Modulation and Detection in Wireless Optical Channels Using Temporal and Spatial Degrees of Freedom

Abstract

<p>Most wireless optica.l modems are able to modulate and detect only the intensity of tho optical carrier. As a result, conventional techniques, designed for radio frequency communications, arc seldom efficient. This thesis designs new efficient modulation and detection techniques for wireless optical communication systems that take into consideration the characteristics and constraints of wireless optical channels.</p> <p>Many degrees of freedom are available at the transmitter and the receiver in both ternpora1 and spatial domains. The main theme of this thesis is to use such degrees of freedom to achieve higher pO\ver/spectral efficiencies and simpler transceivers. A new modulation technique, optical impulse modulation (OIM), is proposed for indoor diffuse wireless infrared channels. Present-day laser diodes have higher pulse rates than the channel bandwidth. OIM utilizes such extra temporal degrees of freedom to satisfy the channel amplitude constraints, while the transmit data are confined to the lowpass region that represents the channel passband. Another modulation technique, halftoned spatial discrete m:ultitone (HSDMT) modulation, is proposed for indoor multi-input/multi-output (MIMO) point-to-point wireless optical links. Current spatial light modulators (SLMs) have higher spatial bandwidth than can be supported by the spatially lowpass MIMO channel. Such bandwidth provides extra spatial degrees of freedom that are employed by HSDMT modulation to decrease the transmitter complexity by considering binary-level SLMs. Finally, a novel detection technique is proposed which employs the receiver spatial degrees of freedom to mitigate the effects of atmospheric turbulence in outdoor free-space optical communications. By using digital micromirror devices, the receiver optically computes linear projections of the turbulence-degraded focal-plane signal distribution onto an orthogonal basis. These projections are used to select the portions of the focal-plane which contain significant energy for symbol detection. Performance improvements are quantified via simulations for all the proposed techniques.</p>