An IEEE Standard for Visible Light Communications

Late last year the IEEE802.15.7 draft standard for VLC [1] was produced.  This standard covers both the physical layer (PHY) air interface and the medium-access control (MAC). We might consider the MAC layer in more detail in later articles but for now I will focus on the physical layer since this is the bit that actually uses the visible light.

The 802.15.7 draft standard is significant for our VLC community because we may now begin to develop products that will be compliant with a future international standard.  It also provides a minimum benchmark for future developments.  If enhancements are to be proposed to the standard, these enhancements must be based on a significant benefit over what is already written into the proposal.  In the following paragraphs I will try to summarise some of the key parameters within the 802.15.7 standard.

The standard is being proposed for a variety of VLC applications relating to Wireless Personal Area Networks (WPAN).  The MAC currently supports three multiple access topologies; peer-to-peer, star configuration and broadcast mode.  The MAC also handles physical layer management issues such as addressing, collision avoidance and data acknowledgement protocols.  The physical layer is divided into three types; PHY I, II & III, and these employ a combination of different modulation schemes.

Data modulation schemes

OOK with Manchester Coding

The OOK modulation scheme uses Manchester Coding

On-off keying (OOK): As the name suggests the data is conveyed by turning the LED off and on.  In its simplest form a digital ‘1’ is represented by the light ‘on’ state and a digital ‘0’ is represented by the light ‘off’ state.  The beauty of this method is that it is really simple to generate and decode.  The 802.15.7 standard uses Manchester Coding to ensure the period of positive pulses is the same as the negative ones but this also doubles the bandwidth required for OOK transmission.  Alternatively, for higher bit rates run length limited (RLL) coding is used which is more spectrally efficient.  Dimming is supported by adding an OOK extension which adjusts the aggregate output to the correct level.

Variable Pulse Position Modulation (VPPM)

Variable Pulse Position Modulation (VPPM) supports dimming

Variable pulse position modulation (VPPM): Pulse position modulation (PPM) encodes the data using the position of the pulse within a set time period.  The duration of the period containing the pulse must be long enough to allow different positions to be identified, e.g. a ‘0’ is represented by a positive pulse at the beginning of the period followed by a negative pulse, and a ‘1’ is represented by a negative pulse at the beginning of the period followed by a positive pulse.  VPPM is similar to PPM but it allows the pulse width to be controlled for light dimming support as shown.

xy Chromaticity Diagram

RGB LEDs can combine different wavelengths for CSK

Colour shift keying (CSK): This can be used if the illumination system uses RGB type LEDs.  By combining the different colours of light, the output data can be carried by the colour itself and so the intensity of the output can be near constant.  The xy chromaticity diagram shows the colour space and associated wavelengths in blue text (units are nm).  Mixing of the red, green & blue primary sources produces the different colours which are coded as information bits.  The disadvantage of this system is the complexity of both the transmitter and the receiver.

The physical layer

Three physical layer options are currently specified:

PHY I: is designed for outdoor, low data rate applications.  It provides data rates in the range 12 – 267 kbit/s.  Convolutional and Reed Solomen codes can be used for forward error correction, and OOK or VPPM are used for modulation.

PHY II: is designed for indoor operation with moderate data rates in the range 1.25 – 96 Mbit/s.  Reed Solomen codes can be used for forward error correction, and OOK or VPPM are used for modulation.  Note that to achieve 96 Mbit/s an optical clock rate of 120 MHz is required which most off the shelf optical devices will not support.  At the more realistic clock rate of 15 MHz a data rate of 9.6 Mbit/s can be achieved.

PHY III: is designed for applications where RGB sources and detectors are available.  It provides data rates in the range 12 – 96 Mbit/s. Again Reed Solomen codes can be used for forward error correction and this time CSK with 4, 8 or 16 colour constellations are used.


This initial draft must be welcomed and will hopefully prompt some thinking around VLC products and applications.  By developing a VLC module that complies even with the low performance end of the specification would allow early stage applications to be developed and testing in the market.  VLC is an enabling technology and I am looking forward to seeing how it will be used and who will be the early adopters.

[1] IEEE802 Part15.7: PHY and MAC standard for short-range wireless optical communication using visible light. (Draft 4), December 2010.

This entry was posted in Uncategorized. Bookmark the permalink.

3 comments on “An IEEE Standard for Visible Light Communications

  1. Hello,

    I am a 5th year MEng student at The University of Glasgow (EEE). I have been trying to find information on the IEEE draft standard for VLC for a project which is using VLC technology. Unfortunately, neither my personal IEEE membership or that of my institution allows access to the standard. This summary is the only relevant information that is freely available on the internet. Is there any more information that it would be possible for me to acquire as it would greatly help me for my project.

    Thank you very much

    Yours Sincerely
    James E. McCann

  2. FAWAD PMI on said:

    I am Fawad pmi, final year student electronics and communication department of Dr. N.G.P college, Coimbatore.I would be grateful if you send me the project title details relevant to IEEE papers above 2009 that you have been training in your company.

  3. Edward Fisher on said:

    Hi Gordon.
    Any idea as to the native (uncoded) bit error rates that the IEEE standardisation effort is aiming at? I couldn’t see it in draft 8-2011?

    Others have mentioned 10^-3 to 10^-6 prior to FECC however I’m interested to see if there have been agreement as to this requirement. BER will dictate signal power and required SNR, however I couldn’t find anything about low level specifications such as this, just the higher level PSY and MAC specs.


Leave a Reply

Your email address will not be published. Required fields are marked *


* Copy This Password *

* Type Or Paste Password Here *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>