Browsing by Subject "Optimization"
Now showing 1 - 20 of 247
Results Per Page
Sort Options
Item Open Access Access pattern-based code compression for memory-constrained systems(Association for Computing Machinery, 2008-09) Ozturk, O.; Kandemir, M.; Chen, G.As compared to a large spectrum of performance optimizations, relatively less effort has been dedicated to optimize other aspects of embedded applications such as memory space requirements, power, real-time predictability, and reliability. In particular, many modern embedded systems operate under tight memory space constraints. One way of addressing this constraint is to compress executable code and data as much as possible. While researchers on code compression have studied efficient hardware and software based code compression strategies, many of these techniques do not take application behavior into account; that is, the same compression/decompression strategy is used irrespective of the application being optimized. This article presents an application-sensitive code compression strategy based on control flow graph (CFG) representation of the embedded program. The idea is to start with a memory image wherein all basic blocks of the application are compressed, and decompress only the blocks that are predicted to be needed in the near future. When the current access to a basic block is over, our approach also decides the point at which the block could be compressed. We propose and evaluate several compression and decompression strategies that try to reduce memory requirements without excessively increasing the original instruction cycle counts. Some of our strategies make use of profile data, whereas others are fully automatic. Our experimental evaluation using seven applications from the MediaBench suite and three large embedded applications reveals that the proposed code compression strategy is very successful in practice. Our results also indicate that working at a basic block granularity, as opposed to a procedure granularity, is important for maximizing memory space savings. © 2008 ACM.Item Open Access Adaptive filtering approaches for non-Gaussian stable processes(IEEE, 1995-05) Arıkan, Orhan; Belge, Murat; Çetin, A. Enis; Erzin, EnginA large class of physical phenomenon observed in practice exhibit non-Gaussian behavior. In this paper, α-stable distributions, which have heavier tails than Gaussian distribution, are considered to model non-Gaussian signals. Adaptive signal processing in the presence of such kind of noise is a requirement of many practical problems. Since, direct application of commonly used adaptation techniques fail in these applications, new approaches for adaptive filtering for α-stable random processes are introduced.Item Open Access Adaptive filtering for non-gaussian stable processes(IEEE, 1994) Arıkan, Orhan; Çetin, A. Enis; Erzin, E.A large class of physical phenomenon observed in practice exhibit non-Gaussian behavior. In this letter, a-stable distributions, which have heavier tails than Gaussian distribution, are considered to model non-Gaussian signals. Adaptive signal processing in the presence of such a noise is a requirement of many practical problems. Since direct application of commonly used adaptation techniques fail in these applications, new algorithms for adaptive filtering for α-stable random processes are introduced.Item Open Access An adaptive, energy-aware and distributed fault-tolerant topology-control algorithm for heterogeneous wireless sensor networks(Elsevier BV, 2016) Deniz, F.; Bagci, H.; Korpeoglu, I.; Yazıcı A.This paper introduces an adaptive, energy-aware and distributed fault-tolerant topology-control algorithm, namely the Adaptive Disjoint Path Vector (ADPV) algorithm, for heterogeneous wireless sensor networks. In this heterogeneous model, we have resource-rich supernodes as well as ordinary sensor nodes that are supposed to be connected to the supernodes. Unlike the static alternative Disjoint Path Vector (DPV) algorithm, the focus of ADPV is to secure supernode connectivity in the presence of node failures, and ADPV achieves this goal by dynamically adjusting the sensor nodes' transmission powers. The ADPV algorithm involves two phases: a single initialization phase, which occurs at the beginning, and restoration phases, which are invoked each time the network's supernode connectivity is broken. Restoration phases utilize alternative routes that are computed at the initialization phase by the help of a novel optimization based on the well-known set-packing problem. Through extensive simulations, we demonstrate that ADPV is superior in preserving supernode connectivity. In particular, ADPV achieves this goal up to a failure of 95% of the sensor nodes; while the performance of DPV is limited to 5%. In turn, by our adaptive algorithm, we obtain a two-fold increase in supernode-connected lifetimes compared to DPV algorithm.Item Open Access Aggregate profile clustering for telco analytics(2013) Abbasoğlu, M.A.; Gedik, B.; Ferhatosmanoğlu H.Many telco analytics require maintaining call profiles based on recent customer call patterns. Such call profiles are typically organized as aggregations computed at different time scales over the recent customer interactions. Customer call