Chirp Index Space Partitioning Based Multiple Access Technique

The rapid development and popularization of Internet-of-Things (IoT) technologies have driven demand for low-power wide-area networks (LPWANs) for years. Long range wide area network (LoRaWAN), which enables long-distance communications with low power consumption, has increased in recent years. Although LoRaWAN is capable of low-power long-distance communication, its data rate is low. Most of the existing research works focus on increasing the data rate of a single device. However, from the viewpoint of LPWAN applications, it is desirable to accommodate a number of devices with a low data rate. This paper aims at increasing the system data rate by dividing the chirp index space into orthogonal subspaces and allocating nonoverlapping subspaces to multiple devices, which enables simultaneous transmission from multiple devices. In addition, in order to demodulate the multiple signals transmitted on the distinct chirp index subspace, it is necessary to understand the receive timing difference between the multiple signals received at GW. In this paper, we also propose a method to estimate the receive timing difference at GW using the preamble sent by LoRa devices. Through computer simulations, we show that the proposed method improves the system data rate by up to 175% compared to the conventional LoRa system.