Selecting WW Events

The rôle of calorimetry in characterising WW events is implicitly assumed in this brief outline of selection. Hadronic WW events are characterised by four hadronic jets, and the total missing energy and momentum are small. Typical preselection of hadronic events would use information on missing energy, multiplicity, spericity and thrust. A final selection would be based on kinematical variables or a multidimensional analysis (e.g. neural network). Selections are highly efficient ($\sim$ 85%) but notably not entirely pure (purity $\sim$ 80%) with the background mostly coming from q$\overline{q}$($\gamma$) events.

The semileptonic events are characterised by two hadronic jets, one isolated high momentum lepton and large missing momentum (using both direction and magnitude information). Choosing the lepton within the event uses lepton identification, in addition to the fact that, at LEP energies not too far above the W-pair threshold, the charged lepton is likely to be the track with the highest momentum component antiparallel to the missing momentum. q$\overline{q}$($\gamma$) events and 4-fermion events are most of the background, but the selection purity, typically 80-95%, is greater than in the hadronic channel.

Fully leptonic WW events have large missing energy and missing p_T and two acoplanar, acollinear leptons. Selection of these events uses lepton identification and event topology. Since there are two neutrinos, one from each W, extracting the W mass from these events relies on using the lepton energy spectra. This non-jet channel is not discussed, although there is a recent ALEPH result for m_W determined using the fully leptonic channel [2,3].