In response to the challenge of signal waveform variation in LoRa communications caused by information randomness, which complicates the establishment of a reference signal for matched filtering in distance measurement, this paper proposes an integrated sensing and communication method utilizing mismatched filtering of LoRa signals. This method employs a reference signal that disregards the waveform variations induced by information randomness. By performing pulse compression with an LFM signal and LoRa signals containing different information, a comprehensive theoretical analysis is conducted to reveal the relationship between LoRa signal information bits and the amplitude and position of the pulse compression peak. This relationship is then utilized to achieve demodulation based on peak amplitude. Furthermore, by determining the correlation between information bit changes and the displacement of the pulse compression peak position, peak position offset correction is achieved, enabling the sensing of LoRa signals under mismatched filtering conditions. Finally, the correctness and effectiveness of this method are validated through simulation experiments.