English Home INDEX
Article-010
Project Report
A Study on Improving the Link Quality
of the Free Space Optical Transmission System


Tomoharu YOKOTA     Kazuhiko ASAKA     Hideki MATSUMOTO
Tsuyoshi YAMAGUCHI     Chiharu KASAI

  1. Introduction    
  2. Configuration of study system
  3. Weather dependency of the Link Quality of the Free Space Optical Transmission System
  4. Simulation of optical propagation characteristics
  5. Conclusion

3 Weather dependency of the Link Quality of the Free Space Optical Transmission System

(1) Weather condition and visibility

The standard visibility is defined as the distance where the black object is discriminated from the other on the horizon in the sunshine in the uniform sphere. Visibility is difficult to be measured directly and is calculated from the data measured by the optical equipment like the visibility meter now used. The distance where the light (white light of color temperature 2700K) is reduced to 5% after the transmission through the free space is defined as MOR (Meteorological Optical Range) and is used as similarly as the visibility of the meteorological term.

(a) Visibility distribution dependent on the weather condition

The change of the visibility affected by weather condition was measured by the weather & visibility meter.

In Table 3-1, the weather condition is divided into four conditions of no rain, haze, fine rain and rain, and the percentage of the visibility under each condition is shown.

In this measurement the visibility less than 1km is only under the condition of rain and the percentage is rare, only 0.32%. By the way the strongest rainfall was 42.6mm/hr and the visibility was 0.16km in this measurement.

Figure 3-1 shows the relation between rainfall and visibility. The visibility gets so shorter as that the rainfall is stronger, but it is difficult to get the visibility directly from the rainfall. Accordingly it is necessary to get the propagation loss of the beam from the visibility in order to make a link design.

Table 3-1  Percentage of Visibility under Weather Condition

Figure 3-1 Relation between Rainfall and Visibility

(b) Result of measurement of visibility, reception power and bit error rate

Figure 3-2 shows the result of the measurement of visibility, reception power and bit error rate in the rain at Mita for a month of December, 2001.

Figure 3-2  Result of Measurement of Visibility, Reception Power and
Bit Error Rate in December, 2001

The visibility changes similarly at Mita and Takanawa macroscopically. The visibility of both points is much reduced at the weak rain (about 3mm/hr in this figure) and the changes of rainfall and visibility coincide with each other. But the visibility is also reduced in no rain and this will be caused by fog and haze. This data verifies the relation between the visibility and the weather condition on the whole.

The anxiety is bit error rate when the Free Space Optical Transmission System is setup. In this measurement the worst bit error rate was 4~10 -8 and it is sufficiently good for the communication link though the weather was relatively fine. But the change of bit error rate does not always coincide with that of rainfall and visibility, and appears rather intermittently and continuously. The reason is not clear now.

(2) Propagation loss and visibility

(a) Estimation formula of propagation loss

The transmission quality of the Free Space Optical Transmission System is affected by aerosol and weather condition in the air. The visibility is generally used in order to estimate the influence and is defined as the distance where the black object is discriminated from the other on the horizon in the sunshine in the uniform sphere. It is measured by weather bureaus in every place and is the data for the visible light. The visibility data for the visible light is used for the estimation of propagation loss because the infrared of about 800nm is near to the visible light. The estimation formula L (dB/km) of the air using the visibility is expressed in two types of formulae. The one is formula (1) using the visibility, the other is formula (2) including the influence of wave length.

Enot including the influence of wave length

(dB/km)   formula (1)

Eincluding the influence of wave length

(dB/km)   formula (2)

q = 0.585 V 1/3
L : propagation loss (dB/km) V : visibility (km) ƒΙ : wave length (nm)

(b) Relation between visibility and reception power

Figure 3-2 shows the relation between visibility and reception power that was measured for about 6 months (October, 2001 to March, 2002) between Mita and Takanawa 1.2km apart.

Figure 3-2 Relation between Visibility and Reception Power (Mita)

The black line means the theoretical curve of formula (1) and the red line means the theoretical curve of formula (2). They were calculated based on transmit power -33dBm and distance 1.2km.

Comparing with the measured data and the theoretical curves, the theoretical curves go along the lower limit value of the measured data. Accordingly it is verified that both formulae are suitable for the estimation.

The results of formula (1) and (2) have not so large difference under the condition of transmission distance 2km and wave length about 800nm, and the theoretical result is the same even if any formula is used for the link design. The simple formula (1) is easy to be used for the estimation of propagation loss.

(3) Propagation loss and transmission quality

(a) Relation between reception power and bit error rate

Figure 3-3 shows the relation between reception power and bit error rate.  The link-down did not happen and the worst bit error rate was 4.3~10 -8 for about three months since October 24, 2001 to January 27, 2002. The link quality is very high for wireless system.

Figure 3-3 Relation between Reception Power and Bit Error Rate

(b) Scintillation

Fluctuation of reception power level caused by the density deviation of the air is scintillation.

Table 3-2 shows the example of scintillation obtained in this measurement.

Date the lowest value
of visibility
weather the highest temperature the lowest temperature the worst value of BER scintillation
2001/12/1 20km (Mita)
20km (Takanawa)
 no rain    100% 12.3Ž 5.9Ž 1.70E-08 4`8dB
2001/12/9 6.5km (Mita)
5.7km (Takanawa)
 no rain    100% 12.6Ž 5.7Ž 1.00E-08 3`8dB
2001/12/1 368km (Mita)
272km (Takanawa)
 no rain    55.6%
 haze       44.4%
13.1Ž 7.0Ž 5.80E-09 1`3dB
2001/12/2 15.1km (Mita)
6.25km (Takanawa)
 no rain    100% 15.2Ž 3.4Ž 4.30E-08 1`2dB

Table 3-2 Measurement of Scintillation

Back Next

 
Mail to   icsa@arib.or.jp
 
Infrared Communication Systems Association (ICSA)
Nittochi Bldg. 14F, 1-4-1, Kasumigaseki, Chiyoda-ku, Tokyo, JAPAN  (100-0013)
TELF03-5510-8596    FAX:03-3592-1103