Development of a Computing System by a Functional Memory Type Parallel Processor.

通过功能存储器型并行处理器开发计算系统。

• 批准号: 07558039
• 负责人: TAMARU Keikichi
• 金额: $ 5.95万
• 依托单位: KYOTO UNIVERSITY
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (A)
• 财政年份: 1995
• 资助国家: 日本
• 起止时间: 1995 至 1997
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-07558039/
• 关键词: Functional Memory; CAM; Vector Quantization; Parallel Processing; VLSI; Low Bit-rate; Image Compression; 超並列処理; 光通信

(a)Development of a Functional Memory for AdditionWe develop a functional memory specialized for addtion. Its memory cell consists of a DRAM,which area is about 10 times smaller than that of an SRAM.A PE consists of two words and a bit-scrial ALU.We have designed and fabricated LSIs to evaluate is functionality and performance.(b)Low Bit-rate Image Compression Using a Functional MemoryHere, we develop a functional memory specialized for vector quantization. It is called "FMPP-VQ", which stands for Functional Memory Type Parallel Processor for Vector Quantization. We have designed and fabricated two LSIs. One is an LSI for evaluation which consists of four PEs. The other has 64 PEs. It can be applied for low bit-rate image compression. We have developed an algorithm to send a QCIF (176*144) video sequence at 10 frames per second via a 29.2kbps line.

(a)加法功能存储器的开发我们开发了一种专门用于加法的功能存储器。它的存储单元由DRAM组成，其面积比SRAM小10倍左右。PE由两个单词和一个串行ALU组成。我们已经设计和制造了lsi来评估其功能和性能。(b)使用功能存储器的低比特率图像压缩在这里，我们开发了一种专门用于矢量量化的功能存储器。它被称为“FMPP-VQ”，即用于矢量量化的功能存储器型并行处理器。我们设计并制造了两个lsi。一种是用于评估的大规模集成电路，由四个pe组成。另一个有64个pe。它可以应用于低比特率的图像压缩。我们开发了一种算法，通过29.2kbps的线路以每秒10帧的速度发送QCIF（176*144）视频序列。

项目成果

K.Tamaru: "A Low Bit-rate Image Compression Algorithm Using a Functional Memory-Type Parallel Processor for Vector Quantization" 10th Karuizawa Workshop. 291-296 (1997)

K.Tamaru：“使用功能存储器型并行处理器进行矢量量化的低比特率图像压缩算法”第 10 届轻井泽研讨会。

K.Kobayashi: "A memory-based parallel processor for vector quantization" 22nd European Solid-State Circuit Conference. 184-187 (1996)

K.Kobayashi：“用于矢量量化的基于存储器的并行处理器”第 22 届欧洲固态电路会议。

K.Tamaru: "Real Time Low Bit-Rate Video Coding Algorithm Using Multi-Stage Hierarchical Vector Quantization(発表予定)" IEEE International Conference on Acoustics,Speech,and Signal Processing. (1998)

K.Tamaru：“使用多级分层矢量量化的实时低比特率视频编码算法（待提交）”IEEE 国际声学、语音和信号处理会议（1998 年）。

K.Tamaru: "Estimation of Short-Circuit Power Dissipation and Its Influence on Propagation Delay for Static CMOS Gates" Proc.of the 1996 IEEE International Symp.on Circuits and Systems. 4. 751-754 (1996)

K.Tamaru：“静态 CMOS 门的短路功耗估计及其对传播延迟的影响”1996 年 IEEE 国际电路和系统研讨会论文集。

K.Tamaru: "An LSI for Low Bit-Rate Image Compression Using Vector Quantization" IEICE Trans.on Electron. E81-C(発表予定). (1998)

K.Tamaru：“使用矢量量化进行低比特率图像压缩的 LSI”IEICE Trans.on Electron（即将推出）。