A Study of Lattice Reduction Detection Techniques for LTE Systems
Liping Wang · Lund University Publications Student Papers (Lund University) · 2016
MIMO (Multiple Transmit and Multiple Receive) antenna techniques are widely used in the most recent wireless communication standards. For example, LTE (Long-Term Evolution), WiMAX (Worldwide Interoperability for Microwave Ac- cess) and IEEE 802.11n have recently been rolled out across the world. In any communication system, the ML (Maximum Likelihood) receiver provides optimal error rate performance, but it turns out to be difficult to implement in system with 4 antennas or more. This thesis studies ways to approach the optimal performance for MIMO chan- nels with M-QAM (4-QAM, 16-QAM and 64-QAM) in Rayleigh MIMO channels with low complexity. While linear detectors like ZF (Zero Forcing) and MMSE (Minimum Mean Square Error) have very low computational complexity, they suffer from noise enhancement and ISI (Inter Stream Interference) respectively. LR (Lattice Reduction) coupled with ZF or MMSE will give us near optimal results close to that of the ML receiver. Same diversity order like ML is also found. Basically, LR is aiming to find a basis of the channel matrix as orthogonal as possible. A basis is composed by a set of linearly independent vectors. A big improvement can be achieved by replacing linear detectors with LR techniques. Correlated channels will be studied for comparison reasons. To evaluate the impact of channel correlation, Kronecker Model with three different correlation factors will be discussed. In future work section, some cutting edge algorithms will be mentioned, for example: BKZ (Block Korkine Zolotarev) Algorithm.