Mekatronik Mühendisliği Bölümü Yayın Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12416/255
Browse
Browsing Mekatronik Mühendisliği Bölümü Yayın Koleksiyonu by Journal "2017 25th Signal Processing And Communications Applications Conference (SIU)"
Now showing 1 - 2 of 2
- Results Per Page
- Sort Options
Book Part Citation Count: Afsin, Mehmet Ertug; Schmidt, Klaus Werner; Schmidt, Ece Guran, "A configurable CAN FD controller: architecture and implementation", 2017 25th Signal Processing And Communications Applications Conference (SIU), (2017).A configurable CAN FD controller: architecture and implementation(IEEE, 2017) Afşin, Mehmet Ertuğ; Schmidt, Klaus Werner; Schmidt, Ece GüranCAN FD is a new standard which provides fast. data rate while preserving the compatibility with CAN (controller area network). In this paper, a Configurable IP core architecture (A-CAN) which is compatible with the CAN FD standard, is proposed. Different than existing CAN/CAN FD controllers, the numbers and sizes of transmit and receive buffers of A-CAN can be configured in run time. To this end, A-CAN enables the best use of single controller hardware for different applications and enables improving the real time communication performance. A CAN communicates with the host device over SPI without any specific interface requirements. A-CAN is implemented on an FPGA Evaluation Board and its functionally is verified at a rate of 2 Mbps.Book Part Citation Count: Leblebicioglu, M. Kemal; Zengin, Yasin; Schmidt, Klaus Werner, "A new multi-agent decision making structure and application to model-based fault diagnosis problem", 2017 25th Signal Processing And Communications Applications Conference (SIU), (2017).A new multi-agent decision making structure and application to model-based fault diagnosis problem(IEEE, 2017) Leblebicioğlu, Kemal; Zengin, Yasin; Schmidt, Klaus WernerA new hierarchical multi-agent decision-making structure has been proposed. There are two phases of the structure. The first phase is the construction phase where the decision making structure consisting of switching and classification agents is built on the training data set generated by the system scenarios. In construction phase, switching and classification agents are trained and made ready for decision making. In the decision phase, which is the second phase, the class of the new data sample is decided. This process is carried out by the transmission of the data sample to the correct classifier agent by the switching agents and the classification by the classifier agent. The proposed structure is applied to a complex fault identification problem and a successful result is obtained. The structure is also adaptable to other big data decision making problems.