Embedded Files
Results were taken according to different conditions of the medium such as muddy water, bubble water ( for turbulence ) and the Bit Error Rate was analyzed in each case.
Results for clear water and mud water
Results for clear water and mud water
Fig 1 : Comparison of BER in clear water and mud water
Mud water with medium concentration was used. It was observed that at the presence of Mud water, noise is increased. The BER has gone up for the higher distances.
BER shows the minimum at 70 cm in both cases and below that value the graph shows the same behavior. This is due to the focusing issues when the camera is being more closer to the transmitter. The effect from the focusing causes more influence in the BER results.
Although mud water was used instead of clear water, the BER was still under 10% up to around 110 cm of transmission distance indicating the system is reliable even in extreme conditions.
Results for different turbulent conditions
Results for different turbulent conditions
Fig 2 : Comparison of BER for turbulence level 1 and still water
Turbulence Level 1 – Introduce a bubble every 20 seconds and disturbing the water surface.
The size of the introduced bubble increases as the pressure reduces when going up to the surface. due to this for longer distance communication link the BER are high.
As before, the plot under turbulence almost behaves similarly like that for still water up to around 70 cm.
In this case the turbulence conditions make BER value above 10%. Thus the reliability is somewhat lower than the mud water case.
Fig 3 : Comparison of BER for turbulence level 2 and still water
Turbulence Level 2 - Introduce a bubble every 10 seconds and disturbing the water surface.
In this case it can be observed that the graph behaves very differently than the prior cases.
With higher transmission distances, the turbulence is high and as a result the BER is very high leading to unacceptable limits. For the transmission distance of 110 cm, the BER is around 30% which is unfavorable for reliable transmission.
Conclusions
Conclusions
For reducing the BER we are proposing these methods.
- Analyze the data to get an average radius when identifying the LED positions in the decoding process. This will ignore the blobs with the errors.
- Use multiple images to identify the final positions of the LED when decoding. This will lead to high accuracy. But the data rates will be less as we are using multiple frames to recreate a frame. there is always a tradeoff between the accuracy and the data rates.
- use more directive LEDs ( Less viewing angle ) and a high tech camera as the receiver
Google Sites
Report abuse