Zonal fairness in wireless infrastructure

Fairness models can be roughly categorized into two groups: proportional fairness based models balancing the relative values of allocated goods, and maxmin fairness based models with reference to absolute valuations. In a specific problem domain it appears that only one fairness model can be used. Here a situation is studied where the users of a wireless infrastructure have in fact conflicting interests with regard to the chosen fairness model. A fair bargaining approach to balance those conflicting interests is presented based on the specification of a binary relation that combines product and lexicographic maxmin and whose maximum set gives the desired solutions to a mixed zonal fairness problem. The application of the approach is presented by analysis of an optimal distinction between users near or far to a base station, and a peak in the total performance related to that maximum could be identified that surpasses both, the allocation by maxmin fairness as well as the allocation by proportional fairness. In addition the approach offers a rich potential of variations that allows application to many other network domain models.