Evaluation of signalling loads in a cognitive network management architecture Kostas Tsagkaris, * † Maria Akezidou, Aristi Galani and Panagiotis Demestichas University of Piraeus, Piraeus 18534, Greece SUMMARY Future networks will need to accommodate a significantly augmented user demand, mainly stemming from the wireless and mobile domains. In general, the emerging radio landscape will comprise multiple, collaborating radio access networks (RANs) able to operate a plethora of diverse radio access technologies (RATs), variant types of mobile terminals (MTs), with the ability to choose among various supported RANs/RATs and, in addition, both devices and networks with dynamic spectrum access capabilities that allow the sharing and/or optimization of spectrum usage among different systems. The above will stress network operators for developing mechanisms to confront the challenges and to leverage the opportunities posed by such a versatile radio environment. In particular, the situation calls for adaptive and flexible management paradigms that are able to dynamically manage network elements and terminals, thus ensuring the great availability and efficient usage of spectrum and other radio resources. Framed within the above, this paper considers a cognitive network management architecture, which is destined for optimized management of future wireless networks operating in versatile radio environments, and presents a performance evaluation methodology, which was set up for measuring the signalling loads that the operation of the architecture will bring to the managed network. The methodology is analytically described, and useful results with respect to the signalling load produced for management signalling purposes in an indicative scenario are presented and analysed. Copyright © 2011 John Wiley & Sons, Ltd. Received 17 September 2010; Revised 28 July 2011; Accepted 31 July 2011 1. INTRODUCTION Future wireless networks will exhibit great levels of heterogeneity, consisting of multiple radio access networks (RANs) with the ability to operate diverse types of radio access technologies (RATs), includ- ing the Long Term Evolution (LTE) system of the 3 rd Generation Partnership Project (3GPP), WiMAX, mobile 2 G/3 G networks and wireless local area networks (WLANs). This operation may change dynamically over time when assuming the deployment of reconfigurable network elements, referred to as flexible base stations (FBSs) [1]. This emerging radio landscape will be complemented by multimode and/or multi-homing enabled mobile terminals (MTs) requiring advanced applications and services for all areas of human activities and content provision. In this context, network operators will struggle to find the appropriate configuration of their FBSs, especially in terms of operating RAT and spectrum and in order to tackle the limited available spectrum as well as the unstable traffic load. At the same time, MTs will strive to find the most suitable RAN/RAT to connect to and spectrum to use, in order to exploit the disposable radio resources in the best possible technological and econom- ical way. On the other hand, the introduction of a flexible spectrum management concept [2,3] and the adoption of cognitive capabilities to both networks and terminals seems to be an efficacious response to this accrued complexity as well as a powerful enabler for the accomplishment of both users’ and operators’ goals. *Correspondence to: Kostas Tsagkaris, Department of Digital Systems, University of Piraeus, 80 Karaoli and Dimitriou St., Piraeus, 18534, Greece. † E-mail: ktsagk@unipi.gr INTERNATIONAL JOURNAL OF NETWORK MANAGEMENT Int. J. Network Mgmt (2011) Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/nem.803 Copyright © 2011 John Wiley & Sons, Ltd.