Distributed heuristic algorithms for RAT selection in wireless heterogeneous networks

In wireless heterogeneous networks, one of the most challenging problems is Radio Access Technology (RAT) selection that must be designed to avoid resource wastage. In this paper we adopt a hybrid model for RAT selection where the system allocates the downlink traffic between two different technologies in order to enhance global performance. We study the case of an integrated hybrid Wireless Local Area Network environment where the challenge we face is the high computational complexity necessary to obtain the global optimal solution. Therefore, we propose four distributed heuristic algorithms for RAT selection, where two of them are based on the distance between the user and the access points (APs), namely, distance based and probabilistic distance based algorithms. While the two others schemes are based on the peak rate that each user receives from these APs (peak rate based and probabilistic peak rate based algorithms). Results show that the proposed algorithms give efficient results compared to the optimal one depending on the spatial users distribution. Moreover these algorithms have a low computational complexity which makes them more advantageous compared to the optimal scheme in presence of a large number of users.