Optimal Pilot Design for Data Dependent Superimposed Training based Channel Estimation in Single/Multi carrier Block Transmission Systems

This paper proposes an optimal training sequence framework for general block transmission systems over spatially correlated multiple-input multiple-output (MIMO) frequency selective channels. The pilot design is based on a formulation that minimizes the Bayesian Cramér-Rao bound (BCRB) for the mean squared error (MSE) of channel estimation, and is thus MSE optimal in nature. The novelty of the proposed work lies in development of a generic pilot design scheme applicable to all the four MIMO block transmission systems, namely single carrier cyclic prefix (SC-CP), single carrier zero padded (SC-ZP), multi-carrier zero padded (MC-ZP), and multi-carrier cyclic prefix (MC-CP) systems. Simulation results are presented to illustrate the superior performance of the proposed technique over the conventional pilot sequences, in terms of both MSE as well as bit error rate (BER).

A time domain synchronous orthogonal frequency division multiplexing (TDS-OFDM) scheme with dual Pseudonoise (PN) sequence is proposed for underwater acoustic (UWA) communication. Instead of using cyclic prefix (CP) or zero padding (ZP) as the guard interval, the proposed TDS-OFDM scheme uses two identical PN sequences as guard interval, and utilizes them for frame synchronization and channel estimation. The proposed TDS-OFDM increases the spectrum efficiency over the conventional CP or ZP OFDM systems where additional pilots have to be inserted for channel estimation. With the dual PN at each frame header, compressed sensing channel estimation is adopted and a rather simple equalization design reduces the receiver complexity. With a low-density parity-check (LDPC) code, the proposed system is simulated in doubly selective underwater acoustic channels with long multipath delay spread and severe Doppler spread. The results show that the TDS-OFDM achieves slightly better BER performance and higher spectrum efficiency than the conventional ZP-OFDM.

Block transmission has recently been considered as an alternative to the conventional continuous transmission technique. In particular, block transmission techniques with zero padding (ZP) and cyclic prefix (CP) are becoming attractive procedures for their ability to eliminate both inter-symbol interference (ISI) and inter-block interference (IBI). In this paper, we present a unified approach to blind channel estimation and interference suppression for block transmission using ZP or CP in both single-carrier (SC) and multicarrier (MC) systems. Our approach uses a generalized multichannel minimum variance principle to design an equalizing filterbank. Channel estimate is then obtained from an asymptotically tight lower bound of the filterbank output power. Numerical examples show that the proposed schemes approach the CRB as the signal-to-noise ratio (SNR) increases. Additionally, they exhibit low sensitivity to unknown narrowband interference and compare favorably with subspace blind channel estimators.

In this paper, a cyclic-prefixed (CP) system with pulse amplitude modulation (PAM) for uplink transmission in light-fidelity (LiFi) is discussed. LiFi supports bi-directional transmission, i.e., downlink and uplink transmissions. However, there have been only a few studies on uplink LiFi despite the fact that the uplink traffic demand increases due to frequent use of, for example, video call and live streaming services. The uplink transmission is designed such that a transmitter has a low peak-to-average ratio as well as a lower computational complexity using a CP in conjunction with PAM. We investigate three different schemes, namely linear equalization (LE) where only a feedforward filter is applied, decision feedback equalization (DFE) where a feedback filter is added at the receiver and Tomlinson-Harashima precoder (THP) where a feedback filter is added at the transmitter. Both feedforward and feedback filters are jointly optimized. In addition, we consider an optical wireless channel with reflections, randomly-located and randomly-oriented user equipment, a human body acting as a reflector and a limited linear dynamic range of a light emitting diode. We conclude that a DFE system is suitable in a low spectral efficiency region, where, compared to an LE system, up to 4 dB gain is achieved. A THP system is suitable for a high spectral efficiency region, where up to 5 dB gain compared to the LE and DFE systems is achieved.

