 Research
 Open Access
Energy and spectral efficient Doppler diversity transmissions in highmobility systems with imperfect channel estimation
 Weixi Zhou^{1}Email author,
 Jingxian Wu^{2} and
 Pingzhi Fan^{1}
https://doi.org/10.1186/s1363801503746
© Zhou et al.; licensee Springer. 2015
 Received: 7 December 2014
 Accepted: 29 April 2015
 Published: 21 May 2015
Abstract
This paper studies energy and spectral efficient Doppler diversity transmissions in the presence of imperfect channel state information (CSI). Fast timevarying fading in highmobility communication systems introduces Doppler diversity that can benefit system performance. On the other hand, it is more difficult to estimate and track fast timevarying channel; thus, channel estimation errors might seriously degrade system performance in highmobility systems. The tradeoffs between channel estimation errors and Doppler diversity are studied by using two precoding schemes, a simple repetition code, and a rate 1 Doppler domain multiplexing (DDM) scheme. The repetition code can achieve the maximum Doppler diversity at the cost of a lower spectral efficiency, and the DDM scheme can achieve the energy and spectral efficient Doppler diversity transmissions. Unlike many other Doppler diversity systems that assume perfect CSI, we explicitly consider the impacts of imperfect CSI on the design and performance of both precoding schemes. Optimum and suboptimum receivers for both schemes are developed by studying the statistical properties of channel estimation errors. The analytical error probabilities of the receivers are expressed as functions of a number of system parameters, such as the maximum Doppler spread, the percentage of pilot symbols for channel estimation, the energy allocation between the pilot and data symbols, etc. The analytical and simulation results indicate that both precoding schemes can achieve the maximum Doppler diversity order through the optimization of the various parameters, even in the presence of imperfect CSI.
Keywords
 Doppler diversity
 Channel estimation error
 Highmobility communications
1 Introduction
With the increasing demands of highspeed railways and aircraft communications, wireless communications in highmobility environment have attracted considerable attentions during the past few years. Signals in highmobility systems could experience large Doppler shifts in the order of kilohertz, while most conventional wireless communication systems are designed for Doppler shifts up to a few hundred hertz. The large Doppler shifts result in fast timevarying fading, which is one of the main challenges for the design of reliable highmobility systems. On the other hand, fast timevarying fading caused by large Doppler shifts in highmobility systems provides Doppler diversity, which can be used to benefit system performance.
Several methods are proposed to exploit the Doppler diversity gain inherent in fast timevarying fading [13], and they provide efficient countermeasures against fading. Most existing works on Doppler diversity assume that perfect channel state information (CSI) can be obtained at the receiver. However, in highmobility systems, it is a nontrivial task to estimate and track the fast timevarying fading with high precision and credibility. Channel estimation errors are usually inevitable and might have significant impacts on system performance when the Doppler frequency is high. The performance of precoded orthogonal frequency division multiplexing (OFDM) systems with channel estimation error is studied in [4]. It is shown in [5] that for a singleinput multipleoutput (SIMO) system with identically and independently distributed (i.i.d.) fading, the conventional maximal ratio combining (MRC) receiver is no longer optimum with imperfect CSI. The results of [4] and [5] cannot be applied to highmobility systems because they both assume the systems experience quasistatic fading channels. The optimum designs of highmobility systems in the presence of imperfect CSI are studied in [6,7] and [8] in terms of different design metrics, such as the bit error rate (BER), symbol error rate (SER), or spectral efficiency. However, Doppler diversity is not considered in these works.

The Doppler diversity transceivers in this paper are developed by explicitly utilizing the properties of imperfect CSI, whereas most existing works on Doppler diversity assume perfect CSI. For example, the DDM scheme is originally proposed in [3] for systems with perfect CSI, and its performance degrades significantly with imperfect CSI, given the fact that conventional receivers are no longer optimum with imperfect CSI. We address this problem by developing new transceiver structures that explicitly consider the impact of channel estimation errors.

The impacts of imperfect CSI on system performance are identified by developing the theoretical error probabilities of the new transceivers. The new analytical results are expressed as functions of the secondorder statistics of the channel estimation errors, and they reveal the tradeoff between Doppler diversity and channel estimation errors.

The new transceiver structure along with the theoretical error probabilities can be used to guide the development of practical Doppler diversity systems. With the analytical and simulation results, the various system parameters that yield the optimum tradeoff between channel estimation errors and Doppler diversity are identified.
