Final published version
Research output: Contribution to Journal/Magazine › Journal article › peer-review
Research output: Contribution to Journal/Magazine › Journal article › peer-review
}
TY - JOUR
T1 - Smart stochastic routing for 6G-enabled massive Internet of Things
AU - Abbas, Ghulam
AU - Abbas, Ziaul Haq
AU - Ali, Zaiwar
AU - Asad, Muhammad Shahwar
AU - Ghosh, Uttam
AU - Bilal, Muhammad
PY - 2021/12/1
Y1 - 2021/12/1
N2 - Faster and energy-efficient data transmission is desired for massive Internet of Things (IoT) applications in sixth-generation networks. In such high speed networks, providing reliable data delivery with low delay, while maintaining energy-efficiency, is a challenging task. In this paper, a deep learning-based stochastic routing approach, called smart stochastic routing (SSR), is presented to address this challenge. SSR takes into account reliability, delays due to transmission, reception and processing of the neighbors’ information, and energy consumption and remaining energy of IoT devices. Through our proposed mathematical model, a dataset is generated to train a deep neural network, which predicts the best routing path from source to destination and achieves substantial accuracy over the mathematically generated dataset. Through simulations, we show the efficacy of SSR over conventional stochastic routing in terms of reduced energy consumption and expected delivery delay.
AB - Faster and energy-efficient data transmission is desired for massive Internet of Things (IoT) applications in sixth-generation networks. In such high speed networks, providing reliable data delivery with low delay, while maintaining energy-efficiency, is a challenging task. In this paper, a deep learning-based stochastic routing approach, called smart stochastic routing (SSR), is presented to address this challenge. SSR takes into account reliability, delays due to transmission, reception and processing of the neighbors’ information, and energy consumption and remaining energy of IoT devices. Through our proposed mathematical model, a dataset is generated to train a deep neural network, which predicts the best routing path from source to destination and achieves substantial accuracy over the mathematically generated dataset. Through simulations, we show the efficacy of SSR over conventional stochastic routing in terms of reduced energy consumption and expected delivery delay.
KW - Deep learning
KW - Energy efficiency
KW - Massive Internet of Things
KW - Stochastic routing
U2 - 10.1016/j.comcom.2021.09.015
DO - 10.1016/j.comcom.2021.09.015
M3 - Journal article
AN - SCOPUS:85116592901
VL - 180
SP - 284
EP - 294
JO - Computer Communications
JF - Computer Communications
SN - 0140-3664
ER -