An intelligent and optimized multihoming approach in real and heterogeneous environments

Over the years, the users’ requirements are increasingly stringent and difficult to fulfil. In response to these challenges, access networks have been suffering a systematic evolution, becoming more robust, more efficient and with a higher capacity. In an overall heterogeneous network, it becomes important to optimize the utilization of all available resources in order to obtain a better user experience and a better network utilization. This paper proposes, implements and evaluates a new multihoming architecture to take the full advantage of each access network in the range of the end-user terminal. We propose a dynamic architecture capable of integrating context-information and, consequently, of maintaining the multihoming optimal rules valid in real-time, without impairing access networks’ performance. To evaluate the proposed architecture, all entities were implemented, integrated with a mobility protocol, and tested in real scenarios with both Wi-Fi and cellular networks. The obtained results demonstrate the advantages of the multihoming approach, which can always achieve the optimal system performance. It is also shown the capability of the proposed architecture to dynamically adapt to environment changes through an on-line approach for the measurement of the available bandwidth. Finally, we demonstrated the good performance of our approach in the face of out-of-order packets.