Code Optimization Techniques
Computers transfer data in a number of different ways. Whether through a serial port, a parallel port, over a modem, over an ethernet cable, or internally from a hard disk to memory, some data will be lost. To compensate for that loss, numerous error detection and correction algorithms have been developed. One of the most common error correction codes is the Reed-Solomon code, which is a special subset of BCH (Bose-Chaudhuri-Hocquenghem) linear cyclic block codes. In the AURA project, an unmanned aircraft sends the data it collects back to earth so it can be analyzed during flight and possible flight modifications made. To counter possible data corruption during transmission, the data is encoded using a multi-block Reed-Solomon implementation with a possibly shortened final block. In order to maximize the amount of data transmitted, it was necessary to reduce the computation time of a Reed-Solomon encoding to three percent of the processor's time. To achieve such a reduction, many code optimization techniques were employed. This paper outlines the steps taken to reduce the processing time of a Reed-Solomon encoding and the insight into modern optimization techniques gained from the experience.
- Research Organization:
- Sandia National Lab. (SNL-NM), Albuquerque, NM (United States); Sandia National Lab. (SNL-CA), Livermore, CA (United States)
- Sponsoring Organization:
- US Department of Energy (US)
- DOE Contract Number:
- AC04-94AL85000
- OSTI ID:
- 760803
- Report Number(s):
- SAND2000-1649C; TRN: AH200037%%286
- Resource Relation:
- Conference: Student Symposium 2000, Albuquerque, NM (US), 08/10/2000; Other Information: PBD: 3 Aug 2000
- Country of Publication:
- United States
- Language:
- English
Similar Records
Final Report for the project titled "Enabling Supernova Computations by Integrated Transport and Provisioning Methods Optimized for Dedicated Channels"
Highly fault-tolerant parallel computation