The rest of this paper is organized as follows. Section 2 introduces the system model and the two precoding schemes. Section 3 shows the channel estimation. The tradeoff between Doppler diversity and channel estimation errors is studied in Section 4 with the repetition code. Section 5 presents the receiver design and performance analysis of the rate 1 DDM scheme. Numerical and simulation results are presented in Section 6, and Section 7 concludes the paper.
2 System model
The system model is presented in this section. At the transmitter, the modulated data symbols are precoded to achieve Doppler diversity transmissions. Pilot symbols are inserted among the precoded data systems to track and estimate the fast timevarying fading channel.
2.1 Precoding
where \({\mathbf \Theta } \in {\mathcal C}^{N_{c} \times N_{s}}\phantom {\dot {i}\!}\) is a precoding matrix with \({\mathcal C}\) being the set of complex numbers, and N _{ c }≥N _{ s } is the length of the codeword. Define the kth precoded data symbol of \(\textbf {c} = [c_{1}, \cdots, c_{N_{c}}]^{T}\phantom {\dot {i}\!}\) as c _{ k }, where k=1,⋯,N _{ c }. The coding rate is \(\rho = \frac {N_{s}}{N_{c}}\) and the precoding matrix satisfies trace(Θ Θ ^{ H })=N _{ s }, with A ^{ H } being the matrix Hermitian operation. Two precoding schemes are considered in this paper.
The first precoding scheme is a simple rate \(\frac {1}{N_{c}}\) repetition code with N _{ s }=1 and \({\mathbf \Theta } = \frac {1}{\sqrt {N_{c}}}\textbf {1}_{N_{c}}\), where \(\textbf {1}_{N_{c}}\) is a length N _{ c } allone vector. Since the repetition code spreads one data symbol over the entire slot, it is guaranteed to achieve the maximum Doppler diversity at the cost of a low spectral efficiency. The repetition code provides the best possible performance in terms of Doppler diversity gain, and its performance can serve as a lower bound for practical precoding schemes [9].
where θ _{ n } is the nth column of Θ. With such a precoding scheme, the nth data symbol is spread out over the entire slot as θ _{ n } s _{ n }, such that the maximum Doppler diversity can be achieved. However, the orthogonality between the columns of the precoding matrix will be destroyed by the fast timevarying fading channel. Therefore, there will be mutual interferences among the data symbols at the receiver. The interference can be partly removed by means of Doppler domain equalization. It should be noted that perfect CSI at the receiver is assumed by [3], and this assumption is usually not true for highmobility systems.
2.2 Pilotassisted transmission
After precoding, N _{ p } equallyspaced pilot symbols are inserted in each slot to assist channel estimation at the receiver. The number of precoded data symbols N _{ c } and pilot symbols N _{ p } are chosen as N _{ c }=N _{ p }(K−1), such that there are K−1 precoded data symbols between a pair of adjacent pilot symbols. The kth pilot symbol is denoted as \(x_{i_{k}}=p_{k}\), where i _{ k }=k K is the index of the kth pilot symbol, for k=1,…,N _{ p }. Similarly, the kth data symbol is denoted as \(x_{n_{k}} = c_{k}\), where \(n_{k} = k + \lfloor \frac {k1}{K1} \rfloor \) is the index of the kth coded data symbol, for k=1,⋯,N _{ c }, where ⌊a⌋ is the largest integer not larger than a. Define \(\textbf {x}_{p} = [x_{i_{1}}, \cdots, x_{i_{N_{p}}}]^{T}\phantom {\dot {i}\!}\) and \(\textbf {x}_{d} = [x_{n_{1}}, \cdots, x_{n_{N_{c}}}]^{T}\phantom {\dot {i}\!} \) as the pilot and coded data vectors, respectively. Pilot and data symbols can be from different modulation alphabet sets. Without loss of generality, pilot symbols are assumed to be constant amplitude symbols, i.e., p _{ k }^{2}=1, which is not necessarily the case for data symbols. With such a slot structure, the pilot percentage can be defined as \(\delta = \frac {N_{p}}{N} = \frac {1}{K}\), with N=N _{ p }+N _{ c }.
where ρ is the precoding code rate. Under a fixed E _{ b }, η, and ρ, increasing the pilot percentage δ will decrease the energy for each pilot symbol; however, it will not affect the energy per coded data symbol.
