Energy Consumption Analysis of LPWAN Technologies and Lifetime Estimation for IoT Application

Abstract

The spectrum of Internet of Things (IoT) applications is exponentially growing, driving the demand for better energy performance metrics. In conjunction, Low Power Wide Area Networks (LPWAN) have evolved as long-range connectivity enabler with low management cost. The integration of LPWAN communication assists in reliable IoT operation with extended lifetime. Notable LPWAN technologies that contend for many of the IoT applications are LoRaWAN, DASH7, Sigfox, and NB-IoT. Most of the end-devices such as sensors and actuators are battery powered, therefore investigating energy consumption becomes crucial. To estimate the consumed power, it is important to analyze the energy consumption in wireless communication. This paper describes an empirical evaluation of energy consumption for LPWAN wireless technologies. We measure the current consumption of LoRaWAN, DASH7, Sigfox, and NB-IoT and derive the respective battery lifetime. These measurements help to quantify the energy performance of different protocols. We observe that LoRaWAN and DASH7 are more energy efficient when compared to Sigfox and NB-IoT. Finally, a case study on energy consumption is done on precision agriculture in the greenhouse, showing that battery lifetime in real applications can drop significantly from the ideal case. These results can be used for increasing the effectiveness of the IoT application by selecting the right technology and battery capacity.

Assessing the Potential of LPWAN Communication Technologies for Near Real-Time Leak Detection in Water Distribution Systems

“…In these cases, sensors with minimal power consumption have to be paired with energy-efficient communication technologies to allow maintenance-free operation without an external power source for multiple years. While most customer smart metering applications rely on highly energy-efficient radio technologies (e.g., versions of wireless M-Bus [ 13 ]) in combination with fixed network solutions [ 3 ], increasing coverage of LPWAN technologies [ 4 , 6 , 22 ] offers water utilities new possibilities to operate larger numbers of pressure-monitoring devices for fast leak detection.…”

“…To assess energy consumption of the EPMD, a simplified energy consumption model is employed, which was derived from multiple sources on the calculation of GPRS [ 73 ], NB-IoT [ 22 , 74 , 75 ], LoRaWAN [ 22 , 76 , 77 , 78 ] and Sigfox [ 22 , 79 ] energy consumption. Simplifications are based on operational experience with the EPMD and recent publications including analysis and field testing of LPWAN sensors and monitoring devices [ 13 , 27 ].…”

“…Measured energy consumptions for the rapidM2M M3 MCU are simplified to mean consumptions for all operating states, as well as mean durations for the operating states monitoring/logging and data transmission ( Figure 3 a). In a similar manner, power consumption and transmission specifications for the three LPWAN technologies, where no measured data was available, were derived from the MCU datasheet [ 71 ] and findings of similar studies [ 22 , 79 ]. By employing this approach, it was possible to include the specifications for all four communication technologies.…”

“…For the EPMD configurations in Figure 4 a, the feasibility was assessed individually for each communication technology, based on constraints of the energy consumption model, the encoding of the payload and transmission performance of the communication modules on the MCUs. Further, constraints and technology-specific limitations on payload or package size [ 22 ], as well as hourly device airtime limits or the maximum number of daily transmissions were considered. Since actual transmission rates may vary based on the conditions encountered at the installation site (e.g., inadequate network coverage), data transmission rates based on the values offered in the MCUs data sheets ( Table 3 ) were assumed.…”

“…Limitations on data transmission for NB-IoT at a frequency of 200 kHz can be contractual and specific to cellular network carriers. Nonetheless, the maximum payload size for a single NB-IoT transmission is limited to 1600 bytes [ 22 ], a constraint, which was considered by splitting respective transmissions into multiple subsequent bursts, similar to the approach for LoRaWAN.…”