Modulations for Terahertz Band Communications: Joint Analysis of Phase Noise Impact and PAPR Effects

The choice of modulation techniques for next-generation communications in the sub-terahertz and terahertz bands remains largely unresolved. A variety of traditional and new schemes show promise, but the question remains open on the illustrative comparison process for realistic terahertz systems. While there are some preliminary studies in this area, we emphasize that the peculiarities of terahertz hardware necessitate a scheme that (i) is resistant to system-wide phase noise (PN) and (ii) has a low peak-to-average power ratio (PAPR). Therefore, in this article, we present a comprehensive methodology to carefully model and jointly study the impacts of PN and PAPR on the performance of candidate modulations for terahertz links. We first deliver a mathematical model for sub-terahertz/terahertz phase noise impairments at 130 GHz, 225 GHz, and 1.02 THz based on measurements from actual terahertz hardware. We then introduce the PAPR penalty - an approach for fair comparison of bit error rate (BER) and spectral efficiency (SE) of modulation schemes with different PAPRs. We finally combine these two effects to comprehensively study the characteristics of single and multi-carrier modulation schemes for terahertz communications. Our study reveals that analyzing PAPR and PN jointly is paramount: accounting for only one leads to major deviations in the numerical results and misleading conclusions on best modulation choice(s). The delivered framework and evaluation should facilitate further studies, leading to a well-motivated selection of the most suitable modulation scheme(s) for prospective sub-terahertz and terahertz radio systems.