Recently, the sharing of network infrastructure using Network Function Virtualisation (NFV) has become an active research topic, being meant to transform the way operators architect their networks [1]. Based on this, in [2] the concept of radio resource virtualisation has been proposed. The goal of this approach is to serve multiple Virtual Network Operators (VNOs) over the same infrastructure, while offering isolation and flexibility in addition to network element abstraction and multi-RAT (Radio Access Technology) support. Instead of splitting available radio resources among VNOs, it is suggested to aggregate and manage them. VNOs request wireless capacity from a set of physical network providers to serve their subscribers, thus not have to deal with the physical infrastrucutre. Applying virtualisation of radio resources increases usage efficiency, while reducing the operational and maintenance costs. It integrates multiple radio access technologies, combines cellular networks and WLANs, and increases the flexibility of handling network traffic changes in RANs. In addition, it offers pay-as-you-go Connectivity-as-a-Service (CaaS) to VNOs, while enabling new business models for network operators and infrastructure providers. Although the virtualisation of RAN may sound similar to active RAN sharing, they correspond to very different concepts. As illustrated in ‎Fig. 1 the key concept of Virtual RAN (V-RAN) corresponds to the one of Virtual Machines (VMs). In RAN virtualisation, in contrast to RAN sharing, the physical infrastructure is not transparent to the clients. By means of isolation, services with different protocols, algorithms, and requirements can be offered over the same physical infrastructure. Since the two virtual entities of a virtual network are completely isolated from each other, the problems of one of them do not affect the other, thus, the network has a lower downtime compared to shared networks.

Fig.1 – Comparison between Virtual Machine and Virtual RAN.

The architecture for the management of virtual radio resources, ‎Fig. 2, is a hierarchal one, consisting of a Virtual RRM (VRRM) entity on the top of the usual RRM ones for heterogeneous access networks [3], Common RRM (CRRM), and local RRMs.

Fig.2 – Architecture of Virtual Radio Access Network.

Virtual Network Operators (VNOs) are placed at the highest level of this hierarchy; a VNO is a network operator that does not own any radio access infrastructure, but providing wireless connectivity for its subscribers. The RAN Provider (RANP), who owns a physical infrastructure, offers a virtualised RAN to VNOs. This V-RAN is also known as RAN-as-a-Service (RANaaS), [4]. The VNO does not have to care about managing a RAN, rather requesting RANPs for capacity to serve each end-user service class. The RANP serves each VNO based on a Service Level Agreement (SLA), which can generally be summarised into three types of contract:
• Guaranteed, in which a VNO is guaranteed a minimum, as well as a maximum, levels of service (e.g., data rate) to be served on-demand. The operator receives the guaranteed capacity regardless of network situation. Subscribers also experience better quality of service.
• Best effort with minimum guarantees is the second type of contract, in which the VNO is served with a minimum guaranteed level of service, but the request for better services (e.g., higher data rates) is performed in a best effort manner. The VNO does not invest as much as the operator with a guaranteed contract, but it can guarantee a minimum QoS to its subscribers. Subscribers experience an acceptable network performance, but not as good as those in a guaranteed VNO do. However, it is expected that they pay relatively lower price for their services, since the RAN costs less to the operator.
• Best effort, in which the VNO is served in a pure best effort manner. This type of contract has the lowest cost for operators. However, operators, and consequently their subscribers, may suffer from low QoS during busy hours.

VRRM, the highest manager in the hierarchy, is in charge of translating VNOs requirements and SLAs into sets of polices for lower levels. VRRM optimises the usage of virtual radio resources, not dealing with physical ones. Nevertheless, reports and monitoring information (e.g., estimated remaining capacity) received from CRRM enable it to improve policies. CRRM, at a lower level, is the intermediate level between VRRM and local RRMs, managing radio resources in heterogeneous access environment. Local RRMs, in the lowest level, are liable for optimising radio resource usage in a single access technology.
In [2, 5], a new model for the management of virtual radio resources is proposed. The model has two main parts: estimating and allocating radio resources. By means of this model, support of different types of SLAs and contracts is possible. The extension of the model to operate in resource shortage situation is presented in [6]. In the same paper, the effect of temporarily changing C-RAN cell layout on VNOs is also studied. In addition, the support for traffic offloading by considering the collision rates in Wi-Fi is presented in [7, 8]. The aim is to prioritise services with higher data rate per session in the offloading procedure, since it leads to a lower collision rate and a higher network throughput.

References
[1] M. Chiosi, D. Clarke, P. Willis, A. Reid, J. Feger, M. Bugenhagen, W. Khan, M. Fargano, C. Cui, H. Deng, J. Benitez, U. Michel, H. Damker, K. Ogaki, and T. Matsuzaki, Network Function Virtualisation: An Introduction, Benefits, Enabler, Challenges, and Call for Action, European Telecommunications Standards Institute, Darmstadt, Germany, Oct. 2012 (http://portal.etsi.org/NFV/NFV_White_Paper.
pdf).
[2] S. Khatibi and L.M. Correia, “Modelling of Virtual Radio Resource Management for Cellular Heterogeneous Access Networks”, in Proc. of PIMRC’14 – IEEE 25th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications, Washington, DC, USA, Sep. 2014.
[3] Pérez-Romero,J., Gelabert,X. and Sallent,O., “Radio Resource Management for Heterogeneous Wireless Access,” in Hossain, E. (ed.), Heterogeneous Wireless Access Networks, Springer, New York, NY, US, 2009.
[4] Carapinha,J. and Parada,C. (eds.), “Reference Scenarios and Technical System Requirements Definition”, Deliverable D2.1, Mobile Cloud Networking Project, Apr. 2013, (http://www.mobile-cloud-networking.eu).
[5] S. Khatibi and L.M. Correia, “Virtualisation of Radio Resources – Next Step in Radio Accesss Virtualisation”, Technical Report TD-14-09014, COST IC1004, Ferrara, Italy, Feb. 2014.
[6] S. Khatibi, L.M. Correia, “A Model for Virtual Radio Resource Management in Virtual RAN“, submitted to EURASIP Journal on Wireless Communications and Networking
[7] S. Khatibi and L.M. Correia, “Modelling Virtual Radio Resource Management with Traffic Offloading Support”, Technical Report TD-14-11026, COST IC1004, Krakow, Poland, Sep. 2014.
[8] S. Khatibi and L.M. Correia, “Modelling Virtual Radio Resource Management with Traffic Offloading Support”, Submmited to EUCUN’15 – European Conference on Networks and Communication, Paris, France, June – July 2015.