Effects of Carbon Dioxide on Different PSK Modulation Formats of Optical Wireless Communications Using USRP

Transcription

1 Effects of Carbon Dioxide on Different PSK Modulation Formats of Optical Wireless Communications Using USRP Jan Vitasek 1, Radek Martinek, Ales Vanderka 1, Lukas Hajek 1, Jan Latal 1, Zhenli Lu 3,4, and Bin Li 4 1 Department of Telecommunications, Faculty of Electrical Engineering and Computer Science, VSB-Technical University of Ostrava, 17. listopadu 15, Ostrava, Czech Republic Department of Cybernetics and Biomedical Engineering, Faculty of Electrical Engineering and Computer Science, VSB-Technical University of Ostrava, 17. listopadu 15, Ostrava, Czech Republic 3 School of Electrical Engineering and Automation, Changshu Institute of Technology, 15500, Changshu, China 4 State key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, , China Tel: , Fax: , jan.vitasek@vsb.cz ABSTRACT This article explores the influence of the concentrated carbon dioxide (CO ) gas on the different Phase Shift Keying (PSK) modulations used for Optical Wireless Communication (OWC) links. In this work, the OWC consisted of a laser diode, a photodetector, and a transmission medium. The Universal Software Radio Peripheral (USRP) unit created and evaluated the PSK modulations. This unit modulated the laser diode on the transmitter side and demodulated the data which were received by the photodetector on the receiver side. The CO gas was emitted as a consequence of sublimation of the solid carbon dioxide in hot water. The measurements were done in a closed box in laboratory. The CO gas is a one element of smog which pollutes cities now and then. Our experimental results with CO gas could help to choose a suitable modulation scheme for OWC which are installed in cities with smog-polluted atmospheric conditions. Keywords: carbon dioxide, optical wireless communications, phase shift keying modulations, smog, USRP unit. 1. INTRODUCTION Optical Wireless Communications (OWC) links are characteristically very similar to radiofrequency (RF) links. The RF links are still used very often, but interferences limit their transmission and capacity. OWC links have the potential to gradually replace some RF systems. They offer a higher data transfer rate without interference, could use non-licensed bandwidths, and are resistant to Electromagnetic Interference (EMI). Additionally, OWC can co-exist with RF links. Other advantages of OWC are: quick installation and information security. The components used in these links are: low cost, small size, and consume less power [1], []. These components emerged from the development of optoelectronics for fiber networks during the last decade. Despite their many qualities, OWC links have their disadvantages. The presence of people or objects between the transmitter and receiver can cause an extinguishment of the Line of Sight (LOS) and can attenuate or even interrupt the received signal. OWC links usually work in an environment where there are other light sources, called as the ambient light. These light sources can cause noise in the photodetector, which in turn limits the data transfer rate and transmission range. Another disadvantage of the OWC links is that their transmission medium, namely the atmosphere, with variable conditions such as fog, rain, snow, and wind. These conditions can drastically attenuate the transmitted signal and consequently reduce their data rate and limit their transmission range. Increasing the optical power in order to overcoming these effects is not a feasible approach as they could cause human safety concerns. Too high optical power could irreversibly damage human tissues [1], []. The greatest problems for OWC links are atmospheric turbulences and fog. Disorderly atmospheric cells originate during turbulences. These cells have different refractive indexes. The changes in refractive indexes in turn cause bending of the light beam, which leads to fluctuations in the received optical power [3], [4], and [5]. Aerosol particles, which mostly create a fog, strongly influence transmission of light beams. The optical power of light beams is scattered by aerosol particles therefore only a part of the transmitted power reaches the receiver. Most of the power is scattered by particles, comparable in size to those of the wavelength of the transmitted light beam [6]. This scattering is called Mie scattering. There is also Rayleigh scattering. Rayleigh scattering occurs when the optical beam interacts with particles which are smaller (10%) than the wavelength. The air molecules (also CO ) have this size [1], []. This scattering usually causes shift of wavelengths but it also attenuates the optical beams because the optical beam is scattered. This article deals with the influences of CO molecules on light beams. The paper is organized as follows. Section briefly discusses a variety of modulations, which are used in OWC links. Section 3 describes some real measurements performed for comparison with PSK modulations in the laboratory. Section 4 presents a summary of the obtained results. The discussion and conclusions evaluate the results /16/$ IEEE 1