profiles are key inputs for analytics targeted at improving operations, marketing, and sales of telco providers. Many of these analytics require clustering customer call profiles, so that customers with similar calling patterns can be modeled as a group. Example applications include optimizing tariffs, customer segmentation, and usage forecasting. In this demo, we present our system for scalable aggregate profile clustering in a streaming setting. We focus on managing anonymized segments of customers for tariff optimization. Due to the large number of customers, maintaining profile clusters have high processing and memory resource requirements. In order to tackle this problem, we apply distributed stream processing. However, in the presence of distributed state, it is a major challenge to partition the profiles over machines (nodes) such that memory and computation balance is maintained, while keeping the clustering accuracy high. Furthermore, to adapt to potentially changing customer calling patterns, the partitioning of profiles to machines should be continuously revised, yet one should minimize the migration of profiles so as not to disturb the online processing of updates. We provide a re-partitioning technique that achieves all these goals. We keep micro-cluster summaries at each node, collect these summaries at a centralize node, and use a greedy algorithm with novel affinity heuristics to revise the partitioning. We present a demo that showcases our Storm and Hbase based implementation of the proposed solution in the context of a customer segmentation application. © 2013 VLDB Endowment.Item Open Access An algorithm based on facial decomposition for finding the efficient set in multiple objective linear programming(Elsevier, 1996) Sayın, S.We propose a method for finding the efficient set of a multiple objective linear program based on the well-known facial decomposition of the efficient set. The method incorporates a simple linear programming test that identifies efficient faces while employing a top-down search strategy which avoids enumeration of efficient extreme points and locates the maximally efficient faces of the feasible region. We suggest that discrete representations of the efficient faces could be obtained and presented to the Decision Maker. Results of computational experiments are reported.Item Open Access Algorithms for efficient vectorization of repeated sparse power system network computations(IEEE, 1995) Aykanat, Cevdet; Özgü, Ö.; Güven, N.Standard sparsity-based algorithms used in power system appllcations need to be restructured for efficient vectorization due to the extremely short vectors processed. Further, intrinsic architectural features of vector computers such as chaining and sectioning should also be exploited for utmost performance. This paper presents novel data storage schemes and vectorization alsorim that resolve the recurrence problem, exploit chaining and minimize the number of indirect element selections in the repeated solution of sparse linear system of equations widely encountered in various power system problems. The proposed schemes are also applied and experimented for the vectorization of power mismatch calculations arising in the solution phase of FDLF which involves typical repeated sparse power network computations. The relative performances of the proposed and existing vectorization schemes are evaluated, both theoretically and experimentally on IBM 3090ArF.Item Open Access Analysis of maintenance policies for M machines with deteriorating performance(Taylor & Francis, 2000) Berk, E.; Moinzadeh, K.In this paper, we consider the maintenance scheduling of a group of M identical machines, the performance of which deteriorates with usage. Examples of such situations are frequently found in the heavy machine tooling, petro-chemical and semi-conductor industries among others. Assuming a limited maintenance resource and that the maintenance times are i.i.d., we propose a dynamic maintenance policy which utilities the information about the number of operating machines and their ages. We analyze the system for the special cases of constant and exponentially distributed maintenance times. We investigate the impact of maintenance time variability on system performance and evaluate the performance of various maintenance policies within the proposed policy class when the expected profit rate is maximized.Item Open Access An analysis of maximum clique formulations and saturated linear dynamical network(Springer-Verlag, 1999) Şengör, N. S.; Çakır, Y.; Güzeliş, C.; Pekergin, F.; Morgül, Ö.Several formulations and methods used in solving an NP-hard discrete optimization problem, maximum clique, are considered in a dynamical system perspective proposing continuous methods to the problem. A compact form for a saturated linear dynamical network, recently developed for obtaining approximations to maximum clique, is given so its relation to the classical gradient projection method of constrained optimization becomes more visible. Using this form, gradient-like dynamical systems as continuous methods for finding the maximum clique are discussed. To show the one to one correspondence between the stable equilibria of the saturated linear dynamical network