Optimal spectrum utilization in joint automotive radar and communication networks

Due to rapid growth of wireless traffic demands in vehicular networks, spectrum scarcity is becoming urgent in the Dedicated Short Range Communication (DSRC) band. One solution is reusing automotive radar bands without degrading radar performance. Despite having massive bandwidths, imaging accuracy of automotive radars is still low due to correlations between sequential target observations of single radar. A solution is that vehicles exchange imaging information through vehicle-to-vehicle communications. Since observations of different vehicles are less correlated, Joint Automotive Radar and Communication (JARC) network is able to improve imaging accuracy. More importantly, some spectrum resources can be left to alleviate the DSRC spectrum scarcity problem. In this paper, we derive the Cramer-Rao bound for parameter estimation in JARC networks. Then, we formulate the spectrum utilization problem as an NP-complete integer quadratic program, to which we propose an optimal (in expectation) algorithm with low complexity. Finally, efficacy of the algorithm is illustrated through numerical results.