Efficient Resource Allocation Optimization in a Multi-User Wireless Communication Systems: Proper Gaussian Signaling vs. Improper Gaussian Signaling

The next-generation wireless communication systems foresee continuous increase in the demand of achievable communication rates and the number of subscribers. This can be achieved by devising resource optimizing algorithms, which can optimize the power allocation or design beamforming solutions to support multiple users in the network. However, owing to the time-varying nature of wireless communication channels, the central processing unit has to re-evaluate the resource allocation parameters frequently, after every channel-coherence time-interval. The existing resource allocation algorithms are computationally complex as they mainly rely on some convex solvers to solve the formulated convex problem, which indirectly degrades the effective communication rates. This project aims to propose computationally efficient resource allocation optimization, which will be based on a closed-form solution at every iteration of the algorithm. This will be achieved by solving a different variant of the objective function compared to the commonly used objectives of sum - throughput ma ximiza tion or minimum users throughput ma ximiza tion (ma x -min throughput optimization). The proposed algorithm will substantially reduce the computational complexity and hence will improve the effective communication rate because the central controller will be able to provide the resource allocation parameters efficiently in a quite smaller training-time. Both proper Gaussian signaling and improper Gaussian signaling will be employed as the latter approach has been shown to outperform the former in t erms of achievable user throughput. The proposed algorithm will be compared with the existing solutions through extensive simulations