and the minima of objective function related to the optimization problem, La Salle's invariance principle has been extended to the systems with a discontinuous right-hand side. In order to show the efficiency of the continuous methods simulation results are given comparing saturated the linear dynamical network, the continuous Hopfield network, the cellular neural networks and relaxation labelling networks. It is concluded that the quadratic programming formulation of the maximum clique problem provides a framework suitable to be incorporated with the continuous relaxation of binary optimization variables and hence allowing the use of gradient-like continuous systems which have been observed to be quite efficient for minimizing quadratic costs.Item Unknown Analytical loading models in flexible manufacturing systems(Elsevier, 1993) Kırkavak, N.; Dinçer, C.It would be difficult to efficiently implement a manufacturing system without solving its design and operational problems. Based on this framework, a system configuration and tooling problem is modeled. The model turns out to be a large mixed integer linear program, so that some alternative optimal seeking and heuristic techniques are used to solve the model for constructing a flow line structured Flexible Manufacturing System. As a result, it may be possible to construct flexible, efficient, simple and easily controllable manufacturing systems. © 1993.Item Unknown Application of energy measures in detection of local deviations in mechanical properties of structural elements(Springer, 2019) Lekszycki, T.; Di Cosmo, F.; Laudato, M.; Vardar, OnurThe identification of local damages in structural elements is considered. In the proposed formulation, the damages are represented as local changes in structural stiffness and they are defined by a set of parameters. The main effort of this work is directed to the use of global measures of energy to indicate the local changes of stiffness. An idea of use of additional design parameters in order to optimize the experiment and to enlarge the sensitivity of objective functional with respect to damage parameters is applied. In order to accomplish this goal, the distribution of energy within the structural domain is optimized. Two cases, namely the eigenproblem and the structure under the static external load, are discussed. Simple illustrative example of identification of damage is discussed, and numerical results are presented.Item Unknown Application placement with shared monitoring points in multi-purpose IoT wireless sensor networks(Elsevier, 2022-11-09) Çavdar, Mustafa Can; Korpeoglu, Ibrahim; Ulusoy, ÖzgürThe main function of a wireless sensor network (WSN) is to gather data from a certain region and transfer the data to a center or remote locations for further processing. The collected data can be of interest for many applications. Therefore, a physical WSN owned by a single provider can be utilized by many customer applications. Additionally, the data of a particular point or sub-region can satisfy the need of multiple applications. Hence, sensing the data only once in such cases is beneficial to reduce the energy consumption, network traffic and acceptance ratio of the applications. We call this as monitoring point based shared data approach. In this paper, we focus on the placement of applications each of which requires several points to be monitored in an area a WSN covers. We first propose such a monitoring point based shared data approach for WSNs that will serve multiple dynamic applications. We also propose two methods for application placement over a shared physical WSN: one greedy method and one genetic algorithm based method called GABAP. We did extensive simulation experiments to evaluate our algorithms. The results show the effectiveness of our methods in fast and close-to-optimum placement of applications over a single network.Item Unknown An archiving model for a hierarchical information storage environment(Elsevier, 2000) Moinzadeh, K.; Berk, E.We consider an archiving model for a database consisting of secondary and tertiary storage devices in which the query rate for a record declines as it ages. We propose a `dynamic' archiving policy based on the number of records and the age of the records in the secondary device. We analyze the cases when the number of new records inserted in the system over time are either constant or follow a Poisson process. For both scenarios, we characterize the properties of the policy parameters and provide optimization results when the objective is to minimize the average record retrieval times. Furthermore, we propose a simple heuristic method for obtaining near-optimal policies in large databases when the record query rate declines exponentially with time. The e ectiveness of the heuristic is tested via a numerical experiment. Finally, we examine the behavior of performance measures such as the average record retrieval time and the hit rate as system parameters are varied.Item Unknown An array of surface-enhanced Raman scattering substrates based on plasmonic lenses(Wiley, 2012-10-01) Kahraman, M.; Cakmakyapan, S.; Özbay, Ekmel; Culha, M.An array of ring-shaped holes is prepared from silver thin films using electron beam lithography. The optimal conditions for high performance as a surface-enhanced Raman scattering (SERS) substrate are investigated. Either the diameter of the hole (0.5, 1.0, 2.0, 3.0 and 4.0 μm) or the slit width (200, 300, 400, 500 and 600 nm) is varied. 