Semiconductor gas sensors have been widely used in environmental monitoring, toxic gas monitoring, respiratory marker monitoring and other fields due to their low price and high sensitivity etc. An easily overlooked parameter, ambient temperature, is a major factor affecting the accuracy of the sensor. In this paper, a new idea to suppress ambient temperature interference is proposed. Dynamic temperature modulation method is used to extend the dimension of output information, so that both ambient temperature and target gas concentration can be expressed simultaneously. The ambient temperature component is separated by coordinate transformation and component elimination. The data inverse transformation is carried out to achieve the purpose of suppressing the ambient temperature interference. Compared with the performance optimization of gas-sensing materials and additional temperature sensors, the first advantage of this method is to reduce the complexity and power consumption of sensor devices. In addition, the use of PCA algorithm also reduces the complexity of the recognition algorithm. At the same time, more intuitive data reduces the cost of parameter optimization.