Performance of LoRa-based Schemes and Quadrature Chirp Index Modulation

This paper proposes new approximations for the bit error probability of several LoRa-based schemes, such as phase-shift keying (PSK) LoRa, in-phase and quadrature (IQ) chirp spread spectrum (CSS), discrete chirp rate keying (DCRK) CSS, and CSS with index modulation (IM). We include a new scheme that combines IQ, CSS, and IM, denoted as IQCIM. Our analysis, which assumes coherent detection in additive white Gaussian noise, compares the LoRa-based schemes also in terms of spectral efficiency, computational complexity, and peak-to-average power ratio. Noncoherent detection of CSS with IM is analyzed as well. Our performance analysis shows that the proposed IQCIM produces a significant increase of spectral efficiency compared to LoRa with similar complexity, at an energy cost of approximately 0.9 dB. When compared to the other LoRa-based schemes, for a fixed decoding latency, the spectral efficiency of IQCIM is still the largest, with reduced computational cost per transmitted bit. In this context, we also propose a maximum-likelihood coherent detector for PSK-LoRa and a novel noncoherent detector for IQCSS.