Study Of Limited Channel Feedback Techniques For Wireless Communications

The compressive sensing theory based techniques of limited channel feedback are deeply studied in this dissertation and the main work includes the following parts. Chapter 2 surveys the existing research on compressive sensing and distributed compressive sensing and radicates the research of this dissertation on limited channel feedback based on compressive sensing. Chapter 3 discusses the compressed sensing based feedback compression method in adaptive OFDM systems. Chapter 4 discusses distributed compressive sensing based channel feedback scheme for massive antenna arrays with spatial-temporal correlation. Chapter 5 discusses limited feedback for cognitive radio networks based on compressed sensing. Chapter 6 discusses spectrum sharing scheme in cognitive radio networks based on primary limited feedback. And Chapter 7 gives summary and prospect.The main contents and contributions of this dissertation are issued as follows:1. Analysis of Limited Channel Feedback TechniquesThe specific limited channel feedback techniques are sorted into different categories. Within each category, the related limited channel feedback techniques are summarized exhaustively. The recent research progress at home and abroad are tracked. The correlation of radio fading channel in space, frequency and time are discussed to be used to reduce feedback furtherly. And such issues are what this dissertation will focus on.2. Analysis of Compressive Sensing Theory and Distributed Compressive SensingThe emerging compressive sensing theory is studied. The relative research progress of sparsity decomposition, measurement matrix, reconstruction algorithms and application are tracked. The distributed compressive sensing and Joint Sparse Mode(JSM) with each reconstruction algorithm are analyzed in detail. The joint sparse models of radio channel for wireless communication are shown. The research of limited channel feedback for wireless communication based on compressive sensing is proved to be the focus of this dissertation.3. Compressed Sensing Based Feedback Compression in Adaptive OFDM SystemsThe required feedback rate can be prohibitively high in adaptive orthogonal frequency division multiplexing(OFDM) systems and so the feedback information needs to be compressed. In this paper, a novel CQI feedback scheme is proposed based on the recently proposed theory of compressed sensing, and its performance is shown to outperform the compression method based on the discrete cosine transform with the throughput increment of 0.1 bit/s/Hz, while the latter is usually considered as the best.4. Limited Feedback for Cognitive Radio Networks Based on Compressed SensingA kind of limited feedback protocol based on compressed sensing is proposed in this dissertation for cognitive radio networks with only a logarithmic scaling in feedback bandwidth. And we numerically study the robustness properties of the least absolute shrinkage and selection operator(LASSO), a popular recovery algorithm, under two error models through simulations. Simulation results indicate that the LASSO recovery algorithm is robust to imperfect channel knowledge and the error tolerance can be increased largely by comparatively lower increase of feedback bandwidth.5. Distributed Compressive Sensing Based Channel Feedback Scheme for Massive AntennaArrays with Spatial-Temporal CorrelationSince the amount of the channel information required by the transmitter is large with massive antennas, the feedback is burdensome in practice, and needs normally to be reduced. In this dissertation, a novel channel feedback reduction scheme based on the theory of distributed compressive sensing is proposed, which permits the transmitter to obtain channel information with acceptable accuracy but with substantially reduced feedback load. Simulation results show that DCS-based compressive scheme has better MSE performance and CDF performance than that of traditional CS-based compressive scheme, and DCS-based channel feedback scheme needs comparatively less feedback resources than CS-based scheme.6. Spectrum Sharing Scheme in Cognitive Radio Networks Based on Primary LimitedFeedbackA spectrum sharing scheme for a pair of secondary users co-existing with multiple primary users is proposed in this dissertation. By overhearing the limited feedback of primary channel quality information, the secondary user accesses the channel with proper transmission power and rate while causing interference to the primary users and so the rate loss of the primary system. Under the primary rate loss constraint, the optimal transmit power and transmission rate for the secondary are obtained to maximize the secondary user throughput. Numerical results show that, with only 3 or 4 bits of feedback from each primary user, the effective throughput of the secondary system is comparable to the case with both primary and secondary transmitters having the perfect CQI of the primary link. The simulation results demonstrate that the proposed spectrum sharing scheme can meet the primary rate loss constraint.