4-Aminothiophenol (ATP) adsorbed on the structures as a self-assembled monolayer (SAM) is used as probe to evaluate the SERS performance of the generated structures. It is found that there is an optimal configuration for ring-shaped holes with a 3.0-μm diameter and 200-nm slit width. The SERS activity on this optimal lens configuration is found to be 13 times greater than that of the activity on the silver thin film. An array of these structures at this optimal configuration can easily be constructed and used in a range of SERS-based sensing applications. An array of ring-shaped holes is prepared from silver thin films using electron beam lithography. The optimal conditions for high performance as a surface-enhanced Raman scattering (SERS) substrate are investigated. It is found that there is an optimal configuration for ring-shaped holes with a 3.0-μm diameter and 200-nm slit with. The SERS activity on this optimal lens configuration is found to be 13 times greater than that of the activity on the silver thin film.Item Unknown Artificial neural network modeling and simulation of in-vitro nanoparticle-cell interactions(American Scientific Publishers, 2014) Cenk, N.; Budak, G.; Dayanik, S.; Sabuncuoglu, I.In this research a prediction model for the cellular uptake efficiency of nanoparticles (NPs), which is the rate that NPs adhere to a cell surface or enter a cell, is investigated via an artificial neural network (ANN) method. An appropriate mathematical model for the prediction of the cellular uptake rate of NPs will significantly reduce the number of time-consuming experiments to determine which of the thousands of possible variables have an impact on NP uptake rate. Moreover, this study constitutes a basis for targeted drug delivery and cell-level detection, treatment and diagnosis of existing pathologies through simulating NP-cell interactions. Accordingly, this study will accelerate nanomedicine research. Our research focuses on building a proper ANN model based on a multilayered feed-forward back-propagation algorithm that depends on NP type, size, surface charge, concentration and time for prediction of cellular uptake efficiency. The NP types for in-vitro NP-healthy cell interaction analysis are polymethyl methacrylate (PMMA), silica and polylactic acid (PLA), all of whose shapes are spheres. The proposed ANN model has been developed on MATLAB Programming Language by optimizing a number of hidden layers (HLs), node numbers and training functions. The datasets are obtained from in-vitro NP-cell interaction experiments conducted by Nanomedicine and Advanced Technology Research Center. The dispersion characteristics and cell interactions with different NPs in organisms are explored using an optimal ANN prediction model. Simulating the possible interactions of targeted NPs with cells via an ANN model will be faster and cheaper compared to the excessive experimentation currently necessary.Item Unknown Atomic and electronic structure of carbon strings(IOP Publishing Ltd., 2005) Tongay, S.; Dag, S.; Durgun, Engin; Senger, R. T.; Çıracı, SalimThis paper presents an extensive study of various string and tubular structures formed by carbon atomic chains. Our study is based on first-principles pseudopotential plane wave and finite-temperature ab initio molecular dynamics calculations. Infinite- and finite-length carbon chains exhibit unusual mechanical and electronic properties such as large cohesive energy, axial strength, high conductance, and overall structural stability even at high temperatures. They are suitable for structural and chemical functionalizations. Owing to their flexibility and reactivity they can form linear chain, ring, helix, two-dimensional rectangular and honeycomb grids, three-dimensional cubic networks, and tubular structures. Metal-semiconductor heterostructures and various quantum structures, such as multiple quantum wells and double-barrier resonant tunnelling structures, can be formed from the junctions of metallic carbon and semiconducting BN linear chains. Analysis of atomic and electronic structures of these periodic, finite, and doped structures reveals fundamentally and technologically interesting features, such as structural instabilities and chiral currents. The double covalent bonding of carbon atoms depicted through self-consistent charge density analysis underlies the chemical, mechanical, and electronic properties.Item Open Access Auto-tuning similarity search algorithms on multi-core architectures(2013) Gedik, B.In recent times, large high-dimensional datasets have become ubiquitous. Video and image repositories, financial, and sensor data are just a few examples of such datasets in practice. Many applications that use such datasets require the retrieval of data items similar to a given query item, or the nearest neighbors (NN or k -NN) of a given item. Another common query is the retrieval of multiple sets of nearest neighbors, i.e., multi k -NN, for different query items on the same data. With commodity multi-core