Power-Imbalance Sparse Code Research Based On Genetic Algorithm

Genetic algorithm is a commonly used heuristic optimization algorithm,which has strong global optimization capabilities and is often used in various engineering practical problems.With the explosive growth of mobile communication services,the future mobile communication system will face the challenge of massive connected devices.Sparse Code Multiple Access(SCMA)is a non-orthogonal multiple access scheme,can provide both lower latency and higher spectral efficiency.In this thesis,the power-imbalance codebook design based on genetic algorithm and the improved multi-user detection algorithm are investigated concretely.The main research contents and achievements are as follows:(1)A new type of power-imbalance codebook is proposed,which is different from the previous power-balance codebook design schemes.A new type of complex number domain mother constellation is obtained by introducing a rotation factor;on the other hand,a more flexible signature matrix is obtained by adjusting the position of the non-zero elements of the Latin structure.(2)The minimum Euclidean distance and the minimum product distance are the key performance indicators on Gaussian channel and Rayleigh fading channel,respectively.In the existing SCMA codebooks,the minimum Euclidean distance is the largest,but the minimum product distance is not necessarily the largest;the minimum product distance is the largest,but the minimum Euclidean distance is not necessarily the largest.In this thesis,the minimum Euclidean distance,the minimum product distance and the two-dimensional vector composed of them are respectively used as objective functions to establish three nonlinear optimization models.By means of genetic algorithm,the power-imbalance SCMA codebooks suitable for Gaussian channel,Rayleigh fading channel and both channels are obtained respectively.Numerical experimental results show that the bit error rate(BER)performance of the three proposed power-imbalance codebooks is significantly improved compared to the existing codebooks.(3)An improved multi-user detection algorithm based on power scheduling is proposed based on original Message Passing Algorithm(MPA).The power-imbalance between user codebooks is fully utilized,and its purpose is to adjust the order of user message update.Numerical results show that both BER performance and convergence speed are significantly improved compared to the parallel MPA scheme and series MPA scheme.