A novel random noise suppression method for SERF magnetometers based on chaotic algorithm

The spin-exchange relaxation free (SERF) magnetometer possesses the highest theoretical sensitivity limit among current magnetic sensors and holds immense potential for the measurement and imaging of extremely weak magnetoencephalography (MEG) signals. However, superconducting quantum interference devices (SQUIDs) remain the dominant sensors for commercial MEG applications. This is primarily because current SERF magnetometers are susceptible to noise interference, resulting in an actual sensitivity that falls significantly short of the theoretical limit, thereby hindering the effective measurement of extremely weak brain magnetic signals. To address the issue where random noise constrains the sensitivity enhancement of SERF atomic magnetometers, this study proposes a random noise suppression method based on a chaotic algorithm. The proposed chaotic algorithm is characterized by its inherent immunity to random noise. Random noise introduced into the SERF system—originating from sources such as photoelectric conversion, analog-to-digital conversion in sampling systems, and environmental vibrations—is effectively suppressed through the novel magnetic measurement method utilizing the proposed chaotic algorithm. Consequently, the sensitivity of the SERF magnetometer is significantly enhanced. These results lay a solid foundation for SERF magnetometers to eventually replace SQUIDs in the ultra-precision measurement and high-resolution imaging of extremely weak brain magnetic signals.