Traffic and Mobility Management in Networks of Femtocells

Abstract

A deployment of femtocells that is harmonic with its environment is a challenging issue. In this respect, interference management has traditionally been in the spotlight. However, architectural improvements for efficient femtocell deployments, despite being equally relevant, have received less attention. This paper presents a system architecture conceived for efficiently deploying femtocells in the form of Networks of Femtocells (NoFs). In this scenario, a group of femtocells in the same administrative domain cooperate towards a global performance improvement. Key to this improvement is the introduction of a new entity called Local Femto Gateway (LFGW) and the modifications in the femtocells in the local network. This allows offloading a high volume of control and data traffic from the core network of the mobile operator to the functional entities in the NoF. In particular, this paper focuses on building blocks related to traffic and mobility management. A two-level routing approach is discussed. The highest level is carried out by the mobile network layer. It is in charge of (1) determining the communication endpoints in the form of GPRS Tunneling Protocol (GTP) tunnel endpoint IDs, and (2) forwarding packets between tunnels belonging to the same Evolved Packet System (EPS) bearer at the appropriate nodes. Solutions for efficient handoff, local breakout, and local location management are presented for this level of routing. On the other hand, the lowest-level routing is carried out by the transport network layer. This level is in charge of finding the path between the above endpoints by efficiently using the local transport network that interconnects the femtocells in the NoF. A distributed routing solution for a large-scale, all-wireless network of femtocells is also presented. Overall, these architectural improvements render NoFs a promising approach for efficient traffic management in large-scale femtocell deployments, hence making them a scalable solution.