The H i 21 cm global signal from the Cosmic Dawn and the Epoch of Reionization (EoR) offers critical insights into the evolution of our Universe. Yet, its detection presents significant challenges, due to its extremely low signal-to-contamination ratio and complex instrumental systematics. In this paper, we examine the effects of the ionosphere and antenna beam on data analysis. The ionosphere, an ionized plasma layer in the Earth’s atmosphere, refracts, absorbs, and emits radio waves in the relevant frequency range. This interaction results in additional spectral distortion of the observed signal, complicating the process of foreground subtraction. Additionally, chromatic variations in the beam can also introduce further contamination into the global spectrum measurement. Notably, the ionospheric effect, being dependent on the direction of incoming light, interacts with the instrumental beam, adding another layer of complexity. To address this, we evaluate three different fitting templates of foreground: the logarithmic polynomial, the physically motivated Experiment to Detect the Global EoR Signature (EDGES) template, and a singular value decomposition (SVD)-based template. Our findings indicate that the EDGES and SVD templates generally surpass logarithmic polynomials in performance. Recognizing the significance of beam chromaticity, we further investigate specific beam distortion models and their impacts on the signal extraction process.