METHOD AND APPARATUS FOR COMPUTING SQUARE ROOTS OF BINARY NUMBERS

A square root is calculated by performing a plurality of iterations of the following: EQU X(n+1)=X(0)+X(n)* X(n)* 2.sup.-0.5*max-1 wherein X(0) is the seed value, max is the weight of the most significant bit of the smallest perfect binary square higher than the most significant bit of the operand A and calculating the square root R as follows: EQU R=2.sup.0.5*max -X(last) The seed value, or X(0) is calculated as follows: EQU x(0)=(2.sup.max -A) * 2.sup.-0.5*max-1 To increase the accuracy and reduce the error, the operand can be initially upscaled and then downscaled in the final operation.