45nm CMOS Tunable Low Pass Filter for On-Chip ECG Sensing

ECG signals are often contaminated with various types of noise and interference that can distort or obscure the underlying information that the signal contains. Examples of noise sources that can affect ECG signals include: 1) Electrical interference from other devices or power sources 2) Muscle activity or movement artifacts from the patient 3) Baseline wander caused by respiration or other body movements 4) High-frequency noise from electrical activity in the environment. To remove these sources of noise and interference, ECG signals are often filtered using a combination of analog and digital signal processing techniques. Design of low Hz range low-pass filter for on-chip biomedical data acquisition systems is an area of continuous research. Some of the biggest hurdles when designing sub-Hz-Hz filters are maintaining a low transconductance gain (Gm), while at the same time maintaining a high tunability range and dynamic range. A single-ended Gm-C integrator is often the basis from which most conventional low pass Gm-C filter structures begin. The performance of single-ended Gm-C integrators can be enhanced by using different techniques. In this work we explore several techniques to reduce transconductance, such as current steering and use of switching transistor. In this paper we realized a comparable design using 45nm CMOS process on cadence virtuoso. Our realized design achieves a cutoff frequency of 120 Hz using a 1V single supply and 500nA bias current. Lastly, we test the performance of our filter by feeding ECG data.