where \({{\mathbf y}_{p}\! = [y(i_{1}), \cdots \!, y(i_{N_{p}})]^{T} \!\in \! {\mathcal C}^{N_{p} \times 1}}\) and \({{\mathbf z}_{p} \!= [z(i_{1}), \cdots \!,} z(i_{N_{p}})]^{T} \in {\mathcal C}^{N_{p} \times 1}\) are the received pilot vector and additive white Gaussian noise (AWGN) vector, respectively, and X _{ p }=diag(x _{ p }) is a diagonal matrix with the main diagonal being x _{ p }, and \(\textbf {h}_{p} = [h(i_{1}), \dots, h(i_{N_{p}})]^{T} \in \mathcal {C}^{N_{p} \times 1}\phantom {\dot {i}\!}\) is the discretetime channel fading vector. The AWGN vector is a zeromean symmetric complex Gaussian random vector (CGRV) with covariance matrix \(N_{0} \textbf {I}_{N_{p}}\phantom {\dot {i}\!}\), where I _{ N } is a size N identity matrix.
where \({{\mathbf y}_{d}\!=\, [y(n_{1}),\! \cdots \!, y(n_{N_{c}})]^{T} \!\!\in \! {\mathcal C}^{N_{c} \times 1}}\) and \({{\mathbf z}_{d} \!= [\!z(n_{1}), \cdots \!,} z(n_{N_{c}})]^{T} \in {\mathcal C}^{N_{c} \times 1}\) are received coded data symbols and AWGN, respectively, X _{ d }=diag(x _{ d }) is a diagonal matrix with the precoded data vector x _{ d } on its main diagonal, and \(\textbf {h}_{d} = [h(n_{1}), \dots, h(n_{N_{c}})]^{T} \in \mathcal {C}^{N_{c} \times 1}\).
3 Channel estimation
where a ^{∗} denotes complex conjugate, \(f_{_{\text {D}}}\) is the maximum Doppler spread of the fading channel, T _{ s } is the symbol period, and J _{0}(x) is the zeroorder Bessel function of the first kind.
In fast timevarying fading, the channel coefficients vary from symbol to symbol by following the time correlation in (7). As a result, the channel coefficients of the pilot symbols, h _{ p }, are different from the channel coefficients of the data symbols, h _{ d }.
where \({\mathbf W}_{d} \in {\mathcal C}^{N_{c} \times N_{p}}\) is the MMSE matrix to minimize \(\frac {1}{N_{c}}{\mathbb E}\left (\\hat {{\mathbf h}}_{d}  \textbf {h}_{d}\^{2}\right)\).
where \({\mathbf W} = \textbf {W}_{p} \textbf {W}_{d} \in {\mathcal C}^{N_{c} \times N_{p}}\) is the MMSE matrix that can minimize \(\frac {1}{N_{c}}{\mathbb E}\left (\\hat {{\mathbf h}}_{d}  \textbf {h}_{d}\^{2}\right)\). In (10), the received pilot symbols are used to estimate the channel coefficients of data symbols through time interpolation; thus, the fast timevariation of the fading coefficients can be accurately tracked.
where A ^{−1} is the matrix inverse operation, \(\textbf {R}_{\textit {dp}} = {\mathbb E}\left [\textbf {h}_{d} \textbf {h}_{p}^{H}\right ] \in {\mathcal R}^{N_{c} \times N_{p}}\) and \(\mathbf {R}_{\textit {pp}}=\mathbb {E}\left [{\mathbf h}_{p} {\mathbf h}_{p}^{H}\right ] \in \mathcal {C}^{N_{p} \times N_{p}}\) with their elements defined in (7).
where \(\gamma _{p} = \frac {E_{p}}{{\sigma _{z}^{2}}}\) is the signaltonoise ratio (SNR) of the pilot symbols, \(\mathbf {R}_{\textit {dd}} = {\mathbb E}\left [\textbf {h}_{d} \textbf {h}_{d}^{H}\right ] \in {\mathcal R}^{N_{c} \times N_{c}}\), and \(\textbf {X}_{p}\textbf {X}_{p}^{H} = \textbf {I}_{N_{p}}\) is used in the derivation of the above equation.
