Designing the time domain training sequences is very critical in multi carrier transmission which degrades the performance as it is contaminated by different blocks in different cells. To improve the spectral efficiency and high accuracy, MU-MIMO needs the sensing matrix to be reduced by using the training sequence design and optimization. Integrating the training sequence design and sparse channel estimation improves the capacity of the system. The capacity can be enhanced by reducing the bit error rate. The system capacity for multi user- multi-input and multi output (MU-MIMO) is studied by proper channel estimation with compressed sensing model. The design and optimization of training sequence is analysed for MU-MIMO model using auto coherence and block coherence matrices. The block coherence matrix is optimized using cuckoo algorithm for obtaining lower coherence value for different sparsity values. The performance improvement in terms of signal to noise ratio is 1 dB for single user- multi-input and multi output (SU-MIMO) using genetic algorithm and the performance of MU-MIMO is observed to be 0.93 dB using cuckoo algorithm.

设计时域训练序列在多载波传输中非常关键，这会降低性能，因为它被不同小区中的不同块污染。为了提高频谱效率和高精度，MU-MIMO需要通过使用训练序列设计和优化来减少感知矩阵。将训练序列设计和稀疏信道估计相结合，提高了系统的容量。可以通过降低误码率来提高容量。多用户多输入多输出 (MU-MIMO) 的系统容量通过使用压缩感知模型的适当信道估计来研究。使用自相干矩阵和块相干矩阵分析了MU-MIMO模型的训练序列的设计和优化。块相干矩阵使用布谷鸟算法进行优化，以获得不同稀疏值的较低相干值。对于使用遗传算法的单用户多输入多输出 (SU-MIMO)，信噪比方面的性能改进为 1 dB，使用布谷鸟算法观察到 MU-MIMO 的性能为 0.93 dB。