Linearly time-varying channel estimation and training power allocation for OFDM/MIMO systems using superimposed training

Abstract

We address the problem of estimating the linearly time-varying (LTV) channel of orthogonal frequency division multiplexing (OFDM)/multiple-input multiple-output (MIMO) systems using superimposed training (ST). The LTV channel is modeled by truncated discrete Fourier bases. Based on this model, a two-step approach is adopted to estimate the LTV channel over multiple OFDM symbols. We also present performance analysis of the channel estimation and derive a closed-form expression for the channel estimation variances. It is shown that the estimation variances, unlike that of the conventional ST-based schemes, approach to a fixed lower-bound as the training length increases, which is directly proportional to information-pilot power ratios. For wireless communication systems with a limited transmission power, we optimize the ST power allocation by maximizing the lower bound of the average channel capacity. Simulation results show that the proposed approach outperforms the frequency-division multiplexed training schemes. CitationZhang H, Dai X H, Pan D R. Linearly time-varying channel estimation and training power allocation for OFDM/MIMO systems using superimposed training.

Superimposed training for channel estimation of OFDM modulated amplify-and-forward relay networks

“…We then examine the performance of separated channel estimation by using both the LS and MMSE estimators for a fixed relay power-allocation factor α=0.4. ST sequences are designed following the optimal criteria (18). As shown in Figure 4, the simulation results are in good agreement with the theoretical derivation, and the MMSE scheme provides a better estimation performance than the LS scheme.…”

“…Remark 2. Note that the first term on the right-hand-side (RHS) of (19) is the extra data interference, which is reviewed as a common problem in the ST-aided literatures [14][15][16][17][18]. Nevertheless, in this paper, one cannot expect to improve the estimation accuracy by averaging the channel estimates (11) over multiple symbol periods since the channel assumed herein is time-varying between consecutive OFDM blocks.…”

“…Nevertheless, the previous methods incur a significant overhead, i.e., an entire training block is required at the source node, thus leading to a significant reduction in transmission efficiency. To save the valuable spectrum resources without entailing high complexity, an alternative approach, referred to as superimposed training (ST) based schemes, has been studied in [14][15][16][17][18]. In such schemes, channel estimation can be performed with no cost of rate with bearable data interference since the training signals are arithmetically added onto the unknown data.…”

“…In accordance with central limit theorem, ifN is large, D(i)can be regarded as a Gaussian-distributed random vector. Assuming that D(i) and V (i) are mutually independent[14][15][16][17][18], R W (i) can be modeled as…”

An efficient sparse channel estimator combining time-domain LS and iterative shrinkage for OFDM systems with IQ-imbalances

“…For that reason, in this paper, we focus on the CS-based sparse channel estimation in OFDM systems with IQ-imbalances at both transmitter and receiver. Through offering an excellent performance of the CS-based channel estimation, this channel estimation is very suitable to achieve a high accuracy of chunk-based resource allocation including bit, subcarrier, and power in OFDM/OFDMA systems [16][17][18][19][20][21].…”