Hegde, Ganapati (2019)
Energy-Efficient and Robust Hybrid Analog-Digital Precoding for Massive MIMO Systems.
Technische Universität Darmstadt
Dissertation, Erstveröffentlichung
Kurzbeschreibung (Abstract)
The fifth-generation (5G) and future cellular networks are expected to facilitate wireless communication among tens of billions of devices with enormously high data rate and ultra-high reliability. At the same time, these networks are required to embrace green technology by significantly improving the energy efficiency of wireless communication to reduce their carbon footprint.
The massive multiple-input multiple-output (MIMO) systems, in which the base stations are equipped with hundreds of antenna elements, can provide immensely high data rates and support a large number of users by employing the precoding at the base stations. However, the conventional precoding techniques - which require a dedicated radio-frequency chain for each antenna element - become prohibitively expensive for massive MIMO systems. To address this shortcoming, the hybrid analog-digital precoding architecture is proposed, which requires fewer radio-frequency chains than the antenna elements. The reduced hardware costs in this novel architecture, however, comes at the expense of reduced degrees of freedom for the precoding, which deteriorates the energy efficiency of the network.
In this thesis, we consider the design of energy-efficient hybrid precoding techniques in multiuser downlink massive MIMO systems. These systems are fundamentally interference limited. To mitigate the interference, we adopt two interference management strategies while designing the hybrid precoding schemes. They are, namely, interference suppression-based hybrid precoding, and interference exploitation-based hybrid precoding. The former approach results in a lower computational complexity - as the resulting precoders remain the same as long as the channel is unchanged when compared to the latter approach. On the other hand, the interference exploitation-based hybrid precoding is more energy efficient due to judicious use of transmit symbol information, as compared to the interference suppression-based hybrid precoding.
In the hybrid analog-digital precoding, analog precoders are implemented in analog radio-frequency domain using a large number of phase shifters, which are relatively inexpensive. These phase shifters, however, typically suffer from artifacts; their actual values differ from their nominal values. These imperfect phase shifters can lead to symbol estimation errors at the users, which may not be tolerable in many applications of future cellular networks. To establish a high-reliable communication under the plight of imperfect phase shifters in the hybrid precoding architecture, in this thesis, we propose an energy-efficient, robust hybrid precoding technique. The designed scheme guarantees 100% robustness against the considered hardware artifacts. Moreover, the thesis demonstrates that the proposed technique can save up to 12% transmit power when compared to a conventional method.
Another critically important requirement of the future cellular networks - apart from ultra-high reliability and energy efficiency - is ultra-low latency. Some envisioned extreme real-time applications of 5G, such as autonomous driving and remote surgery, demand an end-to-end latency smaller than one millisecond. To fulfill such a stringent demand, we devise an efficient implementation scheme for the proposed robust hybrid precoding technique to reduce the required computational time. The devised scheme exploits special structures present in the algorithm to reduce the computational complexity and can compute the precoders in a distributed manner on a parallel hardware architecture. The results show that the proposed implementation scheme can reduce the average computation time of the algorithm by 35% when compared to a state-of-the-art method.
Finally, we consider the hybrid precoding in heterogeneous networks, where the cell edge users typically experience severe interference. We propose a coordinated hybrid precoding technique based on the interference exploitation approach. The numerical results reveal that the proposed coordinated hybrid precoding results in a significant transmit power savings when compared to the uncoordinated hybrid precoding.
