Aiming at the security analysis problem of KATAN48 algorithm， a conditional differential cryptanalysis method of KATAN48 algorithm based on neural distinguishers was proposed. First， the basic principle of multiple output differences neural distinguishers was studied and applied to KATAN48 algorithm. According to the data format of KATAN48 algorithm， the input format and hyperparameters of the deep residual neural network were adjusted. Then， the Mixed-Integer Linear Programming （MILP） model of KATAN48 algorithm was established to search the prepended differential paths and the corresponding constraint conditions. At last， using the multiple output differences neural distinguishers， at most 80-round of the practical key recovery attack results of KATAN48 algorithm were given. Experimental results show that in the single key setting， the number of practical attack rounds of KATAN48 algorithm is increased by 10 rounds， the number of recoverable key bits of KATAN48 algorithm is increased by 22 bit and the data complexity and time complexity of KATAN48 algorithm are reduced from 2³⁴ and 2³⁴ to 2^16.39 and 2^19.68 respectively. Compared to the previous practical attack at the single-key setting， the proposed method can effectively increase the number of attack rounds and recoverable key bits， and reduces the computational complexity of attack.

Aiming at the problem of multiple types of application?layer Distributed Denial of Service （DDoS） attacks， which are difficult to detect simultaneously， an application?layer DDoS attack detection method based on integrated learning was proposed to detect multiple types of application?layer DDoS attacks. Firstly， by using the dataset generation module， the normal and attack traffic was simulated， the corresponding feature information was filtered and extracted， and 47?dimensional feature information characterized Challenge Collapsar （CC）， HTTP Flood， HTTP Post and HTTP Get attacks were generated. Secondly， by using the offline training module， the effective features were processed and input into the integrated Stacking detection model for training， thereby obtaining a detection model that can detect multiple types of application?layer DDoS attacks. Finally， by using the online detection module， the specific traffic type of the traffic to be detected was judged through deploying the detection model online. Experimental results show that compared with the classification models constructed by Bagging，Adaboost and XGBoost，the Stacking integretion model improves the accuracy by 0. 18 percentage points，0. 21 percentage points and 0. 19 percentage points respectively，and has the malicious traffic detection rate reached 98% under the optimal time window. It can be seen that the proposed method has good performance in detecting multi-type application-layer DDoS attacks.

Application of the basic ant colony algorithm for anti-ship missile path planning problem has such shortcomings as slow convergence speed, long computation time, and easily falling into local optimum. Concerning these shortcomings, roulette selection strategy, elite strategy and path optimization strategy were adopted on traditional ant colony algorithm to optimize it, and the optimization algorithm was applied in anti-ship missile route planning. At the same time, by means of limiting the feasible course of anti-ship missile, the maximum search range for route planning was reduced. Simulation results show that the anti-ship missile route planning based on improved ant colony algorithm not only shortens the optimal route length but also speeds up the convergence rate of the optimal route search process.