Yazar "Tunali E.T." seçeneğine göre listele

Listeleniyor 1 - 12 / 12
  • Küçük Resim Yok
    Öğe
  • Küçük Resim Yok
    Öğe
    Adaptive streaming of scalable stereoscopic video over DCCP [Ölçeklenebilir stereo videolarin DCCP üzerinden uyarlamali akişlandirilmasi]
    (2007) Özbek N.; Görkemli B.; Tekalp A.M.; Tunali E.T.
    We propose a new adaptive streaming model that utilizes DCCP in order to efficiently stream multi-view video over the Internet for 3DTV transport. The model effectively allocates the available channel bandwidth, which is calculated by DCCP, among the views. The video rate is adapted to the DCCP rate for each group of pictures (GoP) by adaptive extraction of layers from a scalable multi-view bitstream. The objective of the streaming model is to maximize perceived quality of the received 3D video while minimizing the number of possible display interrupts. Experimental results successfully demonstrate stereo video streaming over DCCP on wide area network.
  • Küçük Resim Yok
    Öğe
    Bandwidth-aware multiple multicast tree formation for P2P scalable video streaming using hierarchical clusters
    (IEEE Computer Society, 2009) Fesci-Sayit M.; Tunali E.T.; Tekalp A.M.
    Peer-to-peer (P2P) video streaming is a promising method for multimedia distribution over the Internet, yet many problems remain to be solved such as providing the best quality of service to each peer in proportion to its available resources, low-delay, and fault tolerance. In this paper, we propose a new bandwidth-aware multiple multicast tree formation procedure built on top of a hierarchical cluster based P2P overlay architecture for scalable video (SVC) streaming. The tree formation procedure considers number of sources, SVC layer rates available at each source, as well as delay and available bandwidth over links in an attempt to maximize the quality of received video at each peer. Simulations are performed on NS2 with 500 nodes to demonstrate that the overall performance of the system in terms of average received video quality of all peers is significantly better if peers with higher available bandwidth are placed higher up in the trees and peers with lower bandwidth are near the leaves. ©2009 IEEE.
  • Küçük Resim Yok
    Öğe
    A comparative study of compressed sensing video encoding GOP patterns for stereo distributed video coding [Stereo dagitik video kodlama i·çi·n sikiştirmali algilamali video kodlama gop desenleri· i·le karşilaştirmali bi·r çalişma]
    (2012) Aslan S.; Tunali E.T.
    In this study, compressed sensing concepts are applied to multi-view video coding. Existing work from single view video is utilized to develop efficient GOP patterns and reference framing for stereo coding. It has been observed that the most typical choice of pattern improved the characteristics 0.4 dB with respect to the model that do not benefit from interview sparsity for all frames. Alternatives for future work are proposed. © 2012 IEEE.
  • Küçük Resim Yok
    Öğe
    A dynamic line rate assignment strategy for MLR WDM optical networks
    (Institute of Electrical and Electronics Engineers Inc., 2018) Algin G.B.; Tunali E.T.
    In this study, we propose a new dynamic line rate assignment strategy that responds to load intensity fluctuations of daily traffic in wavelength-division-multiplexed (WDM) mixed-line-rate (MLR) optical networks. We present Survivable Routing with Rate and Wavelength Assignment (SRRWA) and we propose an efficient shared backup path protection solution for dynamic traffic model. We prepared simulation scenarios to demonstrate the effect of handling load intensity periods over routing performance in terms of communication cost and resource utilization. We compared the results of the proposed strategy with static line rate assignment methods. The simulation results show that proposed strategy has an average performance while highest resource utilization and lowest cost values are obtained by the static rate matching line rate assignment method. © 2017 IEEE.
  • Küçük Resim Yok
    Öğe
    An efficient JSD-based search on interest-based hierarchical clustering of overlay networks
    (2010) Bulut H.; Yardimci A.; Demirci S.; Kaymak Y.; Fesci-Sayit M.; Tunali E.T.
    In P2P networks, peers share contents, especially video files, which represent their interests. However, the underlying P2P topology may not represent this interest distribution. Thus, one important aspect of constructing an efficient P2P network is to exploit the interest similarity among peers. In this paper, we propose a hierarchical clustering mechanism for constructing an overlay network that takes interest similarity among peers into account. By measuring the similarity among interests of peers and clusters, interest-based hierarchical clusters are formed by using Jensen-Shannon Divergence metric. The clustering performance metrics, accuracy and correctness, are reported on PlanetLab. For limited keyword collections, a novel Jensen-Shannon Divergence-based search mechanism is implemented. It has been observed that the integrated mechanism provides an efficient method and better performance as compared to classical keyword-based search. Copyright © IARIA, 2010.
