Slot-by-slot minimum squared error estimator for tag populations in FSA protocols

Framed Slotted ALOHA (FSA) is an anti-collision technique in the currently most popular standard of RF-ID systems in the UHF domain, the ISO/IEC CD 18000-6. The number of tags present in the field is a crucial parameter to apply FSA algorithms in an optimal manner to achieve the highest detection rate possible. Therefore, a lot of effort is spent on the estimation of this parameter and a range of different estimation techniques exist. One of the most promising techniques is the minimum squared error (MSE) estimator for expectation values formulated by Vogt et al. for completely observed frames. This technique has not yet been adapted to an in-frame adjustment of the frame size (i.e. without quitting the interrogation round), as it is favoured in the aforementioned standard. In this work the MSE is reformulated to work on slot-by-slot basis and an efficient algorithm is proposed to exploit the on-the-fly adjustment of the frame size. The results are compared to other common estimation techniques for the tag population with respect to cumulated estimation error and detection rate. Furthermore, it is demonstrated that the performance is approximating the theoretical bound obtained by the maximum likelihood estimator.