A Feasible Approach for QoS Management in Coordinated Heterogeneous Radio Access Networks R. Ferrús  , A. Gelonch  , O. Sallent  , J. Pérez-Romero   Signal Theory and Communications Department, UPC, Barcelona [ferrus, antoni, sallent, jordip]@tsc.upc.edu N. Nafisi  , M. Dohler   Centre for Telecommunications Research, King's College London [nima.nafisi, michael.dohler]@kcl.ac.uk Abstract This paper develops a feasible architecture for end- to-end QoS management over B3G networks where multiple heterogeneous radio access networks (RAN) can be coordinated at radio level by means of supporting Common Radio Resource Management (CRRM) functionalities. The presented approach is claimed to be innovative as long as it is inspired in the most relevant trends being developed in different fields and forums and come up with a global solution capturing such partial concepts. A key point of the architecture is the introduction of a Wireless QoS Broker entity that becomes the link between CRRM and E2E QoS management. In the paper, after introducing the proposed QoS framework, the signaling and procedures to support such architecture in the B3G network are addressed. Over such a basis, a connection establishment procedure is discussed to show how the problem of initial RAN selection can be managed by the proposed architecture. 1. Introduction It is generally acknowledged today that the Beyond 3G network concept encompasses network heterogeneity. A plethora of different radio access networks (e.g. cellular networks as GPRS and UMTS as well as WLAN hotspots) will have to co-exist. These different networks ought to be inter-connected in an optimum manner with the ultimate objective to provide the end-user with the requested services and corresponding QoS (Quality of Service) requirements. In a heterogeneous radio access network (RAN) scenario wireless users can be given access to their services through different available radio access technologies (RAT). A key issue in such complex scenarios where different degrees of interworking may exist (e.g. tight and loose coupling architectures for WLAN and UMTS integration) as well as different usage strategies (e.g. introduction of economical costs criteria) is how QoS can be managed and provisioned in a flexible and affordable way over the whole set of available RATs. Besides, QoS provisioning in the different RATs can be based on different technology-specific mechanisms, thus making even more challenging the need to harmonize them in order to fulfill the B3G network concept. The envisaged beyond 3G scenarios introduces a new dimension into the radio resource management problem, in addition to the need for a proper interworking among RATs through adequate architectures. That is, instead of performing the management of the radio resources independently for each RAT, some form of overall and global management of the pool of radio resources can be envisaged. Common Radio Resource Management (CRRM) is the envisaged process to manage dynamically the allocation and de-allocation of radio resources (e.g. time slots, codes, frequency carriers, etc.) within a single or between different radio access systems for the fixed spectrum bands allocated to each of these systems. In this way, a more efficient usage of the available radio resources can be achieved. 3GPP coped for the UTRAN/GERAN CRRM issue with main contributions in Release 5 [TR 25.881, TR 25.891], though in Release 6 several options were considered but any new functionality has been added yet. In any case, CRRM functionalities should be incorporated someway into the end-to-end QoS management framework in order to encompass critical decision processes such as RAT selection for served connections. Another promising approach to be considered in a B3G QoS management framework is Policy-Based Networking (PBN). PBN is a novel technology that facilitates the management and operation of networks. In today’s highly competitive market, policy-enabled networks appear as a promising approach to reduce the cost of network operation and maintenance while providing a great flexibility to satisfy the demands of complex service provision frameworks. Policy-based management allows operators and networks providers to deploy and correlate business strategies with the overall network actions. Cellular networks are not unaware of this potential and some important efforts have been already overtaken to introduce service policy-based QoS control in the IP