Discrete-Rate Adaptive Modulation with Variable Threshold for Distributed Antenna System in the Presence of Imperfect CSI

• Published in: China communications Vol. 11; no. 1; pp. 31 - 39
• Main Authors: Yu, Xiangbin, Tan, Wenting, Wu, Binbin, Li, Yang
• Format: Journal Article
• Language: English
• Published: China Institute of Communications 2014; National Mobile Communications Research Laboratory, Southeast University, Nanjing 210096, China%Key Laboratory of Radar Imaging and Microwave Photonics, Ministry of Education, College of Electronic and Information Engineering,Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China; Key Laboratory of Radar Imaging and Microwave Photonics, Ministry of Education, College of Electronic and Information Engineering,Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
• Subjects: adaptive modulation; Bit error rate; CSI; distributed antenna system; Estimation error; Fading; imperfect estimation; Modulation; Signal to noise ratio; Switches; variable thresholds; 不完全; 分布式天线系统; 可变; 平均误码率; 离散; 自适应调制; 速率; imperfect estimation; adaptive modulation; distributed antenna system; variable thresholds
• ISSN: 1673-5447
• DOI: 10.1109/CC.2014.7022523

Discrete-rate adaptive modulation （AM） scheme for distributed antenna system （DAS） with imperfect channel state information （CSI） is developed, and the corresponding performance is investigated in composite Rayleigh channel. Subject to target bit error rate （BER） constraint, an improved fixed switching threshold （FST） for the AM scheme is presented by means of tightly-approximate BER expression, and it can avoid the performance loss fxom conventional FST. Based on the imperfect CSI, the variable switching threshold （VST） is derived by utilizing the maximum a posteriori method. This VST includes the improved FST as a special case, and may lower the impact of estimation error on the performance. By the switching thresholds, the spectrum efficiency （SE） and average BER of the system are respectively derived, and resulting closed- form expressions are attained. With these expressions, the system performance can be effectively evaluated. Simulation results show that the derived theoretical SE and BER can match the simulations well. Moreover, the AM with the presented FST has higher SE than that with the conventional one, and the AM with VST can tolerate the large estimation error while maintaining the target BER.