Weighted sum rate maximization on two user Gaussian interference channels with rate splitting

In interference limited networks treating interference as noise leads to performance degradation at high transmit powers. This can be avoided with rate splitting schemes (Han-Kobayashi region) where portions of the interference are decoded and removed at each receiver. We propose an algorithmic framework for solving the weighted sum rate maximization problem for the two user case of Gaussian interference channels. The polyblock outer approximation algorithm is used to solve the problem despite its nonconvexity. Contrary to usual brute force sampling approaches this framework may be extended to more users and other Gaussian interference models such as the X channel.