Please use this identifier to cite or link to this item:
Title: Method for performance-complexity analyses in SOC-based designs
Keywords: performance-complexity analysis, CABAC, CAVLC, H.264/AVC, rate-distortion optimization
Issue Date: 14-Aug-2009
Citation: SHYAM PARIKKAL KRISHNAMURTHY (2009-08-14). Method for performance-complexity analyses in SOC-based designs. ScholarBank@NUS Repository.
Abstract: The analyses of performance versus complexity of all the available algorithms in hardware/software are necessary to study the effectiveness of the implementation in a SoC-based design environment. Several performance-complexity analyses have been conducted, but no standard method has been reported. In this thesis, we propose a Performance-Complexity Index (PCI) to evaluate the cost-effectiveness of implementing one algorithm over other of same type, taking into account trade-offs in performance and complexity. As a demonstration, we analyze the performance and complexity of the two contending entropy coders adopted by H.264/AVC: Context-based Adaptive Binary Arithmetic Coding (CABAC) and Context-based Adaptive Variable Length Coding (CAVLC). Empirical test results using standard sequences show that it is more cost-effective to use CABAC for encoding when the Rate-Distortion Optimization mode is turned off regardless of configurations, in both variable and constant bit-rate implementations. The conclusions based on PCIs are also in total agreement with the empirical results.
Appears in Collections:Master's Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
KrishnamurthySP.pdf1.08 MBAdobe PDF



Page view(s)

checked on May 23, 2019


checked on May 23, 2019

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.