Online and distributed optimization framework for wireless analytics

摘要

A method, computer program product, and computer system directed to an online and distributed optimization framework for wireless analytics. A radio network controller determines a ranking for each of a plurality of received objects using a plurality of similarity graphs. The radio network controller extracts a common structure by collaborative filtering data associated with a plurality of user devices and the plurality of received objects. The common structure is analyzed to infer usage patterns within a time slot. The radio network controller stores a subset of the ranked objects of the plurality of received objects in response to the analysis.

主权项

1. A computer program product residing on a non-transitory computer readable medium having a plurality of instructions stored thereon which, when executed by a processor, cause the processor to perform operations comprising:
determining a ranking for each of a plurality of received objects using a plurality of similarity graphs for a plurality of base stations, wherein the plurality of received objects include data files received from at least one of the plurality of base stations, the ranking calculated byminf∈Rn⁢12⁢fT⁢Lf+M⁢∑i,j⁢ηi,j⁢∑l⁢(ζ⁢Dil∑k⁢Dkl+Dil′∑k⁢Dkl′)⁢Wi⁢Cil-(ζ⁢Djl∑k⁢Dkl+Djl′∑k⁢Dkl′)⁢Wj⁢Cjlsubject⁢⁢tofi-fj≥1-ηij,(1)nij≥0⁢⁢∀i,j(2) where each object Ii has a bandwidth Wi; Cil units of cost is associated with unit transfer of item Ii from the radio network controller to base station Bl; there is a premium demand D′il and a non-premium demand Dil for Ii at base station B1; ζ>1 represents a weight assigned to the premium demand; nij represents a corresponding Lagrangian coefficient L represents a Laplacian matrix of a graph; and M is a coefficient of a regulizer, which is introduced to avoid overfitting; extracting a common structure between a plurality of user devices and the plurality of received objects by collaborative filtering data associated with a plurality of user devices and the plurality of received objects; analyzing the common structure to infer usage patterns within a time slot; generating demand predictions for at least a subset of objects of the plurality of received objects across time slots for the plurality of base stations; storing the subset of the ranked objects of the plurality of received objects in response to the analysis and the demand predictions, including pre-fetching the subset of the ranked objects of the plurality of received objects into a local cache of a base station of the plurality of base stations based upon, at least in part, the demand predictions for the time slot of the base station to reduce response time to access the plurality of received objects; and evicting one or more pre-fetched ranked objects of the subset of ranked objects from the local cache of a base station of the plurality of base stations according to the demand predictions for the time slot of the base station.