Pipelined Distributive Arithmetic-based FIR Filter Using Carry Save and Ripple Carry Adder

FIR filter bank play a vital role in any signal processing systems. The FIR filter is used to implement any frequency response digitally. Usually, the FIR filters contain multipliers, adders and many delays. As we know, the number of MAC operations (multiply and accumulate) makes the FIR filter less efficient and increases hardware complexity. So in this project, we included DA (Distributed Arithmetic) architecture in designing the efficient adaptive FIR filter. The pipelined Distributive Arithmetic architecture provides less power, less area and high values of throughput in creating an adaptive FIR filter. These are used in Echo cancellations, Radio channel equalizers and speech coding. DA architecture also helps in improving the processing speed of the FIR filter by eliminating the number of multipliers. A reconfigurable filter design can be achieved with the help of this Distributive Arithmetic architecture. Here the sum of the partial products which is pipelined is passed as the input values, and it is stored in the LUT (lookup tables) of the distributive architecture. By replacing the adder of the shift accumulation unit with a carry-save adder, the area of the proposed design is reduced. By placing the ripple carry adder, the size and delay are reduced to a further extent. Here we will use the Xilinx ISE tool to verify the simulation result, area required and time consumed.