Research - Wireless Networks Laboratory

Machine Learning and Extreme Value Theory Based Ultra-Reliable Communication for Wireless Control Systems (supported by TUBITAK 2247-A National Outstanding Researchers Program)

Safety critical control applications still cannot leverage the advances in Wireless Internet of Everything (IoE) for two main reasons: 1) Lack of methodologies for guaranteeing ultra-reliable low latency wireless communication based on a statistical model of the environment, 2) challenge of integrating ultra reliable communications into wireless control systems. In this project, our goal is to develop innovative methods based on machine learning and extreme value theory for wireless channel modeling, communication and control systems in the ultra-reliable region in 6G and beyond systems. Extreme value theory (EVT) is a unique statistical discipline that develops techniques and models to describe rare events. On the other hand, machine learning provides efficient solutions for the challenging problems that cannot be solved with traditional analytical approaches in the design of next generation communication systems of complex and heterogenous nature. Large amounts of data are used in the extraction of ultra-reliable channel statistics and the optimization problem in which ultra-reliable communication is integrated into the control systems is very challenging, thus, the use of EVT together with machine learning techniques may enable ultra-reliable control systems.

Communication Aware Dynamic Edge Computing (supported by EU CHIST-ERA)

Internet of things (IoT) is slowly permeating every aspect of our lives; however, we are far from having a truly intelligent IoT. Smart sensors generate massive amounts of data continuously; for instance, an autonomous vehicle is expected to generate about one gigabyte of data per second, but more often than not data is not systematically processed, stored, or analyzed for better inference. In this project, our goal is to develop real edge intelligence by enabling edge nodes to make local decisions rapidly and reliably in a collaborative manner. This will be achieved by developing novel caching, distributed computing and networking methodologies to enable federated/ distributed learning taking into account the network dynamics and physical channel variations. The developed joint computing, caching and communication framework will then be applied to a hierarchical heterogeneous architecture for vehicular ad-hoc networks (VANETs). This will not only enable efficient and reliable learning across mobile nodes, but also improve the security and privacy of autonomous cars by limiting decision making to local neighborhood.

Vehicular Visible Light Communication (supported by Ford Otosan)

Visible light communication (VLC) is a recently proposed alternative communication technology that might be used in achieving a secure communication protocol in vehicular networks by exploiting its distinguishing propagation characteristics. VLC uses modulated optical radiation in the visible light spectrum to carry digital information wirelessly. A VLC system usually uses a light emitting diode (LED) as the transmitting component and a photodiode or CMOS camera as the receiving component. LED has become very common in automotive lighting due to its long service life, high resistance to vibration, and better safety performance. Similarly, CMOS camera is already available in many vehicles as the front or rear camera for lane tracking and parking purposes. The light directivity and impermeability of the optical signal through vehicles and obstacles provide more secure data communication than IEEE 802.11p by limiting the transmission area. In this project, our goal is to design robust VLC based vehicular communication systems by performing wireless channel modeling and designing security protocols.

Radio Frequency Energy Harvesting based Communication Protocol Design for Time Critical Wireless Sensor Networks (supported by TUBITAK)

Time critical wireless sensor networks have been widely used in emergency alert systems and the control of cyber-physical systems due to many advantages, including easy installation and maintenance, low complexity and cost, and flexibility. Extending the battery lifetime of sensor nodes with energy harvesting technologies have initially focused on generating energy from natural sources such as sun, vibration and pressure. However, the dependency of the amount of harvested energy on the natural sources and resulting randomness prevents its usage in time-critical applications. On the other hand, energy harvesting technologies based on inductive or magnetic coupling cannot be practically used for these sensor networks due to their short energy transfer distance. Considering the advantages of having full control on energy transfer, high charging range and small form factor, radio frequency (RF) energy harvesting is the most suitable technology. The design of a large number of RF energy harvesting hardware over the past years is expected to even further extend its usage. In this project, our goal is to propose a novel optimization framework incorporating realistic constraints and scheduling for RF energy harvesting based time critical wireless sensor networks.

Energy Efficient Machine-to-Machine Communications (supported by TUBITAK)

Machine-to-machine (M2M) communications are expected to provide ubiquitous connectivity between machines without the need of human intervention. With the rapid development of the third generation long-term evolution (LTE) cellular networks and its advancement (LTE-Advanced), M2M communications via LTE-Advanced cellular networks with widespread coverage is expected to constitute a significant part of the Internet of Things. For mobile service operators, services through M2M communications have promising and strategic values. For instance, operators can expand their end-to-end information solutions into industries beyond the currently supported by exploiting the different characteristics of M2M services such as little bandwidth usage with minimal impact on the capacity of radio access networks. In this project, our goal is to design M2M communications to maximize the energy efficiency while providing the robustness of the transmissions and study energy efficient resource allocation algorithms for M2M communications by quantifying their utility when utilized together with human-to-human (H2H) communications in LTE advanced cellular networks.

Intra-Vehicular Wireless Sensor Networks (supported by Ford Otosan)

As the number of powerful electronic control units (ECU) and associated distributed sensor and actuator components inside the car keep increasing every year, the manufacturing and installation of wiring harnesses for transmission of data and power for all these components require considerable engineering effort. Eliminating, or greatly reducing, the wires in the harness can potentially provide part cost savings, assembly and maintenance savings, and fuel efficiency due to decreased weight. Moreover, wireless sensing would enable new sensor technologies to be integrated into vehicles (e.g. Tire Pressure Monitoring Systems, Intelligent Tire), which would otherwise be impossible using wired means. The objective of this project is to design the physical (PHY) layer and medium access control (MAC) layer of the communication protocol for intra-vehicular wireless sensor networks.