Font Size: a A A

Robust Signal Processing For Wireless Communication Systems With Erroneous Channel State Information

Posted on:2017-03-31Degree:DoctorType:Dissertation
Country:ChinaCandidate:S MaFull Text:PDF
GTID:1108330488957218Subject:Communication and Information System
Abstract/Summary:PDF Full Text Request
As one of the key technologies of 4G, Multiple Input Multiple Output(MIMO) can exploit spatial degrees of freedom to improve the data transfer rate, without increasing the transmit power and bandwidth conditions. In MIMO systems, beamforming is can achieve directional signal transmission and space division multiple access, interference suppression and higher data rates, which is incorporated in many practical wireless communication applications and research In traditional wireless communication systems, the implementation of beamforming is based on the perfect channel state information(CSI), but the practical CSI is always imperfect and the performance of beamforming is sensitive to the CSI errors. To guarantee the quality of service(Qo S) of users, the dissertation investigates robust beamforming design in cognitive radio networks, two-way relay relay networks and physical-layer security networks for quadratic bounded CSI errors, quadratic unbounded CSI errors and quartic CSI errors. Moreover, as one of the key technologies of 5G, Massive MIMO can save an order of magnitude in transmit power, enhance the reliability of wireless communications and reduce signal processing complexity. However, due to reuse of the same pilot sequences among the different cells, pilot contamination is one bottleneck of Massive MIMO networks.Since the pilot contamination is cause by the users in different cells share the same pilot, the dissertation investigates the optimal pilot assignment in the Massive MIMO networks.To tackle the aforementioned problems, the dissertation study the robust beamforming design in wireless networks, and optimal pilot assignment in Massive MIMO networks. The main work and contributions of this dissertation are listed as follows:1. Considering two practical challenges of interference alignment: imperfect CSI and finite signal-to-noise ratio(SNR), we propose a robust joint signal and interference alignment design for MIMO cognitive radio networks. Specifically, under the assumption of the ellipsoidal CSI errors, the proposed scheme aims to minimize both the leakage of interference signals and that of the desired signals, while maintaining interference to the primary user below a permissible level. The joint worst-case optimization problem is decomposed and reformulated as semidefinite programming(SDP) by usingSlemma, orthogonal relaxation and semi-definite relaxation(SDR). Simulation results verify the effectiveness of the joint design, and robustness of the worst-case design against channel uncertainties.2.Due to CSI errors, the self-interference cannot be completely canceled in two way relay networks. Considering both CSI errors and residual self-interference, we propose a novel restriction and relaxation(RAR) method for robust relay beamforming design in cognitive two-way relay networks. Specifically, the proposed RAR method aims to minimize the transmit power of relay nodes while guaranteeing the worst-case signal-to-interference plus noise ratio(SINR) of secondary users(SUs) as well as satisfying the interference temperature(IT)constraints. In the restriction step of RAR, infinite number of complicated constraints are reformulated into a finite number of linear matrix inequalities(LMIs). In the relaxation step of RAR, the nonconvex robust design problem is transformed into a convex SDP, which can be efficiently solved by interior point methods. Simulation results verify the effectiveness and robustness of the proposed method.3. Considering the unbounded random CSI errors in cognitive radio networks, we propose a novel method for chance constrained robust beamforming problem. Specifically, the proposed method aims to minimize the total SUs’ transmit power under chance constraints corresponding to SINR and IT. Combining use of SDR and two kinds of Bernstein-type inequalities, we transform the chance constraints into deterministic forms, and reformulate the problem as a SDP, which can be solved efficiently using standard interior-point methods.Simulations results verify performance improvements of the proposed method as compared to that based on the worst case method.4. Considering three common CSI errors scenarios in multiple input single output(MISO)wiretap channels, we propose outage constrained robust secure beamforming design framework. Specifically, we seek to maximize the secrecy rate under the transmit power and secrecy rate outage probability constraint. The outage probability constraint requires that the secrecy rate exceeds certain threshold with high probability. Therefore including such constraint in the design naturally ensures the desired robustness. Unfortunately, the presence of the probabilistic constraints makes the problem non-convex and hence difficult to solve.To deal with this difficult, we investigate the outage probability constrained secrecy rate maximization problem using a novel two-step approach. Under a wide range of uncertainty models, our developed algorithms can obtain high-quality solutions, sometimes even exact global solutions, for the robust secure beamformer design problem. Simulation results are presented to verify the effectiveness and robustness of the proposed algorithms.5. For high order CSI errors models, we investigate robust secret amplify-and-forward(AF)relay designs for wireless physical layer security. Specifically, we consider two general CSI scenarios, where the first scenario involve quadratic polynomials of complex Gaussian random variables and the second scenario includes quartic polynomials. In the two scenarios, our goal is to minimize the relay transmit power, while guaranteeing the probability of legitimate’s SINR below a minimum requirement within a given level, and restraining the probability of the eavesdropper’s SINR exceeding maximum allowable threshold below a predetermined level. For the first scenario, the probabilistic constraints are respectively approximated to deterministic forms by properties of the Bernstein-type Inequality. While for the probabilistic constraints involve quartic polynomials, which, to the best of our knowledge, have not been treated from a computational perspective before. We develop a Moment Inequality, which can tackle probabilistic constraints with general Gaussian polynomials.Combining the Moment Inequality and the SDR technique, we propose a new tractable approximation approach for robust beamforming problem with high order CSI errors. Moreover, we analyze the relative tightness of the Bernstein-type Inequality and the Moment Inequality.6.To alleviate the pilot contamination, we propose a joint pilot assignment and user admission scheme to reduce pilot contamination in large MIMO multi-cell networks. Based on the asymptotic SINR, the proposed joint pilot assignment and user admission scheme exploits harmonic SINR utility function to improve the fairness of all users. In particular,the pilot assignment problem is formulated as a minimum weight multi-partite matching problem, which is in general NP-hard. However, for the two-cell network case, the pilot assignment problem can be solved by the Hungarian algorithm, and obtain the optimal pilot assignment scheme within a strongly polynomial time. While for general multi-cell networks case, we propose an iterative algorithm to approximate the optimal solution. After pilot assignment, if some active users leave the networks, the BSs can admit new users. In our proposed joint pilot assignment and user admission scheme, the BSs admit new users for vacant pilot sequences with low computational complexity, which do not need to reassign pilot for all the users. Specifically, the harmonic SINR utility-based user admission problem can be formulated as a supermodular minimization problem. Although, this problems is also in general NP-hard, we propose a greedy algorithm to obtain suboptimal user admission scheme. Moreover, we also present the theoretical gap between the result of proposed greedy algorithm and optimal value. Simulation results verify that our proposed scheme can significantly outperform the conventional scheme.
Keywords/Search Tags:Wireless communications, robust beamforming, CSI errors, cognitive radio networks, relay networks, physical layer security, Massive MIMO networks
PDF Full Text Request
Related items