Rotation measure synthesis allows the estimation of Faraday dispersion via a Fourier transform and is the primary tool to probe cosmic magnetic fields. We show this can be considered mathematically equivalent to the one-dimensional (1D) interferometric intensity measurement equation, albeit in a different Fourier space. As a result, familiar concepts in 2D intensity interferometry designed to correctly account for a range of instrumental conditions can be translated to the analysis of Faraday dispersion. In particular, we show how to model the effect of channel averaging during Faraday reconstruction, which has to date limited the progress of polarimetric science using wide-band measurements. Further, we simulate 1D sparse reconstruction with channel averaging for realistic frequency coverages, and show that it is possible to recover signals with large rotation measure values that were previously excluded from possible detection. This is especially important for low-frequency and wide-band polarimetry. We extended these ideas to introduce mosaicking in Faraday depth into the channel-averaging process. This work thus provides the first framework for correctly undertaking wide-band rotation measure synthesis, including the provision to add data from multiple telescopes, a prospect that should vastly improve the quality and quantity of polarimetric science. This is of particular importance for extreme environments that generate high magnetic fields such as those associated with pulsars and fast radio bursts, and will allow such sources to be accurately used as probes of cosmological fields.