Outage prediction modelling of terrestrial free space optics for tropical region

• Main Author: Muftah Shumani, Mohamed Mahmud
• Format: Thesis
• Language: English; English
• Published: 2019
• Subjects: T Technology (General)
• Online Access: http://eprints.uthm.edu.my/75/1/24p%20MOHAMED%20MAHMUD%20MUFTAH%20SHUMANI.pdf; http://eprints.uthm.edu.my/75/2/MOHAMED%20MAHMUD%20MUFTAH%20SHUMANI%20WATERMARK.pdf; http://eprints.uthm.edu.my/75/

In modern times, Free Space Optical (FSO) communication technology has become an alternative way to radio links and optical cables. FSO links provide gigabit per second data rates and unlimited bandwidth, however its availability can easily be affected by fading due to different meteorological conditions such as fog, haze, snow and rain. In tropical regions like Malaysia, haze and rain are the most common impairment factors to FSO links. These weather conditions limit the visibility, causing high attenuation to the optical signal. This study evaluated the performance of a terrestrial FSO link under tropical climate conditions. The proposed FSO availability prediction model was used to predict the availability of FSO link over path length ranges of up to 5 km was verified under two different weather conditions for four places in Malaysia. The FSO availability prediction model was developed as a function of link distance and signal-to-noise ratio for intensity modulation with direct detection (IM/DD). Benchmarking against ITU-R estimate shows good agreement with data based on tropical climate. The cumulative distribution function (CDF) of the received signal-to-noise ratio (SNR) was used to study the outage performance of the FSO link under hazy weather conditions. The performance analysis was based on 2013, 2014 and 2015 collected visibility data from Malaysian Meteorological Department (MMD) for four stations located at KLIA Sepang, Bayan Lepas, Gong Kedak and Batu Pahat. The study proposes a new link model as a function of SNR and distance. Measured data were analysed to predict attenuation on FSO systems with two categories of visibility, namely; Ijaz Model for visibility of less than 1 km and Kim Model for visibility greater than 1 km were used. Rain intensity was measured with 1-minute integration time at Batu Pahat station for 2015 and 2016. Visibility data were also collected with one hour basis at all four stations. Based on the measurements of rain rate, the percentage of time exceeding 0.01% level corresponded to nearly 108 mm/hr, while the highest of 168 mm/hr was observed at 0.000193%. The outage probability was more than 10-4 when the employed wavelength was 1550 nm and for SNR value equaled 8 dB, whereas for wavelength of 850 nm, the outage probability was more than 10-3. For 99.7% availability, FSO SNRhaze varied from 2 dB to 28 dB over a link distance of 1 km to 5 km under the impact of haze, while the SNRrain varied from 22 dB to 90 dB over the same link distance for 99.99 % under the impact rain. The proposed availability models were compared with the ITU-R availability estimation. Benchmarking the proposed outage probability model with the Basahel model and estimated data from the Malaysia Meteorological Department.