Abstract:
The open nature of radio propagation enables ubiquitous wireless communication,
which allows seamless data transmission. However, unauthorized users may pose
a threat to the security of the data being transmitted leading to network vulnerabilities such as hacking, eavesdropping, and jamming. Physical layer security (PLS)
has been identified as one of the promising security approaches to safeguard the
transmission in a wireless network. The intention is to augment the computationally demanding and complex cryptographic algorithms. Due to the random nature
of the transmission channel such as scattering, it is possible for confidential and
authentic signal transmission in the physical layer. The aim of this thesis is to propose and investigate mathematical optimization techniques for PLS. Firstly, optimal
beamformers for the information signal and the jamming signal were designed with
the aid of an intelligent reflecting surface (IRS) to significantly improve the secrecy
rate of the legitimate transmission in the presence of an eavesdropper. Secondly, the
problem of maximizing the energy harvested by energy receivers while satisfying
the secrecy constraint at the legitimate receiver was considered. This was achieved
through the use of an alternating algorithm to optimize the beamforming of information to the legitimate receiver and the manifold optimization approach to design
optimal phase shifters of the IRS. The results shows that the use of both IRS and
jamming offers an improvement in the secrecy rate by about 28 % to 33 % than in
the conventional PLS schemes. It was also shown that increasing the number of IRS
elements offers an improvement in the secrecy rate level. The second set of results
showed that the use of IRS significantly improves the secrecy rate while achieving
the minimum energy constraint set for the system
