Finding frequent patterns from databases has been the most time consuming process of the association rule mining. Till date, a large number of algorithms have been proposed in the area of frequent pattern generation. However, all of these algorithms produce output only at the completion and are not amenable to the real-time need. The need for real-time frequent pattern mining for online tasks and real-time decision-making is increasingly being felt. In this paper, we describe BDFS(b), an algorithm to perform real-time frequent pattern mining using limited computer memory. Empirical evaluations show that our algorithm can make a fair estimation of the probable frequent patterns and reaches some of the longest frequent patterns much faster than the existing algorithms.