Stabilization of noisy plants over a Gaussian interference channel

Remote stabilization of two noisy linear plants over a symmetric Gaussian interference channel with possibly noisy feedback from controllers to the plants is studied. Each plant is monitored by a separate sensor which communicates its observations to a separate controller over a Gaussian interference channel. The controllers aim at stabilizing the two plants in mean-square sense. The plant noise is assumed to be additive white Gaussian distributed. A delay-free sensing and control scheme is proposed, which is an adaptation of the Schalkwijk-Kailath type coding schemes developed for transmission of reliable information over channels with noiseless feedback. The conditions that guarantee the stabilization of the noisy plants under the proposed scheme are derived, extending the results that are known for mean-square stabilization of noiseless plants over noiseless feedback channels. Numerical results indicate that the stabilizability is not affected by the inclusion of an additive Gaussian noise in the plant models under the proposed scheme.