Regression analysis of refractivity gradient in the lowest 65m near the ground based on relevance vector machine algorithm

The change of the atmospheric refractivity gradient near the ground is an important factor affecting the performance of the radio system, and then, the radio environment in the lowest 65m near the ground is mainly for radio waves propagation such as shipborne and microwave base stations. This paper uses the atmospheric radiosonde data from 2016 to 2020, selects parts of China, uses the relevance vector machine (RVM) algorithm to obtain the refractivity gradient in the lowest 65m near the ground, and compares it with the linear numerical model. The results show that the accuracy of the results obtained by using the RVM algorihm is higher and better than the numerical model, which can better serve the radio system equipment.