The paper considers a dual purpose communication-radar system in the presence of a jammer. The system transmits communication signals and uses their reflections off targets for target tracking. The system uses transmission power P to maintain high throughput for its communication component and high signal-to-interference-plus-noise ratio (SINR) for its radar component, via maximization of a weighted combination of throughput and SINR. The jammer transmits a jamming signal of power J, targeting at reducing a weighted combination of the system throughput and SINR, where the weights are known to the system in statistical terms, via an a priori distribution of possible weights. The problem of selecting P and J is formulated as Bayesian game between the system and the jammer. The waterfilling equation for finding the equilibrium strategy is derived, and the equilibrium uniqueness is proved. The impact of a priori probabilities about jam-mer's preferences on the system anti-jamming strategy is illustrated.