We explore the precision with which the Einstein Telescope ( ET ) will be able to measure the parameters of intermediate–mass–ratio inspirals ( IMRIs ) , i.e. , the inspirals of stellar mass compact objects into intermediate-mass black holes ( IMBHs ) .
We calculate the parameter estimation errors using the Fisher Matrix formalism and present results of Monte Carlo simulations of these errors over choices for the extrinsic parameters of the source .
These results are obtained using two different models for the gravitational waveform which were introduced in paper I of this series .
These two waveform models include the inspiral , merger and ringdown phases in a consistent way .
One of the models , based on the transition scheme of Ori & Thorne ( ) , is valid for IMBHs of arbitrary spin , whereas the second model , based on the Effective One Body ( EOB ) approach , has been developed to cross–check our results in the non-spinning limit .
In paper I of this series , we demonstrated the excellent agreement in both phase and amplitude between these two models for non–spinning black holes , and that their predictions for signal–to–noise ratios ( SNRs ) are consistent to within ten percent .
We now use these waveform models to estimate parameter estimation errors for binary systems with masses 1.4 M _ { \odot } + 100 M _ { \odot } , 10 M _ { \odot } + 100 M _ { \odot } , 1.4 M _ { \odot } + 500 M _ { \odot } and 10 M _ { \odot } + 500 M _ { \odot } and various choices for the spin of the central IMBH .
Assuming a detector network of three ETs , the analysis shows that for a 10 M _ { \odot } compact object ( CO ) inspiralling into a 100 M _ { \odot } IMBH with spin q = 0.3 , detected with an SNR of 30 , we should be able to determine the CO and IMBH masses , and the IMBH spin magnitude to fractional accuracies of \sim 10 ^ { -3 } , \sim 10 ^ { -3.5 } and \sim 10 ^ { -3 } , respectively .
We also expect to determine the location of the source in the sky and the luminosity distance to within \sim 0.003 steradians and \sim 10 \% , respectively .
We also compute results for several different possible configurations of the detector network to assess how the precision of parameter determination depends on the network configuration .