Dynamic Framed Slotted Aloha (DFSA) based tag reading protocols rely on a tag estimation function to calculate the best frame size to use for a given tag set. An inaccurate estimate results in high identification delays and unnecessary energy wastage. This is particularly serious when DFSA based tag reading protocols are used in RFID-enhanced wireless sensor networks (WSNs), where nodes are battery constrained. To this end, this paper presents qualitative and quantitative analysis of five tag estimation functions using Monte Carlo simulations. We iteratively estimate a given set of tags and evaluate the mean error, variability, skew and Kurtosis of each function’s error distribution. Lastly, we compare and identify the most efficient tag estimation function that is suitable for RFID-enhanced WSNs.