An optimal RSSI-based cluster-head selection for sensor networks

Jain, Khushboo; Kumar, Anoop

doi:10.1504/ijais.2019.10030287

Cited by 10 publications

(12 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In place of changing CH's for every round, the authors have introduced an optimal CH‐threshold function and energy‐threshold function to avoid unnecessary CH changes and topology of the sensor network, thus increasing the network lifespan. Extensive simulation proved that EESCT 21 minimizes energy consumption and expands the network‐lifespan when compared with existing protocols over numerous metrics.…”

Section: Related Workmentioning

confidence: 99%

“…A previous study 21 had improved the sensor network lifetime based on the remaining energy intensities of cluster‐heads. The network was initialized by using the LEACH algorithm of Equation , which is stated as:

P_{i} ((), t) goodbreak= ({, \begin{matrix} \frac{o}{n - o true(r italicmod ((), n / o)} {CH}_{i} ((), t) goodbreak= 1 \\ 0 {CH}_{i} ((), t) goodbreak= 0 \end{matrix})

where

n, o, italicand r

are the number of SNs, initial optimal CH range, and the number of rounds, respectively.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

An innovative framework for balanced cluster‐based data aggregation in sensor networks

Jain

Kumar

2022

Int J Communication

Self Cite

View full text Add to dashboard Cite

Summary The essential design concern of a sensor network is to balance energy consumption of sensor nodes (SNs) to prolong the network lifetime. Many research works cited that clustering techniques efficiently utilize the network's energy resource by organizing SNS into groups of clusters and benefits in reduced data transmission. An extensively used category of cluster‐based protocols is the probabilistic clustering technique in which a preset optimum likelihood is used to facilitate the selection procedure of cluster head (CH). These clustering techniques suffer from non‐uniform dissemination of CH, which leads to uneven load balance and uneven energy consumption of SNs during network activities like data transmitting and data receiving. This causes an energy‐hole problem and reduces network lifetime. In order to solve these issues, we have focused on to design a balanced cluster‐based data aggregation and formulate a method that increases the energy efficiency of probabilistic clustering techniques by optimizing the number of clusters and the dissemination of CHs in the sensor network. The simulation analysis proves that the proposed technique accomplishes significantly enhanced than the existing works.

show abstract

Section: Related Workmentioning

confidence: 99%

P_{i} ((), t) goodbreak= ({, \begin{matrix} \frac{o}{n - o true(r italicmod ((), n / o)} {CH}_{i} ((), t) goodbreak= 1 \\ 0 {CH}_{i} ((), t) goodbreak= 0 \end{matrix})

where

n, o, italicand r

are the number of SNs, initial optimal CH range, and the number of rounds, respectively.…”

Section: Related Workmentioning

confidence: 99%

An innovative framework for balanced cluster‐based data aggregation in sensor networks

Jain

Kumar

2022

Int J Communication

Self Cite

View full text Add to dashboard Cite

show abstract

“…Dahl et al developed a multitask deep learning model to predict the chemical structure of molecules, the pharmacophore of the active site, and drug levels toxic to the active site (Lv et al, 2019 ; Lin et al, 2020 ; Zeng et al, 2020b ). Ramsundar et al proposed a deep neural network model that efficiently predicts drug activity and structure (Wallach et al, 2015 ; Jain and Kumar, 2019 ; Zhao et al, 2019 ).…”

Section: Introductionmentioning

confidence: 99%

SE-OnionNet: A Convolution Neural Network for Protein–Ligand Binding Affinity Prediction

et al. 2021

View full text Add to dashboard Cite

Deep learning methods, which can predict the binding affinity of a drug–target protein interaction, reduce the time and cost of drug discovery. In this study, we propose a novel deep convolutional neural network called SE-OnionNet, with two squeeze-and-excitation (SE) modules, to computationally predict the binding affinity of a protein–ligand complex. The OnionNet is used to extract a feature map from the three-dimensional structure of a protein–drug molecular complex. The SE module is added to the second and third convolutional layers to improve the non-linear expression of the network to improve model performance. Three different optimizers, stochastic gradient descent (SGD), Adam, and Adagrad, were also used to improve the performance of the model. A majority of protein–molecule complexes were used for training, and the comparative assessment of scoring functions (CASF-2016) was used as the benchmark. Experimental results show that our model performs better than OnionNet, Pafnucy, and AutoDock Vina. Finally, we chose the macrophage migration inhibitor factor (PDB ID: 6cbg) to test the stability and robustness of the model. We found that the prediction results were not affected by the docking position, and thus, our model is of acceptable robustness.

show abstract

“…In this context, data aggregation 10 has emerged as an intelligent method of collecting data from different SNs at intermediate nodes called cluster heads (CHs), 11 limiting the number of data packets to be transmitted to the BS. In addition, in the architecture of cluster‐based WSNs, the CHs consume more energy than SNs because CHs also perform intracluster data aggregation and then transmit the collected data to the BS 12,13 .…”

Section: Introductionmentioning

confidence: 99%

A lightweight data transmission reduction method based on a dual prediction technique for sensor networks

Jain

Kumar

2021

Trans Emerging Tel Tech

Self Cite

View full text Add to dashboard Cite

An essential design concern in a resource‐constraint sensor network is optimizing data transmission for each sensor node (SN) to prolong the network lifetime. Many research works cited that the dual prediction technique remains the most efficient technique for data reduction. A large amount of redundant data is usually transmitted across the network, leading to collisions, loss of data, and energy dissipation. This article proposes a data transmission reduction method (DTRM) to solve these problems, implemented on the cluster heads and operates in rounds. DTRM is lightweight in processing, has low complexity costs, and needs a limited memory footprint, but it is robust and effective. It can be combined with any form of cluster‐based data aggregation. We have incorporated the proposed DTRM with the data aggregation‐adaptive frame method (DA‐AFM), implemented on the SNs within the clusters. DA‐AFM can eliminate temporal redundancies and correlations in the sensor's time‐series readings. This helps the SN take fewer readings, which improves the efficiency of reducing data transmission and decreases the amount of energy spent during sensing. The proposed DTRM approach decreases the average transmission rates of data while maintaining data quality. This study is evaluated on real data obtained from the Intel Berkeley Lab and compared with three recent data reduction techniques focused on prediction. The results show that DTRM consumes up to 70% less energy while preserving the expected quality of data and reducing transmission.

show abstract

An optimal RSSI-based cluster-head selection for sensor networks

Cited by 10 publications

References 14 publications

An innovative framework for balanced cluster‐based data aggregation in sensor networks

An innovative framework for balanced cluster‐based data aggregation in sensor networks

SE-OnionNet: A Convolution Neural Network for Protein–Ligand Binding Affinity Prediction

A lightweight data transmission reduction method based on a dual prediction technique for sensor networks

Contact Info

Product

Resources

About