Effective pilot assignment schemes in massive MIMO systems

The explosive growth of mobile applications and services over cellular networks poses new challenges to network operators in upgrading existing cellular networks in order to handle huge wireless data transmission. Nonetheless, the fifth generation (5G) launch holds tremendous potential to address these challenges. 5G uses a powerful massive multiple-input-multiple output (MIMO) system between the base station (BS) and the users, which promises high speed, low latency, and massive connectivity. While extremely useful, intercell interference has been identified as one of the major challenges of massive MIMO-enabled cellular systems. When the same pilot sets are reused across adjacent cells to estimate the channel state information (CSI), this causes a so-called pilot contamination problem that saturates the signal to interference plus noise ratio (SINR). Furthermore, this problem cannot be mitigated by increasing the number of serving antennas. In this thesis, an efficient pilot assignment scheme (EPA) is proposed to tackle the pilot contamination problem and consequently improve the uplink data rate of users in multi-cell massive MIMO systems, especially those who suffer from bad channel conditions. This was achieved by using the large-scale characteristics of the fading channel to minimize the amount of outgoing intercell interference at the target cell during the pilot assignment process. Then, a partial pilot assignment scheme (PPA) is developed to reduce the time computational complexity accompanied by the EPA scheme. Specifically, the pilot assignment process is carried out for specific users who are tagged according to comparing their large-scale channel fading coefficients to a specific threshold value. This scheme achieves a data rate that is close to that of the EPA scheme. Furthermore, an effective pilot reuse-PPA scheme (EPR-PPA) is introduced to efficiently mitigate the impact of interference. Not only is the uplink data rate greatly improved, but also the time computational complexity is further reduced. In the EPR-PPA scheme, two pilot sets are used in the network and the PPA algorithm is implemented in cells that cause low interference at the serving cell, which share the same pilot set. Simulation results showed that the proposed schemes outperformed both the existing smart pilot assignment (SPA) and conventional schemes. Herein, different linear receiving detectors are used in evaluating the performance of such proposed schemes. The obtained results ensure that the proposed schemes have significantly improved the system performance in terms of achievable uplink rate and cumulative distribution function (CDF) for both SINR and uplink rate. In particular, the improvements in the uplink data rate are roughly [12% – 78%], compared to the SPA schema. Moreover, the results of the evaluation explain the great improvements in the performance of poor SINR users, with the probability of achieving a higher SINR increasing almost by [20% – 37%], compared to the SPA assuming 64 antenna elements are equipped to the serving BS. The proposed schemes have also proved their high effectiveness and performance even in severe interference environments. In addition, the time computational complexity is reduced by approximately [52% – 72%] compared to the SPA.

Text 114922.pdf Download (858kB) Official URL or Download Paper: http://ethesis.upm.edu.my/id/eprint/18211

Actions (login required)

View Item