Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Multilevel modulation scheme using the overlapping of two light sources for visible light communication with mobile phone camera

Open Access Open Access

Abstract

Visible light communication (VLC) with light emitting diodes (LEDs) is an emerging technology for 5G wireless communications. Recently, using complementary metal-oxide-semiconductor (CMOS) image sensor as VLC receiver is developed owing to its flexibility and low-cost. However, two illumination levels such as on-off keying (OOK) signal are used. To improve the system throughput and reduce complexity of the hardware design, in this paper, we propose and experimentally demonstrate a multilevel modulation scheme for VLC system utilizing the overlapping of two light sources for the first time, and the two light sources are modulated by an OOK and a Manchester signal respectively. At the receiver, a CMOS camera can demodulate the Manchester and the OOK signal simultaneously. Meanwhile, a low-pass filter (LPF) is used to enhance the system performance. The experimental results demonstrate that the proposed multilevel modulation scheme can achieve a net data rate of 4.32 kbit/s.

© 2017 Optical Society of America

1. Introduction

Visible light communication (VLC) as an optical wireless communication technology enables a potential wireless access option for 5G wireless communications [1–3]. The VLC with light-emitting diodes (LEDs) has gained great attention. It benefits from the illumination and communication simultaneously. Meanwhile, the radio spectrum congestion can be resolved in VLC due to the availability of large spectrum resource from 430 THz to 790 THz [4]. VLC using LEDs has some advantages such as immunity to electromagnetic interference, low power consumption, simple installation and low cost [5]. Nowadays, the positive-intrinsic-negative (PIN) and the avalanche photodiode (APD) are used as receiver (Rx) in most of VLC systems. In addition, embedded complementary metal-oxide semiconductor (CMOS) image sensor can be used as a VLC receiver to provide a flexible and low cost wireless communication [6, 7]. The bright and dark information can record by the image frame. In this way, the data rate is mainly constrained by the frame rate. To increase the data rate, the CMOS image sensor using the rolling shutter effect is proposed and the scanline of image sensor can continuously capture the changes about the “on” and the “off” state of the light source on one frame [8]. It is worth noting that there is a frame-to-frame processing time using the rolling shutter effect of CMOS image sensor, and it will result in the loss of the data [6]. Recently, the VLC systems with the COMS image sensor are focused on the two-level modulation [6], [9–15]. However, the two-level modulation such as OOK signal will restrain the data rate. Hence, to improve the data rate, the multilevel modulation is taken into account. For traditional multilevel modulation scheme, it needs multi-voltage amplitude to realize the different intensities of light at the transmitter. And a complex hardware circuits is used to generate the multiple voltage amplitudes.

In this paper, a multilevel modulation scheme utilizing the overlapping of two light sources is proposed and experimentally demonstrated for the first time. The generation of the multi-level voltage amplitudes does not need a complex hardware circuit. At the transmitter, the two light sources are modulated by the OOK and the Manchester signal respectively. At the receiver, there exists 3-level illumination but the data rate is as same as the 4-level modulation. The hybrid signal including the OOK and the Manchester signal can be demodulated by one CMOS image sensor simultaneously. To enhance the system performance, a low-pass filter (LPF) is used to smooth the signal and avoid the effects of the high-frequency noise.

2. Principle

2.1 Multilevel modulation scheme

OOK modulation is used to deliver binary data by changing “on” and “off” state of the light. It is a simple way that using “on” state of light denotes symbol “1” and “off” state of light denotes symbol “0”. But the illumination suffers the effects from the data as there is a series of “0”. However, Manchester coding is used to avoid the flickering effects, and it contains a transition at the middle of each bit period. The bit “1” is represented by a falling edge and the bit “0” is represented by a rising edge [16]. Thus, at a bit period, there are two symbols for the Manchester coding. To improve the data rate, a multilevel modulation scheme can be used [17]. However, it needs multi-voltage amplitude to realize the different intensities of light. To avoid the complexity of hardware circuit, in the paper, a multilevel modulation scheme based on the overlapping of two light sources is proposed. It can be seen from the Fig. 1, LED1 and LED2 are modulated by the OOK and the Manchester signal respectively. After the overlapping of two light sources, the multilevel modulation signal is obtained.

 figure: Fig. 1

Fig. 1 Principle of multilevel modulation.

Download Full Size | PDF

2.2 Rolling shutter characteristics of CMOS camera

In the paper, a mobile phone camera is used as VLC receiver. In general, most of the CMOS cameras are performed at the rolling shutter mode. In the rolling shutter case, as shown in Fig. 2, each row of pixels are sequentially activated and exposed. When the data rate is lower than the rolling shutter speed and higher than the frame rate, the bright and dark stripes standing for the intensity of the data will be captured by the activated rows of pixels at each image frame [18]. There is an overlapping exposure time at the adjacent rows of pixels [19]. Thus, there is a transition band between the changes of the light state, and it can be seen as the inter symbol interference (ISI) [20]. In addition, the rolling shutter is similar to a sample and hold Analog to Digital Converter (ADC). The τreadout is equal to the sampling interval. After the exposure of the last row of the frame, there is a frame processing time as “blind time” for the CMOS camera which cannot capture any information from the VLC transmitter.

 figure: Fig. 2

Fig. 2 CMOS camera based VLC processing schematic diagrams at the receiver.

Download Full Size | PDF

2.3 Multilevel demodulation scheme

For the demodulation of the Manchester signal, it is obvious that the OOK signal don’t change the edge information at a bit period for the hybrid OOK and Manchester signal. Thus, the demodulation of the Manchester signal is based on the edge information of hybrid signal. For the demodulation of the OOK signal, sampling at the middle of a bit period, each symbol is represented by several rows of pixels. Due to the symbol rate far slower than the row scanning speed of the CMOS camera, the exposure time of overlapping is appeared around each row of pixel. At the middle of bit period, there always exists an edge, which is transition band. As the OOK data is “1”, the transition band is brighter than that of the OOK data is “0”. It is due to the influence of overlapping exposure. Thus, the demodulation of OOK can be sampled at the middle of a bit period. Then, an appropriate threshold is used to judge the logic high and logic low for the demodulation of OOK signal. Consequently, the demodulation of the OOK and the Manchester is independent on each image frame.

3. Experimental setup and results

The experimental setup of VLC with mobile phone camera is shown in Fig. 3. Two sets of data modulated by the OOK and the Manchester are offline generated from personal computer (PC). Then, a field-programmable gate array (FPGA) from the Xilinx Spartan 6 series (xc6slx45) is used to control the LED driver circuit (LED Driver). The generated data is loaded into the single port read-only memory (ROM). The general-purpose input/output (GPIO) pin outputs logic high and logic low, which correspond to the data bit “1” and “0” stored in ROM respectively. The transmitted symbol rate is 14.4 kBaud (40 (data bits/frame) × 2 (symbols/data bit)× 60 (frames/second) ×3 (repeat 3 times) = 14.4 kBaud). Meanwhile, the bright state of LED corresponds to the logic high of GPIO, conversely, the logic low. When the GPIO is directly used to drive the LED, the LED cannot be bright enough. It is due to the restricted current drive of GPIO. Thus, after the FPGA, a LED Driver is placed to increase the current drive and control the brightness of LED. The LED1 and the LED2 are modulated by the OOK and the Manchester signal respectively. Subsequently, two sets of data are separately emitted by the white-light LEDs (Cree XLamp XP-E). A LED light diffuser is used to generate the hybrid signal including the OOK and the Manchester signal. At the receiver, with the overlapping of two light sources, a mobile phone (iPhone 6s) is used to capture the hybrid signal. A 2-minute video is then loaded into the Matlab to demodulate. The video is recorded at the frame rate of 60 fps and resolution of 1080×1920. In addition, a digital light meter (Benetech-GM1020) is used to measure the received light.

 figure: Fig. 3

Fig. 3 Experimental setup of VLC system with mobile phone camera.

