Channel Scattering and System Implementation of Underwater Wireless Optical Communication
Gong, Yingda (2020-04-28)
Channel Scattering and System Implementation of Underwater Wireless Optical Communication
(28.04.2020)
Julkaisu on tekijänoikeussäännösten alainen. Teosta voi lukea ja tulostaa henkilökohtaista käyttöä varten. Käyttö kaupallisiin tarkoituksiin on kielletty.
suljettu
Julkaisun pysyvä osoite on:
https://urn.fi/URN:NBN:fi-fe2020062345497
https://urn.fi/URN:NBN:fi-fe2020062345497
Tiivistelmä
In recent years, Underwater Wireless Optical Communication (UWOC) has made great progress and become one of the hotspots of wireless communication research. However, the channel characteristics of underwater optical communication are relatively complicated. Due to the influence of suspended particles, suspended plants and the like in seawater, the movement of photons in water may undergo multiple scattering, which may cause severe energy loss and time dispersion and has a great influence on the performance of the communication system.
This thesis mainly studies the channel scattering and system implementation and improvement of UWOC, which is mainly divided into the following three parts：Firstly, we study the channel characteristics of UWOC. To study the characteristic of multiple scattering channels in UWOC, we simulate this channel by Monte Carlo method. By analyzing the position and energy status of a vast amount of photons during the scattering procedure, we could figure out the channel state information and specify the transmission properties of photons in water. Then we design an optimized orthogonal frequency division multiplexing (OFDM) communication system and apply it in a practical UWOC experiment. A comparison experiment of system BER performance in different water qualities are conducted to verify the influence of multiple scattering in system performance. Besides, we provide improved OFDM underwater optical communication system based on deep neural network (DNN). we use a five-layer deep neural network at the receiver end for direct channel estimation and symbolic output. We also applied the DNN-OFDM system to the experimental environment and test the performance. The experimental results show that under the constraints of specific water quality and communication system conditions, DNN-OFDM system has certain advantages in antichannel scattering, and the system performance is improved to some extent.
This thesis mainly studies the channel scattering and system implementation and improvement of UWOC, which is mainly divided into the following three parts：Firstly, we study the channel characteristics of UWOC. To study the characteristic of multiple scattering channels in UWOC, we simulate this channel by Monte Carlo method. By analyzing the position and energy status of a vast amount of photons during the scattering procedure, we could figure out the channel state information and specify the transmission properties of photons in water. Then we design an optimized orthogonal frequency division multiplexing (OFDM) communication system and apply it in a practical UWOC experiment. A comparison experiment of system BER performance in different water qualities are conducted to verify the influence of multiple scattering in system performance. Besides, we provide improved OFDM underwater optical communication system based on deep neural network (DNN). we use a five-layer deep neural network at the receiver end for direct channel estimation and symbolic output. We also applied the DNN-OFDM system to the experimental environment and test the performance. The experimental results show that under the constraints of specific water quality and communication system conditions, DNN-OFDM system has certain advantages in antichannel scattering, and the system performance is improved to some extent.