Linear Distinguishing Attack on Shannon Algorithm
对Shannon算法的线性区分攻击

A new distinguishing attack is presented on Shannon algorithm. The distinguish attack is built by using linear approximations of both the non-linear feedback shift register and the non-linear filtration function, and 32 distinguishers are derived which the bias is 2-28. Therefore, the Shannon algorithm is distinguishable from truly random cipher after observing 252 keystreams words on average.