Abstract: A method and system for an L.sub.2 norm operation are provided. A sequence of input signals is converted to a sequence of log signals. The sequence of log signals is distributed through a data pipeline to a plurality of processing elements which effectively multiply each log signal by a factor of two to produce a plurality of term signals. The term signals are then converted to inverse-log signals which are summed to produce a feedback signal. The square-root of the feedback signal is computed to generate an output signal. The log/inverse-log conversions of signals are based on estimating a log/inverse-log function using a second-order polynomial. The L.sub.2 norm operation can be performed using the distance between two vectors.