Performance Analysis of GPU Based GNSS Signal Processing

Ulrich Haak, Hans-Georg Büsing, Prof. Peter Hecker

Abstract:	The increasing number of developments of software-based GNSS receivers especially in the research area has been initiated by two major aspects: The first aspect is the lack of interfaces in off the shelf receivers which are nec-essary to realize positioning solutions based on deep sen-sor integration. The second aspect is the increase of com-putational speed due to the developments in the computer sector in the past years which made real-time software receiver applications without additional processing hard-ware like FPGA boards possible. The latest CUDA release offers a profiling toolkit to as-sess the performance of processing implementations. The tracing functionality of the toolkit allows reviewing the processing chain and may expose unnecessary idle times in data processing. The device memory usage and data transfer bandwidth can also be evaluated using the profil-er. Using these tools the software receiver has been opti-mized for deployment in the vehicle PC. In this paper the influence of different hardware and soft-ware configurations as well as optimizing possibilities are presented. Furthermore the detailed profiling results are discussed especially regarding the workload ratio of the individual receiver threads with respect to the data han-dling overhead.
Published in:	Proceedings of the 25th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2012) September 17 - 21, 2012 Nashville Convention Center, Nashville, Tennessee Nashville, TN
Pages:	2371 - 2377
Cite this article:	Haak, Ulrich, Büsing, Hans-Georg, Hecker, Prof. Peter, "Performance Analysis of GPU Based GNSS Signal Processing," Proceedings of the 25th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2012), Nashville, TN, September 2012, pp. 2371-2377.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In