For the design and analysis of the diversity receiver in the presence of imperfect CSI, it is necessary to obtain the statistical properties of \({\hat {\textbf {h}}}_{d}\) by considering the effects of channel estimation errors because the receiver performs detection based on the knowledge of the estimated channel coefficients \({\hat {\textbf {h}}}_{d}\).
where \(\omega _{_{\text {D}}}=2\pi {f_{_{\text {D}}}} T_{s}\), γ _{ b }=E _{ b }/N _{0} is the equivalent SNR of the uncoded bit, and \(\nu =\eta \left (\frac {1}{\delta }1\right)\rho \log _{2} M\).
4 Tradeoff between channel estimation errors and Doppler diversity
The tradeoff between channel estimation errors and the maximum Doppler diversity gain is studied in this section with the help of the simple repetition code.
4.1 Optimum diversity receiver with imperfect CSI
It should be noted that the total energy of the N _{ c } repeated data symbols is normalized to E _{ c } to ensure fair comparison with other precoding schemes and uncoded systems.
The receiver performs detection based on the received data vector y _{ d } and the knowledge of the estimated CSI vector \({\hat {\textbf {h}}}_{d}\). Since both h _{ d } and \({\hat {\textbf {h}}}_{d}\) are zeromean Gaussian distributed, the error vector \(\textbf {e}_{d} =\textbf {h}_{d}  \hat {{\mathbf h}}_{d}\) is zeromean Gaussian distributed. Conditioned on \({\hat {\textbf {h}}}_{d}\), h _{ d } is Gaussian distributed with mean \(\textbf {u}_{d{\hat h}_{d}} = {\mathbb E}[\textbf {h}_{d}{\hat {\textbf {h}}}_{d}] = {\hat {\textbf {h}}}_{d}\) and covariance matrix \(\textbf {R}_{dd{\hat h}_{d}} = {\mathbb E}\left [\left (\textbf {h}_{d}\textbf {u}_{d{\hat h}_{d}}\right)\left (\textbf {h}_{d}\textbf {u}_{d{\hat h}_{d}}\right)^{H} {\hat {\textbf {h}}}_{d}\right ] = \textbf {R}_{\textit {ee}}\).
4.2 Performance analysis
The error performance for systems with Mary phase shift keying (MPSK) modulation and operating with imperfect CSI and the optimum decision rule is derived based on the statistical properties of the estimated CSI \({\hat {\textbf {h}}}_{d}\).
where \(\textbf {Q} = \left (\textbf {R}_{\textit {ee}} + \frac {1}{\gamma _{c}} \textbf {I}_{N} \right)^{1}\).
where t is a dumb variable.
where \(\textbf {V} \in {\mathcal C}^{N_{s} \times N_{s}}\) contains the orthonormal eigenvectors of R _{ ee }, and \({\Lambda } = \text {diag}{[\varphi _{1}, \cdots, \varphi _{N_{c}}]}\phantom {\dot {i}\!}\) is the diagonal matrix containing the corresponding eigenvalues.
where \(\boldsymbol {\Omega } \in {\mathcal R}^{N_{c} \times N_{c}}\phantom {\dot {i}\!}\) is a diagonal matrix with the kth diagonal element being \(\varphi _{k} + \frac {N_{c}}{\gamma _{c}}\phantom {\dot {i}\!}\).
which can be easily evaluated by numerical calculation.
where g _{ n } is the eigenvalue of R _{ dd }, and it is directly related to the Doppler diversity gain.
However, in case of imperfect CSI, the result in (20) shows that MRC is no longer optimal. The presence channel estimation error affects the decision process, and the new optimum decision rule has to take into consideration the statistical properties of the channel estimation error quantified in the matrix R _{ ee }.
The SER expressions in (26) reveal the tradeoff between Doppler diversity and channel estimation errors, and they provide a lower bound on the performance of systems with practical precoders.
5 Doppler domain multiplexing in the presence of imperfect CSI
The design and performance of a practical rate 1 DDM precoding scheme is studied in this section. A suboptimum receiver is developed by studying the statistical properties of the estimated channel coefficients, and the corresponding analytical error performance is derived.
5.1 Doppler domain equalization with imperfect CSI
where \({\hat {\textbf {H}}}_{d} = \text {diag}({\hat {\textbf {h}}}_{d})\) and E _{ d }=diag(e _{ d }) are diagonal matrices with \({\hat {\textbf {h}}}_{d}\) and e _{ d } on their main diagonals, respectively.
