In this paper, results from an ongoing study, where acceleration data acquired with wireless sensors is used for damage assessment, are presented. A time series analysis based damage detection methodology that was previously developed by the authors and tested with wired sensors is employed. The method is based on using ARX models (Auto-Regressive models with exogenous input) for different sensor clusters. The acceleration outputs in each sensor cluster are used as the inputs to the ARX model to predict the output of the reference channel of that sensor cluster. The difference between the fit ratios is used as damage indicating feature. Laboratory experiments using a large-scale bridge model are conducted to investigate the effectiveness of the methodology for damage detection with wireless sensors. Different global (boundary condition change by fixing different supports) and local (local stiffness loss by removing bolts) damage cases are tested. The potential and advantages of the methodology are discussed in light of the analysis results. The limitations and shortcomings of the methodology are also addressed.