为了改善现有RFID系统多标签识别过程中随标签增多而产生的识别速率下降问题,减少数据传输冗余和碰撞次数,在分析研究现有的动态二进制搜索算法的基础上,提出一种基于回溯的不间断轮询算法。该算法利用曼彻斯特编码的特性确定碰撞位置,根据碰撞位置前向搜索同时压入堆栈形成进一步搜索命令,回溯查询堆栈逐一识别相应标签。实验结果表明,该算法在减少数据传输量和搜索迭代次数上要明显优于现有经典二进制防碰撞算法。
In order to improve the rate of recognition multi-tags from existing RFID system, reduce data redundancy and collision frequency. An uninterrupted querying algorithm based on the existing dynamic binary search algorithm is proposed. Firstly this algorithm determined collision position by Manchester encoding, then forward search while the stack is formed further search command. Through optimize the search and response command. Finally backtracking stack to identify the labels one by one; Ex- perimental results show that the algorithm reduces the amount of data transfer and search iterations to be significantly better than existing classical binary anti-collision algorithms.