High Data Rate Efficiency Improvement via Variable Length Coding for LoRaWAN

Abstract

Low-cost options to provide local transactional network coverage are LPWAN (Low Power Wide Area Network), which are simpler to implement and to manage. The emerging technology known as LoRaWAN represents a promising LPWAN option due to its relative long range, as well as the technological development opportunities given its recent introduction and unfinished standard. The main LoRaWAN protocol consists of frames with fixed length encoded data from end-devices, no matter the length of such data; this choice is to simplify coding, transmission and decoding, in order to keep the low-power consumption of the network. In this work we present a modification of the LoRaWAN communications protocol in which payload data are encoded within customised frames which respect the nature and length of such data. The methodology is exemplified considering low-power end-devices (sensors) which generate 9 position, orientation and meteorological variables, and it is carefully implemented in order to simplify the decoding stage in the receiver as well as to optimise the manageability of the network traffic. The main outcome of this frame encoding of variable length technique is a considerable improvement of transmission data rates (with a factor of ${\sim }50{\times }$ for the chosen sensing modules) as the payload field is never filled with zeros up to the fixed size of the corresponding frame (according to original LoRaWAN standard). As the obtained frames are way smaller for this kind of sensors, the maximum number of end-devices supported by the network before its saturation point also increases substantially. This proposal is aimed to represent a viable and inexpensive option to cover the increasing demand in network coverage for end-devices as those used in Internet of Things and smart cities.