2 . MODULATIONS Modulation is a very important consideration in OWC links as it influences the data transmission features (abilities) of the whole system. One of the most popular modulation techniques is Intensity Modulation (IM) in which the time response of the information modulates the immediate transmitted optical power. At the receiver, the optical power is converted to an electrical signal by Direct Detection (DD). The value of the output electrical signal depends on the immediate received optical power. This technique is called IM/DD [1], [], [7], [8], and [9]. Another frequently used modulation format is On-Off Keying (OOK). It is the simplest form of digital modulation, it has a good bandwidth and it is easy to apply. In the OOK modulation scheme, a logical value of 1 turns on a light source and a logical value of 0 turns it off. This modulation often uses the Non-Return to Zero (NRZ) or Return to Zero (RZ) line codes [1], [9], and [10]. Other widely used modulation formats are Phase Shift Keying (PSK), Pulse Width Modulation (PWM), and Pulse Position Modulation (PPM) [11], [1], and [13]. Here we focus on comparison of different PSK modulation formats generated by the Universal Software Radio Peripheral (USRP) unit [15]. 3. MEASUREMENT AND EXPERIMENTAL SET-UP 3.1 Arrangement of the Components Several components were needed to make the experimental measurements possible. The measurements were performed in a plexiglass box with s length of.5 m. The light beam source was a laser diode ML95B45F, with a wavelength 1550 nm and a maximal power of 5 mw. The laser diode was placed inside a Temperature-Controlled Laser Diode Mount TCLDM9. The light beam was collimated by using a lens. The benchtop LD Current Controller LDC0C, and the bench top Temperature Controller TED00C were used for supplying and controlling the laser diode. On the receiver side, there was a photodetector InGaAs Fixed Gain Detector PDA10CF-EC. The photodetector bandwidth was 150 MHz. The distance between the laser diode and the photodetector was 73 cm, and the laser diode and the photodetector were placed inside the box. Other essential building blocks were the USRP-90 Software Radio units, which were connected to the Temperature Controlled Laser Diode Mount (TCLDM9) and to the photodetector (PDA10CF-EC). The first USRP unit generated the modulation signal, which was fed to the TCLDM9 and controlled the output of the laser diode. The other USRP unit processed the signal received by the photodetector. Both USRP units were connected to each other as well as to a PC. The LabVIEW application software was used to control the units. In this application we could select different modulation formats and set parameters like Sampling Rate, Carrier Frequency, and Gain. We selected and gradually generated the following PSK modulations: BPSK, QPSK, 8PSK, 16PSK, π/4 DQPSK. 3. The Evaluation Criterion - MER The parameter Modulation Error Ratio (MER) determines the modulation error. This parameter measures the distance of an actually received symbol to the ideal position of the symbol. The MER is expressed in db. The MER over the number of symbols N is defined as follows [13]: MER = j= 1 N ' ' ( I j + Qj ) j= 1 N ' ' ( I j I j) + ( Qj Qj), (1) where I j is the I component of the j-th symbol received, Q j is the Q component of the j-th symbol received, I j is the ideal I component of the j-th symbol received, and Q j is the ideal Q component of the j-th symbol received. Figure 1. The modulation error ratio (MER).

3 If a constellation diagram is used to plot the landing points of a given symbol over time, the resulting display forms a small cloud of symbol landing points rather than a single point. MER is the ratio of average symbol power to average error power. MER is a measure of how fuzzy the symbol points of a constellation are. It deals for the MER, the higher the number, the better. The MER is somewhat analogous to the baseband signalto-noise ratio. 3.3 Sublimation of Solid Carbon Dioxide The dry ice is a solid phase of carbon dioxide, which emerges from desublimation during cooling under C. Dropping the solid carbon dioxide in hot water causes rapid sublimation of the solid carbon dioxide and very concentrated carbon dioxide (CO ) gas is generated. The CO is heavier than air and sticks to the bottom of the box. The concentration gradually and spontaneously decreases with increasing time [14]. The sublimation of solid carbon dioxide at different levels could be considered as certain kinds of smog (CO is an element of the smog). As many metropolitan cities in the world are burdened by smog pollution, this research could potentially contribute to a better understanding of optical communications in metropolitan areas. 3.4 The Measurement Procedure After completion of the experimental set-up described above, measurements were carried out in the laboratory. The laser diode controllers were turned on, the USRP units generated and received the test signals, and the communication link was established and fully functional. A large container holding 1.7 litters of warm water was placed behind the photodetector. Subsequently, 10 g of the solid carbon dioxide was dropped into the warm water. Sublimation of the dry ice occurred immediately (see Fig. ). A fan blowing air inside the box then dispersed the generated dense CO. Thereby the OWC link was interrupted. As the concentration of the CO gas gradually decreased, the MER parameter increased and the connection became more functional. For each of the modulation formats mentioned above, we measured the time for which the original communication conditions were recovered. Figure. Experimental set-up to perform measurement in the laboratory. 4. MEASUREMENT EVALUATION The LabVIEW application software recorded the MER values together with time. This information enabled us to generate graphs showing the behaviour of the MER under the CO influence. At the beginning of each graph, considerable fall in the value of the MER was evident (see Fig. 3). This sudden fall was caused by the sublimation of the solid carbon dioxide. The MER values increased with gradual decay of the concentrated CO. The time lines of the graphs are not the same because the rise times were different for different PSK modulations. The displayed graphs show how the different PSK modulations respond to an obstacle in communication. The 16PSK modulation scheme was suppressed (attenuated) the shortest time; however the variance of the MER was large. This was caused as this modulation had the largest number of states. The π/4 DQPSK modulation behaved very well under the influence of the CO gas. The restoration began quite soon and the behaviour was smooth. Contrary to other cases, the QPSK modulation was strongly affected by the concentrated CO, and the BPSK modulation regenerated a long time. 3

