Energy Efficient Anti-Collision Algorithm for the RFID Networks

Murukesan Loganathan, Thennarasan Sabapathy, Mohamed Elshaikh, Muzammil Jusoh, R. Badlishah Ahmad, Mohamed Nasrun Osman, Rosemizi Abd Rahim, Mohd Ilman Jais

Abstract


Energy efficiency is crucial for radio frequency identification (RFID) systems as the readers are often battery operated. The main source of the energy wastage is the collision which happens when tags access the communication medium at the same time. Thus, an efficient anti-collision protocol could minimize the energy wastage and prolong the lifetime of the RFID systems. In this regard, EPCGlobal-Class1-Generation2 (EPC-C1G2) protocol is currently being used in the commercial RFID readers to provide fast tag identification through efficient collision arbitration using the Q algorithm. However, this protocol requires a lot of control message overheads for its operation. It is a known fact that the communication subsystem requires much higher energy consumption for its operation as compared to the computational subsystem. Therefore, a protocol with efficient collision arbitration capability and low control message overhead can be a suitable alternative for the commercial readers. In this regard, reinforcement learning based anti-collision protocol (RL-DFSA) is proposed to provide better time system efficiency while being energy efficient through the minimization of control message overheads. The proposed RL-DFSA was evaluated through extensive simulations and compared with the variants of EPC-Class 1 Generation 2 algorithms that are currently being used in the commercial readers. The results show conclusively that the proposed RL-DFSA performs identically to the very efficient EPC-C1G2 protocol in terms of time system efficiency but readily outperforms the compared protocol in the number of control message overhead required for the operation. In fact, RL-DFSA requires an order of magnitude lesser control message overheads as compared to the EPC-C1G2 protocol.

Keywords


Collision avoidance, Dynamic frame slotted Aloha, EPC-C1G2, Q-learning, Reinforcement learning


Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.

Bulletin of EEI Stats