The optimization algorithms of pilot assignment and pilot power allocation to mitigate pilot contamination in massive MIMO systems

Massive MIMO system has been a very popular type of wireless communication technology recently. It plays an important role in 5G and beyond wireless communication systems. In Massive MIMO, base stations have multiple antenna elements to improve spectral and energy efficiency. Multiple antennas also provide a more efficient channel estimation for all users. Each user transmits a set of known symbols to base stations for channel estimation. These sequences are called uplink pilot sequences. Users in different cells can use the same pilot sequences as there is a limited number of pilot sequences to assign. It causes interference between users using basically the same pilots. This issue is called Pilot Contamination in wireless communication. In this study, we develop two innovative approaches to address the problem of Pilot Contamination. The main contributions of this study are summarized as follows:1-A new optimization-based pilot assignment algorithm Dynamic Optimization based Algorithm for Pilot Assignment Scheme (DOAPAS) is demonstrated, which is a combination of comprehensive search and dynamic programming algorithms optimized to reduce Pilot Contamination. This approach dynamically and optimally responds to changes in users' channels. This method finds the optimal user and pilot sequence matches in reduced the user and pilot sequence search space. It performs the optimization by targeting the increase in the cumulative average SINR value during the optimization. Thus, the algorithm output always results in more optimized pilot assignments. In the worst-case scenario (where the pilots are assigned perfectly effectively), the algorithm outputs keeping the existing pilot assignments. The simulation results indicate that the proposed method outperforms the recent popular analytical and deep learning-based pilot assignment approaches. The achieved improvement is even more prominent when all the pilots are fully used. 2- A new and highly efficient pilot power allocation algorithm Exponentially-weighted Based Dynamic Pilot Power Allocation (EWD-PPA) is proposed to increase the overall uplink achievable rate. The presented algorithm approach is based on the exponential function of the SINR difference to determine optimal power allocations. The basic principle of the EWD-PPA algorithm is based on the recognition that each user's Signal Interference plus Noise Ratio (SINR) differs from the average SINR per cell. This approach aims to polarize the SINR values with the intention of worsening the conditions of the lower channels while simultaneously improving the conditions of the superior channels. However, this strategy takes a fair stance rather than completely ignoring users with poor quality channels. This ensures that the allocation not only focuses on optimizing performance, but also recognizes the importance of maintaining a minimum level of service for all users by effectively promoting a balanced network performance. The simulation results clearly show that this strategy outperforms other common approaches in the literature, such as EPPA and WF-PPA methods.