Coded LoRa Performance in Wireless Channels

This paper focuses on the performance of LoRa systems after channel decoding. First, we introduce a matrix model for the coded transmission and reception of LoRa signals. Second, we theoretically evaluate the probability of bit error of coherently and noncoherently detected LoRa signals, in additive white Gaussian noise (AWGN), also assuming hard decoding of the LoRa physical layer Hamming code. Simulation results confirm the correctness of the theoretical expressions. Third, we analyze by simulation the coded performance of LoRa systems, with both hard and soft decoding, assuming a multipath channel model already used in the literature. Our analysis quantifies the performance gain of channel coding with respect to uncoded transmissions, in both AWGN and multipath channels, under the assumption that the coded system uses the same bandwidth with lower bit rate.