19-03-2011, 03:20 PM
presented by:
S.TEJASWI
HIGH-PERFORMANCE DSP CAPABILITY WITHIN OPTIMIZED LOW-COST FPGA ARCHITECTURE
ABSTRACT:
The applications of Digital Signal Processing (DSP) continue to expand, driven by trends such as the increased use of video and still images and the demand for increasingly reconfigurable systems such as Software Defined Radio (SDR). Many of these applications combine the need for significant DSP processing with cost sensitivity, creating demand for high-performance, low-cost DSP solutions.General-purpose DSP chips and FPGAs are two common methods of implementing DSP functions.
The use of DSP techniques will continue to grow at the expense of analog implementations. An analysis of the functions typically used in DSP applications indicates that a combination of multiplier, addition, subtraction and accumulation elements is required. The Lattice ECP devices provide a sophisticated DSP block combined with a low-cost FPGA fabric. Through the implementation of addition, subtraction, accumulation and pipelining within the sysDSP block, performance and LUT utilization are considerably higher than those of alternative low-cost FPGA solutions that provide only basic multiplier capabilities. The speed and utilization advantages of the sysDSP block help users reduce costs through the selection of smaller and lower speed grade devices.
This paper provides an overview of common DSP functions and then explores the differences between the general purpose DSPs and FPGAs. This is followed by a description of the LatticeECP™-DSP (EConomy Plus Digital Signal Processing) architecture and a comparison of the LatticeECP-DSP to existing FPGA solutions.
S.TEJASWI
HIGH-PERFORMANCE DSP CAPABILITY WITHIN OPTIMIZED LOW-COST FPGA ARCHITECTURE
ABSTRACT:
The applications of Digital Signal Processing (DSP) continue to expand, driven by trends such as the increased use of video and still images and the demand for increasingly reconfigurable systems such as Software Defined Radio (SDR). Many of these applications combine the need for significant DSP processing with cost sensitivity, creating demand for high-performance, low-cost DSP solutions.General-purpose DSP chips and FPGAs are two common methods of implementing DSP functions.
The use of DSP techniques will continue to grow at the expense of analog implementations. An analysis of the functions typically used in DSP applications indicates that a combination of multiplier, addition, subtraction and accumulation elements is required. The Lattice ECP devices provide a sophisticated DSP block combined with a low-cost FPGA fabric. Through the implementation of addition, subtraction, accumulation and pipelining within the sysDSP block, performance and LUT utilization are considerably higher than those of alternative low-cost FPGA solutions that provide only basic multiplier capabilities. The speed and utilization advantages of the sysDSP block help users reduce costs through the selection of smaller and lower speed grade devices.
This paper provides an overview of common DSP functions and then explores the differences between the general purpose DSPs and FPGAs. This is followed by a description of the LatticeECP™-DSP (EConomy Plus Digital Signal Processing) architecture and a comparison of the LatticeECP-DSP to existing FPGA solutions.