System-Level, Model-Based Power Estimation of IoT Nodes

Ozkaya, Ozen; Örs, Berna

doi:10.1109/wf-iot51360.2021.9595622

Cited by 1 publication

(7 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As depicted in Figure 5, a sensor node is composed of four modules: a data processing module, which includes a micro-controller unit (MCU); a communication module, which includes a radio transceiver; a sensing/actuator module, which includes sensors and actuators; and a power module, which includes a battery and, on occasions, energy harvesting [12,26,27]. Models in this category are those whose metrics reflect the performance of individual sensor nodes [8].…”

Section: Node Modelsmentioning

confidence: 99%

“…These models are subdivided into energy efficiency, timing, and coverage, as shown in Figure 6. The energy efficiency models analyze the sensor node metrics to determine how much energy the nodes consume [12,26]. The timing model is related to time synchronization metrics [28].…”

Section: Node Modelsmentioning

confidence: 99%

“…The sensor node requires energy to perform multiple tasks, such as sensing, processing and data transmission [11,26]. Data processing, communication and sensing blocks typically consume the most energy; hence, reducing energy consumption in these blocks is the primary goal for prolonging the lifetime of the sensor node.…”

Section: Energy Efficiency Modelsmentioning

confidence: 99%

“…Furthermore, MCUs have different operation states in which multiple combinations of MCUs' modules are active [8]. The most basic approach to modeling data processor energy is counting the operating system (OS) ticks represented by σ, which is the period of execution task scheduling and other OS services [26]. The OS tick amount is multiplied by the average power consumption (P Oper (t)) value in this approach.…”

Section: Data-processing Modelsmentioning

confidence: 99%

“…An ADC is a device used to collect data from analog sensors. A communication interface, such as a serial peripheral interface (SPI), an inter-integrated circuit (I2C), or a universal asynchronous receiver transmitter (UART) are typically used to read digital sensors [26]. Furthermore, the data capture methodology can be synchronous, in which sensing occurs in a periodic time interval (T s ), or asynchronous, in which sensing is event-driven and can be modeled as a probabilistic distribution for the event (P r (e)).…”

Section: Sensor Energy Modelsmentioning

confidence: 99%

See 4 more Smart Citations

On Wireless Sensor Network Models: A Cross-Layer Systematic Review

Ojeda

Méndez

Fajardo

et al. 2023

JSAN

View full text Add to dashboard Cite

Wireless sensor networks (WSNs) have been adopted in many fields of application, such as industrial, civil, smart cities, health, and the surveillance domain, to name a few. Fateway and sensor nodes conform to WSN, and each node integrates processor, communication, sensor, and power supply modules, sending and receiving information of a covered area across a propagation medium. Given the increasing complexity of a WSN system, and in an effort to understand, comprehend and analyze an entire WSN, different metrics are used to characterize the performance of the network. To reduce the complexity of the WSN architecture, different approaches and techniques are implemented to capture (model) the properties and behavior of particular aspects of the system. Based on these WSN models, many research works propose solutions to the problem of abstracting and exporting network functionalities and capabilities to the final user. Modeling an entire WSN is a difficult task for researchers since they must consider all of the constraints that affect network metrics, devices and system administration, holistically, and the models developed in different research works are currently focused only on a specific network layer (physical, link, or transport layer), making the estimation of the WSN behavior a very difficult task. In this context, we present a systematic and comprehensive review focused on identifying the existing WSN models, classified into three main areas (node, network, and system-level) and their corresponding challenges. This review summarizes and analyzes the available literature, which allows for the general understanding of WSN modeling in a holistic view, using a proposed taxonomy and consolidating the research trends and open challenges in the area.

show abstract