Wajid, Imran (2012)
Robust Algorithms for Downlink Beamforming in the Conventional and Cognitive Radio Networks with Erroneous Channel State Information.
Technische Universität Darmstadt
Dissertation, Erstveröffentlichung
Kurzbeschreibung (Abstract)
In the last couple of decades, there has been an explosive growth in the demand of wireless services. This growth has urged the researchers to find new and efficient ways of increasing the data rates in wireless networks while providing improved coverage and reliability. Multi-antenna techniques have emerged as a promising solution to these new challenges due to the huge potential these techniques offer in improving the throughput in a wireless network. The use of multiple antennas at the transmitter and/or receiver provides us with additional diversity to fight against the channel fading. Multi-antenna techniques allow us to use beamforming at the transmitter and/or receiver that can result in significant gains in a wireless network.
Recently, it has been identified that the current fixed spectrum assignment policies result in the underutilization of the available radio spectrum. To solve this problem, the concept of cognitive radio (CR) has gained a lot of popularity in the recent years. CR allows the spectrum sharing between a primary network (PN), to which the spectrum is licensed, and a secondary network (SN). The SN is granted the opportunistic use of the underutilized spectrum of the PN under the condition that it does not harm the transmissions in the PN. In this regard, beamforming can be applied at the transmitter of the SN to control the interference leaked to the PN.
In this thesis, we address the problems of robust multiuser downlink beamforming for both the conventional and CR wireless networks. In both cases, the transmitter is assumed to have erroneous covariance-based channel state information (CSI) where the bounds on the mismatch in the channel covariance matrices are assumed to be available. The goal of the robust problem in the conventional downlink scenario is to minimize the total transmitted power under the worst-case quality-of-service (QoS) constraints for the receivers. The robust downlink beamforming problem for CR network is mathematically similar to the conventional case where the QoS constraints apply to the users of the SN. However, additional constraints exist in CR case that limit the interference leaked to the primary users below a given interference threshold.
In our first approach of solving the robust downlink beamforming problems, we assume the uncertainties in the channel covariance matrices to be confined in ellipsoids of given sizes and shapes. The ellipsoids are modelled using weighted Frobenius norm, where the coefficients of weighting matrices depend on the statistical distribution of the CSI mismatch. In both the conventional and CR downlink beamforming problems, we obtain exact reformulations of the worst-case QoS and interference constraints based on Lagrange duality, avoiding the coarse approximations used by previous solutions. The final problem formulations are converted to convex forms using semidefinite relaxation (SDR). We also solve the aforementioned robust problems by replacing the worst-case based constraints with the probabilistic constraints and prove the equivalence between the worst-case and probabilistic robust designs. Additionally, for the robust CR scenario, we propose an iterative approach based on the uplink-downlink duality that combines the benefits of the proposed worst-case based solution with the computationally efficient non-robust iterative technique. Computer simulations verify the performance improvements of the proposed techniques over earlier robust downlink beamforming techniques for both conventional and CR scenarios.
Finally, for the case of conventional networks, we propose an alternative approach to the robust downlink beamforming problem. We consider the fact that the mismatch in the CSI is a combination of mismatches originating from different sources such as estimation errors, feedback quantization or finite sampling effects. The traditional approach is to put a bound on the cumulative error without taking the specific error sources into account and, therefore, exact error bounds for these approaches are generally difficult to obtain. In our alternative approach, we propose to bound the CSI mismatches resulting from different sources independently. Our mismatch model takes into account the physical phenomena, that cause the errors in the CSI, which helps in deriving meaningful threshold values for the errors from various sources. In this approach, we again use Lagrange duality for the reformulation of the worst-case QoS constraints and finally SDR is applied to convert the resulting problem formulation to a convex form. The simulation results verify the effectiveness of the proposed technique.
