Physical waveform research for beyond 52.6 GHz in 5G NR networks

Historically, in order to fulfil all the requirements for the new generations, the frequency bands have been expanded from generation to generation. In particular for the fifth generation new radio (5G NR), where the use of millimetre wave (mmWave) frequencies can offer higher bandwidths, communications in frequencies beyond 52.6 GHz seem really promising and are now under discussion in the 3rd Generation Partnership Project (3GPP) standardisation for the 5G NR future releases. More concretely, both academia and industry are doing research for the frequency range between 52.6 GHz and 114.25 GHz. The reasons why communications beyond 52.6 GHz are interesting is because in those frequencies, high data rate and low latency can be provided due to the large and contiguous channel bandwidth that is available. Also, new use cases can be explored in this frequency range since high accuracy positioning is possible at higher carrier frequencies, such as Orthogonal Frequency Division Multiplexing (OFDM) radar sensing, that allows new kinds of services. New challenges appear at higher frequencies, or other implementation issues that were not critical in lower frequencies start to become dominant and have to be taken into consideration while defining the new modulations and comparing the possible candidates. The main problems that have to be faced at higher frequencies are the poor propagation conditions (propagation losses are higher than in frequencies below 52.6 GHz), and the radio frequency (RF) impairments that electronic components may have, especially the lower power amplifier (PA) efficiency. Therefore, in order to have a good signal quality, if the peak to average power ratio (PAPR) of the original signal is high, the back-off should be high to make the PA work in the linear region. Thus, the waveform design has to be focused on generating signals with “nearly constant” envelope in order to be able to work closer to the saturation zone of the amplifier without distorting the signal. Also, another problem that has to be taken into account is the large phase noise (PN) present at these frequencies. The main goal of this work is the comparison between different modulations for discrete Fourier transform (DFT) Spread OFDM (DFTs-OFDM) in order to find a suitable candidate that can be part of the 5G NR communications for carrier frequencies beyond 52.6 GHz, and targeting specially low spectral efficiency (between 1 and 2 bps/Hz). Therefore, the main modulation references are pulse shaped π/2- binary phase shift keying (BPSK) and quadrature phase shift keying (QPSK) supported in 5G NR Release 15 up link (UL). In this Thesis, several modulation candidates have been tested under realistic conditions by using a 3GPP 5G NR compliant radio link simulator in Matlab. In order to find the best candidate, the waveforms should be able to present good characteristics that can overcome the problems present in mmWave communications. The main contribution of this thesis is to propose a new "constrained" phase shift keying (PSK) modulation, called CPSK, which applies a constraint to the symbols that are transmitted in order to reduce the PAPR of the signal. The results have shown that under the mmWave communications conditions (such as low PA efficiency and high PN), the new CPSK modulations can provide significant improvement with the evaluated PA model when compared to QPSK modulation, and together with extensive link level performance evaluations, a clear link budget gain can also be shown for specific CPSK modulation candidates and pulse shaped π/2-BPSK