Cell-Free massive multiple-input multiple-output (CF-mMIMO) systems have emerged as a promising architecture for next-generation wireless networks because they provide uniform coverage, high spectral efficiency, and improved user fairness by eliminating conventional cell boundaries. By enabling cooperation among a large number of geographically distributed access points (APs), CF-mMIMO systems exploit macro-diversity and spatial multiplexing gains. However, their practical deployment is constrained by several critical challenges, including pilot contamination during channel estimation, high computational complexity, fronthaul signalling limitations, and the need to support user mobility.
This thesis proposes an integrated framework for contamination-aware channel estimation and distributed precoding in CF-mMIMO systems with mobility support. A multi-stage contamination mitigation strategy is developed that combines spatial pilot assignment, successive interference cancellation, and contamination-aware linear minimum mean-square error (LMMSE) weighting. This approach explicitly accounts for the pilot-contamination structure and significantly improves channel-estimation accuracy compared with conventional LMMSE estimation.
To enable scalable and practical downlink transmission, a distributed minimum mean square error (MMSE) precoding scheme with signal-to-noise ratio (SNR)-adaptive regularization is proposed. The precoding is performed locally at each AP using partial channel state information, reducing fronthaul requirements while achieving performance comparable to centralized MMSE precoding. In addition, power allocation strategies are investigated to balance sum spectral efficiency, energy efficiency, and user fairness under per-AP power constraints.
User mobility is addressed through a temporal channel-correlation model and a dynamic access point selection and handover mechanism. The proposed mobility framework maintains service continuity while limiting handover frequency and load imbalance across APs. System performance is evaluated through extensive Monte Carlo simulations under realistic propagation conditions, including dual-band operation at sub-6 GHz and millimeter-wave frequencies.
Simulation results show that the proposed framework achieves substantial pilot-contamination reduction, near-centralized precoding performance, and stable operation under user mobility. These findings indicate that contamination-aware estimation combined with distributed MMSE precoding offers an effective and practical approach for deploying CF-mMIMO systems in future wireless networks.
| Date | 2 Apr 2026 |
|---|
| Original language | American English |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Michel Kadoch (Supervisor) |
|---|
Shajari Kohan, M. (Author),
Kadoch (Supervisor),
2 Apr 2026Student thesis: Master's thesis › Master in Engineering: Electrical Engineering