No abstract

LoRaWAN is a promising low power long range wireless communications technology for the Internet of Things. An important feature of LoRaWAN gateways is related to so-called capture effect: under some conditions the gateway may correctly receive a frame even if it overlaps with other ones. In this paper, we develop a pioneering mathematical model of a LoRaWAN network which allows finding network capacity and transmission reliability taking into account the capture effect.

No abstract

No abstract

LoRaWAN infrastructure has become widely deployed to provide wireless communications for various sensor applications. These applications generate different traffic volumes and require different quality of service (QoS). The paper presents an accurate mathematical model of low-power data transmission in a LoRaWAN sensor network, which allows accurate validation of key QoS indices, such as network capacity and packet loss ratio. Since LoRaWAN networks operate in the unlicensed spectrum, the model takes into account transmission attempt failures caused by random noise in the channel. Given QoS requirements, we can use the model to study how the performance of a LoRaWAN network depends on the traffic load and other scenario parameters. Since in LoRaWAN networks the transmissions at different modulation and coding schemes (MCSs) typically do not collide, we use the model to assign MCSs to the devices to satisfy their QoS requirements.

No abstract

In the Internet of Things scenarios, it is crucially important to provide low energy consumption of client devices. To address this challenge, new Wi-Fi standards introduce the Target Wake Time (TWT) mechanism. With TWT, devices transmit their data according to a schedule and move to the doze state afterwards. The main problem of this mechanism is the clock drift phenomenon, because of which the devices cease to strictly comply with the schedule. As a result, they can miss the scheduled transmission time, which increases active time and thus power consumption. The paper investigates uplink transmission with two different TWT operation modes. With the first mode, a sensor transmits a packet to the access point (AP) after waking up, using the random channel access. With the second mode, the AP polls stations and they can transmit a packet only after receiving a trigger frame from the AP. For both modes, the paper studies how the average transmission time, the packet loss rate and the average energy consumption depend on the different TWT parameters. It is shown that when configured to guarantee the given packet loss rate, the first mode provides lower transmission time, while the second mode provides lower energy consumption.

No abstract