Comparison of iterative signal detection algorithms in massive MİMO systems

Nowadays, with the increase of the number of users making instant communication and the increase of the number communicating with each other has led to an increase in the density of the spectrum used for communication systems. This density caused latency, increase interference with different systems and on the power consumption. To solve these problems, new solutions are proposed for especially 5G and beyond. Multiple Input Multiple Output (MIMO) which is one of these ideas is examined in this scope of this thesis. BER versus SNR performances of ZF, MMSE, ML, Neumann, GS, CG, OCD and ADMIN algorithms are examined by using BPSK, QPSK, 16QAM, 16APSK and 64QAM modulations. The Channel State Information (CSI) is assumed to be known exactly in literature. However, it is not possible to know CSI exactly due to system's noises, transmission area etc. while making implementation. Besides, the correlation occurs between the antennas due to the antenna placements. In this scope of thesis, six different situations are examined which are modulation types, CSI, system dimension ratio, iteration number, algorithms' performances and running time. The results obtained from the simulation show that ADMIN has the best algorithm performance and its' best performance is observed the situation that BPSK is used, CSI is known exactly, system dimension ratio is least 4 and iteration number is least 5. In addition, the channels being related with each other and/or known with noise makes it difficult to detect the signals in all system sizes and in all used algorithms.