Merging massive black hole binaries (MBHBs) are important gravitational wave (GW) sources for the future space-based observatory LISA. The GW signal from the merger will be detected throughout the entire Universe. Characterization of the GW signal allows us to infer masses and spins of MBHs, the position of the source in the sky and the distance.  This information will allow us to understand the mechanism of MBHs formation and their evolution through cosmic history. 
   The binary could be surrounded by a gaseous disk accreting on each companion, which causes a variability in the X-ray emission correlated with the GW signal. Observation of this feature is a "smoking gun" of a binary system present in the active galactic nuclei. An accurate measure of the MBHB sky position during the inspiral will facilitate simultaneous multimessenger (GW and e/m) observation of merging MBHBs. 
   In this project, we consider early detection of MBHBs. The inspiral part of the signal is significantly fainter than the merger, nonetheless, we still can detect some sources days to weeks before the merger. The sky localisation of the source in the sky could be improved if we detect higher modes of the GW signal (breaking degeneracies in the parameter space).  In addition, we should be able to predict the merger time with an accuracy of 1-2 hours a few days before the merger and trigger a "protection period" that puts on hold any scheduled maintenance or manoeuvres. 
    The main objective of this stage is to demonstrate early detection of inspiralling MBHBs with higher modes. We will estimate the evolution of the source sky localisation as a function of time to merger and assess accuracy in determining the merger time. The side product of this project is the estimation of the MBHB parameters (masses, spins).