The system can be considered as an equivalent MIMO system with the equivalent channel matrix being \(\hat {{\mathbf H}}_{d} {\mathbf \Theta }\), which introduces interference among the symbols in the frequency domain. In addition, interference is introduced by the channel estimation error.
We propose to detect the transmitted symbols by using an iterative softinput softoutput (SISO) block decision feedback equalizer (BDFE) [14]. Unlike conventional BDFE that assume perfect CSI at the receiver, the BDFE in this paper is developed by considering the statistical properties of the estimated channel coefficients and the channel estimation errors.
The input to the SISOBDFE equalizer is the a priori loglikelihood ratio (LLR) of the information bits, and the output of the SISOBDFE equalizer is the a posteriori LLR of the information bits. The a priori LLR at the ith iteration is the a posteriori LLR at the (i−1)th iteration. Based on the a priori LLR of the bits, the a priori mean, \({\bar s}_{n}\), and variance, \({\sigma _{n}^{2}}\), of the symbol s _{ n } can be calculated as described in [14].
where the soft output \({\tilde {\textbf {s}}}_{n}\) is used for the detection of s _{ n }.
Since B _{ n } is strictly upper triangular, the detection is performed in a reverse order, that is, s _{ n+1} is detected before s _{ n } and the hard decision of s _{ n+1} is fed back to facilitate the detection of s _{ k } for k<n+1.
where \(\phantom {\dot {i}\!}{\mathbf R}_{\textit {ss}} = \text {diag}\left [\sigma _{1}^{2}, \cdots, \sigma _{N_{c}}^{2}\right ]\) with \({\sigma _{n}^{2}}\) being the a priori variance of s _{ n }, \(\textbf {T} = \textbf {E}_{d}\left ({\mathbf \Theta } {\mathbf R}_{\textit {ss}} {\mathbf \Theta }^{H}\right) \textbf {E}_{d}^{H}\phantom {\dot {i}\!}\). Since E _{ d } is diagonal, the (m,n)th element of T is \((\textbf {T})_{m,n} = e_{m} e_{n}^{*} \bar {v}_{\textit {mn}}\), where e _{ m } is the mth element of the channel estimation error vector e, and \(\bar {v}_{\textit {mn}}\) is the (m,n)th element of the matrix \(\overline {\textbf {V}} = {\mathbf \Theta } {\mathbf R}_{\textit {ss}} {\mathbf \Theta }^{H}\phantom {\dot {i}\!}\). Thus \({\mathbb E}\left [(\textbf {T})_{m,n}\right ] = (\textbf {R}_{\textit {ee}})_{\textit {mn}} \cdot v_{\textit {mn}}\). Therefore, the matrix \({\mathbb E}\left (\textbf {T} \right) = \textbf {R}_{\textit {ee}}\odot {\overline {\textbf {V}}}\phantom {\dot {i}\!}\), where \((\overline {\textbf {A}} \odot \overline {\textbf {B}})_{m,n} = (\overline {\textbf {A}})_{m,n} (\overline {\textbf {B}})_{m,n}\phantom {\dot {i}\!}\) is the elementwise multiplication between two matrices.
where \({\mathbf U}_{\xi } \in {\mathcal C}^{N_{c} \times N_{c}}\) is an upper triangular matrix with unit diagonal elements, \({\mathbf D}_{\xi } \in {\mathcal R}^{N_{c} \times N_{c}}\) is a diagonal matrix, and \({\mathbf L}=\sqrt {{\mathbf D}_{\xi }} {\mathbf U}_{\xi }\). With the Cholesky decomposition described in (33), the feedback matrix B _{ n } can be calculated as \(\textbf {B}_{n} = \textbf {U}_{\xi }\textbf {I}_{N_{c}}\phantom {\dot {i}\!}\). Consequently, the error covariance matrix of the BDFE equalizer is \(\mathbf {\Phi }_{\epsilon \epsilon } = \textbf {D}_{\xi }^{1}\).
