This part of IEC 61280-2 defines a procedure for calculating the root-mean-square error vector magnitude of optical n-APSK signals from a set of measured symbols. It specifically defines the normalization of the reference states and a procedure for optimal scaling of the measured symbol states. The procedure described in this document applies to single-polarized optical signals as well as to conventional polarization-multiplexed signals with independently modulated polarization tributaries. In general, it is not advisable to apply these procedures without modification to signals, in which optical amplitude, phase, and polarization state are simultaneously modulated to encode the information data. This document does not specify any signal processing steps for extracting the symbols from the received optical signals, because these steps depend on the optical receiver and can vary with the type of the transmitted n-APSK signal. These and optional additional signal processing steps are defined in application-specific documents.