Rodents and Felidae whiskers are highly sensitive, detecting extrinsic inputs like airflow or contact and intrinsic inputs such as base vibrations or self-induced motion. Building effective artificial whisker sensors faces a challenge due to the intricate coupling of responses at the whisker base. There’s a research gap in understanding whisker sensors’ responses to intrinsic and extrinsic inputs. To address this, we propose two methods, using base acceleration as a reference input, 1) employing frequency domain adaptive filtering (FDAF) and 2) introducing the base vibration response model (BVRM), mathematically representing the whisker sensor’s behavior as a linear time-invariant (LTI) system. Validation of FDAF and BVRM is conducted through simulation and experimentation. The BVRM excels in both simulation and experiment with an SNR value of 35.63, proving superior. BVRM also shows potential in filtering sensor responses for independent use in terrain identification, flow sensing, and surface profile identification. Separating responses to extrinsic and intrinsic inputs without discarding either makes whisker sensors more versatile and multipurpose.