Narrow Band Internet of Things (NB-IoT) has many kinds of secondary synchronization signals, which need a lot of space to store. Therefore, the requirement of low power consumption is difficult to be satisfied. In order to solve this problem, we propose a low complexity cell Identity (ID) detection algorithm. Firstly, the algorithm splits the secondary synchronization signal, and reduces the multiplication operation by using the law of phase offset changing with frame number. Then it uses the symmetry of Zadoff-Chu (ZC) sequence in the secondary synchronization signal to calculate the autocorrelation value of the received signal. Combined with the cross-correlation between the local signal and the received signal, the secondary synchronization signal can be detected and the cell ID number can be calculated. By comparing with other algorithms, the proposed algorithm can greatly reduce the complexity of the algorithm while maintaining the acceptable performance.