Algorithm for Deriving Optimal Frame Size in Passive RFID UHF Class1-Gen2 Standard Using Combinatorial Model Boundaries

Abstract

Original scientific paper MAC level efficiency constraints of the RFID Reader-Tag communication require proper frame size selection in order to achieve maximum throughput. To set optimal frame size, one needs to estimate the number of tags. Deriving optimal frame size results in reducing time for identification of large RFID tag populations, and thus increases throughput of RFID system.In accordance with the standard Dynamic Framed Slotted ALOHA (DFSA) protocol, we propose model for frame size adaptation using number of collisions (C) and successful slots (S) within the frame length L. We found that expected number of tags N , linearly depends on S in the frame of length L, for fixed C. Proposed model for estimation of the number of tags can be described with 4 characteristic points, which can be calculated from proposed combinatorial model. Algorithm implementation in the first step estimates number of tags, and then in the second step, using the estimates, set the optimal frame size.In order to compare proposed system performance with the other approaches, we use standard measures, System Efficiency and Collision Ratio. Simulation results show that our method provide better performance than the Qselection algorithm, specified within the UHF-Class1-Gen2 standard.Key words: Dynamic framed slotted ALOHA, Transmission control strategy, Q-selection algorithm, RFID identification, Tags estimation method (TEM)Algoritam za definiranje optimalne duljine okvira u pasivnoj UHF Class1-GEN2 tehnologiji korištenjem ograničenja iz kombinatornog modela. Ograničenja MAC sloja u komunikaciji RFIDčitač-tag zahtjevaju ispravno odabiranje veličine okvira s ciljem postizanja maksimalne propusnosti. Da bi se postavila optimalna duljina okvira, potrebno je procjeniti broj tagova koji se nalaze u područjučitanja. Odabir optimalne duljine okvira rezultira skraćivanjem vremena za identifikaciju velikog broja RFID tagova, te povećanja propusnosti sustava.U skladu s standardnim Dynamic Frame Slotted ALOHA (DFSA) protokolom, predlaže se model za adaptaciju duljine okvira korištenjem broja kolizijskih mjesta C i uspješno pročitanih mjesta S unutar okvira duljine L. Istraži-vanjem je zaključeno da očekivani broj tagova N linearno ovisi o S u okviru duljine L za fiksni C. Predstavljeni model za procjenu broja tagova se može opisati sa 4 karakteristične točke, koje se mogu izračunati iz predloženog kombinatornog modela. Implementacija algoritma u prvom koraku procjenjuje broj tagova, te u sljedećem koraku korištenjem procjene, postavlja optimalnu duljinu okvira.Da bi usporedili rezultate sa drugim pristupima, koristimo standardne mjere efikasnosti sustava i omjera kolizija. Rezultati simulacije pokazuju da naša metoda omogućava bolje performanse nego Q-Selection Algoritam, specificiran u UHF-Class1-Gen2 standardu.Ključne riječi: dynamic framed slotted ALOHA, strategije kontrole prijenosa, Q-selection algoritam, RFID identifikacija, procjena broja tagova (TEM)

Early Frame Break Policy for ALOHA-Based RFID Systems

“…The critique on existing tag estimate methods are provided in [12], where authors claim that formula (8) provides the best estimation because previously published works [3]- [6], [13] mainly contain errors due to their approaches which do not consider the slot type dependence. Considering authors simulations, and rationale described in the paper [12], we conclude that given formula provides the correct joint probability distribution of finding tags among slots.…”

“…ILCM is the ML tag estimate [8], based on (1) where is the frame size, stands for the number of ways to distribute tags to slots, and describes the number of ways in how to distribute the remaining tags to slots. Denominator stands for the total number of ways to distribute tags in slots.…”

A Reduced Complexity of Vahedi's Tag Estimation Method for DFSA

“…When there is a collision, the reader will not be able to distinguish these signals and identify each individual tag correctly. To overcome this collision problem, several anti-collision algorithms have been proposed in literature [5][6][7][8][9], some of which are presented in the following section in greater details.…”