Typ des Eintrags: | Dissertation | ||||
---|---|---|---|---|---|
Erschienen: | 2019 | ||||
Autor(en): | Hegde, Ganapati | ||||
Art des Eintrags: | Erstveröffentlichung | ||||
Titel: | Energy-Efficient and Robust Hybrid Analog-Digital Precoding for Massive MIMO Systems | ||||
Sprache: | Englisch | ||||
Referenten: | Pesavento, Prof. Dr. Marius ; Masouros, Prof. Dr. Christos | ||||
Publikationsjahr: | 24 Oktober 2019 | ||||
Ort: | Darmstadt | ||||
Datum der mündlichen Prüfung: | 22 Oktober 2019 | ||||
URL / URN: | https://tuprints.ulb.tu-darmstadt.de/9210 | ||||
Kurzbeschreibung (Abstract): | The fifth-generation (5G) and future cellular networks are expected to facilitate wireless communication among tens of billions of devices with enormously high data rate and ultra-high reliability. At the same time, these networks are required to embrace green technology by significantly improving the energy efficiency of wireless communication to reduce their carbon footprint. The massive multiple-input multiple-output (MIMO) systems, in which the base stations are equipped with hundreds of antenna elements, can provide immensely high data rates and support a large number of users by employing the precoding at the base stations. However, the conventional precoding techniques - which require a dedicated radio-frequency chain for each antenna element - become prohibitively expensive for massive MIMO systems. To address this shortcoming, the hybrid analog-digital precoding architecture is proposed, which requires fewer radio-frequency chains than the antenna elements. The reduced hardware costs in this novel architecture, however, comes at the expense of reduced degrees of freedom for the precoding, which deteriorates the energy efficiency of the network. In this thesis, we consider the design of energy-efficient hybrid precoding techniques in multiuser downlink massive MIMO systems. These systems are fundamentally interference limited. To mitigate the interference, we adopt two interference management strategies while designing the hybrid precoding schemes. They are, namely, interference suppression-based hybrid precoding, and interference exploitation-based hybrid precoding. The former approach results in a lower computational complexity - as the resulting precoders remain the same as long as the channel is unchanged when compared to the latter approach. On the other hand, the interference exploitation-based hybrid precoding is more energy efficient due to judicious use of transmit symbol information, as compared to the interference suppression-based hybrid precoding. In the hybrid analog-digital precoding, analog precoders are implemented in analog radio-frequency domain using a large number of phase shifters, which are relatively inexpensive. These phase shifters, however, typically suffer from artifacts; their actual values differ from their nominal values. These imperfect phase shifters can lead to symbol estimation errors at the users, which may not be tolerable in many applications of future cellular networks. To establish a high-reliable communication under the plight of imperfect phase shifters in the hybrid precoding architecture, in this thesis, we propose an energy-efficient, robust hybrid precoding technique. The designed scheme guarantees 100% robustness against the considered hardware artifacts. Moreover, the thesis demonstrates that the proposed technique can save up to 12% transmit power when compared to a conventional method. Another critically important requirement of the future cellular networks - apart from ultra-high reliability and energy efficiency - is ultra-low latency. Some envisioned extreme real-time applications of 5G, such as autonomous driving and remote surgery, demand an end-to-end latency smaller than one millisecond. To fulfill such a stringent demand, we devise an efficient implementation scheme for the proposed robust hybrid precoding technique to reduce the required computational time. The devised scheme exploits special structures present in the algorithm to reduce the computational complexity and can compute the precoders in a distributed manner on a parallel hardware architecture. The results show that the proposed implementation scheme can reduce the average computation time of the algorithm by 35% when compared to a state-of-the-art method. Finally, we consider the hybrid precoding in heterogeneous networks, where the cell edge users typically experience severe interference. We propose a coordinated hybrid precoding technique based on the interference exploitation approach. The numerical results reveal that the proposed coordinated hybrid precoding results in a significant transmit power savings when compared to the uncoordinated hybrid precoding. |
||||
Alternatives oder übersetztes Abstract: |
|
||||
URN: | urn:nbn:de:tuda-tuprints-92107 | ||||
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 18 Fachbereich Elektrotechnik und Informationstechnik > Institut für Nachrichtentechnik > Nachrichtentechnische Systeme |
||||
Hinterlegungsdatum: | 10 Nov 2019 20:55 | ||||
Letzte Änderung: | 10 Nov 2019 20:55 | ||||
PPN: | |||||
Referenten: | Pesavento, Prof. Dr. Marius ; Masouros, Prof. Dr. Christos | ||||
Datum der mündlichen Prüfung / Verteidigung / mdl. Prüfung: | 22 Oktober 2019 | ||||
Export: | |||||
Suche nach Titel in: | TUfind oder in Google |
Frage zum Eintrag |
Optionen (nur für Redakteure)
Redaktionelle Details anzeigen |