The use of magnetic induction (MI) based transmissions in challenging environments has been investigated in various works. Recently, a system model has been proposed, which explains how the pathloss of the magnetic induction based transmissions depends on the system parameters. It is frequently assumed, that perfect channel state information (CSI) is available at the transmitter and at the receiver, such that the optimal set of system parameters can be determined in order to maximize the overall data rate. However, in practical systems this knowledge may not always be easily acquired. In addition, a permanent feedback signaling is needed, in order to update the CSI at the transmitter. This feedback signaling usually occupies several time slots and therefore reduces the bandwidth efficiency. An interesting aspect, which has been overlooked in the past, is the channel estimation and prediction of the received signal within the MI transmitter circuit without explicit feedback signaling of CSI. In this paper, we investigate the potential of this technique for the wireless underground sensor networks.