2000 Conference Proceedings

Go to previous article 
Go to next article 
Return to 2000 Table of Contents

Assistive Communication Systems for Disabled Individuals using Visible Lighting

Steven B. Leeb
George B. Hovorka,
Elmer .C. Lupton,
Roderick T. Hinman
Massachusetts Institute of Technology
Talking Lights, LLC, Boston, MA

Billie Louise Bentzen,
Randolph D. Easton,
Lisa Lashell
Boston College, Chestnut Hill, MA

Abstract A transceiver system is developed to allows conventional fluorescent lights to be used as transmitters for broadcasting audio, textual, graphical, and position information. Battery-powered, hand-held electronic devices receive the transmissions from the lights. This technologyique provides high quality voice transmissions or data transmissions without causing visible interference visible in the lights.


Commercial and public enterprises, museums, schools, medical facilities and mass transit systems typically provide signs that list facility organization by floor, possibly with a map or schematic representation of the store or system. The Americans with Disabilities Act was enacted to ensure the accessibility of public facilities to people with disabilities. Accessible information is an essential aspect of an accessible environment. . In "Enabling America,'' the authors note that most "environments can be restrictive in their accessibility for people with [disabilities]",'' and while many technologies, e.g., wireless transmission to hearing aids, captioning, etc., can be used, "all of these technologies ... can be improved"'' to maximize cost efficiency, ease of installation and flexibility of data transmission [1].

We have developed an information transfer system that can transmit text, graphics, binary computer data, and audio signals. Theis Talking Lights™ System uses fluorescent lights to transmit information to a portable, battery powered optical receiver. The system requires no wiring other than that used for existing fluorescent lamp fixtures. It can be used anywhere a fluorescent light can be installed: inside or outside a building; in a bright sunlight or underground in a subway station; or inside a vehicle likes a bus or plane. The receiver is designed with an arbitrarily narrow acceptance angle. That is, the receiver will "lock on" to the transmission of a light directly in front of it, ignoring the transmissions of other lights in a room. A room can be filled with many lights, each broadcasting a different message. The receiver will decode the messages of each individual lights as it is panned around the room. The system can therefore be used for wayfinding or guiding a user through a facility as. This is illustrated by the drawings in Figure 1, which shows Talking Lights users receiving textual and audio information. It also can broadcast the same message on all lights, providing enhanced aural information to the hard of hearing.

How Does it Work?

A Talking Lights transceiver set has two parts: a fluorescent light transmitters and an optical receiver.

Fluorescent Lights as Transmitters:

A fluorescent light is a phosphor-coated tube of glass , with a wire or electrode at each end, filled with special gases at a specific pressure. An electric cur-rent passes between the electrodes, causing the gases to ionize and emit energy. This energy is absorbed by phosphors on the glass, which emit light [2]. A "ballast" controls the current, ensuring the lamp starts or "strikes" and that the current is not too large or too small, providing desired illumination with maximum lamp life. The ballast drives the bulb with a rapidly alternating current that balances the tasks of each electrode as an emitter and collector of current, again to maximize lamp life.

Figure 1: Talking Lights Text and Audio Receiver Over half of the artificial light produced in the United States comes from fluorescent lamps [2]. Fluorescent lighting fixtures are generally designed to flood an area with light, ensuring good access for a communication channel. The choice of fluorescent lamps as data transmitters is appealing in comparison to other techniques, such as dedicated infra-red or radio transmitters, because the lamps are already installed in most locations and provide large broadcast power since they are designed for bright illumination. Modern electronic lamp ballasts typically alternate the current in the lamp between 20 and 100 kHz. This alternation is too rapid to be perceived by the human eye. Hence, the lamp light appears to be a steady.

In a Talking Lights ARCLight Ô ballast, the frequency of alternation of current (i.e.the flickering of the lamp light ) is varied to encode information in the lamp. Current frequency may be varied smoothly and continuously to encode an audio signal, or the frequency may move among a limited number of discrete frequencies for digital data. Several schemes have been proposed in the past to use fluorescent lights as transmitters [3-6], but the Talking Lights approach is the first to use modern band-pass FM for transmit-ting analog and digital information [7]. The ARCLight ballast has been carefully designed to eliminate visually perceptible flicker, regardless of data transmitted. It can even transmit more than one channel of information using a single lamp, allowing several broadcasts from the same bulb, analagous to radio stations or TV channels.

