Operators are deploying small cells to complement their macro cells to increase the coveragernand capacity of their mobile networks. Backhauling the small cells trafï¬c to therncore network is a challenge to operators as they are usually placed in difï¬cult to accessrnlocations. To overcome this challenge, wireless solutions have been proposed as a costeffectivernalternative to wired solutions. There are, however, many wireless technologies tornrealize this: point-to-point/point-to-multipoint, line-of-sight/non-line-of-sight (NLOS) andrnmillimeter wave/microwave bands. This research studies practical wireless conï¬gurations ofrnthis technologies, and proposes a planning method an operator can determine which technologyrnto use on each link of the small cell backhaul network so that the total cost of ownershiprn(TCO) is minimized. To this end, a mixed integer linear programming (MILP) is formulatedrnwith a range of constraints for wireless technologies and backhaul requirements.rnTo demonstrate the effectiveness of the method, the MILP was used to plan a small cellrnbackhaul network for different scenarios, in a relatively built up area of Addis Ababa. Thernscenarios were created based on different small cell demands, and on whether NLOS linksrnare allowed in the network. The path loss values were predicted by ray tracing to determinernthe feasibility of a link conï¬guration of each technology between nodes of the backhaulrnnetwork. The results showed the reduction of TCO by inclusion of more technologies as canrnbe seen with the need for more new aggregation nodes when NLOS was not considered. Byrnaccepting an optimality gap of 5 %, the computation time was also reduced to less than twornhours which is acceptable for the purpose of planning. Using this method, an operator canrnplan a cost-effective backhaul from available practical wireless technologies.