In the context of dynamic spectrum access, spectrum sharing among multiple operators has recently emerged as a promising paradigm to improve the efficiency of resource usage. Several theoretical evaluations have proven the benefits offered by pooling the available frequencies so as to tune the capacity offered by the operators according to their different needs, especially the service demands from their users. However, practical aspects concerning the application of sharing techniques are rarely studied, and deserve more detailed investigations. This paper aims at tackling this problem, in particular investigating the impact of asymmetries and dynamics of the user demands on the implementation of spectrum sharing techniques and the resulting performance, especially in terms of fairness among the users, which seems to be often neglected by many studies. We show that in variable traffic conditions, a constantly monitored and updated sharing of frequency bands performs much better than a static allocation simply based on average traffic loads. However, it is possible to choose the update rate of the spectrum allocation so that it does not represent a heavy computational and signaling burden, while retaining most of the improvements brought by the spectrum sharing paradigm.
Fairness evaluation of practical spectrum sharing techniques in LTE networks
GUIDOLIN, FRANCESCO;BADIA, LEONARDO;ZORZI, MICHELE
2014
Abstract
In the context of dynamic spectrum access, spectrum sharing among multiple operators has recently emerged as a promising paradigm to improve the efficiency of resource usage. Several theoretical evaluations have proven the benefits offered by pooling the available frequencies so as to tune the capacity offered by the operators according to their different needs, especially the service demands from their users. However, practical aspects concerning the application of sharing techniques are rarely studied, and deserve more detailed investigations. This paper aims at tackling this problem, in particular investigating the impact of asymmetries and dynamics of the user demands on the implementation of spectrum sharing techniques and the resulting performance, especially in terms of fairness among the users, which seems to be often neglected by many studies. We show that in variable traffic conditions, a constantly monitored and updated sharing of frequency bands performs much better than a static allocation simply based on average traffic loads. However, it is possible to choose the update rate of the spectrum allocation so that it does not represent a heavy computational and signaling burden, while retaining most of the improvements brought by the spectrum sharing paradigm.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.