Filtreler

Filtreleri Sıfırla

Ayarlar

Bölüm: Işık University, Faculty of Engineering, Department of Electrical-Electronics Engineering × Konu: Optimization ×

Arama Sonuçları

Listeleniyor 1 - 10 / 27
  • Küçük Resim
    Yayın
    Cognitive cooperative MAC with one primary and two secondary users: achievable rates and optimal power control
    (IEEE-INST Electrical Electronics Engineers Inc, 2014-11) Edemen, Çağatay; Kaya, Onur
    We consider a three-user fading cognitive cooperative multiple access channel (MAC) with one primary and two secondary transmitters. We propose two encoding/decoding strategies with varying levels of cooperation, based on block Markov superposition encoding and backward decoding. The first is an overlay model, where the secondary users (SUs) aid the transmission of the primary user (PU) by causally decoding the PU message and forwarding it while also cooperating among each other. The second is an underlay model, where the SUs cooperate by decoding and forwarding each other's messages while treating the signal received from the PU as noise. In either case, the PU is guaranteed to operate at its maximum achievable single user rate. We characterize the achievable SU rate region for both models and maximize this region as a function of the transmit powers. The simulation results show that the SU rate region can be significantly enlarged, particularly using the overlay model.
  • Küçük Resim
    Yayın
    Optimal primary-secondary user pairing and power allocation in cognitive cooperative multiple access channels
    (IEEE, 2014) Bakşi, Saygın; Kaya, Onur
    We develop jointly optimal power control and primary-secondary user partnering strategies for a cognitive cooperative multiple access channel with K primary and K secondary users. For each primary user, a cooperating secondary user is assigned. We consider both underlay and overlay modes for cognition/cooperation. In overlay mode, each secondary user decodes and relays part of its assigned primary user's message, and simultaneously transmits its own independent message, while ensuring the primary user achieves at least its single user capacity with power control. The encoding is based on channel adaptive block Markov superposition coding, where the powers assigned to primary and secondary user codewords are optimized so as to maximize either the system's sum rate, or the sum of secondary users' rates. In underlay mode, each secondary user employs independent signalling and allocates its power to maximize its own rate, without decreasing its assigned primary user's rate. The partnering problem for either mode is reduced to a maximum weighted matching (MWM) problem on a bipartite graph, and solved jointly optimally with the power allocation problem.
  • Küçük Resim
    Yayın
    Design and implementation of wideband microwave amplifiers based on Normalized Gain Function
    (Institute of Electrical and Electronics Engineers Inc., 2014) Köprü, Ramazan; Kılınç, Sedat; Aksen, Ahmet; Yarman, Bekir Sıddık Binboğa
    In this work, we introduce the design and implementation of wideband microwave amplifiers based on "Normalized Gain Function (NGF)" method. Normalized Gain Function is defined as the ratio of desired shape or frequency response of the gain function of the amplifier to be designed and shape of the transistor forward gain function. Synthesis of input/output matching networks (IMN/OMN) of the amplifier require target gain curves as the functions of normalized gain function to be tracked in two sequential nonlinear optimization processes. A prototype low power amplifier circuit is produced and measured to show the usability of the design approach.
  • Küçük Resim
    Yayın
    Rate-distortion and complexity joint optimization for fast motion estimation in H.264 video coding
    (IEEE, 2006) Ateş, Hasan Fehmi; Kanberoğlu, Berkay; Altunbaşak, Yücel
    H.264 video coding standard offers several coding modes including inter-prediction modes that use macroblock partitions with variable block sizes. Choosing a rate-distortion optimal mode among these possibilities contributes significantly to the superior coding efficiency of the H.264 encoder. Unfortunately, searching for optimal motion vectors of each possible subblock incurs a heavy computational cost. In this paper, in order to reduce the complexity of integer-pel motion estimation, we propose a rate-distortion and complexity joint optimization method that selects for each MB a subset of partitions to evaluate during motion estimation. This selection is based on simple measures of spatio-temporal activity within the MB. The procedure is optimized to minimize mode estimation error at a certain level of computational complexity. Simulation results show that the algorithm speeds up the motion estimation module by a factor of up to 20 with little loss in coding efficiency.
  • Küçük Resim
    Yayın
    Power control for fading cooperative multiple access channels
    (IEEE, 2007-08) Kaya, Onur; Ulukuş, Şennur
    For a fading Gaussian multiple access channel with user cooperation, we obtain the power allocation policies that maximize the average rates achievable by block Markov superposition coding, subject to average power constraints. The optimal policies result in a coding scheme that is simpler than the one for a general multiple access channel with generalized feedback. This simpler coding scheme also leads to the possibility of formulating an otherwise non-concave optimization problem as a concave one. Using the perfect channel state information available at the transmitters to adapt the powers, we demonstrate gains over the achievable rates for existing cooperative systems.
