One of the main channels which allows for a large rate of neutralino dark matter annihilation in the early Universe is via the pseudoscalar Higgs A -resonance .
In this case , the measured dark matter abundance can be obtained in the minimal supergravity ( mSUGRA ) model when \tan \beta \sim 50 and 2 m _ { \widetilde { Z } _ { 1 } } \sim m _ { A } .
We investigate the reaction pp \rightarrow b \phi \rightarrow b \mu ^ { + } \mu ^ { - } + X ( where \phi = A or H ) at the CERN LHC where requiring the tag of a single b -jet allows for amplification of the signal-to-background ratio .
The rare but observable Higgs decay to muon pairs allows for a precise measurement of the Higgs boson mass and decay width .
We evaluate signal and background using CalcHEP , with muon energy smearing according to the CMS detector .
We find that the Higgs width ( \Gamma _ { A } ) can typically be determined with the accuracy up to \sim 8 \% ( \sim 17 \% ) for m _ { A } \sim 400 ( 600 ) GeV assuming 10 ^ { 3 } fb ^ { -1 } of integrated luminosity .
Therefore , the pp \rightarrow b \phi \rightarrow b \mu ^ { + } \mu ^ { - } + X process provides a unique possibility for direct \Gamma _ { A } measurement at the LHC .
While the Higgs width is correlated with the parameter \tan \beta for a given value of m _ { A } , extracting \tan \beta is complicated by an overlap of the A and H peaks , radiative corrections to the b and \tau Yukawa couplings , and the possibility that SUSY decay modes of the Higgs may be open .
In the case where a dilepton mass edge from \widetilde { Z } _ { 2 } \rightarrow \ell ^ { + } \ell ^ { - } \widetilde { Z } _ { 1 } is visible , it should be possible to test the relation that 2 m _ { \widetilde { Z } _ { 1 } } \sim m _ { A } .

PACS numbers : 14.80.Ly , 12.60.Jv , 11.30.Pb