CPUs becoming more and more widespread at lower costs, developing parallel algorithms for these search problems has become increasingly important. While the core nearest neighbor search problem is relatively easy to parallelize, it is challenging to tune it for optimality. This is due to the fact that the various performance-specific algorithmic parameters, or "tuning knobs", are inter-related and also depend on the data and query workloads. In this paper, we present (1) a detailed study of the various tuning knobs and their contributions on increasing the query throughput for parallelized versions of the two most common classes of high-dimensional multi-NN search algorithms: linear scan and tree traversal, and (2) an offline auto-tuner for setting these knobs by iteratively measuring actual query execution times for a given workload and dataset. We show experimentally that our auto-tuner reaches near-optimal performance and significantly outperforms un-tuned versions of parallel multi-NN algorithms for real video repository data on a variety of multi-core platforms. © 2013 Springer Science+Business Media New York.Item Open Access Automatic selection of compiler optimizations by machine learning(IEEE - Institute of Electrical and Electronics Engineers, 2023-08-28) Peker, Melih; Öztürk, Özcan; Yıldırım, S.; Uluyağmur Öztürk, M.Many widely used telecommunications applications have extremely long run times. Therefore, faster and more efficient execution of these codes on the same hardware is important in critical telecommunication applications such as base stations. Compilers greatly affect the properties of the executable program to be created. It is possible to change properties such as compilation speed, execution time, power consumption and code size using compiler flags. This study aims to find the set of flags that will provide the shortest run time among hundreds of compiler flag combinations in GCC using code flow analysis, loop analysis and machine learning methods without running the program.Item Open Access Autonomous multiple teams establishment for mobile sensor networks by SVMs within a potential field(2012) Nazlibilek, S.In this work, a new method and algorithm for autonomous teams establishment with mobile sensor network units by SVMs based on task allocations within a potential field is proposed. The sensor network deployed into the environment using the algorithm is composed of robot units with sensing capability of magnetic anomaly of the earth. A new algorithm is developed for task assignment. It is based on the optimization of weights between robots and tasks. The weights are composed of skill ratings of the robots and priorities of the tasks. Multiple teams of mobile units are established in a local area based on these mission vectors. A mission vector is the genetic and gained background information of the mobile units. The genetic background is the inherent structure of their knowledge base in a vector form but it can be dynamically updated with the information gained later on by experience. The mission is performed in a magnetic anomaly environment. The initial values of the mission vectors are loaded by the task assignment algorithm. The mission vectors are updated at the beginning of each sampling period of the motion. Then the teams of robots are created by the support vector machines. A linear optimal hyperplane is calculated by the use of SVM algorithm during training period. Then the robots are classified as teams by use of SVM mechanism embedded in the robots. The support vector machines are implemented in the robots by ordinary op-amps and basic logical gates. Team establishment is tested by simulations and a practical test-bed. Both simulations and the actual operation of the system prove that the system functions satisfactorily. © 2012 Elsevier Ltd. All rights reserved.Item Open Access Available bit rate traffic engineering in MPLS networks with flow-based multipath routing(Institute of Electronics Information and Communication Engineers, 2004) Akar, N.; Hökelek, İ.; Karasan, E.In this paper, we propose a novel traffic engineering architecture for IP networks with MPLS backbones. In this architecture, two link-disjoint label switched paths, namely the primary and secondary paths, are established among every pair of IP routers located at the edges of an MPLS backbone network. As the main building block of this architecture, we propose that primary paths are given higher priority against the secondary paths in the MPLS data plane to cope with the so-called knock-on effect. Inspired by the ABR flow control mechanism in ATM networks, we propose to split traffic between a source-destination pair between the primary and secondary paths using explicit rate feedback from the network. Taking into consideration the performance deteriorating impact of packet reordering in packet-based load balancing schemes, we propose a traffic splitting mechanism that operates on a per-flow basis (i.e., flow-based multipath routing). We show via an extensive simulation study that using flow-based multipath traffic engineering with explicit rate feedback not only provides consistently better throughput than that of a single path but is also void of out-of-order packet delivery.