Carbon monoxide (CO) is a toxic gas, and environmental pollutant. Its detection and control in residential and industrial environments are necessary in order to avoid potentially severe health problems in humans. In this review paper, we discuss the importance of furthering research in CO sensing technologies for finding the proper material with low-range detection ability in very optimum condition. We build our discussion through the perspective of a cyber-physical system (CPS) modeling framework, because it provides a comprehensive framework to model and develop automated solutions for detection and control of poisonous chemical compounds, such as the CO. The most effective CO sensors, then, can be used in CPS network to provide a pathway for real-time monitoring and control in both industrial and household environment. In this paper, first, we discuss the necessity of CO detection, the proposal of a basic CPS framework for modeling and system development, how the CPS-CO model can be beneficiary to the environment, and a general classification of the various CO detection mechanisms. Next, a broad overview emphasizes the sensitivity, selectivity, response and recovery time, low concentration detection ability, effects of external parameters and other specifications that characterize the performance of the sensing methods proposed so far. We will discuss recent studies reported on the use of metal oxide semiconductor (MOS) sensing technologies for the detection of CO. MOS based micro-sensors play an important role in the measurement and monitoring of various trace amounts of CO gas. These sensors are used to sense CO through changes in their electrical properties. In addition to MOS based sensors, optical sensing methods have recently become popular, due to their increased performance. Hence, a brief overview of newly proposed optical based CO detection methods is provided as well.