Detail View

The number of transmit and receive antennas in multi-input multi-output (MIMO) systems is increasing rapidly to enhance the throughput and reliability of next-generation wireless systems. Benefits of large size MIMO systems, however, can be realized only when the quality of estimated channels is ensured at the transmitter and receiver side alike.
One of major issues in realizing the MIMO-OFDM systems is that the amount of pilot signals needed for channel estimation is proportional to the number of the transmit antennas. When the number of transmit antennas is large, pilot signals occupy significant portion of downlink resources, eating out the data throughput significantly. One can consider the reduction of pilot signal density but this is undesirable since it will simply cause the degradation of channel estimation quality, affecting link performance and data throughput eventually.
When the pilot resources are depleted, one can naturally consider the option of using the data signals for the channel estimation purpose. This approach, often referred to as decision-directed channel estimation (DD-CE), uses the decision on data symbols in the re-estimation of channels. Two main concerns of the DD-CE for the MIMO-OFDM systems, not thoroughly addressed in the previous efforts, are the reliability of the detected data symbols and the interstream interference caused by the MIMO transmission. First, the quality of the estimated channel would not be appealing if the soft statistics of the data tones used for channel estimation are not reliable. Second, unlike the pilot symbols, the data symbols in the MIMO-OFDM systems are transmitted simultaneously through multiple transmit antennas so that channel estimation is interfered by the data symbols coming from other transmit antennas. Clearly, proper control of these interstream interferences is crucial for effective decision-directed channel estimation.
This dissertation investigates efficient receiver techniques for massive MIMO communications. In the first part of the dissertation, we introduce a new decision-directed channel estimation technique dealing with pilot shortage in the MIMO-OFDM systems. The proposed channel estimator uses soft symbol decisions obtained by iterative detection and decoding (IDD) scheme to enhance the quality of channel estimate. Using the soft information from the decoders, the proposed channel estimator selects reliable data tones, subtracts interstream interferences, and performs re-estimation of the channels. We analyze the optimal data tone selection criterion, which accounts for the reliability of symbol decisions and correlation of channels between the data tones and pilot tones. From numerical simulations, we show that the proposed channel estimator achieves considerable improvement in system performance over the conventional channel estimators in realistic MIMO-OFDM scenarios.
In the second part of the dissertation, we propose a linear receiver technique that maintains the linear scaling law of transmission capacity even under the imperfect channel state information (CSI). Recent works on ad hoc network study have shown that achievable throughput can be made to scale linearly with the number of receive antennas even if the transmitter has only a single antenna. The key feature of proposed algorithm to make our approach effective is to exploit the estimation of the receiver weight using the receiver samples in the normal transmission period motivated by the transmission rate loss of conventional approaches due to the training period for estimating the sample covariance. We show that the linear scaling law holds true even under the practical scenario where the receiver weight is estimated. In fact, by incorporating the desired channel information on top of the observations including interference and noise only, the proposed method achieves large fraction of the optimal MMSE transmission capacity without transmission rate loss. Simulation results on the realistic ad hoc network system show that the proposed non-parametric linear MMSE receiver brings substantial performance gain over existing multiple receive antenna algorithms.
In the third part of the dissertation, we propose an interference aware node activation strategy that efficiently controls the data transmission and back off based on the measured interference power. The key ingredients of the proposed method is a simple yet effective interference aware node activation algorithm which activates a node pair only when the interference level measured by the receiver is smaller than a predefined threshold for improving the rate of node pairs with good SINR. From the SINR analysis as well as transmission capacity simulations in realistic ad hoc network system, we show that the proposed receiver method brings substantial performance gain over existing multiple receive antenna algorithms.

Abstract 1
1 Introduction 4
1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2 Outline and Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.3 Common Notations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2 Iterative Channel Estimation using Virtual Pilot Signals for MIMO-OFDM
Systems 11
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2 System description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.2.1 MIMO-OFDM Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.2.2 Iterative Detection and Decoding . . . . . . . . . . . . . . . . . . . . . . 16
2.2.3 Channel Estimation for MIMO-OFDM Systems . . . . . . . . . . . . . . 17
2.3 Channel Estimation using Virtual Pilot Signal . . . . . . . . . . . . . . . . . . . 19
2.3.1 Channel Re-Estimation using Virtual Pilot Signal . . . . . . . . . . . . 20
2.3.2 Virtual Pilot Signal Selection Criterion . . . . . . . . . . . . . . . . . . . 25
2.3.3 Comments on Complexity . . . . . . . . . . . . . . . . . . . . . . . . . . 28
2.3.4 Derivation of (2.25) and (2.26) . . . . . . . . . . . . . . . . . . . . . . . 30
2.3.5 Derivation of Error Covariance Matrix Ceg,eg . . . . . . . . . . . . . . . 31
2.3.6 Derivation of (2.38) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.4 Simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.4.1 Simulation Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.4.2 Simulation Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.5 Chapter Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 42
3 An Efficient Linear MMSE Receiver for Wireless Ad Hoc Networks 50
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
3.2 Ad Hoc Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.2.1 Background and Problem Statement . . . . . . . . . . . . . . . . . . . . 51
3.2.2 System Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.2.3 Performance Metrics and Max SINR Receiver . . . . . . . . . . . . . . . 54
3.3 Non-Parametric Linear MMSE Receiver . . . . . . . . . . . . . . . . . . . . . . 56
3.3.1 The Non-Parametric Linear MMSE . . . . . . . . . . . . . . . . . . . . . 57
3.3.2 The Scaling Law of the Non-parametric Linear MMSE . . . . . . . . . . 60
3.4 Simulation Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . 64
3.4.1 Simulation Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
3.4.2 Simulation Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.5 Chapter Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 68
4 Interference Aware Node Activation for Wireless Ad Hoc Networks 70
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
4.2 Background and Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . 71
4.3 Interference Aware Node Activation . . . . . . . . . . . . . . . . . . . . . . . . 73
4.3.1 Interference Aware Node Activation . . . . . . . . . . . . . . . . . . . . 73
4.3.2 Expected Delay Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 74
4.3.3 Comments on Transmission Capacity . . . . . . . . . . . . . . . . . . . . 79
4.4 Simulation Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . 82
4.4.1 Simulation Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
4.4.2 Simulation Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.5 Chapter Summary and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 87
5 Conclusions 89
Acronyms 91
Bibliography 93