Title :
Hybrid Receive Antenna Selection Scheme for V-BLAST
Author :
Lu, Di ; So, Daniel K C
Author_Institution :
Univ. of Manchester, Manchester
Abstract :
Recent research show that antenna selection schemes can reduce the hardware cost and power consumption of multiple antenna systems by using less radio frequency (RF) chains. The selected subset of antennas can achieve the diversity of a full system and a significant portion of the full capacity. In this paper, a new interpretation of the multiple-input multiple- output (MIMO) capacity formula is derived in term of mean square error (MSE) of different detection stages for V-BLAST specifically. Performance based and capacity based antenna selection schemes are compared in terms of computational complexity as well as their optimality in minimizing the system error probability. In addition, a hybrid scheme is proposed to combine the strength of both performance based and capacity based approaches. Simulation results show that performance based selection has the best performance yet with a higher complexity than the other two schemes for most combinations of antennas. The hybrid scheme, which has the least complexity among all the three schemes has a slightly worse block error rate (BLER) than the performance based selection, but better than the capacity based scheme. Robustness of selection schemes under channel estimation errors is investigated and verified via computer simulations.
Keywords :
MIMO communication; channel estimation; computational complexity; error statistics; mean square error methods; receiving antennas; V-BLAST; block error rate; channel estimation errors; computational complexity; error probability; hybrid receive antenna selection; mean square error; multiple antenna systems; multiple-input multiple-output capacity; radiofrequency chains; Capacity planning; Computational complexity; Costs; Energy consumption; Error probability; Hardware; MIMO; Mean square error methods; Radio frequency; Receiving antennas;
Conference_Titel :
Global Telecommunications Conference, 2007. GLOBECOM '07. IEEE
Conference_Location :
Washington, DC
Print_ISBN :
978-1-4244-1042-2
Electronic_ISBN :
978-1-4244-1043-9
DOI :
10.1109/GLOCOM.2007.830