Abstract
Powerful and reliable estimation of transmission parameters is an indispensable task in each receiver unit — not only for radio frequency, but also for optical wireless communication systems. In this context, the signal-to-noise ratio (SNR) plays an eminent role, especially for adaptive scenarios. Assuming a bandlimited optical intensity channel, which requires a unipolar waveform design, an algorithm for SNR estimation is developed in this paper, which requires no knowledge of the transmitted data. This non-data-aided approach benefits to a great extent from the fact that very long observation windows of payload symbols might be used for the estimation process to increase the accuracy of the result; this is in striking contrast to a data-aided approach based on pilot symbols reducing the spectral efficiency of a communication link. Since maximum likelihood, moment-based or decision-directed algorithms are not considered for complexity and performance reasons, an expectation-maximization solution is introduced whose error performance is close to the Cramer-Rao lower bound as the theoretical limit, which has been derived as well.
Originalsprache | englisch |
---|---|
Aufsatznummer | 802 |
Seitenumfang | 13 |
Fachzeitschrift | Sensors |
Jahrgang | 23 |
Ausgabenummer | 2 |
DOIs | |
Publikationsstatus | Veröffentlicht - Jan. 2023 |
ASJC Scopus subject areas
- Analytische Chemie
- Information systems
- Instrumentierung
- Atom- und Molekularphysik sowie Optik
- Elektrotechnik und Elektronik
- Biochemie
Fields of Expertise
- Information, Communication & Computing
Treatment code (Nähere Zuordnung)
- Theoretical