In this paper, the channel estimation in uplink multiple access relay networks (MARNs) with analog network coding protocol has been researched. We apply the superimposed training (ST) scheme where each relay puts a separate training sequence on the top of the received one before forwarding to destination, and design a maximum likelihood based channel estimation algorithm for the composite source-relay-destination and individual relay-destination links. The optimal training sequences as well as the superimposed training power are also derived in closed forms. To make our study more complete, the channel estimation in the time-selective fading environment is further considered, and a correlation-based channel estimation (CBCE) algorithm is developed by taking advantage of time-domain channel autocorrelation nature. Simulation results show that the presented ST scheme can effectively improve the performance of multi-user detection in MARNs, and the proposed CBCE algorithm significantly outperforms the existing channel estimation methods.

In this paper, we consider bandwidth efficient design and channel estimation over time-varying frequency-selective environments, where the basis expansion model (BEM) is used to describe the time-varying channel. Superimposed training based channel estimation based affine precoding model is designed, where the data interference is eliminated from received data prior to performing channel estimation, while the training sequence is completely out in the symbol detection. Simulations results show that the proposed scheme can yield better detection performance with high bandwidth efficiency.

Block transmission with zero padding (ZP) and cyclic prefix (CP) are becoming attractive procedures for their ability to eliminate both inter-symbol interference (ISI) and inter-block interference (IBI). In this paper, we present joint blind channel estimation and interference suppression for single carrier block transmission using both ZP and CP. The proposed estimators take the same form, which gives flexibility for application to either one with minor changes. Our approach uses a generalized multichannel minimum variance principle to design an equalizing filterbank. Channel estimate is then obtained by deriving an asymptotically tight lower bound of the filterbank output power. Numerical examples show that the proposed schemes exhibit low sensitivity to unknown narrowband interference and compare favorably with sub-space blind channel estimators

This paper explores asynchronous interference suppression in cyclic prefix communication systems such as orthogonal frequency division multiplexing (OFDM). An asynchronous interferer is defined as a signal whose cyclic prefix does not line up in time with the desired signal's cyclic prefix. First, this paper presents an analysis of frequency-domain approaches to show that traditional circular-convolution based approaches cannot perfectly suppress the asynchronous interferer unless the number of receive elements is greater than the number of spatially distinct multipath rays. Second, this paper exploits the cyclic redundancy to design semi-blind space-time weights that are very effective at combating both synchronous and asynchronous interference. Simulation results demonstrate that frequency-domain weights have difficulty in suppressing asynchronous interference and also show the effectiveness of the proposed semi-blind suppression technique.

In this paper we study a new semi-blind channel estimation algorithm in orthogonal frequency division multiplexing (OFDM) systems. The proposed scheme is an extension of a recently reported subspace-based blind channel estimation algorithm in cyclic prefix systems which requires very few received blocks. The semi-blind estimation algorithm is devised by using the information obtained both from the blind channel estimation method and a pure pilot-assisted method. The proposed algorithm uses a small amount of received data and can be applied to any types of communication constellations. Simulation results show that, with the same number of pilot samples, the semi-blind algorithm has a clear improvement in system performance over the pure pilot-assisted method. To achieve the same bit- error-rate performance, the proposed semi-blind algorithm uses fewer pilot samples.

A single carrier cyclic prefix assisted CDMA system (SCRIP-CDMA) is proposed to solve the potential problems of DS-CDMA and MC-CDMA for high data rate transmission in wireless environment. This paper proposes a data dependent pilot structure for downlink transmission of SCRIP-CDMA to enhance pilot assisted channel estimation in frequency domain. The pilot symbols are computed by assuming constant envelope in frequency domain. This structure gives an easy method for channel estimation. The performance of the proposed pilot structure is compared with common pilot structure, which is data independent. Extensive computer simulations were conducted for both pilot structures.

