Cooperative Wireless Communications in the Presence of Limited Feedback

Stefan CEROVIC
Thesis defended on September 25, 2019, 10:00 AM at CentraleSupelec (Gif-sur-Yvette)

Cette soutenance aura lieu à CENTRALESUPELEC, Gif-sur-Yvette, Bâtiment Bouygues -Amphi sc.071 

Composition du jury:

Antoine BERTHET          CentraleSupélec                           Directeur de thèse

Raphaël VISOZ               Orange Labs                                 Encadrant de thèse

Karine AMIS                    IMT Atlantique Bretagne               Rapporteur

Didier LE RUYET            CEDRIC/LAETITIA CNAM            Rapporteur

Raymond KNOPP           EURECOM                                   Examinateur

Samson LASAULCE       CNRS                                           Examinateur

 

 

Abstract : 

A constant need for improved quality of wireless services has pushed the wireless technology and development of wireless networks to the point where they have became an integral part of our modern society. Exploiting an innovative concept such as cooperative communications is one possible avenue for answering the increasingly challenging demands from users, which is the main subject of this thesis. Its principle idea is to allow devices to share their available resources in power and/or bandwidth in order to mutually improve their transmission and reception.

Cooperation techniques have been studied for Multiple Access Multiple Relay Channel (MAMRC), consisted of at least two sources which communicate with a single destination with the help of at least two nodes which perform relaying functions (relaying nodes). A relaying node can be either a dedicated relay, which does not have its own message to transmit, or a source itself, which does have its own message and that can relay the messages of the other sources in some cases. All relaying nodes are assumed to operate in half-duplex mode, while all the channels experience slow (quasi-static) fading. Time Division Multiplexing (TDM) is assumed. First, the link adaptation algorithm is performed at the scheduler which is located at the destination. Sources transmit in turns in consecutive time slots during the first transmission phase. The second phase consists of a limited number of time slots for retransmissions. In each time slot, the destination schedules a node (being a source or a relay) to transmit redundancies, implementing a cooperative Hybrid Automatic Repeat reQuest (HARQ) protocol. Bidirectional limited control channels are available from sources and relays towards the destination to implement the necessary control signaling of HARQ protocols.

In the first part of the thesis, the focus is on design of centralized scheduling (node selection) strategies for the second phase. The scheduling decisions are made based on the knowledge of the correctly decoded source sets of each node, with the goal to maximize the average spectral efficiency under the given constraint of fairness. A scheduled node uses Joint Network and Channel Coding (JNCC) on its decoded source set. An information outage analysis is conducted and Monte-Carlo (MC) simulations are performed, which show that these strategies outperform the state of the art one based on the minimization of the probability of the common outage event after each time-slot.

In the second part of the thesis, a slow-link adaptation algorithm is proposed which aims at maximizing the average spectral efficiency under individual QoS targets for a given modulation and coding scheme (MCS) family. The defined utility metric is conditional on the node selection strategy that is used in the second phase. Channel Distribution Information (CDI) is reported to the destination in order to derive the source rates on a long-term basis, which is adapted to the scenario of fast changing radio conditions. Two variants of the algorithm are proposed, one where CDI, and the other where Channel State Information (CSI) of each link is reported to the destination in order to derive the source rates. They are adapted to scenarios with fast and slow changing radio conditions, respectively.

Discrete source rates are first determined using the ``Genie-Aided'' assumption, which is followed by an iterative rate correction algorithm. The resulting scheduling and link adaptation algorithm yields performance close to the exhaustive search approach as demonstrated by MC simulations. In addition, a fast-link adaptation algorithm is proposed, adapted to the scenario where the CSI of all links is reported to the destination.

In the third part of the thesis, performances of three different cooperative HARQ protocols are compared, with the goal to identify the one which offers the best trade-off between performance and complexity. Incremental Redundancy (IR) HARQ with single-user and multiuser encoding are considered, as well as the Chase Combining HARQ with single-user encoding. MC simulations demonstrate that IR-HARQ with single-user encoding offers the best trade-off between performance and complexity for a small number of sources in our setting. Additionally, a practical encoding and decoding scheme is proposed for a scenario where relaying nodes implement single-user encoding, and its performance has been evaluated using MC simulations. The encoding/decoding scheme is based on a turbo code in conjunction with the proposed link adaptation algorithm. The algorithm operates with a family of practical MCSs, where circular buffer is implemented to form transmission messages with desired coding rate.

Keywords: Relay-assisted cooperative communications, multi-source multi-relay wireless networks, centralized scheduling, node selection, slow-link adaptation, iterative rate correction, spectral efficiency, chase combining, incremental redundancy, HARQ, turbo codes.