Skip to Main content Skip to Navigation

Study and improvement of long range communication technologies for wireless sensor networks

Abstract : The progress in low-energy, low-cost communication technologies have revolutionized remote sensing and monitoring applications. Internet of Things (IoT) has promised an ecosystem of connected devices across a wide range of applications such as in smart cities.Currently, many competing standards and technologies are attempting to seize the IoT, particularly in the area of remote sensing and communication technologies. LoRa (Long Range) is one of those technologies that is gaining popularity and attraction in the Wireless Sensor Networks (WSN) applications. The ability to make long-distance communications with relatively simple nodes, minimal infrastructure, reduced power requirements, and the use of unlicensed ISM bands provides a significant competitive advantage. Although the communication range in LoRa can exceed 15 kilometers in line of sight, the maximum bit rate that can be achieved is limited to few kilobits per second. Additionally, when a collision occurs in LoRa, the throughput is further reduced due to frame losses and retransmissions. The work of this thesis deals with the problem of collisions in LoRa that may occur under heavy load, and which degrade the performance of the network.First, we consider the context for LoRaWAN uplink communications. We study the context of fully synchronized colliding LoRa signals, where each end-device has to retransmit its entire colliding frame after a collision occurs in LoRa. This behavior decreases the overall throughput, and increases the energy consumption of the end-devices, and the delay of the frames. Therefore, in order to mitigate the damaging effects of collisions, we proposed a decoding algorithm to resolve synchronized colliding LoRa signals, in a saturated and confirmed network traffic. We substituted the conventional retransmission model of LoRa by having end-devices transmitting bitmaps instead of retransmitting whole frames to determine the correct symbols of each colliding frame. Our algorithm was able to significantly improve the overall throughput of the LoRaWAN MAC layer based on LoRa, and to decrease the energy consumption of the transmitters and the delay of the frames.Second, we consider the context for LoRaWAN downlink communications. We noticed that the downlink in LoRa is a bottleneck. Hence, we worked on the gateway selection by the network server and its impact on the throughput, the energy consumption and the delay. We studied three types of gateway deployment and we show that the system performance depends on this deployment. We showed that balancing the number of end-devices per gateway (also known as load) improves the throughput compared to choosing the gateway with the highest signal quality. Moreover, we showed that combining load and signal quality does not further improve the throughput. In addition, we showed that choosing the gateway with the highest signal quality decreases the delay and energy consumption compared to choosing the gateway with the lowest load.
Complete list of metadata
Contributor : Abes Star :  Contact
Submitted on : Friday, December 18, 2020 - 4:38:45 PM
Last modification on : Wednesday, February 24, 2021 - 4:24:03 PM


Version validated by the jury (STAR)


  • HAL Id : tel-03082631, version 1



Samira Abboud. Study and improvement of long range communication technologies for wireless sensor networks. Networking and Internet Architecture [cs.NI]. Université Clermont Auvergne, 2020. English. ⟨NNT : 2020CLFAC028⟩. ⟨tel-03082631⟩



Record views


Files downloads