statistical/unit selection TTS system for morphologically rich languages is proposed. The proposed hybrid system aims at improving the quality of the baseline HTTS system while keeping the memory footprint small. First, the motivation of the proposed hybrid system is given after the comparison of both systems. Then the proposed hybrid system is presented along with the details of the baseline HTTS system. In order to assess the performances of proposed and baseline systems, the subjective and objective tests are conducted. Intelligibility and quality scores of the baseline system are comparable to the MOS scores of English reported in the Blizzard Challenge tests. Results of the AB preference tests revealed the listeners' preference for the hybrid system over the baseline system.
Hizmet olarak hata maskeleme
In service-oriented architectures, composite services depend on a set of partner services to perform their tasks. These partner services may become unavailable due to system and/or network faults, leading to an increased error rate for the composite service. In this dissertation, we propose an approach to prevent the occurrence of errors that result from the unavailability of partner services. We introduce an external Web service, FAS (Fault Avoidance Service), to which composite services can register at will. After registration, FAS periodically checks the partner links, detects unavailable partner services, and updates the composite service with available alternatives. Thus, in case of a partner service error, the composite service will have been updated before invoking the partner service. We provide mathematical analysis regarding the error rate and the ratio of false positives with respect to the monitoring frequency of FAS for various partner service availability rates. We obtain empirical results regarding these metrics based on several tests we performed using the Amazon Elastic Compute Cloud. We use these results to evaluate our mathematical analysis. We also introduce an industrial case study for improving the quality of a service-oriented system from the broadcasting and content delivery domain.
Metinden bağımsız konuşmacı doğrulama sistemlerinin çok sınıflı, yarı otomatik bir tanıma senaryosunda karşılaştırılması
Performance of the speaker veri cation systems is typically measured based on their binary decision accuracy. Soft outputs of the systems are used mostly for calibration or multiple system combination purposes. However, in speaker veri cation applications where close to 100% accuracy is required, such as the systems that are used in the call centers of nance companies, it is not possible to rely on the binary decisions of the existing veri cation systems. Still, in such cases, multi-class veri cation outputs (for example, high, medium and low veri cation score) returned by the speaker veri cation systems can be used by a human agent to either reduce the veri cation time and/or increase the veri cation accuracy compared to a human-only scenario. In this thesis, an overview of a speaker veri cation system is given explaining in detail the algorithms that are implemented. Particularly the details about a classi- er, GDA, which was rstly used by us for a veri cation purpose are given. It does relatively better job than state of the art algorithms for non-linear data like in our case. In the experiments section, some of the most popular speaker veri cation systems are compared in terms of the classical performance metric used in the literature. Then, multi-class output performance of them is compared when a human agent is assumed to be in the veri cation loop. Performance is measured by the reduction in the number of questions used by the human agent for verifying the identity of the caller without compromising the security. Experiments are performed using the NIST 2006 and 2008 databases. Eight and one conversation sides (5 minutes each) enrollment data and 1 side and 10 seconds veri cation data conditions are used.