This work is motivated by modern monitoring and control infrastructures appearing in smart homes, urban environments, and industrial plants. These systems are characterized by multiple sensor and actuator devices at different physical locations, communicating wirelessly with each other. Desired monitoring and control performance requires efficient wireless communication, as the more information the sensors convey the more precise actuation becomes. However wireless communication is constrained by the inherent uncertainty of the wireless medium as well as resource limitations at the devices, e.g., limited power resources. The increased number of wireless devices in such environments further necessitates the management of the shared wireless spectrum with direct account of control performance. To address these challenges, the goal of this work is to provide control-aware and resource-aware communication policies. This is first examined in the fundamental problem of allocating transmit power resources for wireless closed loop control. Opportunistic online adaptation of power to plant and wireless channel conditions is shown to be essential in achieving the optimal tradeoff between control performance and power utilization. Optimal structural properties of channel access mechanisms are also considered for the problem of guaranteeing multiple control performance requirements over a shared wireless medium. This includes scheduling mechanisms implemented by central authorities, as well as decentralized mechanisms implemented independently by the wireless devices with emerging wireless interferences. Again the mechanisms exhibit an opportunistic adaptation to varying wireless channel conditions, especially designed to explore the tradeoffs between different communication links and meet control performance requirements. The structural characterization is augmented with tractable optimization algorithms to compute these channel access mechanisms. Finally, as control is naturally a dynamic task that requires a long term planning, appropriate dynamic algorithms adapting to the varying control system states are examined. Besides adapting dynamically, the proposed algorithms provide guarantees about long term control performance and resource utilization by construction.