Heuristic approach for forecast scheduling

Hind Zaaraoui (1) , Zwi Altman (1) , Sana Ben Jemaa (2) , Eitan Altman (3) , Tania Jimenez (4)

1 Orange Labs [Issy les Moulineaux]
2 Orange Labs [Paris]
3 NEO - Network Engineering and Operations
4 LIA - Laboratoire Informatique d'Avignon

Forecast Scheduling (FS) is a scheduling concept that utilizes rate prediction along the users' trajectories in order to optimize the scheduler allocation. The rate prediction is based on Signal to Interference plus Noise Ratio (SINR) or rate maps provided by a Radio Environment Map (REM). The FS has been formulated as a convex optimization problem namely the maximization of an α−fair utility function of the cumulated rates of the users along their trajectories [1]. This paper proposes a fast heuristic for the FS problem based on two FS users' scheduling. Furthermore, it is shown that in the case of two users, the FS problem can be solved analytically, making the heuristic computationally very efficient. Numerical results illustrate the throughput gain brought about by the scheduling solution.

Mots clés: Forecast scheduler; high mobility; alpha-fair; geo-localized measurements; Radio Environment Maps

Domaines: Réseaux et télécommunications [cs.NI]

Hind Zaaraoui, Zwi Altman, Sana Ben Jemaa, Eitan Altman, Tania Jimenez. Heuristic approach for forecast scheduling. IWSON 2018 - 7th International Workshop on Self-Organizing Networks, Apr 2018, Barcelona, Spain. pp.1-6. ⟨hal-01705829⟩