RSS stands for "Root-Sum-Square", which means “taking the square root of the sum of the squares of all the elements of a dataset”.

It can be employed to determine the total uncertainty in a measurement result caused by several independent error sources which are of similar magnitude (see NOTE 1). For each source of error the maximum possible value of deviation is squared in order to obtain a valid representation of the absolute worst case condition. This method is beneficial since it does not require that all error sources, at their extreme values, be of the same ‘direction’, i.e. all positive or all negative, so as to be able to sum together constructively. In many instances Spirent employ RSS and express the result in dB. The RSS method cannot be applied directly to logarithmic units and thus any such sources of error must first be expressed in linear units and the result converted back to logarithmic units. This is achieved as follows:-

NOTES:
1. When combining uncertainties of differing orders of magnitude using the RSS method then uncertainties are first separated and grouped by order of magnitude. The RSS method is applied to each individual group and the result of each calculation added together to obtain the overall total uncertainty for the measurement.