4 Figure 3. The MER values for different PSK modulations. Another evaluation criterion could be the rise time. The rise time is determined as the time required for the increase from 10% to 90% of the maximal amplitude. Table 1 summarizes the results. From this table it is evident that the rise time gradually decreased when the modulations were changed. The simpler the modulation, the longer the rise time. Table 1. Rise times. Modulation format Total attenuation (s) Rise time (s) BPSK QPSK PSK PSK π/4 DQPSK DISCUSSION The most resistant modulations in the presence of smog could be the 16PSK or π/4 DQPSK. The π/4 DQPSK modulation had the shortest rise time of all the PSK modulations. It could be concluded that simpler modulations showed worse results, which were unexpected. We believe that these results are caused by different constellation diagrams of modulations. In the BPSK, QPSK, and 8PSK modulations, a change of state means a zero crossing on the real and imaginary axes. The π/4 DQPSK modulation does not have the zero crossing (see Fig. 4). It is obvious from Fig. 3 that the π/4 DQPSK modulation is more resistant to concentrated CO gas than the M-PSK modulations. Figure 4. Comparison of the QPSK and π/4 DQPSK modulations. 4

5 Based on the observed results, we plan further experimentation with shifted M-PSK modulations, which could be more resistant to the concentrated CO based on the assumptions of the zero crossing on the real and imaginary axes. 6. CONCLUSIONS In this article we explored the influence of concentrated CO gas on different PSK modulations. The concentrated CO gas was generated by the sublimation of solid carbon dioxide in hot water. The concentrated CO gas could be partly considered as smog. It was unexpected to observe that simpler modulations exhibited worse resistance under the influence of concentrated CO compared to the more complex modulations. This research will continue by performing more tests and comparing other modulation formats. ACKNOWLEDGEMENTS The authors would like to acknowledge the financial support of the Ministry of Education, Youth and Sports of the Czech Republic under Projects No. SP016/149, SP016/146 and SP016/151 of VSB Technical University of Ostrava, Czech Republic. Our research was also supported by Projects No. VG , TA , and TA The work has been partially supported by Project No. CZ.1.07/.3.00/ This work was also supported by the National Science Foundation of China under Grant and by Chinese State Key Laboratory of Robotics under Grant 014-O08. REFERENCES [1] R. Ramirez-Iniguez, S. M. Idrus, and Z. Sun: Optical Wireless Communications: IR for Wireless Connectivity, CRC Press, Boca Raton, 008. [] A. K. Majumdar and J. C. Ricklin: Free-Space Laser Communications, Springer, New York, 008. [3] S. Rajbhandari, et al.: Experimental error performance of modulation schemes under a controlled laboratory turbulence FSO channel, J. of Lightwave Technol., vol. 33, no. 1, pp , Jan [4] B. Barua and D. Barua: Evaluate the performance of FSO communication link with different modulation technique under turbulent condition, in Proc. 14th International Conference on Computer and Information Technology (ICCIT), Dhaka, Bangladesh, Dec. 011, pp [5] T. Y. Elganimi: Studying the BER performance, power- and bandwidth-efficiency for FSO communication systems under various modulation schemes, in Proc. IEEE Jordan Conference on Applied Electrical Engineering and Computing Technologies (AEECT), Amman, Jordan, Dec. 013, pp [6] A. Vaderka et al.: Design, simulation and testing of the OOK NRZ modulation format for free space optic communication in a simulation box, Advances in Electrical and Electronic Engineering, vol. 1, no. 6, pp , Dec [7] W.O. Popoola and Z. Ghassemlooy: BPSK subcarrier intensity modulated free space optical communications in atmospheric turbulence, J. of Lightwave Technol., vol. 7, no. 8, pp , Apr [8] J. Li et al.: Optical communication using subcarrier PSK intensity modulation through atmospheric turbulence channels, IEEE Transaction on Communication, vol. 55, no. 8, pp , Aug [9] S. Arnon et al.: Advanced Optical Wireless Communication Systems, Cambridge University Press, Cambridge, 01. [10] Y. Cheng et al.: Combination of subcarrier intensity modulation and spatial modulation for OWC environment, in Proc. International Conference on ICT Convergence (ICTC), Jeju, South Korea, Oct. 013, pp [11] A. M. Mbah et al.: Performance evaluation of digital pulse position modulation for wavelength division multiplexing FSO systems impaired by interchannel crosstalk, IET Optoelectronics, vol. 8, no. 6, pp , Dec [1] R. Agalliu and M. Lucki: Benefits and limits of modulation formats for optical communications, Advances in Electrical and Electronic Engineering, vol. 1, no., pp , Jun [13] National Instruments: Modulation Error Ratio (MER) and Error Vector Magnitude (EVM), [14] A. Zimmerman Jones, What is Dry Ice? [15] R. Martinek and J. Zidek: The real implementation of NLMS channel equalizer into the system of software defined radio, Advances in Electrical and Electronic Engineering, vol. 10, no. 5, pp , Dec