SUMMARYTwo practical channel estimation schemes, the moment‐based first‐and‐second moments and the simplified maximum likelihood estimators, are proposed for the MIMO/on–off keying system with square envelope detection applied for wireless sensor networks. Here, both the channel response and noise power are estimated simultaneously in comparison with other approaches in which the noise quantity is assumed to be known at the receiver. Hence, the developed estimators are more practical than those estimators without noise power estimation. Simulation results reveal that the system with both proposed schemes can achieve an excellent BER performance in a wide signal‐to‐noise ratio (SNR) range. More specifically, we observed that the simplified maximum likelihood estimator performed as well as the moment‐based first‐and‐second moments estimator for SNR greater than 7.5 dB, yet had much more decline at low SNRs. This study also investigated the effects of the numbers of receive antennas and transmit antennas on the system performance. Simulation results demonstrated that, at the BER of 10−3, the 5 × 5 system had an improvement of 7 dB in SNR compared with the 3 × 3 system. Copyright © 2011 John Wiley & Sons, Ltd.

No abstract

In this paper, we propose a novel Kalman-alike scheme for error compensation in enhanced V -BLAST schemes. The propose method can mitigate the signal decision error caused by error propagation and improve system performance. In the method an optimal gain is obtained by minimizing the error covariance function. Then the decision signal is modified for achieving the goal of reducing BER performance. Simulations of enhanced V BLAST with the proposed scheme are performed, and results are compared with that with the iterative detect and decoding (lDD) approach. Our simulation results show that the Kalman-alike approach can improve the system performance and also has simple computation in comparison with the IDD approach.

Multiple input multiple output orthogonal frequency division multiplexing (MIMO-OFDM) is very sensitive to the carrier frequency offset (CFO), which not only affects the channel estimation but also the selection of a minimum error probability based precoder for peak-to-average power ratio (PAPR) reduction such that the system performance will degrade. This paper presents the joint scheme of CFO) and channel estimation in the MIMO-OFDM systems with the minimum error probability based precoder for reducing P APR.First, a minimum error probability based precoding matrix to reduce the PAPR of OFDM signals is designed. The CFO and CIR estimators based on the expectation maximization (EM) algorithm with an iterative scheme are developed to execute the estimations. To reduce computational complexity, a simpler coarse CFO estimator is carried out before the fine estimation so that a more accurate and faster convergent estimate can be attained. Simulation results show that the CFO estimator with a coarse initial estimate not only has an excellent performance at reduced computational complexity, but also makes the CIR estimation achieve almost ideal performance.

No abstract