RSSI (Received Signal Strength Indication) localization techniques using Wi-Fi presents substantial advantages compared to others. They are light weight both in terms of computation and energy consumption. RSSI localization techniques are mostly used indoors, where many APs (Access Points) are present and no GPS is available. Recently, APs are getting deployed outdoors as well, and urban canyon phenomenon degrades the capability of GPS localization even in outdoor environments, which makes RSSI localization techniques attractive as an outdoor localization solution as well. The downside of RSSI localization is that it is polarized, which means it has either high performance and economic cost or low cost and poor accuracy. Both cases are inadequate for a general deployment; high-cost algorithms can only be deployed in heavily populated area for cost feasibility and the accuracy of low-cost algorithms is nowhere near credible. In this paper, we propose a range-based RSSI localization algorithm that has reasonable accuracy yet has very low cost. The proposed algorithm consists of DB-assistance, ration base algorithm, and an elementary machine learning algorithm. This helps achieving the qualities that can provide a feasible RSSI localization solution that can be employed in a much wider area.