Research On The Application Of Compressed Sensing In UWB Channel Estimation

The ultra-wide band impulse radio(IR-UWB)technology is a short-range high-speed wireless communication technology that has drawn much attention in recent years.It has the advantages of high data transmission rate,spectrum sharing with other wireless systems,high security,strong anti-interference ability,strong penetrability and simple structure.However,due to its extremely wide signal bandwidth,it requires very high sampling rate of the analog-to-digital converter(ADC)on the receiving end,making the design and implementation of the ADC very difficult.Compressed sensing theory provides a new idea to solve this problem.it is a new sampling theory: if the signal is sparse in a domain,then you can reconstruct the original signal through sampling points less than Nyquist sampling theorem required.The purpose of this paper is to study the application of compressed sensing theory in IR-UWB communication as well as its optimization and improvement,so that it can improve the performance of compressed sensing based UWB communication system and further improve its practicability.The specific content of the article is as follows:(1)The system model of IR-UWB channel estimation is elaborated,and an IR-UWB channel estimation scheme based on compressed sensing is introduced.Several key problems to be solved in this scheme are pointed out.(2)Three main methods of sparse representation of UWB signals are analyzed.Compared with the other two dictionaries,the multipath diversity dictionary has higher reconstruction accuracy and no channel samples are needed during its construction.So it is more suitable for channel estimation.(3)Study the optimization method of Gaussian measurement matrix when using multipath diversity dictionary.Firstly,the advantages and disadvantages of several current measurement matrix optimization schemes are introduced,and then the concept of the correlation coefficient of the measurement matrix is explained.An adaptive step-size method for measurement matrix iterative optimization is proposed.Then,the optimization model and the optimization algorithm of this paper are given.Simulation results show that the optimization algorithm proposed in this paper is superior to the current algorithms in the aspects of correlation coefficient,operation time and reconstruction accuracy,and this advantage is more obvious in UWB applications.The new algorithm uses Barzilai-Borwen method and Armijo criterion to calculate and control the iteration step.The simulation results show that in the IR-UWB channel estimation based on compressed sensing,the new algorithm outperforms several existing algorithms in terms of correlation coefficient,computation time,and reconstruction accuracy.The optimized Gaussian measurement matrix has better performance than the Bernoulli random matrix,partial Hadamard matrix,sparse random matrix,and other measurement matrices.(4)Analyze the particularity of IR-UWB signal reconstruction and improve the generalized orthogonal matching pursuit(GOMP)algorithm.The new algorithm tries to avoid selecting atoms with similar indexes in the same iteration to effectively reduce the possibility of atom misselection.In addition,the iterative stopping condition of the GOMP algorithm is also improved so that it can be used even when the sparsity is unknown.The simulation results show that compared with several other matching pursuit algorithms,the improved GOMP algorithm can perform well in both runtime and reconstruction accuracy.