A new detection algorithm for direct sequence spread spectrum (DSSS) signals using peak-to-average power ratio (PAPR) is proposed. The received signals are divided into several blocks. Then we calculate peak-to-average power ratio of each block and we use the average of peak-to-average power ratios as test statistics. The advantage of the proposed algorithm is no need to estimate noise variance. Additionally, it could confirm low computational complexity. The proposed algorithm shows advanced performance as more samples of received signal and blocks without SNR wall.