Bilgisayar Mühendisliği Bölümü Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.11779/1940
Browse
17 results
Search Results
Research Project İmece-depo: İşbirlikçi Hücresel Ağlarda Veri Önbellekleme için Cihazdan Cihaza Iletişim ile Dağıtık Depolama, Optimale Yakın Kodlama ve Protokol Tasarımı.(2023) Haytaoğlu, Elif; Pourmandı, Massoud; Kaya, Erdi; Arslan, Şefik Şuayb; 02.02. Department of Computer Engineering; 02. Faculty of Engineering; 01. MEF UniversityHücresel ağlarda popüler dosyaların cihazlarda önbelleklenmesi ile, cihazlar arası etkileşim baz istasyonu (Bİ) üzerine düşen iletişim yükünü oldukça azaltmaktadır. Dağıtık veri önbellekleme işlemi popüler bir dosyanın parçalarının kodlanmamış orijinal haliyle ya da herhangi bir silinti kodu kullanılarak kodlanmış halinin mobil cihazlar içerisinde dağıtık bir şekilde depolanması yardımıyla gerçekleştirilir. Dosyanın herhangi bir parçası, komşu mobil cihazlardan ya da mümkün değilse, doğrudan Bİ?lerden, yüksek bir iletişim maliyeti pahasına indirilebilir. Bir hücresel ağda, rastgele zamanlarda bazı düğümlerin hücreye katıldığı ve bazılarının ayrıldığı göz önüne alındığında, performans için Bİ ile iletişimin minimum düzeyde olmasını sağlayacak akıllı veri onarım yöntemlerine ihtiyaç duyulacaktır. Tek bir veya birden fazla Bİ?nin sisteme katılımı, önceki onarım paradigmalarına, özellikle de işbirlikçi düğüm onarım süreçlerine farklı bir boyut eklemektedir. Bunun nedeni, çalışma protokolü kurallarının yanı sıra iletişim kısıtlamalarının da değişmesidir. Literatür, bu durum için temel bant genişliği/depolama ödünleşim uzayını inceleyen bir çalışma içermemektedir. Yeni hücre mimarileri buna göre, yeni silinti kod yapılarını, verimli protokol tasarımlarını, veri erişim gecikmesi, gerçekçi kuyruk modelleri ve gerçekçi benzetim platformları dahil ancak bunlarla sınırlı olmamak üzere farklı tasarım değerlendirmelerini gerektirmektedir. Bu projede ilk olarak, daha önceki hiçbir çalışmada düşünülmemiş işbirliği yapan Bİ?lerin cihaz ayrılışlarında yaşanan kayıp verinin onarımı için bant genişliği ve depolama kapasitesinin iyileştirilmiş teorik sınırlarının veri akış diyagramları kullanılarak elde edilmesi amaçlanmıştır. Bununla beraber, bant genişliği ve depolama alanını en iyi kullanan kod yapılarından esinlenerek, veri önbellekleme işlemini optimale yakın bir maliyetle gerçekleştirecek tamamen özgün çizge tabanlı kod yapıları ve bu yeni kodlar için daha önce düğüm onarım problemine uygulanmamış genetik algoritma, optimize edilmiş artık veri dağıtımı gibi yeni yaklaşımlar kullanılarak önceden düşünülmemiş düğüm onarım algoritmaları önerilecektir. Ayrıca, düğümlerin hücreye katılma ve ayrılma süreçleri için, bant genişliği ve veri depolaması gereksinimlerini en aza indirmeye yardımcı olacak enerji tüketimi odaklı son derece özgün protokoller önerilecektir. Bu protokoller, düğümlerin bir hücreden diğerine geçiş yapabileceği ve hücre içi kaynakların etkin kullanılmasına yardımcı olmak için Bİ?lerin işbirliği yapmasını sağlayan geçiş senaryoları ile güçlendirilecektir. Bu durum, iki onarım işlemi arasındaki sürenin ayarlanması, veri erişim maliyetlerinin azaltılması, hücreye katılan düğüm içeriğinin kullanımı, artık veri kullanımı v.s. gibi yenilikleri içerecektir. Son olarak, önerilen kod yapıları ve protokol mimarisinin performansını analitik olarak türetmek için bilinen çeşitli ve daha gerçekçi kuyruklama modelleri değerlendirilecektir. Analitik sonuçlarımızı doğrulamak için daha sonra hücresel ağ tabanlı büyük ölçekli benzetimler yapılıp sayısal yöntemler ile toplam iletişim ve dosya onarım işlemlerinin maliyet hesaplamaları ve karşılaştırmaları yapılacaktır. MEF Üniversitesi öğretim üyesi Dr. Şuayb Arslan?ın yürütücüsü olduğu ve 36 ay sürecek projede, Pamukkale üniversitesi Bilgisayar Mühendisliği Bölümü öğretim üyesi Dr. Elif Haytaoğlu araştırmacı olarak görev alacaktır. Projede, iki doktora, iki yüksek lisans ve son iki senemizde iki lisans öğrencisi bursiyer olarak görev alacaktır.Patent Artificial Intelligence Augmented Iterative Product Decoding(2023) Arslan , Şuayb Şefik; Göker, TurguyA method for product decoding within a data storage system includes receiving data to be decoded within a first decoder; performing a plurality of decoding iterations to decode the data utilizing a first decoder and a second decoder; and outputting fully decoded data based on the performance of the plurality of decoding iterations. Each of the plurality of decoding iterations includes (i) decoding the data with the first decoder operating at a first decoder operational mode to generate once decoded data; (ii) sending the once decoded data from the first decoder to the second decoder; (iii) receiving error information from the first decoder with an artificial intelligence system; (iv) selecting a second decoder operational mode based at least in part on the error information that is received by the artificial intelligence system; and (v) decoding the once decoded data with the second decoder operating at the second decoder operational mode to generate twice decoded data; and outputting fully decoded data based on the performance of the plurality of decoding iterations.Patent Joint Multi-Nanopore Sequencing for Reliable Data Retrieval in Nucleic Acid Storage(2023) Arslan , Şuayb Şefik; Göker, Turguy; Doerner, DonA nucleic acid storage system (100) that uses nanopore sequencing to read data values chemically embedded in oligonucleotides includes a membrane (102), a voltage source (108), and a nucleic acid strand (110). The membrane (102) has a plurality of nanopores (104) that are stacked upon one another in a multi-nanopore arrangement. The voltage source (108) is configured to direct voltage across the plurality of nanopores (104). The nucleic acid strand (110) including the oligonucleotides is threaded through each of the plurality of nanopores (104) within the membrane (102). A separate base signal (118) is generated from the nucleic acid strand (110) being threaded through each of the plurality of nanopores (104), and Recursive Neural Networks can be used to estimate a signal shape for each oligonucleotide. Recurrent Convolutional Neural Networks and noise predictive data detection algorithms can be used based on the estimated signal shapes to sequence the oligonucleotides.Article What Is the Effective Resolution of the Retinal Image of a Distant Face?(Vision Sciences Society Annual Meeting Abstract, 2023) Arslan , Şuayb Şefik; Arslan, Şefik Şuayb; Sinha, Pawan; 02.02. Department of Computer Engineering; 02. Faculty of Engineering; 01. MEF UniversityWe consider the following question: What is the effective resolution of a face image projected on the retina, when the face is at a specified distance from the eye? Though simple to state, this is a surprisingly challenging issue to resolve. The mapping between viewing distance and effective resolution cannot be readily derived based on the contrast sensitivity, Snellen acuity, or even the packing density of photoreceptors in the fovea. With initial guidelines derived from theoretical considerations, images of varying resolution were presented across a range of viewing distances. For each distance, participants were required to perform an ‘odd one out’ task. This involved detecting the one that was different from the rest in a 2x2 grid, with image resolution being the only dimension of variation. As the experiment progressed, the viewing distance decreased monotonically, and participants were able to detect increasingly subtle resolution differences between the three standard images and the outlier. The collected data have allowed us to establish the upper/lower bounds on the effective available resolution for typical human vision as a function of viewing distance. Interestingly, we find that humans perform significantly better, particularly at short ranges, than what a theoretical model predicts based on projected image size, cone density, and foveal extent. Accordingly, we suggest that the non-uniform in-fovea density, as well as less sharp fall-off in the acuity density function outside the fovea, need to be integrated into future theoretical models to translate viewing distance to perceived image characteristics. A pragmatic benefit of the <distance : effective-resolution> mapping is that it enables a direct comparison of human face recognition performance as assessed across blur and viewing distance. Additionally, it allows us to systematically compare human performance on face recognition at varying distances with that of machine vision systems using the common axis of resolution.Article Comparing Humans and Deep Neural Networks on Face Recognition Under Various Distance and Rotation Viewing Conditions(Journal of Vision, 2023) Fux, Michal; Arslan, Şefik Şuayb; Jang, Hojin; Boix, Xavier; Cooper, Avi; Groth, Matt J; Sinha, Pawan; 02.02. Department of Computer Engineering; 02. Faculty of Engineering; 01. MEF UniversityHumans possess impressive skills for recognizing faces even when the viewing conditions are challenging, such as long ranges, non-frontal regard, variable lighting, and atmospheric turbulence. We sought to characterize the effects of such viewing conditions on the face recognition performance of humans, and compared the results to those of DNNs. In an online verification task study, we used a 100 identity face database, with images captured at five different distances (2m, 5m, 300m, 650m and 1000m) three pitch values (00 - straight ahead, +/- 30 degrees) and three levels of yaw (00, 45, and 90 degrees). Participants were presented with 175 trials (5 distances x 7 yaw and pitch combinations, with 5 repetitions). Each trial included a query image, from a certain combination of range x yaw x pitch, and five options, all frontal short range (2m) faces. One was of the same identity as the query, and the rest were the most similar identities, chosen according to a DNN-derived similarity matrix. Participants ranked the top three most similar target images to the query image. The collected data reveal the functional relationship between human performance and multiple viewing parameters. Nine state-of-the-art pre-trained DNNs were tested for their face recognition performance on precisely the same stimulus set. Strikingly, DNN performance was significantly diminished by variations in ranges and rotated viewpoints. Even the best-performing network reported below 65% accuracy at the closest distance with a profile view of faces, with results dropping to near chance for longer ranges. The confusion matrices of DNNs were generally consistent across the networks, indicating systematic errors induced by viewing parameters. Taken together, these data not only help characterize human performance as a function of key ecologically important viewing parameters, but also enable a direct comparison of humans and DNNs in this parameter regimeArticle Cooperative Network Coding for Distributed Storage Using Base Stations With Link Constraints(arXiv, 2021) Arslan, Şuayb Şefik; Arslan, Şefik Şuayb; Haytaoğlu, Elif; 02.02. Department of Computer Engineering; 02. Faculty of Engineering; 01. MEF UniversityIn this work, we consider a novel distributed data storage/caching scenario in a cellular setting where multiple nodes may fail/depart at the same time. In order to maintain the target reliability, we allow cooperative regeneration of lost nodes with the help of base stations allocated in a set of hierarchical layers. Due to this layered structure, a symbol download from each base station has a different cost, while the link capacities connecting the nodes of the cellular system and the base stations are also limited. In this more practical and general scenario, we present the fundamental trade-off between repair bandwidth cost and the storage space per node. Particularly interesting operating points are the minimum storage as well as bandwidth cost points in this trade-off curve. We provide closed-form expressions for the corresponding bandwidth (cost) and storage space per node for these operating points. Finally, we provide an explicit optimal code construction for the minimum storage regeneration point for a given set of system parameters.Conference Object Residual Data Usage in LDPC Codes(IEEE, 2022) Arslan, Şefik Şuayb; Pourmandi, Massoud; Haytaoglu, Elif; Arslan, Şefik Şuayb; 02.02. Department of Computer Engineering; 02. Faculty of Engineering; 01. MEF UniversityIn distributed storage systems/coded caching systems, padding operations should be performed when the encoded data cannot be divided by the number of storage nodes evenly. Thus, extra zero values are stored in one of the nodes to balance each node's storage content. In this study, distribution of data to storage nodes with no padding was investigated for distributed caching context in which a base station and devices both store the coded data. In other words, no redundancy (no-padding) is included into the encoded data. This approach is named as residual data distribution. LDPC codes are selected as the erasure code due to their low complexity encode/decode operations. Moreover, performance comparisons were conducted between using traditional data distribution approach (with padding) and using residual data (use of no-padding) (standard) in terms of repair time. In our work, the effect of no-padding