Fog Networking for Machine Health Prognosis: A Deep Learning Perspective

Ensuring machine health and predicting failures beforehand is of utmost importance. Fog networking is an emerging paradigm of computing. Advances in neural networks research and their benefits in accurate classification make them best candidate for application to prognostics. Ubiquity of smartphones, wearables and sensors is giving rise to tremendous amount of data necessitating intelligent data analytics at the data center. Various researches have focused on compression of large deep models for inferencing on smart devices in order to seek answers to issues like reducing the enormity of computation, decreasing the power and memory requirements on these devices. At this point in time, state of the art in inference such as recurrent neural networks (RNN) seem impossible to be implemented on resource constrained devices. Cloud based inference incurs cost in terms of power as more energy is depleted to access infrastructure via technologies such as 3g/4g networks while also witnessing latency issues due to network congestion. A few real world situations offload inferencing to network edge as it is more beneficial to solve many issues like security, privacy of users, redundancy rather than uploading enormous volume of data on cloud. The paradigm, known as Fog Computing, is a hot topic in research facilitating edge analytics by overcoming constraints and shortcoming of cloud enabling quality of service (QoS) required in some real world applications such as Industrial Internet of Things (IoT). Fog acts as a medium between edge device and cloud. In this paper, we propose a hybrid approach of using Fog in conjunction with deep neural nets for inferencing. We propose how neural nets exploited by Fog/Cloudlet can provide computational and storage services to nearby devices and it will also put a barrier by filtering the voluminous data to a reduced form for cloud.

[1]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[2]  Meikang Qiu,et al.  Health-CPS: Healthcare Cyber-Physical System Assisted by Cloud and Big Data , 2017, IEEE Systems Journal.

[3]  Guangquan Zhao,et al.  Research advances in fault diagnosis and prognostic based on deep learning , 2016, 2016 Prognostics and System Health Management Conference (PHM-Chengdu).

[4]  Eunhyeok Park,et al.  Compression of Deep Convolutional Neural Networks for Fast and Low Power Mobile Applications , 2015, ICLR.

[5]  Mi-Young Lee,et al.  Hierarchical Compression of Deep Convolutional Neural Networks on Large Scale Visual Recognition for Mobile Applications , 2016 .

[6]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[7]  Ivan Stojmenovic,et al.  The Fog computing paradigm: Scenarios and security issues , 2014, 2014 Federated Conference on Computer Science and Information Systems.

[8]  Tao Zhang,et al.  Fog and IoT: An Overview of Research Opportunities , 2016, IEEE Internet of Things Journal.

[9]  Zhuo Chen,et al.  Edge Analytics in the Internet of Things , 2015, IEEE Pervasive Computing.

[10]  Adam Coates,et al.  Deep Voice: Real-time Neural Text-to-Speech , 2017, ICML.

[11]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[12]  Kay Chen Tan,et al.  Multiobjective Deep Belief Networks Ensemble for Remaining Useful Life Estimation in Prognostics , 2017, IEEE Transactions on Neural Networks and Learning Systems.

[13]  J. Wenny Rahayu,et al.  Mobile cloud computing: A survey , 2013, Future Gener. Comput. Syst..

[14]  Sateesh Addepalli,et al.  Fog computing and its role in the internet of things , 2012, MCC '12.

[15]  Ruqiang Yan,et al.  Learning to Monitor Machine Health with Convolutional Bi-Directional LSTM Networks , 2017, Sensors.

[16]  Nicholas D. Lane,et al.  DeepX: A Software Accelerator for Low-Power Deep Learning Inference on Mobile Devices , 2016, 2016 15th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN).

[17]  Trishul M. Chilimbi,et al.  Project Adam: Building an Efficient and Scalable Deep Learning Training System , 2014, OSDI.

[18]  Ying Peng,et al.  Current status of machine prognostics in condition-based maintenance: a review , 2010 .

[19]  Xiaoli Li,et al.  Deep Convolutional Neural Network Based Regression Approach for Estimation of Remaining Useful Life , 2016, DASFAA.

[20]  Fei-Fei Li,et al.  Visualizing and Understanding Recurrent Networks , 2015, ArXiv.