Data transmission over wireless networks is challenging due to the occurrence of burst errors, and packet loss caused by such errors seriously limits the maximum achievable throughput of wireless networks. To tailor efficient transmission schemes, it is essential to develop a wireless error model that can provide insight into the behavior of wireless transmissions. In this study, we investigate the wireless error model of Bluetooth networks. We study the FHSS feature of Bluetooth using both ordinary hopping kernels and Adaptive Frequency Hopping (AFH) kernels, and design analytical error models accordingly to capture the channel behavior of Bluetooth networks. We evaluate the proposed models by comparing the analytical results to the simulation results obtained by Markov Chain Monte Carlo (MCMC) algorithms. The results show that our analytical models can represent the channel behavior of Bluetooth networks in all test cases.