Short: Achieving Energy Efficiency in Dense LoRaWANs through TDMA

Abstract

Long Range Wide Area Networks (LoRaWANs) have recently emerged as a hot research topic for their capability to collect sporadic data from a great number of widely spread low power devices. By enabling low-cost low-traffic wireless communications at large scale, such networks can be adopted in many application domains, including smart agriculture, logistics, and emergency detection among others. LoRaWAN employs a pure ALOHA default medium access scheme, that limits the maximum achievable throughput to 18%. Increasing the number of terminals would lead to more frame collisions, and eventually to network collapse. The easiest way to allow bigger network sizes is to synchronize devices, so that they can discretize time into slots, and use them to schedule frame transmissions. Herein, a Time Division Multiple Access (TDMA) scheme on the top of LoRaWANs is highly desired. However, since keeping synchronization can lead to heavy energy consumption, the design of such scheme should enforce energy-efficiency. This contribution describes the technical issues related to the implementation of such mechanism, that as a matter of fact represents the cornerstone on which more sophisticated random access protocols or even scheduling techniques can be designed. In dense networks, the proposed scheme combined with a very simple slotted ALOHA mechanism has also been shown to outperform the default LoRaWAN access protocol in terms of expected energy consumption.

TREMA: A traffic-aware energy efficient MAC protocol to adapt the LoRaWAN capacity

“…Several works in the literature have proposed channel access improvements to increase the scaling capabilities of LoRa, many of them based on synchronization and the consequent possibility to slice time into slots. Such enhancement allows to increase the maximum throughput, but we showed in a previous contribution [4] that the additional energy consumption induced by the process only makes it beneficial for high traffic situations.…”

“…The figure shows that for this deployment, Class A is capable of offering a maximum of 220 B • s −1 , while the scheduled access peaks at 440 B • s −1 . It therefore provides a 100% capacity increase, which is better than the 88% increase offered by slotted ALOHA over Class S in [4].…”

“…In a previous contribution, we proposed to rely on the LoRaWAN beacons for synchronization. In that, we defined Class S as an extension of Class B in order to introduce uplink transmission timeslots fitting the longest frame time-onair of the network [4]. The beacons were also leveraged in [15], with the goal of providing a contention-free access during a sub-portion of the beacon period.…”

“…The proposed time-synchronized resource scheduling protocol is built upon Class S, a TDMA scheme already introduced in a previous contribution [4]. Class S takes advantage of the Class B synchronization beacons broadcast by the gateways to define transmission timeslots.…”

LoRaWAN for Smart Campus: Deployment and Long-Term Operation Analysis

“…It has been found that introduction of LBT has both positive (e.g., the improved delivery rate for dense network and the potential for increasing the throughput, especially relevant for downlink) and negative consequences (e.g., the increased energy consumption and potentially lower performance in the multigateway environment). The use of slotted ALOHA and schedule-based (i.e., time-division multiple access (TDMA)) protocols have also been suggested [ 51 , 54 , 55 , 56 ]. The latter ones have been found especially beneficial for industrial use cases requiring guaranteed latency and higher reliability [ 55 , 56 ].…”