Orthogonal frequency-division multiplexing (OFDM), with the help of a cyclic prefix, enables low complexity frequency domain equalization, but suffers from a high crest factor. Single carrier with cyclic prefix (SC-CP) has the same advantage with similar performance, but with a lower crest factor and enhanced robustness to phase noise. The cyclic prefix is overhead, so we put more information in it by implementing this cyclic prefix as a training sequence (TS). This new training aided frequency domain equalized single carrier (TASC) scheme offers us additional known symbols and enables better synchronization and (potentially) channel estimation, with the same performance as SC-CP.

In this thesis we consider the performance degradation of three telecommunication transceivers with respect to parameters uncertainties. The major contribution of our analysis is the derivation of simple and informative approximations to the excess mean square error, which uncover the basic factors that determine the performance of each transceiver under the pres- ence of parameter uncertainties. In Chapter 1, we brie°y present the channel models and the transceiver structures used in this thesis. In Chapters 2 and 3, we consider the case of the Multiple-Input Multiple-Output (MIMO) °at fading channel and two widely known transceiver schemes, i.e., the trans- mit MIMO Wiener ?ltering and the Tomlinson-Harashima (TH) precoding. For the case of the transmit MIMO Wiener ?ltering, degradation is due to channel estimation errors and time-variations, and noise second-order statistics estimation errors. For the case of the TH precoder, degradation is due to channel estimation errors and time-variations. For both cases, our ?nal expressions uncover the factors that determine the performance degradation in practice, and the relative importance between di®erent error sources. In Chapter 4, we consider the case of the Single-Input Single-Output (SISO) frequency selective channel and the minimum mean square error (MMSE) linear equalizer. The error sources that cause the system degradation is the channel and CFO estimation errors. Our aim is to uncover the relative importance of these error sources. It turns out that the CFO estimation error is much more important than the channel estimation error. This fact leads to useful conclusions concerning the optimal training sequence design for joint CFO and channel estimation.

With the wireless communications growing rapidly, we need great-performance and high-efficiency channel estimation. An estimation method based on the quadri-phase complete complementary sequences (CCs) is proposed for the Multiple-Input and Multiple-Output (MIMO) systems. A time slot structure that using CCs with optimal periodic autocorrelations as cyclic prefix (CP) and zero padded (ZP) suffix is investigated. With the pilot blocks of CCs, optimal least square (LS) estimation in the minimum mean square error (MMSE) is got. In addition, with the appropriate length CCs, higher accuracy channel estimation will be achieved. To reduce the waste of frequency spectrum that pilot blocks usage, the same blocks would include the training sequences and the data. Analytical and simulation results demonstrate that the method is available and obtainable.

Orthogonal frequency division multiplexing (OFDM) combined with time division multiplexing (TDM), called OFDM/TDM, can bridge the conventional OFDM and single-carrier (SC) transmission by using frequency-domain equalization (FDE). Accurate channel estimation is required for FDE. A well-known channel estimation scheme is pilot-based channel estimation that uses periodically transmitted pilot signals. If a channel estimation error is present, intersymbol interference (ISI) is produced in OFDM/TDM. We apply a Gaussian approximation to the channel estimation error and theoretically investigate the impact of imperfect channel estimation on the average bit error rate (BER) performance of OFDM/TDM. It is shown by numerical evaluation that the channel estimation error degrades the BER performance more when OFDM/TDM approaches SC.

Generalized frequency division multiplexing (GFDM) is a promising candidate waveform for next-generation wireless communication systems. However, GFDM channel estimation is still challenging due to the inherent interference. In this paper, we formulate a pilot design framework with linear minimum mean square error (LMMSE) channel estimation for GFDM, and propose a novel pilot design to achieve interference precancellation during pilot generation with the fixed transmit sample values at selected frequency bins. Numerical results demonstrate that the proposed method reduces the channel estimation mean square error and the symbol error rate (SER) in high signal-to-noise ratio (SNR) regions, compared with the conventional methods.