5.2 Error performance analysis
The pairwise error probability (PEP) and a BER lower bound of the rate 1 DDM system with imperfect CSI is developed in this subsection.
where X _{ α }=diag(c _{ α }) and X _{ β }=diag(c _{ β }).
where \(Q_{z} = \ {\mathbf y}_{d}  \sqrt {{E_{s}}} {\mathbf X}_{\beta } \hat {{\mathbf h}}_{d} \^{2}  \ {\mathbf y}_{d}  \sqrt {{E_{s}}} {\mathbf X}_{\alpha } \hat {{\mathbf h}}_{d} \^{2}\).
with \({\mathbf R}_{z_{1}} = E_{s} ({\mathbf X}_{\alpha }{\mathbf X}_{\beta }) {\mathbf R}_{\hat {d}\hat {d}} ({\mathbf X}_{\alpha }{\mathbf X}_{\beta })^{H} + \textbf {R}_{w_{\alpha }}\), \(\textbf {R}_{w_{\alpha }} = {\mathbb E}\left [\textbf {w}_{\alpha } \textbf {w}_{\alpha }^{H}\right ] = E_{s} \textbf {X}_{\alpha } {\mathbf R}_{\textit {ee}} \textbf {X}_{\alpha }^{H} + {\sigma _{z}^{2}} {\mathbf I}_{N_{c}}\).
where λ _{ i } is the ith nonzero eigenvalue of the rank N _{ w } matrix \({\mathbf D}_{z}^{\frac {1}{2}} {\mathbf U}_{z}^{H} {\mathbf K}{\mathbf U}_{z} {\mathbf D}_{z}^{\frac {1}{2}}\).
where \(\mu _{i} = \prod _{n=1, n \neq i}^{N_{w}} \frac {\lambda _{i}}{\lambda _{i}  \lambda _{n}}\).
where u(x) is the unit step function, and sgn(x)=1 if x≥0 and −1 otherwise.
where λ _{ n } for \(n=1, \dots, N_{w}^{}\) are the negative eigenvalues.
For a pair of information vectors, s _{ α } and s _{ β }, their Hamming distance are defined as D _{ H }(s _{ α },s _{ β })=∥s _{ α }−s _{ β }∥_{0}, where ∥a∥_{0} is the l _{0} norm operator that returns the number of nonzero elements in the vector a. Intuitively, the BER can be reduced by assigning codeword pairs with smaller PEP to information vector pairs with larger Hamming distance, such that the probability of error events with a large number of bit errors is small. Equivalently, when D _{ H }(s _{ α },s _{ β }) is small, a good precoding scheme should yield a relatively large PEP P(X _{ α }→X _{ β }), where X _{ α } and X _{ β } are the codewords of s _{ α } and s _{ β }, respectively.
In the equation above, the outer summation with respect to α is used to average over all \(2^{N_{c}}\phantom {\dot {i}\!}\) possible values of s _{ α }. When N _{ c } is large, the averaging operation is timeconsuming. However, it can be evaluated by using a large number of randomly generated s _{ α } instead. It should be noted that our analysis can be easily extended to any MPSK modulated system by considering the dominant error events.
6 Simulation results
Analytical and simulation results are presented in this section to study the tradeoff between Doppler diversity and channel estimation errors and to validate the performance of the two precoding schemes in the presence of imperfect CSI. All systems employ a symbol rate of 0.1 Msym/s and operating at 1.9 GHz. The block length is N=100. When the movement speed is between 56.8 and 568.4 km/hr, the corresponding range of Doppler spread is between 100 Hz (\(f_{_{\text {D}}} T_{s} = 10^{3}\)) to 1 KHz (\(f_{_{\text {D}}} T_{s} = 10^{2}\)).
7 Conclusions
The maximum Doppler diversity transmissions for highmobility systems in the presence of channel estimation errors have been studied in this paper. The tradeoff between Doppler diversity and channel estimation errors has been studied by using a repetition code and a rate 1 Doppler domain multiplexing scheme. The analytical performance of both systems have been obtained by analyzing the statistical properties of the channel estimation errors, and they quantitatively identify the impacts and interactions of a number of system parameters, such as the pilot percentage, the maximum Doppler spread, and the energy allocation factor between pilot and data symbols, etc. It has been shown that the error probability is quasiconvex in Doppler spread and monotonically decreasing in the pilot percentage. The performance of systems with a sufficiently high pilot percentage can approach that of a system with perfect CSI. On the other hand, if the pilot percentage is too low, the benefits of Doppler diversity are offset by channel estimation error such that a system with a lower Doppler spread could get a better performance.