data usage on the repair time and the ratios of storage savings have been also demonstrated.Conference Object Citation - WoS: 2Citation - Scopus: 2Base Station-Assisted Cooperative Network Coding for Cellular Systems With Link Constraints(IEEE, 2022) Arslan, Suayb S.; Arslan, Şefik Şuayb; Haytaoglu, Elif; 02.02. Department of Computer Engineering; 02. Faculty of Engineering; 01. MEF UniversityWe consider a novel distributed data storage/caching scenario in a cellular network, where multiple nodes may fail/depart simultaneously To meet reliability, we allow cooperative regeneration of lost nodes with the help of base stations allocated in a set of hierarchical layers1. Due to this layered structure, a symbol download from each base station has a different cost, while the link capacities between the nodes of the cellular system and the base stations are also constrained. Under such a setting, we formulate the fundamental trade-off with closed form expressions between repair bandwidth cost and the storage space per node. Particularly, the minimum storage as well as bandwidth cost points are formulated. Finally, we provide an explicit optimal code construction for the minimum storage regeneration point for a special set of system parameters.Article Citation - WoS: 12Citation - Scopus: 20Compress-Store on Blockchain: a Decentralized Data Processing and Immutable Storage for Multimedia Streaming(Springer, 2022) Arslan, Şuayb Şefik; Arslan, Şefik Şuayb; Goker, Turguy; 02.02. Department of Computer Engineering; 02. Faculty of Engineering; 01. MEF UniversityDecentralization for data storage is a challenging problem for blockchain-based solutions as the blocksize plays a key role for scalability. In addition, specific requirements of multimedia data call for various changes in the blockchain technology internals. Considering one of the most popular applications of secure multimedia streaming, i.e., video surveillance, it is not clear how to judiciously encode incentivization, immutability, and compression into a viable ecosystem. In this study, we provide a genuine scheme that achieves this encoding for a video surveillance application. The proposed scheme provides a novel integration of data compression, immutable off-chain data storage using a new consensus protocol namely, Proof-of-WorkStore (PoWS) in order to enable fully useful work to be performed by the miner nodes of the network. The proposed idea is the first step towards achieving greener application of a blockchain-based environment to the video storage business that utilizes system resources efficiently.Article Citation - WoS: 4Citation - Scopus: 7On the Distribution Modeling of Heavy-Tailed Disk Failure Lifetime in Big Data Centers(IEEE, 2021) Arslan, Şuayb Şefik; Arslan, Şefik Şuayb; Zeydan, Engin; 02.02. Department of Computer Engineering; 02. Faculty of Engineering; 01. MEF UniversityIt has become commonplace to observe frequent multiple disk failures in big data centers in which thousands of drives operate simultaneously. Disks are typically protected by replication or erasure coding to guarantee a predetermined reliability. However, in order to optimize data protection, real life disk failure trends need to be modeled appropriately. The classical approach to modeling is to estimate the probability density function of failures using nonparametric estimation techniques such as kernel density estimation (KDE). However, these techniques are suboptimal in the absence of the true underlying density function. Moreover, insufficient data may lead to overfitting. In this article, we propose to use a set of transformations to the collected failure data for almost perfect regression in the transform domain. Then, by inverse transformation, we analytically estimated the failure density through the efficient computation of moment generating functions, and hence, the density functions. Moreover, we developed a visualization platform to extract useful statistical information such as model-based mean time to failure. Our results indicate that for other heavy-tailed data, the complex Gaussian hypergeometric distribution and classical KDE approach can perform best if the overfitting problem can be avoided and the complexity burden is overtaken. On the other hand, we show that the failure distribution exhibits less complex Argus-like distribution after performing the Box–Cox transformation up to appropriate scaling and shifting operations.