Title :
Most Dispersed and Greedy Tree Growing Algorithm for Designing LBG Initial Codebook
Author :
Liang, Yanxia ; Yang, Jiawei ; Li, Ye ; Liu, Wei
Author_Institution :
Broadband Wireless Commun. Lab., Xidian Univ., Xi´´an, China
Abstract :
The codebook generated by LBG algorithm depends on an initial codebook, thus a new LBG initial codebook generation algorithm was developed on the basis of Greedy Tree Growing Algorithm (GTGA) and Most Dispersed Codewords in Initialization (MDCI) algorithm. A fundamental codebook was generated by GTGA first, and then an initial codebook was obtained from this fundamental codebook by MDCI algorithm. The computation of this new algorithm is lower than usual Split algorithm, and the run time decreases too. Compared with these too basic algorithms, both the Average Distortion and Average Spectral Distortion are reduced.
Keywords :
greedy algorithms; trees (mathematics); vector quantisation; GTGA; LBG initial codebook generation algorithm; Linde-Buzo-Gray algorithm; MDCI algorithm; average spectral distortion; dispersed growing algorithm; greedy tree growing algorithm; most dispersed codeword in initialization algorithm; split algorithm; vector quantization; Algorithm design and analysis; Distortion measurement; Prediction algorithms; Signal processing algorithms; Size measurement; Training; Variable speed drives;
Conference_Titel :
Vehicular Technology Conference (VTC Spring), 2011 IEEE 73rd
Conference_Location :
Yokohama
Print_ISBN :
978-1-4244-8332-7
DOI :
10.1109/VETECS.2011.5956747