Cost based Heterogeneous Access Management in Multi-Service, Multi-System Scenarios

This paper covers the issue of how mobile users of different service classes should be assigned to a set of radio access technologies (RATs) with overlapping coverage from a practical perspective. The aim is to allocate users with minimum rate constraints to all RATs that the weighted sum of assignable users is maximized. We introduce the concept of resource costs to formulate an optimization problem, which is very expensive to solve (NP-complete). Using the approach of continuous relaxation, we develop a suboptimal radio access selection algorithm which is designed for practical applications and converges close to the optimum. Simulation results show considerable gains of the utility in comparison to a standard Load Balancing Strategy.