Pipelined implementation of fixed point square root in FPGA using modified non-restoring algorithm

Square root is one of the fundamental arithmetic operations in signal and image processing algorithms. This article presents a novel pipelined architecture to implement N-bits fixed point square root in FPGA using non-restoring algorithm. Pipelining hazards were avoided by modifying the non-restoring algorithm resulting in a 30% improved latency time. Furthermore, the proposed architecture is flexible and can be modified as per the need of an application. The performance of the proposed system, as a function of execution time and power consumption per operation, has been compared with other floating point pipelined implementations. It is demonstrated that the proposed system is ~ 2 times efficient compared to its counterparts.