Evaluation of High Speed Hardware Multipliers - Fixed Point and Floating point

Abstract

There is a huge demand in high speed arithmetic blocks, due to increased performance of processing units. For higher frequency clocks of the system, the arithmetic blocks must keep pace with greater requirement of more computational power. Area and speed are usually conflicting constraints so that improving speed results mostly in larger areas. In our research we will try to determine the best solution to this problem by comparing the results of different multipliers. Different sized of two algorithms for high speed hardware multipliers were studied and implemented ie. Parallel multiplier, Bit serial multiplier. The workings of these two multipliers were compared by implementing each of them separately in VHDL. A number of high speed adder designs are developed and algorithm and design of these adders are discussed. The result of this research will help us to choose the better option between serial and parallel multipliers for both fixed point and floating point multipliers to fabricate in different systems. As multipliers form one of the most important components of many systems, analysing different multipliers will help us to frame a better system with area and better speed. DOI: http://dx.doi.org/10.11591/ijece.v3i6.4185