Optical Receivers:

An electronic photodetector is used to receive the light from a fluorescent lamp transmitter. The receiver decodes frequency variations in the lamp light to recover the transmitted information. Different circuits are used for recovering text and audio information. These can be mixed and matched to provide a "multimedia" receiver. A block diagram of a Talking Lights text receiver is shown in Figure 2. With appropriate modifications, similar circuitry can decode audio signals.

A Talking Lights receiver provides the best overall system performance when used with Talking Lights transmitters. To support existing installations, a Talking Light receiver could not be used in principle to also decode the transmissions of other optical information systems like remote infrared audible signage systems. Figure 2: Text Receiver Transceiver Set

Together, a fluorescent lamp transmitter and optical receiver form a complete Talking Lights transceiver set. We are exploring a number of approaches for configuring the transceiver set. For example, preprogrammed lights in a building could endlessly transmit simple numerical location codes or serial numbers. An intelligent receiver could use these codes to cue the presentation of visual or audio information stored or programmed in the receiver. In this scenario, a user might program the receiver when entering a building, perhaps from a CD-ROM or from light transmissions in the entryway. Relevant information would be presented as the user passed different light fixtures.

The lights could also be configured to broadcast custom information received from a wired input or a power-line carrier modem. In this case, the receiver would require little in the way of sophisticated data storage. Instead, it would simply present information received from the transmitters. This technique was used in the experimental trials to provide audio directional guidance to blind users.

Use for Wayfinding by Blind Participants

Eight blind participants were trained to find a mounted Talking Lights (TL) transmitter. They then traveled two indoor routes using information provided by TLs and two routes using verbal route descriptions (VD). All procedures were conducted individually.


Subjects:. Eight persons who had no more vision than light projection participated in this project. All were active, independent travelers. Each received an honorarium of $40.00 for their participation. Participants ranged in age from 32 to 54 with a mean of 44; 5 were male and 3 female; 7 used a long cane and one used a dog guide; 6 had been blind more than 4 years and 2 less than 4 years; 6 rated their travel capability as "excellent" and 2 as "good."

Materials: . The evaluation was conducted on the 2nd floor of the building in which Talking Lights, Inc. is a tenant. It is an old factory building with multiple and changing uses. The 2nd floor has irregular wall surfaces, a wide range of door and door frame styles,, and numerous pipes, ducts and conduits along the walls. In short, it is unusual, unpredictable building for persons without vision.

Four routes used for the evaluation; all had the same start. Nine TLs mounted on moveable wooden supports were used to label the environment with wayfinding information necessary to travel the four routes. Routes varied from 44.5’ to 76.2’ in length, required two to four turns, and required the use of three to four TL messages. Braille room numbers or labels (women/men) were provided at a height of 60" for all rooms along the route. The most complex series of messages was to the rest rooms:


"Room 1one is on the left. Rooms 10 through 20 and the restrooms are on the right."

At a T shaped intersection:

"Room 12 is on the left. Rooms 14 through 19 and the restrooms are on the right"

At a T-shaped intersection:

"Room 19 is on the left. Rooms 14 through 17 and the restrooms are straight ahead"

Between the Women’s/Men’s rest room:

"Women’s room to the left, men’s room to the right"

The corresponding restroom VD was: "Turn right. Take the next hall going right. The restrooms are in front of you at the end of the hall. Women’s room on the left, men’s room on the right."


Participants were familiarized with traveling a route using TL. They were positioned at home base and told they would be traveling to Room #14. The practice route was 62.3' long, had three turns, and used of four TLs. Participants were given verbal and physical assistance in using the TL system to travel this route.

Following familiarization with route travel using TL, participants traveled each of the four experimental routes, using TL for two and VD for two. Across the entire evaluation, each route was used four times with each of the two wayfinding conditions. Order of routes and wayfinding conditions were counterbalanced across participants. All routes began with participants in the same position. at the top of the stairs

For each condition the following measures were made:. 1) The time between the experimenter's stating the destination, and participant reaching the destination, i.e. touching the room door or partition. 2) The time between the experimenter's stating the destination, and until a participant placing a hand on the Braille sign identifying the destination. 3) N The number of times a participant "gave up" and asked to be taken back to home base to start again.


Because participants often took considerable time to locate by hand the Braille sign once in the vicinity of the destination, two travel time analyses were conducted: time to reach the destination and total time including finding the Braille sign to confirm the correct destination. Mean travel times to reach a destination are presented in Table 1 for each of the four routes for both the verbal directions and TL conditions. Corresponding total time means are presented in Table 2. Note that travel times are generally consistent with the varying length of each route (i.e., greater time for longer routes). Because of counter balancing measures used, a given participant traveled only two of the routes with verbal directions and two with TL. Thus for purposes of analysis a one-way repeated measures ANOVA was conducted to assess the effect of VD vs TL collapsed across the four routes. Mean times to reach the destination were 90 sec (se=30 sec) for VD and 117 sec (se=22 sec) for TL , which was not a statistically significant effect, F(1,7) = p< 1. Mean total times (including time to find the Braille sign at the destination) were 149 sec (se=27 sec) for VD and 136 sec (se=28 sec) for TL, again a non significant effect, F(1,7)) = p< 1.

It is important to note that for VD seven trials resulted in subjects having to restart a trial because of disorientation or lack of memory for the route’s verbal directions. In contrast, for TL only one trial required a restart. The implication of this finding is twofold. First, the travel times reported above include added time for the re-traveled routes (the clock was stopped while the participant was guided back to home base and then restarted once travel began anew). Thus the analysis of travel times above incorporates the restart measure. Second, negligible restarts for TL reflect the fact that no memory for the route is required on the part of the participant; the light’s messages were constantly available when a participant was in the vicinity of the light. Thus disorientation is unlikely.

Table 1: Mean Travel Time (in sec.) to reach destination. Table 2: Mean Total Time (including finding Braille sign,) in sec.)


Overall the finding of no difference in travel time for verbal directions vs.. Talking Lights is quite note-worthy. Verbal directions are only available to blind people if other people are present. Additionally, the routes and verbal directions used here were relatively simple and short. Longer routes and directions would presumably impose memory loads sufficiently high to result in travel disorientation and more restarts, a pattern of results beginning to emerge in the present data as noted above. The transceiver system using conventional fluorescent lights to transmit verbal information for wayfinding enabled blind users who were blind to travel indoor routes with no need for individual assistance.


The authors gratefully acknowledge the valuable advice and support of Deron K. Jackson whose. His assistance in creating figures for this paper was essential. The authors also wish to acknowledge the support of the U.S. Department of Education through grant 99-ED-3928, the National Science Foundation through grant 9860329 and the National Eye Institute through grant 1R43EY12470-01

Literature cited

[1] E. Brandt and A. Pope, eds., "Enabling America,'' E. Brandt and A. Pope, eds., Nat.ional Academy Press, Washington, D.C., 1997, pp. 135 --136

[2] J. Waymouth, Electric Discharge Lamps, MIT Press, Cambridge, Massachusetts, 1971.

[3] M. Dachs, "Optical Communication System,"'' U.S. Patent #3900404, August 1975.

[4] K. King, R. Zawislak, and R. Vokoun, "Boost-Mode Energization and Modulation Circuit for an Arc Lamp,"'' U.S. Patent #5550434, August 1996.

[5] M. Smith, "Modulation and Coding for Transmission using Fluorescent Lamp Tubes,"'' U.S. Pat.ent #5657145, August 1997.

[6] K. Uehara and K. Kagoshima, "Transceiver for Wireless In-Building Communication Sytem [sic],"'' U.S. Patent#5424859, June 1995.

[7] Buffaloe, T.K., D.K. Jackson, S.B. Leeb, M.F. Schlecht, and R.A. Leeb, "Fiat Lux: A Fluorescent Lamp Transceiver,"'' Applied Power Electronics Conference, Atlanta, Georgia, 1999.7

Go to previous article 
Go to next article 
Return to 2000 Table of Contents 
Return to Table of Proceedings

Reprinted with author(s) permission. Author(s) retain copyright.