| Presentation | 2003/5/22 A High-Performance Processor Dedicated to Codebook Design for Vector Quantization Chiaki TAKAGI, Ryusuke EGAWA, Kentaro SANO, Tadao NAKAMURA, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Vector quantization(VQ) is an attractive technique for lossy data compression, which has been a key technology for data storage and/or transfer. So far, various algorithms have been proposed to design optimal codebooks presenting quantization with minimized errors. However, the amount of computation for codebook design is extremely huge for large data sets, restricting applications of optimal vector quantization. This paper provides massively parallel architecture for a VQ processor. This VQ processor performs modified Kohonen learning where delayed update of codebooks is introduced. Software simulation gave results denoting that the delayed update does not introduce serious increase of errors. The VQ Processor consists of code-vector cells on a 2D grid. The synopsys design compiler provided the critical path of 15.00 ns for the code-vector cell designed by using the ROHM 0.35 μm CMOS process technology. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | vector quantization / codebook design / functional memory / massively parallel processing |
| Paper # | NC2003-6 |
| Date of Issue |
| Conference Information | |
| Committee | NC |
|---|---|
| Conference Date | 2003/5/22(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Neurocomputing (NC) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | A High-Performance Processor Dedicated to Codebook Design for Vector Quantization |
| Sub Title (in English) | |
| Keyword(1) | vector quantization |
| Keyword(2) | codebook design |
| Keyword(3) | functional memory |
| Keyword(4) | massively parallel processing |
| 1st Author's Name | Chiaki TAKAGI |
| 1st Author's Affiliation | Graduate School of Information Science, Tohoku University() |
| 2nd Author's Name | Ryusuke EGAWA |
| 2nd Author's Affiliation | Graduate School of Information Science, Tohoku University |
| 3rd Author's Name | Kentaro SANO |
| 3rd Author's Affiliation | Graduate School of Information Science, Tohoku University |
| 4th Author's Name | Tadao NAKAMURA |
| 4th Author's Affiliation | Graduate School of Information Science, Tohoku University |
| Date | 2003/5/22 |
| Paper # | NC2003-6 |
| Volume (vol) | vol.103 |
| Number (no) | 92 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |