Rate and Performance Analysis of Indoor Optical Camera Communications in Optical Wireless Channels
4 likes891 views
This document presents a dissertation written by Willy Anugrah Cahyadi for the degree of Doctor of Philosophy at Pukyong National University in November 2018. The dissertation focuses on improving the data rate and performance of optical camera communications (OCC) by investigating downlink and proposing uplink solutions. Key contributions include achieving a downlink rate of 11,520 bits/s using a split-frame technique and a rate of over 2 Mbits/s using high-density modulation with a neural network. An uplink scheme using a smartphone display is also presented achieving an effective data rate of 360 bits/s. The aim is to address critical limitations of OCC and provide complementary solutions within the standardized use cases.
Rate and Performance Analysis of Indoor Optical Camera Communications in Optical Wireless Channels
- 1. Mobile Transmission System Laboratory Partial fulfillment for the Degree of Doctor of Philosophy Willy Anugrah Cahyadi Advisor: Prof. Yeon Ho Chung Department of Information and Communications Eng. Pukyong National University November 23th, 2018
- 2. Mobile Transmission System Laboratory Outline 2 Introduction Motivation and Research Objectives Main Contributions • Rate of Downlink OCC • Rate of Uplink OCC • Performance of OCC Conclusions Future Scope
- 3. Mobile Transmission System Laboratory3 INTRODUCTION Optical Wireless Communications Optical Camera Communications
- 4. Mobile Transmission System Laboratory Optical Wireless Communications Transmitting information using light without fiber – Light propagation through air/gas, liquid, solid (diffuser), and vacuum Advancements of LED light transmitter – Miniaturized and highly efficient for generating light – Can be flickered with a transition period < ms 4
- 5. Mobile Transmission System Laboratory Optical Wireless Communications Light is more efficient for line-of-sight (LOS) compared with radio wave – Directed LOS transmission – Robustness against weather – Applicable for underwater communication Unregulated optical spectrum – 10000× more bandwidth compared to radio wave spectrum – Imposes less health risks compared to radiation from radio waves 5
- 6. Mobile Transmission System Laboratory Electromagnetic Spectrum 6 Radio waves bandwidth ≈ 300 GHz IR theoretical bandwidth ≈ 12.5 THz
- 7. Mobile Transmission System Laboratory Electromagnetic Spectrum 7 Visible light theoretical bandwidth ≈ 320 THz
- 8. Mobile Transmission System Laboratory Electromagnetic Spectrum 8 10nm 200nm 100nm 280nm Vacuum UV 315nm 400nm UV-C UV-B UV-A Completely absorbed by ozone layer UV theoretical bandwidth ≈ 75 PHz
- 9. Mobile Transmission System Laboratory Classification of OWC 9 Indoor Outdoor
- 10. Mobile Transmission System Laboratory10 INTRODUCTION (Optical Camera Communications)
- 11. Mobile Transmission System Laboratory Principles Pragmatic version of VLC – OCC utilizes a camera receiver instead of PD – Everybody practically carries a camera everywhere – An available receiver for everyone to use Main theme of this dissertation 11
- 12. Mobile Transmission System Laboratory Principles Camera receiver – Single or multiple cameras Transmitters: – Single (illumination) LED – Array of LEDs – Digital display Advantage – 2D image capture Disadvantage – Image capture is relatively slow 12
- 13. Mobile Transmission System Laboratory Standards: IEEE 802.15.7 IEEE 802.15 Working group of IEEE – Specifies the WPAN standards – IEEE 802.15.7 is a Task Group in charge of Standards for Visible Light Communication – There are 15 subgroups of IEEE 802.15 • TG7r1: subgroup for Optical Wireless Communications TG7r1 task group handles revisions to IEEE 802.15.7-2011 Standard – LED-ID: wireless light ID – OCC: Image sensor communications
- 14. Mobile Transmission System Laboratory IEEE 802.15.7r1: LED-ID Technology LED-ID – Wireless light ID system using LED lights
- 15. Mobile Transmission System Laboratory IEEE 802.15.7r1: Concept of OCC Modulate LED light source with data bits Received by a camera – That decodes the bits and extracts the data High-precision positioning solution – Using smart device camera 15
- 16. Mobile Transmission System Laboratory IEEE 802.15.7r1: Use Cases LED QR/2D Color Code – LED based barcode as a tag and smartphone camera as the reader P2P Tx/Rx & Relay Application – Device-to-device short range communication 16
- 17. Mobile Transmission System Laboratory IEEE 802.15.7r1: Use Cases Signage/Display – Tag: Display/Signage – Reader: Smartphone camera Exhibition and Store Service In-flight Service 17
- 18. Mobile Transmission System Laboratory IEEE 802.15.7r1: Use Cases Underwater / Seaside communication Smart living: Smart office and smart home 18
- 19. Mobile Transmission System Laboratory Critical Issues in OCC Information carrier in 2D image capture – Spatial coordinate, intensity, color, and shape Image processing – Generally has a slow capture rate • Millions of pixels processed by the camera (Megapixels) • Cameras in general: 30-60 fps • Slow-motion cameras: 240-480 fps • High-speed cameras: 10000-1 million fps (reduced resolution) – Compared to PD in VLC up to 10 GHz sampling rate OCC data rate is somehow limited – tens of bit/s ~ few Kbit/s 19
- 20. Mobile Transmission System Laboratory Critical Issues in OCC OCC has been studied for either LOS link or NLOS link only, not both Uplink channels for OCC have never been explored – It is essential to support the uplink in indoor wireless communications Flickering LEDs or displays – Cause serious focus issues on camera receiver Static cameras are only considered in the OCC 20
- 22. Mobile Transmission System Laboratory Motivation Why OCC? – OCC has been formulated in the 802.15.7r1 standardization activity for the past 7 years • The most recent update will be released at the end of Nov 2018 – OCC is the most practical or viable indoor OWC scheme • Camera receiver is available on daily used smartphones – OCC has much potential and variation, but its rate and performance still need to be enhanced significantly for practical or commercial applications in the near future. • The rate is somewhat limited to a few kbps in the literature • The performance is also not competitive, compared with its RF counterparts
- 23. Mobile Transmission System Laboratory Objectives Providing improved and complementary solutions – To address the critical disadvantage (rate, performance) of OCC – Within the scope of the standardized OCC use cases Investigating data rate improvements – Both downlink and uplink channels in OCC – Higher than existing studies in OCC Proposing uplink solutions for OCC – Undocumented in the literature Investigating complementary solutions – Maintaining illumination provision – Providing a wide orientation in OCC • Both LOS and NLOS 23
- 24. Mobile Transmission System Laboratory24 RATE OF DOWNLINK OCC • Split-frame Technique • High-density Modulation with Neural Network
- 25. Mobile Transmission System Laboratory Split-frame Technique An alternate method to increase the camera capture rate – Utilizing dual camera • Multiple cameras are common in smartphones – Capturing the transmitter on each camera – Splits the capturing process 2× data rate increase Experiment distance Transmitter LCD Data generator computer Data processing computerDual camera on smartphones (receivers)
- 26. Mobile Transmission System Laboratory Split-frame Technique Acquires half of the capture frame – Reduces the frame capture period by half 26 Rolling Shutter Demodulation (RSD) Split-frame
- 27. Mobile Transmission System Laboratory Experiment Parameters Parameter Values Capture device Android-based smartphone Video capture resolution 1280×720 pixels Camera lens and sensor size f/2.2 aperture, 31 mm lens, 64° diagonal FOV, and 1/3” sensor size. Video capture rate 30 and 60 fps (standard shutter) 30, 60, and 120 fps (split-frame) Frame period 17, 20, 25, 33, 40, 50, 67, and 100 ms Transmitter flicker rate (LCD refresh rate) 60, 50, 40, 30, 25, 20, 15, and 10 Hz Transmitter size 30×30 cm2 Experimented distances 100 – 300 cm (20 cm increments) 27 Split-frame enables faster capture rate – 60 fps 120 fps
- 28. Mobile Transmission System Laboratory Fitting the Transmission Frame Full-fit frame Partial-fit frame 28 The requirement for split-frame – Equally divides the transmission frame – Full-fit frame: TX frame fills the camera capture frame fully – Partial-fit frame: TX frame fills the camera capture frame partially
- 29. Mobile Transmission System Laboratory Achievable Data Rate Data rate depends on the transmitter frame period 11,520 bit/s data rate was achieved – Effective capture rate of 120 fps
- 30. Mobile Transmission System Laboratory BER vs Distance
- 31. Mobile Transmission System Laboratory Comparison 31 Parameters RSD Split-frame Transmission distance 50 cm 200 cm Increases resolution + data rate + transmission distance + data rate Increases number of cells/LEDs + data rate − transmission distance + data rate Frame fit Strictly full-fit Partial-fit or full-fit Transmitter type LED only LED and Digital displays
- 32. Mobile Transmission System Laboratory Increasing Data Rate in OCC Common methods to increase data rate – Increase camera capture rate, although it is impractical for existing devices, such as smartphones – Increase the number of LEDs/cells, color, and intensity 32 126.72 Kbit/s @ 1.4 m distance 330 fps FPGA-controlled camera Huang, W, et al., “Design and implementation of a real-time CIM-MIMO OCC system,” Optics Express, vol. 24, 2016.
- 33. Mobile Transmission System Laboratory Increasing Data Rate in OCC 33 Boubezari, S, et al., “Smartphone camera based visible light communication,” Journal of Lightwave Technology, vol. 34, 2016. 112.5 Kbit/s 15 cm distance Monochrome 75×50 cells
- 34. Mobile Transmission System Laboratory High-density Modulation with NN A denser modulation is proposed to achieve the Mbit/s data rate efficiently – Practical smartphone camera – Existing screen-based transmitter High-density modulation (HDM) – Combines 4 modulation entities: 1. Cell (group of pixels in digital display) 2. Color 3. Intensity 4. Shape – Requires an NN for demodulation A data rate of 2.66 Mbit/s for a distance of up to 20 cm – Device-to-device OCC 34
- 36. Mobile Transmission System Laboratory Experiment Setup 36 Screen size: 7.2 cm×12.7 cm
- 37. Mobile Transmission System Laboratory Structure of HDM Scheme 37 0 0 0 0 0 0 0 1 0 0 1 0 0 0 1 1 0 1 0 0 0 1 0 1 0 1 1 0 0 1 1 1 1 0 0 0 1 0 0 1 1 0 1 0 1 0 1 1 1 1 0 0 1 1 0 1 1 1 1 0 1 1 1 1 4 bits 9 bits • 3 bits of R • 3 bits of G • 3 bits of B 16 shapes 512 color intensities Shape modulation Cell formulation (78×44 cells) Color + intensity modulation Data bits
- 38. Mobile Transmission System Laboratory Structure of TX Frame 38 78×44 cells 20 anchors – 2 on each corner – 2 on each side – 4 on center
- 39. Mobile Transmission System Laboratory Anchors – Color calibration Anchors for color calibration 39 Color Chromaticity: 1. Black 2. Red 3. Green 4. Blue 5. White 6. Magenta 7. Yellow 8. Cyan Intensity: 12.5% gray 25% gray 37.5% gray 50 % gray 62.5% gray 75% gray 87.5% gray 100% gray (white) The CIE 1931 color space chromaticity diagram
- 40. Mobile Transmission System Laboratory TX RX Bits Color composition Bits 8 levels of Red Anchor identification 8 levels of Green 8 levels of Blue 512 color intensity palette HDM Anchors for color chromaticity Anchors for intensity TX Frame Perceived color chromaticity Perceived intensity Calibrated 512 color intensity palette References HDD Color Calibration Scheme Color calibration Camera imperfections Display screen imperfections Optical channel
- 41. Mobile Transmission System Laboratory Trained NN Structure 41 Neural Network (NN) Structure Layer 1: Layer 2: 36 bits Unique bit patterns (16 bits) 36 bits 36 bits NN33×33 pixels 16 shapes 1 38 2 1 2 16 Output: 16 bits Input: 36 bits
- 42. Mobile Transmission System Laboratory System Diagram 42 TX Random bits Received bits Color intensity modulation Frame generation LCD DisplayDisplay RX Shape modulation HDM Camera capture 2X resolution Synchronized framesHDD Color calibration Smartphone NN Filtering Optical Channel
- 43. Mobile Transmission System Laboratory Experiment Parameters No. HDM format (NE, NC, NS) Transmitter Receiver TX screen resolution Flickering rate Camera resolution Camera capture rate 1 78×44, 512, 16 TX1 (2560×1440) 60 Hz 1280×720 120 fps 2 78×44, 512, 16 TX2 (1280×720) 60 Hz 1280×720 120 fps 3 78×44, 512, 16 TX1 (2560×1440) 30 Hz 1920×1080 60 fps 4 39×22, 512, 16 TX1 (2560×1440) 60 Hz 1280×720 120 fps 5 39×22, 512, 16 TX1 (2560×1440) 30 Hz 1920×1080 60 fps 43 TX 1 screen size: 7.2 cm×12.7 cm (smartphone) TX 2 screen size: 9.6 cm×15.1 cm (tablet)
- 44. Mobile Transmission System Laboratory Experiment Results 44 Config. 2: – Larger screen Config. 3: – Higher camera resolution – Reduced capture rate Config. 4: – Reduced cells Config. 5: – Reduced cells – Higher camera resolution
- 45. Mobile Transmission System Laboratory Experiment Results 45 PSNR is utilized to evaluate reception quality A minimum PSNR of 22dB is required – Adequate transmission quality
- 46. Mobile Transmission System Laboratory46 RATE OF UPLINK OCC • Low Rate Display-based Solution • High Rate Near-infrared-based Solution
- 47. Mobile Transmission System Laboratory Display-based Uplink Scheme for OCC 47
- 48. Mobile Transmission System Laboratory Display-based Uplink Scheme for OCC 2×2 RGB cells displayed on smartphone screen – Resolution: 2560×1440 pixels – Size: 50 mm × 50 mm – Flicker rate: 30 Hz Chamber dimension (cm): 100×100×200 48
- 49. Mobile Transmission System Laboratory Display-based Uplink Scheme for OCC Camera – Capture rate: 60 fps – Resolution: 1280×720 pixels – FOV: 60° Effective data rate: 360 bit/s 49
- 50. Mobile Transmission System Laboratory Display-based Uplink Scheme for OCC 50 FEC limit
- 51. Mobile Transmission System Laboratory Display-based Uplink Scheme for OCC Considerable data rate limitation – Smartphone screen size is too small for practical uplink distance transmission – Glare from the smartphone screen is significant – 360 bit/s hardly extendable – 90 cm distance 60° orientation angle Smartphone screen is dedicated for communication – Inconvenience for the user 51
- 52. Mobile Transmission System Laboratory High Rate Near-infrared-based Solution Infrared-based proximity sensors on smartphones (with near-IR LEDs) – IR is invisible to human eyes non-disruptive – Visible to the camera • Especially if the IR-filter is removed A potential uplink transmitter 52 IR LEDs
- 53. Mobile Transmission System Laboratory Infrared LED Camera Comm. (ICC) Transmitter: IR LEDs with wavelength of 940 nm Receiver: a modified webcam – Removed IR blocking filter – IR band-pass filter (Hoya R-72) 53 distance Rx Processing PC Tx Tx MCU (modulation) IR LEDs (940 nm) Data packet generation Rx Processing PC (demodulation) Camera (Removed IR blocking filter) Retrieved data packet LED driver Diffuser + IR band pass filter Fresnel lens
- 54. Mobile Transmission System Laboratory Transmitter and Receiver Units 54 Flexible arm IR LEDs Fresnel lens Connected to LED driver and MCU Camera holder Camera unit IR bandpass filter (940nm) Diffuser + holder Modified camera IR blocking filter removed Custom designed mounting
- 55. Mobile Transmission System Laboratory MCU + LED Driver Schematic 55
- 56. Mobile Transmission System Laboratory Data Packet 56 PayloadHeader 4-bit 28-bit Cyclic prefix 1-bit zero gap Header is a cyclic prefix of the payload – Important for synchronization
- 57. Mobile Transmission System Laboratory Flowchart of ICC Positioning 57 START Acquire frame + calculate mean intensity Fix camera exposure time to 1/4096 s Mean intensity of pixel rows (mR) Mean intensity of pixel columns (mC) Estimate Tx position LY=(max (mR) – σR) LX=(max (mC) – σC) μX= mean (mC)
- 58. Mobile Transmission System Laboratory58 Demodulation END Start demodulation on payload data Yes NoHeader acquired? Normalize mF based on the threshold MLE Acquire the pixel columns having intensity < μX Mean intensity of filtered pixel rows (mF) AIC Flowchart of ICC Sync + Demodulation
- 60. Mobile Transmission System Laboratory MLE Algorithm 60 Correlation Pixel rows 𝛾 x = 𝑘=𝑥 𝑥+𝐻−1 𝑟 𝑘 𝑟 𝑘 + 𝑁
- 61. Mobile Transmission System Laboratory Coverage Test 61 Parameter Values Room dimension 130 cm×81 cm×200 cm Video capture resolution 640×480 pixels Camera FOV 60° diagonal FOV Video capture rate 60 fps Exposure period 1/4096 s Ambient illuminance 800-900 lx (from illumination)
- 62. Mobile Transmission System Laboratory Coverage Test 62 SNR(dB) Offset from the center (cm) Wider FOV of the camera – More consistent SNR for uplink on 3 exemplary positions of the transmitter compared to a PD solution
- 63. Mobile Transmission System Laboratory Fixed Data Rate of 6.72 Kbit/s 63 BER Transmission distance (cm) 10-4 SNR(dB) 5 10 15 20 25 30 35 40 45 20 40 60 80 100 120 140 160 180 2000 10-3 10-2
- 64. Mobile Transmission System Laboratory Data Rates with Target BER of 10-3 64 Transmission distance (cm) 20 40 60 80 100 120 140 160 180 200 Datarate(bit/s) 7000 6000 5000 4000 3000 2000 1000 0 SNR(dB) 5 10 15 20 25 30 35 40 45
- 65. Mobile Transmission System Laboratory Infrared-based Indoor Positioning Indoor positioning system (IPS) is an important breakthrough – GPS does not work indoors – Uplink channel in OCC utilizes an IR LED • a viable IPS solution IR LED An invisible beacon utilized for IPS Surveillance camera is often available in indoor environments – The camera is sensitive to IR light (switchable IR blocking filter) 65 Day mode Night mode
- 66. Mobile Transmission System Laboratory Infrared-based Indoor Positioning 66 The surveillance camera is set to night-mode – Captures both visible light and IR light – Requires an algorithm to minimize the ambient light interference – Wide FOV of 170 Designed beacon – Similar to IR LED on the proximity sensors of the smartphone
- 67. Mobile Transmission System Laboratory Infrared-based Indoor Positioning 67 Experiment setup
- 68. Mobile Transmission System Laboratory Infrared-based Indoor Positioning 68 ROI for Measurement Area
- 69. Mobile Transmission System Laboratory Algorithm of Positioning Scheme 69 Start RGB channel mixing (monochrome) Fixed exposure period Surveillance image Red channel retrieval (monochrome) Frame capture start Intraframe positioning End
- 70. Mobile Transmission System Laboratory Intraframe Positioning Scheme 70 𝑥 𝑝 𝑥𝑖 = 𝑆 2𝑑 tan−1 1 2 𝛼 𝑣 𝑤 𝑦𝑝 𝑦𝑖 = 𝑆 2𝑑 tan−1 1 2 𝛽 𝑣ℎ (𝑥𝑖, 𝑦𝑖): acquired beacon’s coordinate (𝑥 𝑝, 𝑦𝑝): physical beacon’s coordinate 𝛼: horizontal FOV 𝛽: horizontal FOV 𝑆: scaling constant 𝑑: distance between camera and the experi- ment plane
- 71. Mobile Transmission System Laboratory Experiment Parameters 71 Parameter Values Room dimension 350 cm × 390 cm × 220 cm Flicker rate of the beacon 2 KHz Operating voltage of the beacon 5V Operating power of the beacon 150 mW Ambient illuminance (minimum – maximum) 480 - 1120 lx Exposure periods of the camera 1/500 s, 1/1000 s, 1/2000 s and 1/4000 s Video resolution of the camera 1920 × 1080 pixels Capture rate of the camera 30 fps FOVs of the camera (diagonal, horizontal, and vertical) 170°, 168°, and 164° Fixed height of the beacon 100 cm
- 72. Mobile Transmission System Laboratory Various Exposure Periods 72 1/500 s 1/1000 s 115 cm 37 cm AccuracyCaptured frame
- 73. Mobile Transmission System Laboratory Various Exposure Periods 73 1/2000 s 1/4000 s 1 cm 1 cm AccuracyCaptured frame
- 74. Mobile Transmission System Laboratory RGB Channel Mixing The exposure period is fixed to 1/2000 s – Limit the interference – Captured images are darker Mixing the RGB channels – Produces brighter images for surveillance 74 RGB channel mixing Captured frame Frame for surveillance
- 75. Mobile Transmission System Laboratory Results: Static Beacon Positions 75 Errors: due to the interference light near the beacon – The camera captures both visible and IR light
- 76. Mobile Transmission System Laboratory Results: Static Beacon Positions 76 Static positions Mean positioning errors Misidentified frames Position 1 2 cm 0% Position 2 4 cm 1% Position 3 2 cm 5% Position 4 6 cm 5% Position 5 2 cm 2% Position 6 3 cm 3% Position 7 2 cm 1%
- 77. Mobile Transmission System Laboratory Results: Moving Beacon 77
- 78. Mobile Transmission System Laboratory Moving speed (approximate) Exposure period Mean positioning errors Misidentified frames 50 cm/s 1/2000 s 3 cm 3% 50 cm/s 1/4000 s 2 cm 1% 100 cm/s 1/2000 s 4 cm 3% 100 cm/s 1/4000 s 3 cm 1% Results: Moving Beacon 78 Shorter exposure period more accurate positioning with less misidentified frames Intraframe algorithm does not analyze the relation between several frames
- 79. Mobile Transmission System Laboratory IR-based IPS Experimentally proven – Centimeter scale positioning – Both static and moving beacon – Mean positioning error is limited to 6 cm Further improvements in the future – Real-time image analysis • Dynamic exposure period and threshold instead of fixed ones – Multiple cameras • Increased accuracy + coverage behind obstacles – Interframe positioning algorithm 79
- 80. Mobile Transmission System Laboratory80 PERFORMANCE OF OCC • Focus and Light Metering with Additional Illumination LEDs • Wide Orientation Transmission with Illumination
- 81. Mobile Transmission System Laboratory Performance of OCC Data rate is only one performance factor There are performance factors important to OCC – Camera automatically obtains the optimum light metering based on its focus • LED / screen is flickering camera – Strict orientation of the camera (receiver) Performance enhancements of OCC is important for indoor environments – Practical schemes are required – Maintain illumination provision is also preferred – OCC is either investigated for LOS only or NLOS only
- 82. Mobile Transmission System Laboratory I-OCC Scheme Employs additional white illumination LEDs – References for both focus and light metering • Help focus the camera while the transmitting LEDs are flickering – Ensuring the illumination provision • Illumination is not affected by the transmitting LEDs LED driver PC Smartphone ServerDot matrix LED array
- 83. Mobile Transmission System Laboratory Block Diagram of I-OCC Rx Rotation compen- sation Smartphone camera Data evalua- tion Demap -ping Tx Mapping LED driver + dot matrix LED Data genera- tion Optical channel
- 84. Mobile Transmission System Laboratory Keyframe Structure 84 Rotation marker Alignment marker (𝒌 − 𝟏) framesKF 𝒌 frames (𝒌 − 𝟏) framesKF 𝒌 frames - - - - 𝑵 − 𝟏 framesKF 𝑵 frames (full transmission frames Keyframe 𝑭 (1 frame) Keyframe 𝑭 𝟏 (1 frame) Keyframe 𝑭 𝒏 (1 frame) Aperiodic Periodic
- 85. Mobile Transmission System Laboratory Alleviating the Blooming Effect 85 Misfocused Correctly focused
- 86. Mobile Transmission System Laboratory Experimental Parameters Parameters Values Capture devices 2013 Android-based smartphone (Cam 1) 2010 iOS mobile device (Cam 2) Video capture resolution Cam 1: 1920×1080 pixels Cam 2: 1280×720 pixels Capture rate Cam 1: 60 fps, variable capture rate Cam 2: 30 fps, constant capture rate Total captured frame 25440 frames for Cam 1, and 11160 frames for Cam 2 Frame period 25, 50, 67, and 100 ms Experimented distances 5 cm, 10 cm, 20 cm, and 30 cm Ambient illumination 6 lx Illumination LED 3.143 V, 22.37 mA Dot matrix LED 4.31 V, 7.44 mA 86
- 87. Mobile Transmission System Laboratory Achievable Data Rate (𝐷 𝑅) 𝐿 𝑁: number of LEDs for data transmission 𝑡 𝑟: LED flickering rate (LFR) 𝐹𝑝: reduction of data rate due to keyframe insertion T: transmission period 87 𝐷 𝑅 = 𝐿 𝑁 𝑡 𝑟 − 𝐹𝑝 𝐹𝑝 = 𝐿 𝑁 𝑇 , 𝑎𝑝𝑒𝑟𝑖𝑜𝑑𝑖𝑐 𝑘𝑒𝑦𝑓𝑟𝑎𝑚𝑒 10𝐿 𝑁, 𝑝𝑒𝑟𝑖𝑜𝑑𝑖𝑐 𝑘𝑒𝑦𝑓𝑟𝑎𝑚𝑒
- 88. Mobile Transmission System Laboratory Results of Simulation and Experiment Experimented data rate of 1280 bit/s – 30 fps camera capture rate 88
- 89. Mobile Transmission System Laboratory Results of Experiment (BER) 89 Distance LFR Cam 1 Cam 2 5 cm 10 14 20 0.0 0.0 0.0 0.0 0.0 0.0 10 cm 10 14 20 0.0 0.0 0.0 0.0 0.0 0.0 20 cm 10 14 20 0.0 0.0 0.0 0.0 0.0 0.0 30 cm 10 14 20 0.0016 0.0047 0.0813 - - -
- 90. Mobile Transmission System Laboratory Data Detection Region-of-Interest is set initially by identifying keyframe Differential detection threshold – Quantized intensity binary thresholding 90 Data RoI Red Color Channel Quantized Intensity
- 91. Mobile Transmission System Laboratory Illuminance Measurement 91 Distance Illumination (lx) Dot matrix (lx) Illumination + dot matrix (lx) 5 cm 2330 - 2340 206 - 254 2190 – 2240 10 cm 1312 - 1313 146 – 161 1369 – 1389 20 cm 597 – 598 116 – 126 624 - 631 30 cm 303-304 96-100 305-312 Large difference of the illuminance level – Dot matrix LEDs (flickering red light) do not effect illumination (white light)
- 92. Mobile Transmission System Laboratory Wide Receiver Orientation (WRO) Existing OCC researches consider either LOS only or NLOS only WRO scheme proposes a wide orientation for both LOS and NLOS – A specially designed transmitter – Utilizing diffuse reflection for the NLOS link – RSD based demodulation for reception – Illumination is still provided 92 Source: T. Nguyen, A. Islam, T. Hossan and Y. M. Jang, “Current Status and Performance Analysis of Optical Camera Communication Technologies for 5G Networks,” in IEEE Access, vol. 5, pp. 4574-4594, 2017.
- 93. Mobile Transmission System Laboratory Wide Receiver Orientation (WRO) 93
- 94. Mobile Transmission System Laboratory Block Diagram of The WRO Scheme 94 Tx MCU (Modulation) Proposed transmitter Data packet generation Rx Offline demodulation Smartphone camera Retrieved data packet
- 95. Mobile Transmission System Laboratory Experimental Setup 95 Photo of the chamber Wall materials
- 96. Mobile Transmission System Laboratory Specially Designed Transmitter 96 4 directional LEDs (angled) – Distribute illumination equally to all direction 1 central LED Diffuser: polymorph plastic – Blend the light emission
- 97. Mobile Transmission System Laboratory Intensity of Each Pixel (𝑁𝑝) 𝐾 : Calibration constant 𝑓𝑠 : lens aperture 𝑆 : ISO sensitivity of the camera 𝐸𝜌/𝜋 : a total amount of luminance entering the camera from a diffused reflection 𝐸 : illuminance that falls on the sensor surface 𝜌 : reflectance of a surface causing diffuse reflection 97 𝑁𝑝 = 𝐾 𝑡𝑆 𝑓𝑠 2 𝐸𝜌 𝜋
- 98. Mobile Transmission System Laboratory Adaptive Blooming Mitigation (ABM) 98
- 99. Mobile Transmission System Laboratory ABM: Filtered Pixel Rows 99
- 100. Mobile Transmission System Laboratory Normalized Intensity 100 Output of ABM Final algorithm is MLE
- 101. Mobile Transmission System Laboratory Experimental Parameters Parameter Values Chamber dimension 40 cm × 40 cm × 40 cm LED flickering rate 4 KHz ( 2 bits / cycle) Data rate 6.72 Kbit/s (fixed) Excluding the header bits and zero gap Camera capture resolution 1920×1080 pixels (video capture) ISO value 2700 Exposure period 1/6000 s 101
- 102. Mobile Transmission System Laboratory Experimental Parameters Parameter Values MCU ATMega328P with an operating clock of 16 MHz Center LED Rated power: 3 W Operating voltage: 3.4 V Color temperature: 6000K Directional LEDs Rated power: 3 W Operating voltage: 3.4 V Color temperature: 6000K Smartphone camera LG V10 (F600L) Operating system: Android 7.0 Lens aperture: f/1.8 FOV: 78° 102
- 105. Mobile Transmission System Laboratory Illumination Distribution 105 Equally distributed illuminance of 80 lx and 70 lx – Distance of 60 cm and 70 cm
- 106. Mobile Transmission System Laboratory Reflectance 𝜌 : ratio of the reflected illuminance to incident illuminance 𝐸𝑟: reflected illuminance 𝐸𝑖: incident illuminance 106 𝜌 = 𝐸𝑟 𝐸𝑖
- 107. Mobile Transmission System Laboratory Measured Reflectance Material Illuminance reflectance White wood panel 0.5771 White paper 0.6129 Glossy PVC wallpaper 0.4158 107 Higher reflectance: more diffused reflection – White paper has the highest reflectance – Measurement distance: 10 cm from the wall – Both incident and reflection angles are 45°
- 109. Mobile Transmission System Laboratory Results: Multiple Height 109
- 110. Mobile Transmission System Laboratory Results: Multiple Orientations 110
- 111. Mobile Transmission System Laboratory Results: Multiple Orientations 111
- 112. Mobile Transmission System Laboratory Results: Moving Receiver 112
- 114. Mobile Transmission System Laboratory Conclusions The present study has considered the two critical issues of rate and performance for OCC to evolve further. The data rate has been investigated on both downlink and uplink channel. We have obtained a higher data rate than the existing OCC schemes – ≈ 10 times the data rate of existing schemes for downlink Mbit/s rate – Up to 6.72 Kbit/s for uplink Performance improvements have also been investigated in terms of focus and light metering as well as wide orientation transmission. 114
- 115. Mobile Transmission System Laboratory Conclusions In addition, an infrared based indoor positioning scheme has been investigated. – Novel approach using infrared – Non-disruptive and practical positioning with centimeter-scale accuracy 115
- 116. Mobile Transmission System Laboratory Future Scope – OCC vision Rapid development of smartphone camera. Higher data rate for OCC in the future is envisioned. NPU (or NN) implemented on smartphones. OCC is expected to be very viable and pragmatic short-range indoor wireless communication. IEEE standard will be announced in January 2019. 116
- 117. Mobile Transmission System Laboratory Future Scope – Further works Channel model formulations through experiments Increasing efficiency of the existing modulations Investigating multiple camera schemes – Beneficial for increasing data rate and robustness Investigating advanced selective capture – Improvement of the split-frame technique 117
- 118. Mobile Transmission System Laboratory118 LIST OF PUBLICATIONS
- 119. Mobile Transmission System Laboratory Journal Papers (1st Authorship) Willy Anugrah Cahyadi and Yeon Ho Chung, “Wide Receiver Orientation Using Diffuse Reflection in Camera-based Indoor Visible Light Communication,” Optics Communications, vol. 431, pp. 19-28, 2018. (SCI) Willy Anugrah Cahyadi and Yeon Ho Chung, “Smartphone Camera based Device-to-device Communication Using Neural Network Assisted High-density Modulation,” Optical Engineering, vol. 57, no. 9, p. 096102, 2018. (SCI) Willy Anugrah Cahyadi and Yeon Ho Chung, “Experimental Demonstration of Indoor Uplink Near-infrared LED Camera Communication,” Optics Express, vol. 26, No. 15, pp. 19657- 19664, 2018. (SCI)
- 120. Mobile Transmission System Laboratory Journal Papers (1st Authorship) Willy Anugrah Cahyadi and Yeon Ho Chung, “Experimental Demonstration of VLC based Road Wetness Detection Techniques for Preventing Danger of Hydroplaning,” The Journal of Korean Institute of Communications and Information Science s (J-KICS), vol. 42, no. 08, Sept 2017. Willy Anugrah Cahyadi, Yong-hyeon Kim, and Yeon-ho Chung, “Dual Camera based Split Shutter for High-rate and Long-distance Optical Camera Communications,” Optical Engineering, vol. 55, no. 11, p. 110504, 2016. (SCI) Willy Anugrah Cahyadi, Yong Hyeon Kim, Yeon Ho Chung, and Chang-Jun Ahn, “Mobile Phone Camera-Based Indoor Visible Light Communications with Rotation Compensation,” IEEE Photonics Journal, vol. 8, no. 2, pp. 1-8, 2016. (SCIE)
- 121. Mobile Transmission System Laboratory Journal Papers (2nd Authorship) Tahesin Samira Delwar, Willy Anugrah Cahyadi, and Yeon-ho Chung, “ Visible Light Signal Strength Optimization Using Genetic Algorithm In Non-line-of-sight Optical Wireless Communication,” Optics Communic ations, vol. 426, pp. 511-518, 2018. (SCI) Shivani Teli, Willy Anugrah Cahyadi, and Yeon-ho Chung, “High-Speed Optical Camera V2V Communications Using Selective Capture,” Photo nic Network Communications, pp. 1-7, 2018. (SCI) Shivani Teli, Willy Anugrah Cahyadi, and Yeon Ho Chung, “Trained Ne urons-based Motion Detection in Optical Camera Communications,” Optical Engineering, vol. 57, no. 4, p. 040501, 2018. (SCI) Arsyad Ramadhan Darlis, Willy Anugrah Cahyadi, and Yeon Ho Chung, “Shore-to-Undersea Visible Light Communication,” Wireless Personal Communications, pp. 1-14, December 2017. (SCIE)
- 122. Mobile Transmission System Laboratory Journal Papers (2nd Authorship) Shivani Teli, Willy Anugrah Cahyadi, and Yeon Ho Chung, “Optical Ca mera Communication: Motion Over Camera,” IEEE Communications Magazine, vol. 55, no. 8, pp. 156-162, August 2017. (SCI) Yong Hyeon Kim, Willy Anugrah Cahyadi, and Yeon Ho Chung, “Experi mental Demonstration of VLC-Based Vehicle-to-Vehicle Communicatio ns Under Fog Conditions,” IEEE Photonics Journal, vol. 7, no. 6, pp. 1-9 , 2015. (SCIE) Durai Rajan Dhatchayeny, Willy Anugrah Cahyadi, and Yeon Ho Chung , “An Assistive VLC Technology for Smart Home Devices Using EOG,” Wireless Personal Communications, pp. 1-9, August 2017. (SCIE) Trio Adiono, A. Pradana, Rachmad V. W. Putra, Willy Anugrah Cahyadi , and Yeon Ho Chung, “Physical Layer Design with Analog Front End for Bidirectional DCO-OFDM Visible Light Communications,” Optik, Int ernational Journal for Light and Electron Optics, March 2017. (SCI)
Editor's Notes
- #5: https://www.dial.de/en/blog/article/efficiency-of-ledsthe-highest-luminous-efficacy-of-a-white-led/
- #10: FSO is an alternate fiber optics communication without the fiber Laser-based transmitter coherent light Point-to-point communication Visible Light Communication is a specific type of FSO Mostly for indoor usage distributed light for illumination Shared connection LED-based transmitter
- #14: IEEE 802.15 is a working group of the Institute of Electrical and Electronics Engineers (IEEE) IEEE 802 standards committee which specifies wireless personal area network (WPAN) standards. There are 10 major areas of development, not all of which are active.
- #15: IEEE 802.15 is a working group of the Institute of Electrical and Electronics Engineers (IEEE) IEEE 802standards committee which specifies wireless personal area network (WPAN) standards. There are 10 major areas of development, not all of which are active.
- #42: NN is trained using the samples containing 16 variants of shape
- #47: The investigations specific to this chapter are dedicated to address the uplink channels since the existing OCC schemes do not provide uplink communication [4, 16].
- #60: AIC
- #76: 7 different positions 420 frames
- #77: 7 different positions 420 frames
- #78: 7 different positions 420 frames
- #79: 7 different positions 120 frames
- #81: How to improve the implementation of OCC while also maintaining the illumination provision
- #111: Fixed height: 40 cm
- #114: How to improve the implementation of OCC while also maintaining the illumination provision