Download Full Size | PDF

In the case of Manchester coding, each symbol stands for 1/2 bit. Due to the same sampling clock for both GPIO1 and GPIO2 ports, each bit of OOK signal sends two times consecutively at bit period. Figure 4 depicts the structure of data packet. Each packet consists of three sets of Headers and Payloads. The Header is used for synchronization and clock extraction, and the Payload is the effective data. For Manchester coding, it only exists two consecutive same pulses. Thus, in the Header, four consecutive “1” is enough to synchronize for the combination of the OOK and the Manchester signal. Meanwhile, the “0101” of Manchester coding following the synchronization pulse are used to clock recovery. In addition, at each frame, the Payload from the LED1 and the LED2 consists of 36 data bits. Thus, at the receiver, 72 data bits are captured at each frame by mobile phone camera. The measured frame processing time as “blind time” is about 38% of frame duration. Therefore, the Header and the Payload transmit 3 times successively so that each image frame can contain a complete data packet as shown in Fig. 4.

 figure: Fig. 4

Fig. 4 The structure of data packet.

Download Full Size | PDF

The system process diagram of VLC system with mobile phone camera is shown in Fig. 5. The recorded video is imported into a PC, and processed by offline Matlab. Every image frame from the video is extracted to jpg file that is available for image processing. Then, the image file is converted into gray-scale intensity image. The 256 levels of gray from 0 to 255 represent the brightness from complete dark to complete bright. The received image frames at different illuminance are inserted in Figs. 5(a)-5(c). The experimental scene is illustrated in Fig. 6. All rows of pixels start and end exposure row by row. Hence, a column of gray-scale is selected to demodulate as shown in Fig. 7. In addition, a LPF with filter coefficient h = [1/3, 1/3, 1/3] is used to smooth the data and avoid the appearance of the sharp impulse. The Header information can be found at each image frame. The partial enlarged drawing including the Header is inserted in Fig. 5(c). The decoding process for the OOK and the Manchester signal is executed to obtain the original binary data simultaneously. Finally, the bit error rate (BER) performance is evaluated.

 figure: Fig. 5

Fig. 5 System process diagram of VLC system with camera receiver. (a-c) Image frames received at different illuminances.

Download Full Size | PDF

 figure: Fig. 6

Fig. 6 Experimental scene.

Download Full Size | PDF

 figure: Fig. 7

Fig. 7 Column matrix of grayscale values with LPF.

Download Full Size | PDF

For the decoding of the Manchester signal, acquisition of the rising or the falling edge is important. There are 256 levels in a column of pixels and the selected column gray-scale is defined as y(n) (1n1080). To obtain the edge information, we can take an approximation of derivatives using first order difference approximation as e(n) = y(n + 1) – y(n). There is a rising or a falling edge when e(n) is greater or smaller than 0. Finally, the original binary data is recovered according to the sampling clock extracted from the Header. For the demodulation of the OOK signal, a threshold derived from the third order polynomial fitting as the red line is shown in Fig. 7. It is utilized for the decision of the OOK signal. The data bit is regarded as “1” or “0” according to the sampling point greater or smaller than the threshold. From the Fig. 7, it can be seen that the symbol “1” adjacent to the symbol “2” can achieve a larger gray-scale than that of adjacent to “0”. It exists overlapping exposure as the row of pixels is activated sequentially in rolling shutter mode. Thus, the symbol will be affected by the adjacent symbols.

