Performance Analysis of LoRa Modulation with Residual Frequency Offset

The rapid development of Internet of Things (IoT) puts forward much requirements for wireless communication technology. Low Power Wide Area Networks (LPWAN) are designed for low bandwidth, low power, long range and large number of connected IoT applications. As one of the LPWAN, Low Power Long Range Transceiver (LoRa) described as a Frequency Shift Chirp Modulation (FSCM), is widely concerned and studied. This paper analyzes the physical layer of LoRa, presents a basic mathematical signal processing model, and proposes a demodulation algorithm based on Fast Fourier Transform (FFT). The simulation performance of the demodulation algorithm is compared with the theoretical performance one of quadrature modulation incoherent detection. The results show that the performance derived through these two approaches is consistent. In addition, the performance of LoRa system with residual frequency offset under different spread factor (SF) is analyzed, and verified that the performance of the system decreases with the increase of SF. So we have to adopt better algorithm to correct frequency offset, otherwise the coverage will decrease significantly.