Satellite communications (SATCOM) are prone to both intentional and unintentional interference, which can significantly degrade the reliability of packet transmissions. Here, we investigate different approaches for interference mitigation in SATCOM based on dynamic frequency hopping (DFH). We consider a star topology, where multiple LEO satellites transmit packets to a common GEO satellite. The FH pattern of each LEO-GEO link is adjusted according to an outcome of out-of-band proactive sensing scheme, carried out by a cognitive radio (CR) module that resides in the GEO satellite. Based on sensing results, channels with high predicted interference are replaced with better channels, without disrupting the communications of other satellites in the network. In searching for replacement channels, we aim to ensure that all satellite links are assigned channels such that their SINR requirements are met. At the same time, the total transmission power and the communication overhead resulting from altering the FH patterns are minimized. We formulate the problem as a multi-objective minimization problem. Continuous-time Markov chain analysis is used to predict future channel conditions. The proposed scheme is compared with two other schemes, namely best-channel-to-closest- user and best-channel-to-farthest-user. Finally, we use simulations to study the effects of different system parameters on the performance of the proposed DFH design.