A Survey on Energy Efficient Narrowband Internet of Things (NBIoT): Architecture, Application and Challenges

Abstract

The advancement of technologies over years has poised Internet of Things (IoT) to scoop out untapped information and communication technology opportunities. It is anticipated that IoT will handle the gigantic network of billions of devices to deliver plenty of smart services to the users. Undoubtedly, this will make our life more resourceful but at the cost of high energy consumption and carbon footprint. Consequently, there is a high demand for green communication to reduce energy consumption, which requires optimal resource availability and controlled power levels. In contrast to this, IoT devices are constrained in terms of resources-memory, power, and computation. Low power wide area (LPWA) technology is a response to the need for efficient utilization of power resource, as it evinces characteristics such as the capability to proffer low power connectivity to a huge number of devices spread over wide geographical areas at low cost. Various LPWA technologies, such as LoRa and SigFox, exist in the market, offering a proficient solution to the users. However, in order to abstain the need of new infrastructure (like base station) that is required for proprietary technologies, a new cellular-based licensed technology, narrowband IoT (NBIoT), is introduced by 3GPP in Rel-13. This technology presents a good candidature to handle LPWA market because of its characteristics like enhanced indoor coverage, low power consumption, latency insensitivity, and massive connection support towards NBIoT. This survey presents a profound view of IoT and NBIoT, subsuming their technical features, resource allocation, and energy-efficiency techniques and applications. The challenges that hinder the NBIoT path to success are also identified and discussed. In this paper, two novel energy-efficient techniques ''zonal thermal pattern analysis'' and energy-efficient adaptive health monitoring system have been proposed towards green IoT. INDEX TERMS Internet of Things (IoT), narrowband Internet of Things (NBIoT), low power wide area network (LPWAN), green communication, smart agriculture, smart health.

Machine Learning Methods for Monitoring of Quasiperiodic Traffic in Massive IoT Networks

“…Here, a number of pre-5G IoT systems for massive connectivity have been developed and are currently being deployed. These include the low power wide area networks (LPWANs): SigFox, LoRaWAN, NB-IoT and LTE-M [3], [4], [5]. A central problem of wireless IoT connectivity is monitoring and status detection for a massive number of connected devices, which provides insights into the status of the links and devices and can potentially lead to corrective actions [1], [6], [7].…”

“…2 while α i * has not converged do 3 Estimate η i m i for α i * . 4 Update the hypothesis for α i * based on η i m i . 5 Calculate KPIs based on app parameters.…”

“…Hypothesize α i * for an app i, given T x * RX . 3 if α i * is significant then 4 Label data points that best fit α i * and extract β i * and T i * RX .…”

“…2 if max(φ i ) > 0 then 3 Add the transmission T x M X to the app j which had the best fit. 4 Update the parameters, α i * , β i * .and the KPIs for i. 5 else 6 Add the transmission to the set of unlabeled transmissions T x * RX .…”

Computation Offloading for IoT in C-RAN: Optimization and Deep Learning

“…Internet-of-things (IoT) has a great potential to impact our lives in the future by providing solutions related to multiple sectors of industry, smart homes, transportation, etc. It is predicted that there will be about 50 billion IoT devices by 2020 [1], [2]. The deployment of a large-scale IoT ecosystem requires the IoT devices (IoTDs) with a small physical size to be built from costefficient hardware components, which results in major challenges due to their limited battery life and computational capability.…”