Wideband Wireless Transmitter Identification Based On Hammerstein-Wiener Model

With the development of technology,the wireless network faces an increasingly serious security threat.One of the most serious threats is that a malicious user impersonates or pretends to be a legitimate user accessing a wireless network.In the wireless network,the existence of non-cooperative users will lead to serious security risks.In this thesis,a Hammerstein-Wiener model is adopted for modeling the wideband wireless transmitters,and the identification of the same type of wideband wireless transmitters is studied from the perspective of system parameter identification.Combined with the idea of context awarenesss,and allowed the system to make different choices in different environments.The main work of this thesis is as follows:1.The wideband wireless transmitter identification,system identification and the research status of context awareness are introduced.2.The wideband wireless transmitter is modeled as a Hammerstein-Wiener model.For the identification of the model,the mesh adaptive direct search(MADS)algorithm,the adaptive niche genetic algorithm and the step-by-step iterative algorithm are introduced respectively.3.Because the adaptive niche genetic algorithm's local search ability is limited,the identification accuracy is not ideal,the gradient optimization is used to reduce the time complexity in the population initialization.And the recognition degree of eel algorithm and the approach of learning are introduced.In the paper an improved algorithm is proposed.Which makes up the deficiency of local optimization of genetic algorithm(GA)and makes the GA achieve the balance between global optimization and local optimization,which increases the accuracy of parameter identification and improves the recognition rate of the transmitter.The algorithm is used to identify the wideband wireless transmitter and compare it with the step-by-step iterative algorithm and the adaptive niche genetic algorithm.4.According to the different environment of mobile receiver,using the corresponding strategy to identify the transmitter.A wideband wireless transmitter identification technique based on context awarenesss is proposed.Predicting the position state of the mobile receiver based on the Hidden Markov Model,in which the parameters of the model are optimized by Baum-Welch training,the Viterbi decoding can get the mobile receiver position state.Through the method proposed an optimal identification scheme corresponding to the predicted motion profile of the wireless receiver can be obtained.