  • Küçük Resim
    Yayın
    Cooperative multiple access under energy harvesting constraints
    (Institute of Electrical and Electronics Engineers Inc, 2015) Su, Nuğman; Kaya, Onur; Ulukuş, Şennur; Koca, Mutlu
    We consider a cooperative multiple access channel (MAC) with two energy harvesting transmitters. The transmitters perform delay constrained transmission, i.e., every information block is encoded, transmitted and decoded between two consecutive energy harvests. We aim to maximize the achievable departure region over a finite transmission duration. We formulate the departure region maximization as a convex optimization problem. We propose an iterative algorithm which uses a directional waterfilling strategy to calculate the optimal power components. The departure region obtained by cooperation is shown to be significantly larger than that of a MAC without cooperation under the same energy arrival patterns. As a special case, we also analyze an energy harvesting relay channel with full duplex cooperation.
  • Küçük Resim
    Yayın
    Optimal nonlinear controller design for flexible robot manipulators with adaptive internal model
    (INST Engineering Technology-IET, 2007-05) Doğan, Mustafa; İstefanopulos, Yorgo
    Developing nonlinear adaptive and robust controllers for a two-link flexible robot arm is the main objective of this research. The dynamic state feedback controller is used to achieve robust regulation of the rigid modes as well as suppression of elastic vibrations. However, the control of highly nonlinear multi-link flexible arms is subject to uncertainties caused by backlash, payload changes and external disturbances. Therefore adaptive and robust control of multi-link flexible arms is a challenging problem. The internal model approach is adaptively tuned up for unknown disturbances, parallel with a robust stabiliser. The stabiliser part of the controller is optimised with a new evolutionary algorithm.
  • Küçük Resim
    Yayın
    A numerical real frequency broadband matching technique based on parametric representation of scattering parameters
    (IEEE, 1998-11-24) Aksen, Ahmet; Çimen, Ebru Gürsu; Yarman, Bekir Sıddık Binboğa
    In this study, a computer, aided broadband matching technique based on the parametric representation of scattering parameters is proposed. The new technique combines the advantageous of the so called parametric approach utilizing the impedance formalizm and the real frequency scattering technique. With the reformulation of these real frequency techniques, some laborous numerical computational steps are omitted. The descriptive network functions are obtained in an explicit form without any need to the polynomial factorizations and the solution of equation sets. The application of the technique and the comparision of the results with those of the other real frequency approaches is discussed on examples.
  • Küçük Resim
    Yayın
    Mixed element wideband microwave amplifier design via simplified real frequency technique
    (IEEE Computer Society, 2014) Kılınç, Sedat; Köprü, Ramazan; Aksen, Ahmet; Yarman, Bekir Sıddık Binboğa
    In this study, we illustrate the design and implementation of a wideband microwave small-signal amplifier composed of mixed elements. The design is based on Simplified Real Frequency Technique (SRFT). A design of low power amplifier circuit is completed and its simulations are performed in success. The circuit is designed with lumped elements, however, some of the lumped elements are converted to distributed elements for their convenience in production. In this way, a mixed element wideband microwave amplifier comprised of input/output matching networks with lumped and distributed elements has been formed. Layout work and also post layout simulation is given with satisfying results.
  • Küçük Resim
    Yayın
    Circuit model for given reflectance data constructed with mixed lumped and distributed elements for high speed/high frequency communication systems
    (IEEE, 2005) Yarman, Bekir Sıddık Binboğa; Şengül, Metin; Kılınç, Ali; Aksen, Ahmet
    In this paper, a reflectance-based "non linear interpolation method" is presented to model the measured or computed data, obtained from a "passive one-port physical device" using mixed lumped and distributed elements. Mixed element model is constructed with cascade connection of series inductors [L], commensurate transmission lines or so called Unit Elements [UE] and shunt capacitors[C]. Basis of the new model rests on the numerical generation of the scattering parameters of the lossless two-port constructed with cascade connection of simple [L]-[UE]-[C] elements which describes a lossless 2-port in Darlington sense. The new modeling technique does not require direct optimization of the circuit elements of the selected topology. Rather, two-variable reflection coefficient is directly determined by means of a non linear but "convergence guaranteed" interpolation process to best fit the given data. A low-pass filter input reflection coefficient modeling example is included to exhibit the utilization of the proposed modeling method.