  • Küçük Resim Yok
    Öğe
    The mapping of an adaptive robot control algorithm onto RM, a reconfigurable machine for highly parallel real-time applications
    (Institute of Electrical and Electronics Engineers Inc., 1993) Erdogan S.S.; Tunali E.T.; Kuo N.
    The paper describes a configuration of the reconfigurable machine (RM) specifically optimized to handle inverse dynamics computation for robot control and to implement a neural network control strategy in a parallel platform. The difficulties of operating on a reconfigurable platform are overcome by defining a fixed processing element (PE) model with multiple processing components and by using the PE's local memory for storing micro-instructions to drive the parallel processors. Traditionally, reconfigurable platforms have been used to contain all the execution sequences and a local memory is usually provided to store the necessary data. VHDL description of the model has been used for simulation, synthesis and optimization for mapping to XILINX 4005 FPGA technology. © 1993 IEEE.
  • Küçük Resim Yok
    Öğe
    New Results for Identifiability of Nonlinear Systems
    (1987) Tunali E.T.; Tarn T.-J.
    The parameter identifiability problem of deterministic non-linear control systems is studied. Relations between nonlinear observability, nonlinear functional expansions, the uniqueness theorem of nonlinear realization theory, and identifiability are investigated. By using these relations, necessary and sufficient conditions for identifiability are obtained for the first time. These results provide insight into the identifiability problem for nonlinear systems. Copyright © 1987 by The Institute of Electrical and Electronics Engineers, Inc.
  • Küçük Resim Yok
    Öğe
    A novel study on developing a model-driven visual dictionary [Model-güdümlü görsel sözlük gelistirme üzerine yenilikçi bir çalisma]
    (Institute of Electrical and Electronics Engineers Inc., 2016) Aslan S.; Akgul C.B.; Sankur B.; Tunali E.T.
    We designed SymPaD framework, a model-driven visual dictionary construction and description method, with new shape models and quantized shape library. We demonstrate that, with this new design, the most current model-driven dictionary construction method is outperformed with even smaller dictionary in object recognition and image retrieval tasks. © 2016 IEEE.
  • Küçük Resim Yok
    Öğe
    On optimal receiver buffer size in adaptive Internet video streaming
    (2006) Ozbek N.; Tunali E.T.
    The effect of receiver buffer size on perceived video quality of an Internet video streamer application was examined in this work. Several network conditions and several versions of the application are used to gain understanding of the response to varying buffer sizes. Among these conditions local area versus wide area, bandwidth estimation based versus non-bandwidth estimation based cases are examined in detail. A total of 1000 min of video is streamed over Internet and statistics are collected. It was observed that when bandwidth estimation is possible, choosing larger buffer size for higher available bandwidth yields quality increase in perceived video.
  • Küçük Resim Yok
    Öğe
    Robust quality adaptation for internet video streaming
    (2005) Tunali E.T.; Kantarci A.; Ozbek N.
    Internet video streaming is a widely popular application however, in many cases, congestion control facilities are not well integrated into such applications. In order to be fair to other users that do not stream video, rate adaptation should be performed to respond to congestion. On the other hand, the effect of rate adaptation on the viewer should be minimized and this extra mechanism should not overload the client and the server. In this paper, we develop a heuristic approach for unicast congestion control. The primary feature of our approach is the two level adaptation algorithm that utilizes packet loss rate as well as receiver buffer data to maintain satisfactory buffer levels at the receiver. This is particularly important if receiver has limited buffer such as in mobile devices. When there is no congestion, to maintain best buffer levels, fine grain adjustments are carried out at the packet level. Depending on the level of congestion and receiver buffer level, rate shaping that involves frame discard and finally rate adaptation by switching to a different pre-encoded video stream are carried out. Additive increase multiplicative decrease policy is maintained to respond to congestion in a TCP- friendly manner. The algorithm is implemented and performance results show that it has adaptation ability that is suitable for both local area and wide area networks. © 2005 Springer Science + Business Media, Inc.
  • Küçük Resim Yok
    Öğe
    SymPaD: Symbolic patch descriptor
    (SciTePress, 2015) Aslan S.; Akgül C.B.; Sankur B.; Tunali E.T.
    We propose a new local image descriptor named SymPaD for image understanding. SymPaD is a probability vector associated with a given image pixel and represents the attachment of the pixel to a previously designed shape repertoire. As such the approach is model-driven. The SymPad descriptor is illumination and rotation invariant, and extremely flexible on extending the repertoire with any parametrically generated geometrical shapes and any desired additional transformation types. Copyright © 2015 SCITEPRESS - Science and Technology Publications All rights reserved.