Compressed sensing (CS) is applied in wireless body sensor network (WBSN) to reduce the data rate and minimize the power consumption of the sensor nodes. However, as the CS encoder and decoder are tightly coupled, a model of the overall acquisition chain is required in the first stages of development and validation. To overcome this issue, we propose a virtual prototyping of WBSN based on CS with SystemC AMS 1.0. The proposed model consists of three sensor nodes which capture electrocardiogram (ECG), electromyogram (EMG) and respiration (RESP) signals. The proposed virtual prototype had allowed a functional verification of WBSN at system level and a rapid exploration of the impact of compression ratio on the quality of reconstruction. Results show how to tailor the measurement matrix for a best tradeoff between the compression ratio, the quality of reconstruction, and the energy consumption.