| Summary: | 博士 === 國立臺灣科技大學 === 資訊管理系 === 101 === In Radio Frequency Identification (RFID) systems, the reader recognizes tags through communication over a shared wireless channel. When multiple tags send their IDs simultaneously, their signals collide, causing the identification delay. Thus, tag anti-collision algorithms have long been an important issue in RFID systems. Recently, some researchers have considered that the RFID system performs on the scenarios of re-identification, capture effect and adopting bit tracking technology.
In re-identification, the reader repeatedly identifies the same staying tags which stayed in reader’s range. Existing anti-collision algorithms, e.g. adaptive binary splitting (ABS) and single resolution blocking (SRB), can rapidly identify the staying tags by remembering the order in which the tags were recognized. We propose adaptive frame ABS (AFA) and dynamic blocking adaptive binary splitting (DBA) to cope with re-identification. AFA utilizes the dynamic regulation that estimates the numbers of staying tags and newly-arriving tags and optimally adjusts the numbers of slots allocated for them to reduce idle slots when many recognized tags leave and reduce collision slots when many newly-arriving tags enter. Following the regulation, multiple staying tags may share the same slot and cause a collision among them. Thus, an efficient ordering splitting is designed to deterministically split the collided staying tags according to the order in which they were recognized. DBA based on the blocking mechanism, which prevents the newly-arriving tags from colliding with the staying tags. Moreover, DBA utilizes a dynamic condensation technique to reduce the number of idle slots produced when recognized tags leave. Following the condensation process, multiple staying tags may be required to share the same slot, and thus may cause collisions among them. Accordingly, an efficient ordering binary tree mechanism is proposed to split the collided tags deterministically.
When capture effect occurs, the reader decodes a tag ID even when multiple tags simultaneously transmit signals, and causing some tags, which transmitted their IDs but not be identified in this slot, do not transmit IDs again in the current identification. By referring generalized query tree (GQT), we propose generalized binary tree (GBT) that separates the identification process into several binary tree (BT) cycles to solve the capture effect. Unrecognized tags, hidden by the capture effect in a BT cycle, will join subsequent cycles repeatedly until there are all identified.
When a RFID system adopts bit tracking technology, the reader to detect the locations of collided bits in a collision slot, collision tracking tree algorithm (CTTA), enhanced anti-collision algorithm (EAA) and new enhanced anti-collision algorithm (NEAA) all adopt this technology. We also propose optimal query tracking tree protocol (OQTT) and optimal binary tracking tree protocol (OBTT) to improve the performance of tag identification by utilizing the bit tracking technology. First, they adopt bit estimation to estimates the number of tags based on the locations of collided bits, and then they use optimal partition to determine the number of the initial sets. Finally, OQTT uses query tracking tree and OBTT uses binary tracking tree to splits a set of collided tags into two subsets using the first collided bit.
This dissertation formally analyzes the efficiency AFA, DBA, GBT, OQTT and OBTT, The analytical and simulation results show that our algorithms are better than previous algorithms: AFA outperforms ABS, DBA outperforms SRB, GBT outperforms GQT, OQTT and OBTT outperform than CTTA, EAA and NEAA. Thus, we can conclude these algorithms are fast, and efficiency.
|
|---|
