Markov Chain Monte Carlo Methods for a Low Complexity LTE-Advanced Joint Detector

Start Page
1
End Page
5
Citation
IEEE 87th Vehicular Technology Conference, VTC Spring 2018: 1-5 (2018)
Contribution to Conference
IEEE 87th Vehicular Technology Conference, VTC Spring 2018  
Publisher DOI
10.1109/VTCSpring.2018.8417572
Scopus ID
2-s2.0-85050967387
Publisher
IEEE
ISBN of container
978-1-5386-6355-4
|978-1-5386-6354-7
978-1-5386-6356-1
To meet the goal of tenfold increase in spectral efficiency, interference cancellation and multiuser detection are expected to be important tasks of fifth-generation (5G) radio access systems. Both tasks can be realized by joint detection algorithms. However, joint detection algorithms such as maximum likelihood (ML) detection have a high computational complexity. Previous works have shown that joint detectors based on Markov chain Monte Carlo (MCMC) methods can achieve similar results compared to ML detection with a large reduction in the computational complexity for systems with a large number of streams or users. The purpose of this work is to present a MIMO joint detector based on MCMC methods and evaluate it within the constraints of LTE Advanced (LTE-A), namely, using at most 8 transmit antennas and 64-QAM modulation. The evaluation is done separately from the channel decoder. Moreover, the complexity of the presented algorithm is compared to the one of an ML detector. The results show that the proposed MIMO detector offers a similar detection error rate compared to an ML detector. Furthermore, it was observed that the complexity reduction is significant for systems with more than six transmit antennas.