Based on frame slotted ALOHA algorithm

a grouped adaptive allocating slots (GAAS) anti-collision algorithm is presented to solve the problem of collision between the reader and multi-tag in radio frequency identification (RFID) system. First

the reader needs to obtain the time slots chosen randomly by tags and send the results to each tag; then the tags rectify the time according to the instruction; moreover

the reader skips free and collision time slots

and adaptively distributes valid ones; finally

the tags are quickly recognized in GAAS. When the number of unidentified tags is very large

the tags are grouped and the frame sizes are adjusted dynamically to reduce the processing time. The simulation results show that GAAS has higher identification efficiency and stability

and lower cost of communication. Particularly

when the number of tags is over 1000

the throughput rate still maintains above 71%. Compared with the framed slotted ALOHA-256 algorithm and the grouped dynamic framed slotted ALOHA algorithm

the proposed algorithm enhances the system efficiency by 300% and 97.2% respectively.