- Mobile Ad Hoc Networks
- Vehicular Ad Hoc Networks (VANETs)
- Opportunistic and Delay-Tolerant Networks
- Energy Efficient Wireless Sensor Networks
- IoT and Edge/Fog Computing
- Security in Wireless Sensor Networks
- Digital Transformation in Industry
- UAV Applications and Optimization
- Network Security and Intrusion Detection
- Blockchain Technology Applications and Security
- Fault Detection and Control Systems
- Machine Fault Diagnosis Techniques
- IoT-based Smart Home Systems
- Advanced MIMO Systems Optimization
- Probabilistic and Robust Engineering Design
- Advanced Neural Network Applications
- Video Surveillance and Tracking Methods
- Autonomous Vehicle Technology and Safety
- Distributed Control Multi-Agent Systems
- Indoor and Outdoor Localization Technologies
- Human Mobility and Location-Based Analysis
- Reliability and Maintenance Optimization
- Software Engineering and Design Patterns
- Advanced Data and IoT Technologies
- Media Discourse and Social Analysis
Marmara University
2014-2024
United States Naval Academy
2016
Istanbul Technical University
2006-2012
Naval Academy
2003-2010
Sensor nodes in a Wireless Network (WSN) can be dispersed over remote sensing area (e.g., the regions that are hardly accessed by human beings). In such kinds of networks, datacollectionbecomesoneofthemajorissues. Getting connected to each sensor node and retrieving information time introduces new challenges. Mobile sink usage-especially Unmanned Aerial Vehicles (UAVs)-is most convenient approach covering accessing large-scale WSN. However, operation UAV depends on some parameters, as...
IoT-based networks are vulnerable to jamming attacks due their large-scale deployment and shared communication environment. Resource constraints the low computational power of IoT devices make it harder implement high-performance ML-based architectures for detection. In this work, effects on a Wi-Fi network presented novel real-time detection mechanism is devised which can identify multiple channels in 2.4 GHz bandwidth simultaneously. The experiments conducted lab environment by generating...
Wireless sensor networks (WSNs) consist of battery–limited nodes which have the ability sensing environment, communicating with other and processing data. Large number node deployment over a geographical area imposes some constraints on retrieval The use mobile sinks (e.g., unmanned aerial vehicle, UAV) is an effective solution method for such large–scale networks. However, depending path altitude UAV, type radios in use, coverage problem arises where cannot get connected to UAV. In this...
The Internet of Things (IoT) is increasingly used in every application field. Wi-Fi-based communication the most preferred choice wireless IoT systems due to its accessibility, high bandwidth, affordability, and versatility. However, security vulnerabilities devices pose considerable risks, especially industrial critical applications. This paper presents a novel technique detect jamming IoT, focusing on environmental dynamics settings. To better examine conditions production environment, it...
In wireless sensor networks, energy-aware and network-lifetime prolonging techniques are proposed, as the energy becomes most valuable resource. Most of these use complex calculations failing to be easy implement. Another approach is mobile-sink node enhance performance metrics. However, not scalable large scale networks. Scalability a routing protocol for networks mainly affected by topological changes number nodes in network. multiple-sinks (gateways) should used provide scalability. this...
As the energy becomes most valuable resource in wireless sensor networks, to prolong lifetime of network, energy-aware protocols are proposed which use minimization techniques and multi-hop paths. However, paths complex calculations at nodes introduce delay processing overhead. Providing energy-efficiency a scalability problem. Scalability routing protocol for networks is mainly affected by topological changes number network. In large scale with nodes, multiple-sinks (gateways) should be...
The nature of military operations introduces additional requirements on sensor and ad-hoc networks such as reliability operating in real-time. Although there has been many techniques providing real-time data communication the literature, their implementations challenge with resource limitations peculiar to Wireless Sensor Networks (WSN). Real-time demand objective minimization energy consumption. Moreover, most these require topological information deployed network which communications...
The proliferation of Internet Things (IoT) app-lications across various domains has raised concerns about the security wireless communication, particularly against jamming attacks. This paper presents an in-depth analysis four attack types-constant, random, pulsed, and barrage-on IoT network performance in noisy noiseless environments. study focuses on a server-client application using HTTP protocol. impact attacks is evaluated by collecting analyzing crucial parameters, such as delay,...
Sensor nodes in a Wireless Network (WSN) can be dispersed over remote sensing area e.g. the regions that cannot accessed by human beings (inaccessible regions). In such kind of networks, data collection becomes one major issues. Getting connected to each sensor node and retrieving information time introduces new challenges. Mobile sink usage, especially Unmanned Aerial Vehicle (UAV), is most convenient approach cover access large scale WSN. However, operation UAV depends on some parameters...
