Load Balancing in Hybrid VLC and RF Networks Based on Blind Interference Alignment

Visible light communications (VLC) are proposed for increasing the spectral efficiency and the number of devices served in indoor environments, while providing illumination through light emitting diodes (LED). For VLC, each optical access point (AP) provides a small and confined area of coverage. Since several sources of light are usually deployed in overlapping fashion in order to provide satisfactory illumination, VLC are limited by inter-cell interference. Moreover, transmission from a specific optical AP can be blocked by the elements of the scenario. On the other hand, radio-frequency (RF) systems such as WiFi are usually available in most of the indoor scenarios. In this work, we first propose a dynamic cell formation method for grouping the optical APs in multiple optical cells that cover a footprint each minimizing the inter-cell interference. After that, we use transmission based on blind interference alignment (BIA) in each optical cell. Considering the coexistence with RF systems based on orthogonal frequency division multiplexing (OFDM), a load balancing algorithm is proposed for managing the resources of the resulting hybrid VLC/RF network and determining the user association to each system. However, the complexity of this optimization problem is excessively high for practical VLC/RF networks. In order to obtain a suboptimal but tractable solution, we propose a decentralized optimization method based on Lagrangian multipliers. Simulation results show that the proposed scheme outperforms other approaches for user grouping and managing the resources of hybrid VLC/RF networks.