Browsing by Author "Zaker, Maryam"

Now showing 1 - 3 of 3

Citation - WoS: 15
Citation - Scopus: 18
A Nano-Design of a Quantum-Based Arithmetic and Logic Unit for Enhancing the Efficiency of the Future Iot Applications
(Aip Publishing, 2025) Ahmadpour, Seyed Sajad; Zaker, Maryam; Navimipour, Nima Jafari; Misra, Neeraj Kumar; Zohaib, Muhammad; Kassa, Sankit; Hakimi, Musawer
The Internet of Things (IoT) is an infrastructure of interconnected devices that gather, monitor, analyze, and distribute data. IoT is an inevitable technology for smart city infrastructure to ensure seamless communication across multiple nodes. IoT, with its ubiquitous application in every sector, ranging from health-care to transportation, energy, education, and agriculture, comes with serious challenges as well. Among the most significant ones is security since the majority of IoT devices do not encrypt normal data transmissions, making it easier for the network to breach and leak data. Traditional technologies such as CMOS and VLSI have the added disadvantage of consuming high energy, further creating avenues for security threats for IoT systems. To counter such problems, we require a new solution to replace traditional technologies with a secure IoT. In contrast to traditional solutions, quantum-based approaches offer promising solutions by significantly reducing the energy footprint of IoT systems. Quantum-dot Cellular Automata (QCA) is one such approach and is an advanced nano-technology that exploits quantum principles to achieve complex computations with the advantages of high speed, less occupied area, and low power consumption. By reducing the energy requirements to a minimum, QCA technology makes IoT devices secure. This paper presents a QCA-based Arithmetic Logic Unit (ALU) as a solution to IoT security problems. The proposed ALU includes more than 12 logical and arithmetic operations and is designed using majority gates, XOR gates, multiplexers, and full adders. The proposed architecture, simulated in QCADesigner 2.0.3, achieves an improvement of 60.45% and 66.66% in cell count and total occupied area, respectively, compared to the best of the existing designs, proving to be effective and efficient.
A New Design of Arithmetic and Logic Unit for Enhancing the Security of Future Internet of Things Devices Using Quantum-Dot Technology
(Pergamon-Elsevier Science Ltd, 2025) Zaker, Maryam; Ahmadpour, Seyed Sajad; Navimipour, Nima Jafari; Zohaib, Muhammad; Misra, Neeraj Kumar; Kassa, Sankit; Alsaleh, Omar I.
The Internet of Things (IoT) is a network of interconnected devices that collect, monitor, analyze, and exchange data. This technology plays a crucial role in the smart city infrastructure by seamlessly interconnecting various nodes. The extensive application and recognition of IoT across multiple city domains, such as healthcare, transportation, energy, education, and agriculture, bring significant challenges, with security among the most pressing. Traditional hardware technologies like Complementary Metal Oxide Semiconductor (CMOS) and Very Large Scale Integration (VLSI) suffer from limitations such as high power consumption and insufficient scalability, which hinder secure and sustainable IoT deployment. Such limitations have prompted the need to seek other technologies that would serve the dual purpose of providing security as well as energy. Quantum-based technologies can become adequate candidates offering promising solutions to make IoT devices and sustainable systems more secured. Quantum-dot Cellular Automata (QCA) has been proposed as a nanotechnology with the potential of consuming ultra-low powers, less area, and high-speed operation. QCA enhances security through sustainable computing objectives by minimizing energy usage. To improve the future security and efficiency of IoT hardware, this paper suggests a QCA-based Arithmetic Logic Unit (ALU). This ALU can generate more than 12 logical and arithmetic operations. Designed together with the majority gates, XOR gates, multiplexers, and full adders, the ALU is simulated using the QCA-Designer 2.0.3. Simulated results indicate improvements in the number of cells and reduced occupied area relative to the earlier designs. These results indicate the potential of QCA technology in enabling secure, energy-efficient, and compact computing architecture applicable in the future IoT.
Sosyal Medya Kullanıcılarında Depresyon ve Anksiyete Bozukluklarının Şiddetini Derin Öğrenme Tabanlı Bir Modelle Öngörmeye Yönelik Çok Boyutlu Bir Teknik
(2025) Zaker, Maryam; Navimipour, Nima Jafari
Modern dünyada, depresyon ve kaygı gibi ruhsal bozukluklar önemli bir halk sağlığı sorunu haline gelmiştir. Belirtilerin erken tanımlanması ve müdahalesi hayati öneme sahip olsa da, geleneksel klinik değerlendirmeler maliyetli olup etiketleme eksikliğinden muzdariptir. Instagram gibi sosyal medya platformları; görseller, altyazılar, biyografi bilgileri, profil fotoğrafları, etkileşim istatistikleri ve paylaşım örüntüleri gibi, ruh sağlığı belirtilerini gösteren zengin çok modlu veriler sunmaktadır. Ancak mevcut tanı yöntemleri genellikle büyük, tek modlu veri kümelerine dayanmakta ve yalnızca küçük, klinik olarak doğrulanmış kohortlar mevcut olduğunda yeterli genelleştirilebilirlik sağlayamamaktadır. Bu çalışmada, Instagram kullanıcıları arasında depresyon ve kaygı şiddetini tahmin etmek için çok modlu few‑shot sınıflandırma çerçevesi (FS‑MMN) önerilmektedir. PHQ‑9 ve GAD‑7 anketlerini doldurup aydınlatılmış onam veren ve Instagram hesaplarına erişim izni sağlayan 137 yetişkin gönüllü işe alındı. Metin, görsel ve davranışsal modlarda 'küçük ancak kapsamlı' bir veri seti elde etmek üzere özenli ön işleme ve özellik mühendisliği uygulandıktan sonra, epizodik makro F1 hedefiyle optimize edilmiş, prototip‑tabanlı çok kollu füzyon modeli tasarlandı. Deneysel sonuçlar, veri kullanım verimliliğinin ve modelin genelleştirilebilirliğinin yüksek olduğunu göstermektedir. PHQ‑9 ölçeğine dayalı depresyon şiddeti tahmininde FS‑MMN modeli; 1, 2, 4, 8 ve 16‑shot senaryolarında sırasıyla 0.565, 0.627, 0.719, 0.825 ve 0.851 makro F1 değerleri elde ederek temel yöntemlerin üzerinde performans göstermiştir. GAD‑7 ölçeğine dayalı kaygı şiddeti tahmininde de aynı shot senaryolarında sırasıyla 0.536, 0.593, 0.683, 0.760 ve 0.764 makro F1 değerleri yakalayarak ağaç tabanlı yöntemleri. Bu bulgular, FS‑MMN'in dikkat‑tabanlı füzyon ve prototip‑tabanlı kenar eğitimi ile çok kollu tasarımının, sosyal ağ verileri üzerinde çok örnekli öğrenme ile depresyon ve kaygı şiddeti tahmininde yeni bir standart sunduğunu doğrulamaktadır.