Massive Multiple Input Multiple Output (mMIMO) is a key technology for wireless communications. The concept behind the technology is using a large number of antennas at the base station which servers multiple user equipment simultaneously. The main advantage of this technology is that it provides excellent spectral and energy efficiency; however, at the same time, it poses significant computational demanding order to process the data. To overcome this problem various researchers have developed dedicated architectures such as three-angle complex rotation and triangular systolic array (TACR/TSA), low-complexity complex givens rotation (LC-CGR), tournament-based complex Givens rotation (T-CGR), and very large scale integration (VLSI)architectures to meet the required computational needs of the MIMOsystem. This thesis focuses on evaluating the performance of MIMOimplementation on heterogeneous many-core architecture Epiphany, how it fares against TACR/TSA, LC-CGR, T-CGR and VLSI architectures. The results obtained from the dedicated architectures TACR/TSA,LC-CGR, T-CGR, and VLSI, provide a base for comparison. To achieve this, Zero Forcing algorithm combined with Householder Transformations(HH) and Givens Rotations (GR) algorithms of QR decomposition techniques were used. The algorithms are parallelized and run simultaneously over the cores of Epiphany. The speedup for HH is12.19x with the timely execution of 114ms for the 16x16 matrix. The comparison of 16-cores Epiphany and the architectures mentioned in this thesis gives more information on why Epiphany does not currently meet the required standards to be used in MIMO systems. The future scope is to propose possible improvements in the Epiphany architecture to make it suitable for massive MIMO applications without compromising energy efficiency.