Abstract

Cognitive radio (CR) represents a key solution to the existing spectrum scarcity problem. Under the scenario of CR, spectrum sharing systems allow the coexistence of primary users (PUs) and secondary users (SUs) in the same spectrum as long as the interference from the secondary to the primary link stays below a given threshold. In this thesis, we propose a number of adaptive transmission schemes aiming at improving the performance of the secondary link in these systems while satisfying the interference constraint set by the primary receiver (PR). In the proposed techniques, the secondary transmitter (ST) adapts its transmission settings based on the availability of the channel state information (CSI) of the secondary and the interference links. In this context, these schemes offer different performance tradeoffs in terms of spectral efficiency, energy efficiency, and overall complexity.

In the first proposed scheme, power adaptation (PA) and adaptive modulation (AM) are jointly used with switched transmit diversity in order to increase the capacity of the secondary link while minimizing the average number of antenna switching. Then, the concept of minimum-selection maximum ratio transmission (MS-MRT) is proposed as an adaptive variation of maximum ratio transmission (MRT) in a spectrum sharing scenario in order to maximize the capacity of the secondary link while minimizing the average number of transmit antennas. In order to achieve this performance, MS-MRT assumes that the secondary's CSI (SCSI) is perfectly known at the ST, which makes this scheme challenging from a practical point of view. To overcome this challenge, another transmission technique based on orthogonal space time bloc codes (OSTBCs) with transmit antenna selection (TAS) is proposed. This scheme uses the full-rate full-diversity Alamouti scheme in an underlay CR scenario in order to maximize the secondary's transmission rate.

While the solutions discussed above offer a considerable improvement in the performance of spectrum sharing systems, they generally experience a high overall system complexity and are not optimized to meet the tradeoff between spectral efficiency and energy efficiency. In order to address this issue, we consider using spatial modulation (SM) in order to come with a spectrum sharing system optimized in terms spectral efficiency and energy efficiency. Indeed, SM can be seen as one of the emerging and promising new technologies optimizing the communication system while reducing the energy consumption thanks to the use of a single radio frequency (RF) chain for transmission. In this context, we propose the adaptive spatial modulation (ASM) scheme using AM in order to improve the spectral efficiency of SM. We also extend ASM to spectrum sharing systems by proposing a number of ASM-CR schemes aiming at improving the performance of these systems in terms of spectral efficiency and energy efficiency.

While the use of a single RF-chain improves the energy efficiency of the above schemes, the RF-chain switching process between different transmissions comes with additional complexity and implementation issues. To resolve these issues, we use the concept of reconfigurable antennas (RAs) in order to improve the performance of space shift keying (SSK). In this context, employing RAs with SSK instead of conventional antennas allows for implementing only one RF chain and selecting different antenna-states for transmission without the need for RF switching. Moreover, the reconfigurable properties of RAs can be used as additional degrees of freedom in order to enhance the performance of SSK in terms of throughput, system complexity, and error performance. These RAs-based schemes are also extended to spectrum sharing systems in order to improve the capacity of the secondary link while reducing the energy consumption and the implementation complexity of the SU.

In summary, we propose in this thesis several adaptive transmission schemes for spectrum sharing systems. The performance of each of these schemes is confirmed via Monte-Carlo simulations and analytical results and is shown to offer different tradeoffs in terms of spectral efficiency, energy efficiency, reliability, and implementation complexity. In this context, these proposed schemes offer different solutions in order to improve the performance of underlay cognitive radio systems.