Typ des Eintrags: | Dissertation | ||||
---|---|---|---|---|---|
Erschienen: | 2012 | ||||
Autor(en): | Wajid, Imran | ||||
Art des Eintrags: | Erstveröffentlichung | ||||
Titel: | Robust Algorithms for Downlink Beamforming in the Conventional and Cognitive Radio Networks with Erroneous Channel State Information | ||||
Sprache: | Englisch | ||||
Referenten: | Pesavento, Prof. Dr. Marius ; Larsson, Prof. Dr. Erik G. | ||||
Publikationsjahr: | 5 Juli 2012 | ||||
Ort: | Darmstadt, Germany | ||||
Datum der mündlichen Prüfung: | 15 Oktober 2012 | ||||
URL / URN: | http://tuprints.ulb.tu-darmstadt.de/3454 | ||||
Kurzbeschreibung (Abstract): | In the last couple of decades, there has been an explosive growth in the demand of wireless services. This growth has urged the researchers to find new and efficient ways of increasing the data rates in wireless networks while providing improved coverage and reliability. Multi-antenna techniques have emerged as a promising solution to these new challenges due to the huge potential these techniques offer in improving the throughput in a wireless network. The use of multiple antennas at the transmitter and/or receiver provides us with additional diversity to fight against the channel fading. Multi-antenna techniques allow us to use beamforming at the transmitter and/or receiver that can result in significant gains in a wireless network. Recently, it has been identified that the current fixed spectrum assignment policies result in the underutilization of the available radio spectrum. To solve this problem, the concept of cognitive radio (CR) has gained a lot of popularity in the recent years. CR allows the spectrum sharing between a primary network (PN), to which the spectrum is licensed, and a secondary network (SN). The SN is granted the opportunistic use of the underutilized spectrum of the PN under the condition that it does not harm the transmissions in the PN. In this regard, beamforming can be applied at the transmitter of the SN to control the interference leaked to the PN. In this thesis, we address the problems of robust multiuser downlink beamforming for both the conventional and CR wireless networks. In both cases, the transmitter is assumed to have erroneous covariance-based channel state information (CSI) where the bounds on the mismatch in the channel covariance matrices are assumed to be available. The goal of the robust problem in the conventional downlink scenario is to minimize the total transmitted power under the worst-case quality-of-service (QoS) constraints for the receivers. The robust downlink beamforming problem for CR network is mathematically similar to the conventional case where the QoS constraints apply to the users of the SN. However, additional constraints exist in CR case that limit the interference leaked to the primary users below a given interference threshold. In our first approach of solving the robust downlink beamforming problems, we assume the uncertainties in the channel covariance matrices to be confined in ellipsoids of given sizes and shapes. The ellipsoids are modelled using weighted Frobenius norm, where the coefficients of weighting matrices depend on the statistical distribution of the CSI mismatch. In both the conventional and CR downlink beamforming problems, we obtain exact reformulations of the worst-case QoS and interference constraints based on Lagrange duality, avoiding the coarse approximations used by previous solutions. The final problem formulations are converted to convex forms using semidefinite relaxation (SDR). We also solve the aforementioned robust problems by replacing the worst-case based constraints with the probabilistic constraints and prove the equivalence between the worst-case and probabilistic robust designs. Additionally, for the robust CR scenario, we propose an iterative approach based on the uplink-downlink duality that combines the benefits of the proposed worst-case based solution with the computationally efficient non-robust iterative technique. Computer simulations verify the performance improvements of the proposed techniques over earlier robust downlink beamforming techniques for both conventional and CR scenarios. Finally, for the case of conventional networks, we propose an alternative approach to the robust downlink beamforming problem. We consider the fact that the mismatch in the CSI is a combination of mismatches originating from different sources such as estimation errors, feedback quantization or finite sampling effects. The traditional approach is to put a bound on the cumulative error without taking the specific error sources into account and, therefore, exact error bounds for these approaches are generally difficult to obtain. In our alternative approach, we propose to bound the CSI mismatches resulting from different sources independently. Our mismatch model takes into account the physical phenomena, that cause the errors in the CSI, which helps in deriving meaningful threshold values for the errors from various sources. In this approach, we again use Lagrange duality for the reformulation of the worst-case QoS constraints and finally SDR is applied to convert the resulting problem formulation to a convex form. The simulation results verify the effectiveness of the proposed technique. |
||||
Alternatives oder übersetztes Abstract: |
|
||||
URN: | urn:nbn:de:tuda-tuprints-34540 | ||||
Sachgruppe der Dewey Dezimalklassifikatin (DDC): | 600 Technik, Medizin, angewandte Wissenschaften > 620 Ingenieurwissenschaften und Maschinenbau | ||||
Fachbereich(e)/-gebiet(e): | 18 Fachbereich Elektrotechnik und Informationstechnik 18 Fachbereich Elektrotechnik und Informationstechnik > Institut für Nachrichtentechnik > Nachrichtentechnische Systeme 18 Fachbereich Elektrotechnik und Informationstechnik > Institut für Nachrichtentechnik |
||||
Hinterlegungsdatum: | 16 Jun 2013 19:55 | ||||
Letzte Änderung: | 16 Jun 2013 19:55 | ||||
PPN: | |||||
Referenten: | Pesavento, Prof. Dr. Marius ; Larsson, Prof. Dr. Erik G. | ||||
Datum der mündlichen Prüfung / Verteidigung / mdl. Prüfung: | 15 Oktober 2012 | ||||
Export: | |||||
Suche nach Titel in: | TUfind oder in Google |
Frage zum Eintrag |
Optionen (nur für Redakteure)
Redaktionelle Details anzeigen |