“…Ideally, in a differential measurement, two identical electrodes should cancel-out their half-cell potentials [15]. In practice, a random differential electrode offset, up to hundred millivolts, can appear [4], [13], [16], [17]. Moreover, the situation is worsened when the recording functionality is combined with the brain stimulation, where a build-up of charge occurs between electrodes [18]- [20], as well as undesired in-band stimulation artifacts [16], [17].…”