The BER performance at different illuminance is illustrated in Fig. 8. The OOK and Manchester is derived from the hybrid signal in Fig. 8. The LED1 is modulated by the OOK signal and the LED2 is modulated by the Manchester signal simultaneously. At the receiver, the two kinds of signal can be demodulated using a mobile phone camera simultaneously. Thus, the Payload at each image frame including the OOK and the Manchester is 72 data bits. It indicates that as the increase of illuminance, the better BER performance is obtained. It is due to the signal noise ratio enhanced with the increasing of illuminance. Based on the hybrid signal including the OOK and the Manchester coding, the BER of the OOK signal from the LED1 and the Manchester signal from the LED2 is calculated respectively. From the Fig. 8, it can be seen that the BER performance of the OOK is better than that of the Manchester at the same illuminance. The reason is that, the bandwidth of the Manchester is twice wider than that of the OOK. And the edge is easily interfered by the CMOS image noise and channel noise. Moreover, from the Fig. 8, it can be seen that, the hybrid signal including the OOK and the Manchester can achieve better performance with LPF than that of without LPF. Thus, the LPF can be used to improve the BER performance effectively. In addition, as the illuminance over 3500 illuminance (lux) with the LPF, the BER of the hybrid signal below the 20% forward error correction (FEC) limit can be obtained. Furthermore, in our proposed system, the net data rate can be up to 4.32 kbit/s (72 (bits per frame) ×60 (frames per second) = 4.32 kbit/s).

 figure: Fig. 8

Fig. 8 BER versus the different illuminance.

Download Full Size | PDF

4. Conclusion

In this paper, we proposed and experimentally demonstrated a multilevel modulation scheme for VLC system with mobile phone camera. The two LED light sources were modulated by the OOK and the Manchester signal, respectively, and a mobile phone was used to receive the light overlapped by the two LEDs. In addition, the OOK and the Manchester signal could be demodulated independently. The experimental results showed that the VLC system could achieve a net data rate of 4.32 kbit/s under the 20% FEC limit. Moreover, a LPF was used to enhance the system BER performance effectively. It is believed that the flexible and low-cost system can be applied in vehicle-to-vehicle communication and near field communication.

Acknowledgments

This work was supported in part by National Natural Science Foundation of China under Grant 61377079, Grant 61307087 and Grant 61571188; in part by the Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry of China; and in part by Science and Technology Project of Hunan Province (2016GK2011).

References and links

1. C. X. Wang, F. Haider, X. Gao, X. H. You, Y. Yang, D. Yuan, H. M. Aggoune, H. Haas, S. Fletcher, and E. Hepsaydir, “Cellular architecture and key technologies for 5G wireless communication networks,” IEEE Commun. Mag. 52(2), 122–130 (2014). [CrossRef]  

2. L. U. Khan, “Visible light communication: applications, architecture, standardization and research challenges,” Digit. Commun. Netw. 3(2), 78–88 (2017). [CrossRef]  

3. S. Wu, H. Wang, and C. H. Youn, “Visible light communications for 5G wireless networking systems: from fixed to mobile communications,” IEEE Netw. 28(6), 41–45 (2014). [CrossRef]  

4. D. Tsonev, S. Videv, and H. Haas, “Towards a 100 Gb/s visible light wireless access network,” Opt. Express 23(2), 1627–1637 (2015). [CrossRef]   [PubMed]  

5. X. Huang, S. Chen, Z. Wang, J. Shi, Y. Wang, J. Xiao, and N. Chi, “2.0-Gb/s Visible light link based on adaptive bit allocation OFDM of a single phosphorescent white LED,” IEEE Photonics J. 7(5), 1–8 (2015). [CrossRef]  

6. C. W. Chow, C. Y. Chen, and S. H. Chen, “Enhancement of signal performance in LED visible light communications using mobile phone camera,” IEEE Photonics J. 7(5), 1–7 (2015). [CrossRef]  

7. K. Liang, C. W. Chow, Y. Liu, and C. H. Yeh, “Thresholding schemes for visible light communications with CMOS camera using entropy-based algorithms,” Opt. Express 24(22), 25641–25646 (2016). [CrossRef]   [PubMed]  

8. C. Danakis, M. Afgani, G. Povey, I. Underwood, and H. Haas, “Using a CMOS camera sensor for visible light communication,” in 2012 IEEE Globecom Workshops, 2012, pp. 1244–1248.