Clustering algorithms improve network manageability through several topology partitioning techniques. In some particular cases, such as vehicular ad hoc (VANETs) communications, significant performance improvements can be introduced via clustered networking solutions whereas merging clusters for the sake of scalability may lead to degraded stability. this paper, we explore impact clusters, and furthermore based on these results, propose a new clustering technique, namely Relatively Stable...
A novel source-initiated geographical data flow technique, called stateless weighted routing (SWR), is presented in this paper. Nodes keep only their own virtual position and require no local topological information. Each node calculates the weight of its own. Initially, value relative distance to sink. decides retransmit or drop packet received by comparing sender sink, i.e, destination, which are contained packet. The comparison actually provides mean for routing. Although parameter...
Classical ad-hoc network algorithms become impractical for Vehicular Ad Hoc Networks when density of vehicles is low in rural areas and during the periods vehicle traffic cities (such as at night). End-to-end connectivity requirement these causes deflection. Connections get broken a suitable element isn't available data transfer. This problem which occurs dense partially connected hoc can be solved by using delay tolerant approaches routing algorithms. By store-carry-forward method, carried...
Real-time objection detection is becoming more important and critical in all application areas, including Smart Transport City. From safety/security to resource efficiency, real-time image processing approaches are used than ever. On the other hand, low-latency requirements available resources present challenges. Edge computing integrated with cloud minimizes communication delays but requires efficient use of due its limited resources. For example, although deep learning-based object methods...
DECICE is a Horizon Europe project that developing an AI-enabled open and portable management framework for automatic adaptive optimization deployment of applications in computing continuum encompassing from IoT sensors on the Edge to large-scale Cloud / HPC infrastructures. In this paper, we describe architecture. Furthermore, highlight use-cases evaluation: intelligent traffic intersection, magnetic resonance imaging, emergency response.
The technology of object detection is widely utilized in Intelligent Transportation Systems (ITSs) to ensure safe and dependable road experiences. However, adverse weather conditions can pose significant challenges for applications ITSs. This challenge arises from the traditional machine learning assumption that training testing sets have similar data distributions, which may not hold some real-world situations. In this paper, YOLOv5 model trained tested on DAWN dataset, a diverse dataset...
Jamming attacks presents a significant challenge to the security and reliability of wireless communication networks, especially in context IoT applications. This study introduces novel approach detecting jamming by utilizing upper-layer network parameters from application transport layers. An experimental testbed is developed consisting Wi-Fi-based server-client collect data. The are gathered both noiseless noisy environment conditions examine performance variations different deep-learning...
In this study, the road surface irregularities (pits, bumps, holes, speed breakers, etc.) is intended to be detected and identified with a system placed on vehicles. For purpose; vibration-sensitive designed consisting of sensors, processor, GPS receiver as hardware algorithms used for analyzing obtained data. With use 3-dimensional acceleration sensors in system, measurements at vehicles (vehicle speed/acceleration, severity concussion location information) are transferred database then...
There is a number of ways to estimate the people in particular area. From image processing movement pattern analysis, sheer methods exist literature. An interesting one among these estimating with use signal processing. Because it much cheaper compared other literature, more commonly implemented. In this study, area estimated by analyzing WiFi signals captured from mobile devices.
V2X communications is one key feature in the 5G standards that enables vehicles get connected and access cellular services anywhere anytime. It essential to build a realistic simulation environment for large scale demonstration of system under test including varying number vehicles, application/network services, road network infrastructure new solution approaches/methods proposed enhance or make improvements on performance. This paper describes developed this purpose also evaluates...
In Vehicular Ad Hoc Networks (VANETs), efficient use of the control channel affects both safety applications that and other service channels. Its efficiency deteriorates due collisions on accessing to or merging which happen when fast moving vehicles same allocated this paper, we propose an algorithm predicts possible overtaking vehicles. It within 2-hops neighborhood recommends vehicle acquire a new time slot in channel. Through simulations, it is shown proposed approach reduces collisions.
The packet delivery ratio and throughput in Vehicular Ad Hoc Networks (VANETs) depend on the network connectivity degrade as decreases. A delay–tolerant approach increases partially connected networks. In previous studies, lifetime vehicle's carry time are determined to maximise VANETs. However, IPv6 uses hop count Hop Limit field instead of using seconds TTL field, IPv4. Similarly, although IPv4 packets designed this is used for practice. case, there no mechanism determine how long vehicles...