Programmable Logic Device with an 8-stage cascade of 64K-bit Asynchronous SRAMs

نویسندگان

  • Kazuyuki NAKAMURA
  • Tsutomu SASAO
  • Munehiro MATSUURA
  • Katsumasa TANAKA
  • Kenichi YOSHIZUMI
  • Hui QIN
چکیده

The first implementation of a new programmable logic device using LUT(Look-Up Table) cascade architecture is developed in 0.35um CMOS logic process. Eight 64Kb asynchronous SRAMs are simply connected to form an LUT cascade with a few additional circuits. Benchmark results show that it has a competitive performance to FPGAs. 1. Introduction RAMs and PLAs (Programmable Logic Array) are used for programmable logic devices that realize multiple-output combinational logic functions. However, when the number of inputs and/or outputs for the target function is large, these devices require excessive amounts of hardware. Thus, FPGAs (Field Programmable Gate Arrays) are often used. However, the area and delay for interconnections among logic cells are much larger than those for logic elements. Therefore the prediction of the performance of the FPGA is difficult without complete physical design. To solve these problems, an LUT cascade architecture that is composed of a serial connection of large-scale memories has been developed [1]. The LUT cascade uses relatively larger LUTs (1kb-1Mb), and the interconnections between LUTs are limited to the adjacent cells in the cascade. This is quite different from the two-dimensional structure of FPGAs with smaller (16b-64b) LUTs. The large area for the interconnections in an FPGA is absorbed in the larger LUTs in the cascade. So, the cascade is re-configured by only changing the contents of the LUTs. 2. Design of LUT Cascade LSI An LUT block is mainly composed of an asynchronous 64kbit SRAM with 13bit address inputs and 8bit data I/Os. Since the memory should be operated as a data-path in LUT cascade architecture, an asynchronous SRAM is employed. In our design, each LUT block has 17 inputs: 8 bits are connected to the outputs of the preceding LUT, and 9 bits are from the external inputs (X). Then 13 of the 17 inputs are selected by crossover switches to form the actual address inputs of an LUT, and the unselected 4 signals can be used as intermediate outputs through Y terminals. This configuration enables to extend the number of allowable input/output signals from 48/8 (without crossover switches) to 76/36 (with crossover switches) for an 8-LUT cascade. This achieves an increase in the number of functions realizable with the LUT cascade. The LUT cascade LSI is simply realized by a cascade connection of LUT blocks. Each LUT block also has connections to the common address lines used for programming and testing. When a TEST (Program) …

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

A Memory-Based Programmable Logic Device Using Look-Up Table Cascade with Synchronous Static Random Access Memories

A large-scale memory-technology-based programmable logic device (PLD) using a look-up table (LUT) cascade is developed in the 0.35-mm standard complementary metal oxide semiconductor (CMOS) logic process. Eight 64K-bit synchronous SRAMs are connected to form an LUT cascade with a few additional circuits. The features of the LUT cascade include: 1) a flexible cascade connection structure, 2) mul...

متن کامل

Small World Communications MAP 04 T Very High Speed MAP Decoder

MAP04T Features 16 state soft–in–soft–out (SISO) maximum a posteriori (MAP) triple interleaved error control decoder and systematic recursive convolutional encoder Up to 95 Mbit/s decoding speed Rate 1/2, 1/3, or 1/4 with optional punctured inputs 6–bit received data, 8–bit soft–in and soft–out data for information and parity bits for all rates 8–bit branch metric inputs for rate 1/2 Optional c...

متن کامل

Division Algorithm Design Using Field Programmable Gate Array

This project is to design eight bit division algorithm program by using Xilinx ISE 10.1 software for simulation algorithm circuit partitioning through hardware Field Programmable Gate Array (FPGA). The algorithms are divide 8-bit dividend by 8-bit divisor for input and get the result 16-bit for the output. Circuit partitioning algorithms eight bits used to implement the distribution process for...

متن کامل

LM628/LM629 Precision Motion Controller

The LM628/LM629 are dedicated motion-control processors designed for use with a variety of DC and brushless DC servo motors, and other servomechanisms which provide a quadrature incremental position feedback signal. The parts perform the intensive, real-time computational tasks required for high performance digital motion control. The host control software interface is facilitated by a high-lev...

متن کامل

A programmable BIST architecture for clusters of multiple-port SRAMs

This paper presents a BIST architecture, based on a single micro-programmable BIST Processor and a set of memory Wrappers, designed to simplify the test of a system containing many distributed multi-port SRAMs of different sizes (number of bits, number of words), access protocol (asynchronous, synchronous), and timing.

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2005