9. B. Lin, Z. Ghassemlooy, C. Lin, X. Tang, Y. Li, and S. Zhang, “An indoor visible light positioning system based on optical camera communications,” IEEE Photonics Technol. Lett. 29(7), 579–582 (2017). [CrossRef]  

10. R. Boubezari, H. L. Minh, Z. Ghassemlooy, and A. Bouridane, “Smartphone camera based visible light communication,” J. Lightwave Technol. 34(17), 4121–4127 (2016). [CrossRef]  

11. Z. Ong and W. Y. Chung, “Long range VLC temperature monitoring system using CMOS of mobile device camera,” IEEE Sens. J. 16(6), 1508–1509 (2016). [CrossRef]  

12. Y. Liu, H. Y. Chen, K. Liang, C. W. Hsu, C. W. Chow, and C. H. Yeh, “Visible light communication using receivers of camera image sensor and solar cell,” IEEE Photonics J. 8(1), 1–7 (2016). [CrossRef]  

13. T. Nguyen, C. H. Hong, N. T. Le, and Y. M. Jang, “High-speed asynchronous optical camera communication using LED and rolling shutter camera,” in 2015 Seventh International Conference on Ubiquitous and Future Networks, 2015, pp. 214–219. [CrossRef]  

14. M. Liu, K. Qiu, F. Che, S. Li, B. Hussain, L. Wu, and C. Patrick Yue, “Towards indoor localization using visible light communication for consumer electronic devices,” in 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2014, pp. 143–148. [CrossRef]  

15. Y. Liu, C. W. Chow, K. Liang, H. Y. Chen, C. W. Hsu, C. Y. Chen, and S. H. Chen, “Comparison of thresholding schemes for visible light communication using mobile-phone image sensor,” Opt. Express 24(3), 1973–1978 (2016). [CrossRef]   [PubMed]  

16. A. D. Cheok and L. Yue, “A novel light-sensor-based information transmission system for indoor positioning and navigation,” IEEE Trans. Instrum. Meas. 60(1), 290–299 (2011). [CrossRef]  

17. J. F. Li, Z. T. Huang, R. Q. Zhang, F. X. Zeng, M. Jiang, and Y. F. Ji, “Superposed pulse amplitude modulation for visible light communication,” Opt. Express 21(25), 31006–31011 (2013). [CrossRef]   [PubMed]  

18. T. H. Do and M. Yoo, “Analysis on visible light communication using rolling shutter CMOS sensor,” in 2015 International Conference on Information and Communication Technology Convergence (ICTC), 2015, pp. 755–757. [CrossRef]  

19. C. W. Chow, C. Y. Chen, and S. H. Chen, “Visible light communication using mobile-phone camera with data rate higher than frame rate,” Opt. Express 23(20), 26080–26085 (2015). [CrossRef]   [PubMed]  

20. T. H. Do and M. Yoo, “Performance analysis of visible light communication using CMOS sensors,” Sensors (Basel) 16(3), 309 (2016). [CrossRef]   [PubMed]  

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (8)

Fig. 1
Fig. 1 Principle of multilevel modulation.
Fig. 2
Fig. 2 CMOS camera based VLC processing schematic diagrams at the receiver.
Fig. 3
Fig. 3 Experimental setup of VLC system with mobile phone camera.
Fig. 4
Fig. 4 The structure of data packet.
Fig. 5
Fig. 5 System process diagram of VLC system with camera receiver. (a-c) Image frames received at different illuminances.
Fig. 6
Fig. 6 Experimental scene.
Fig. 7
Fig. 7 Column matrix of grayscale values with LPF.
Fig. 8
Fig. 8 BER versus the different illuminance.
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.