Optical Spatial Modulation (OSM) is a technique that uses (i) index of the LED, and (ii) intentional variations on the output light intensity of the LED to send information. Advantages of OSM are its simplicity of implementation and the ability to use all available LEDs to send data without interfering with each other.

Imagine there are two LEDs located on the ceiling of a room. Each LED can be set at either ‘ON1’ (half brightness) or ‘ON2’ (full brightness) and ‘OFF’ (dark). Additionally we impose a constraint that only one LED can be ‘ON’ at any given time. Now, as shown in table 1, there are four possible states that the LEDs can be set in.

Now for example let us assume we wish to transmit bits ‘10’ using this setup. The first bit ‘1’ indicates that LED ‘b’ should be ‘ON’. The second bit ‘0’ indicates that the LED must be set at ‘ON1’ (half brightness). Thus simply setting LED ‘b’ to half brightness transmits these 2 bits.

In order to receive this data, we need at least two photodiodes. In a typical handheld device, these photodiodes would be placed very close to each other. Additionally, the distance between the photodiodes would be very small as compared to the distance between the LEDs and the photodiodes. This implies that when either LED ‘a’ or ‘b’ is set at brightness level, the output currents from the photodiodes are very similar and the receiver cannot figure which LED is ON.

Instead, if we used a camera-like receiver, each LED would form its image at a different location on the sensor and the receiver can then use that information to detect which LED is ON.

A technical report on performance of OSM with 'camera-like' receiver can be found here