Theses
The Computer Engineering Group offers topics for theses in the area of integrated circuits, FPGAs, and video communications. For further information, please contact the academic staff directly.
Open Topics
Video decoding hardware implementation
- Title: Implementing a video decoding algorithm in hardware
- Type: Bachelor Thesis
- Supervisor: Dr.-Ing. Christian Herglotz
- Description: Video files tend to be very large. In order to use them practically, compression standards like H.264 have been defined allowing for significantly reduced bitrates at acceptable visual qualities. Unfortunately, compression is highly complex. The objective of this topic is to get an overview of how this standard works, test it on a software level, and later accelerate it by implementing on a hardware platform (specifically FPGA). The performance of both software and hardware should be evaluated, so that the acceleration of the hardware solution becomes evident.
Required Skills:
- Solid knowledge of basic hardware design (ideally with FPGAs).
Completely open-source hardware custom GPU
- Title: Obtaining memory traces out of a soft-core GPU (FGPU)
- Type: Master Thesis
- Supervisor: M. Eng. Hector Gerardo Munoz Hernandez
- Description:
The FGPU is a soft-core GPU running on an FPGA and has a working openCL support. The objective of this work is to get memory trace information from an already running hardware design, and use other available tools to obtain useful information about how to improve such memory accesses, thus tackling the memory bottle-neck.
Required skills:
- Basic knowledge of hardware design
- Basic knowledge of FPGAs and GPUs