Efficiency of the FPGA Implementation of Dataflow Computations in Directly Connected Systems
Zhabin V.I., Zhabina V.V., Bezginsky M.A.
Possibility of reduction of the necessary FPGA resource for realization of data flow computing systems with direct connections between computing modules is considered. It is shown that the use of the computing modules, which process operands digit-by-digit with the use of redundant numerical system, enables to perform data flow calculations as well as to reduce the number of connections between computing modules. In comparison with the use of parallel devices the demanded number of FPGA functional cells and pins can be reduced with data processing high speed maintenance. There are shown examples of polynomial calculator immersion into FPGA and investigated its parameters.