Declarations
Acknowledgements
The work of Weixi Zhou and Pingzhi Fan was supported by the Chinese 973 Program (No. 2012CB316100), the NSFC project (No. 61471302), the 111 Project (No. 111214), the MoE Key Grant Project (No. 311031100), and the Young Innovative Research Team of Sichuan Province (No. 2011JTD0007). The work of Jingxian Wu was supported in part by the U.S. National Science Foundation (NSF) under Grants ECCS1202075 and ECCS1405403.
Authors’ Affiliations
References
 AM Sayeed, B Aazhang, Joint multipathDoppler diversity in mobile wireless communications. IEEE Trans. Commun. 47(1), 123–132 (1999). doi:10.1109/26.747819.View ArticleGoogle Scholar
 X Ma, GB Giannakis, Maximumdiversity transmissions over doubly selective wireless channels. IEEE Trans. Inform. Theory. 49(7), 1832–1840 (2003). doi:10.1109/TIT.2003.813485.View ArticleMATHMathSciNetGoogle Scholar
 J Wu, in Proc. IEEE GLOBECOM. Exploring maximum Doppler diversity by Doppler domain multiplexing (San Fransico, USA, 2006), pp. 1–5. doi:10.1109/GLOCOM.2006.897.Google Scholar
 KI Ahmed, C Tepedelenlioglu, A Spanias, Performance of precoded OFDM with channel estimation error. IEEE Trans. Signal Process. 54(3), 1165–1171 (2006). doi:10.1109/TSP.2005.862898.View ArticleGoogle Scholar
 J Wu, C Xiao, Optimal diversity combining based on linear estimation of Rician fading channels. IEEE Trans. Commun. 56(10), 1612–1615 (2008). doi:10.1109/TCOMM.2008.060598.View ArticleGoogle Scholar
 S Savazzi, U Spagnolini, On the pilot spacing constraints for continuous timevarying fading channels. IEEE Trans. Commun. 57(11), 3209–3213 (2009). doi:10.1109/TCOMM.2009.11.080024.View ArticleMathSciNetGoogle Scholar
 N Sun, J Wu, Maximizing spectral efficiency for high mobility systems with imperfect channel state information. IEEE Trans. Wireless Commun. 13(3), 1462–1470 (2013). doi:10.1109/TWC.2014.012314.130772.View ArticleGoogle Scholar
 N Sun, J Wu, in Proc. IEEE GLOBECOM. Maximizing spectral efficiency with imperfect channel information in high mobility systems (Atlanta, USA, 2013). doi:10.1109/GLOCOM.2013.6831591.Google Scholar
 W Zhou, J Wu, P Fan, in Proc. IEEE ICC. Maximizing Doppler diversity transmisssions for high mobility systems with imperfect channel state information (Sydney, Australia, 2014). doi:10.1109/ICC.2014.6884267.Google Scholar
 J Wu, N Sun, Optimum sensor density in distortion tolerant wireless sensor networks. IEEE Trans. Wireless Commun. 11, 2056–2064 (2012). doi:10.1109/TWC.2012.041612.110127.View ArticleGoogle Scholar
 N Sun, J Wu, Optimum sampling in spatialtemporally correlated wireless sensor networks,. EURASIP J. Wireless Commun. Networking. 2013, 5 (2013). doi:10.1186/1687149920135.View ArticleGoogle Scholar
 N Sun, J Wu, in Proc. IEEE MILCOM. Minimum error transmissions with imperfect channel information in high mobility systems (San Diego, USA, 2013). doi:10.1109/MILCOM.2013.160.Google Scholar
 S Siwamogsatham, MP Fitz, JH Grimm, A new view of performance analysis of transmit diversity schemes in correlated Rayleigh fading. IEEE Trans. Inform. Theory. 48(4), 950–956 (2002). doi:10.1109/18.992794.View ArticleMATHMathSciNetGoogle Scholar
 J Wu, YR Zheng, Low complexity softinput softoutput block decision feedback equalization. IEEE J. Sel. Areas Commun. 26(2), 281–289 (2008). doi:10.1109/JSAC.2008.080205.View ArticleGoogle Scholar
 A Stamoulis, GB Giannakis, A Scaglione, Block FIR decisionfeedback equalizers for filterbank precoded transmissions with blind channel estimation capabilities. IEEE Trans. Commun. 49(1), 69–83 (2001). doi:10.1109/26.898252.View ArticleMATHGoogle Scholar
Copyright
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited.