The increasing demand for wireless sensing systems has led to the exploration of alternative technologies to overcome the spectrum scarcity of traditional approaches based on radio frequency (RF) waves or microwaves. Incoherent light sources, such as light-emitting diodes, offer potential as an attractive option for wireless sensing. Unlike coherent light or lasers, incoherent visible and infrared light is low-intensity and safe for human eyes and skin. However, the application of incoherent light in wireless sensing is still an emerging research topic. This dissertation proposal explores the immense potential of incoherent light in wireless sensing through its various indoor and outdoor applications.

First, we propose a new system model for incoherent light-wave sensing (LWS), where incoherent light is transmitted to the subject and the modulated reflected light is captured using a light sensor. Signal processing techniques and machine learning are employed to analyze the collected data, providing valuable insights about the subject in a non-contact and safer manner. Additionally, a few applications of LWS from the literature including vehicular speed estimation, vitals monitoring, glucose sensing, occupancy estimation and structural health monitoring are reviewed in brief. Second, the applications of machine learning in various stages of non-contact vitals monitoring using RF and camera-based sensing have been thoroughly reviewed from recent literature. Third, we present a non-contact respiratory anomaly detection system using infrared sensing that can classify normal breathing, 6 different types of abnormal breathing and erroneous data with up to 96.75% accuracy. Robot-simulated respiration validates the proof of concept, and the system performs effectively even with training data collected at variable distances within 1.5m from the LWS setup. Finally, we explore another application, hand gesture recognition using incoherent light, where infrared sensing differentiates among 8 different gestures within a 20-35cm range with 96% average accuracy. Incoherent light has the potential to supersede other wireless sensing technologies like RF, laser and camera, by providing additional benefits including easy implementation, reusable frequency, minimum interference, enhanced privacy and simpler data processing. This dissertation sets the foundation for extensive investigation into the advantages and constraints of leveraging incoherent light in the realm of wireless sensing.