Fog Computing in Wireless

Sensor Networks: Concepts,

Applications and Issues

Presented by Tian Wang

Huaqiao University, Xiamen, Fujian

2015/12/26

Outline

Fog Computing

Fog in Wireless Sensor Networks

Applications

Open Issues

1/32

Internet of Things (IoT)

• Connect both inanimate and living things

• Use sensors for data collection

• Everything have a digital identity and connectivity

2/32

Impact of IoT

• By 2020:

−50 billion connected things

−40 or 50 trillion GB data

• How to deal with them?3/32

IoT and The Cloud

• Cloud Computing:

−Connect all things to the Cloud

−Provide management, storage and computing services and

applications

4/32

Problems in the Cloud

• Communication bottleneck

−Some devices don’t have enough high bandwidth, especially

for cell phones, tablets and other portable devices

• Low real-time

−The cloud is actually deployed far away from connected

devices, which means a long-distance communication

through unstable networks with low real-time response

• Security risks

−The cloud is a centralized center with all data together and

designed to be convenient for various connection, which

becomes vulnerable for attackers

5/32

The Fog Framework

• The Fog:

−A intermediate layer between the cloud and end devices

−Provides storage and computing services in short distance

Centralized Core: the Cloud

Distributed Edge: the Fog

Various End: Things

Core

Edge

Location

6/32

Fog Computing

• An architecture that uses one or a collaborative

multitude of end-user clients or near-user edge

devices to carry out a substantial amount of storage

(rather than stored primarily in cloud data centers),

communication (rather than routed over the internet

backbone), and control, configuration, measurement

and management.

7/32

Features of Fog Computing

• Edge location, location awareness and low latency

• Geographical distribution

• Compatibility with various end devices

• Support for mobility and wirelesses

• Real-time interactions

• Heterogeneity

8/32

Other Computing Patterns

• Edge Computing pushes applications, data, and servicesaway from centralized nodes to the logical extremes of anetwork, leveraging resources that may not be continuouslyconnected to a network such as laptops, smartphones,tablets and sensors.

• Grid Computing is a form of distributed computingwhereby a “super virtual computer” is composed of manynetworked loosely coupled computers acting together toperform large tasks. It is the collection of computerresources from multiple locations to reach a common goal.

• Fog Computing is a kind of extension about cloudcomputing and based on the cloud, which is different fromlocal computing or grid computing that can be independentwith the cloud.

9/32

Smart Grid

• The fog process the data generated by grid sensors anddevices and send control commands to actuators

• The fog can implement energy load balancingapplications to switch alternative energies automatically

10/32

Smart Traffic Lights

• The fog interacts locally with sensors, detects presence of

pedestrian and bikers and measures the distance and speed of

approaching vehicles

• According to video camera, the fog can automatically change

street lights to open lanes for an ambulance to pass by11/32

Outline

Fog Computing

Fog in Wireless Sensor Networks

Applications

Open Issues

12/32

Wireless Sensor Networks (WSNs)

• Sensors:

−Generates a lot of data from surrounding environment

−Limited in sources and abilities such as energy, computing

and storage

Sink

Sensors

Manager

13/32

Sensor-Cloud

• Sensor-Cloud architecture is an integration of cloudcomputing and WSNs. The cloud provides vast storagecapacity and processing capabilities, enabling collectingthe huge amount of sensor data by the gateways onboth sides

Sink

Sensors

The Cloud

14/32

Problems in Sensor-Cloud

• Communication burden

−Sensors are limited in energy as well as communication ability

with a low bandwidth

• Weak real-time

−The cloud cannot respond to sensors’ real-time requests in

some scenarios

• Low dependability

−Sensors are inclined to faults with unstable connection and

inaccuracy data

• Security problem

−Data from sensors and in the cloud are all vulnerable

15/32

Advantages of the Fog in WSNs

• Efficient connectivity

−The fog can be deployed closely to the WSNs and even has

a direct connection to them

• High dependability

−The fog can adjust transmission burden and cycle duty of

sensors in order to optimize their energy efficiency and even

recover data

• High real-time

−The fog is close to WSNs and is able to response

immediately to sensors' real-time requests

• Strong Security

−The fog can use strong encryption towards connections to

the cloud

16/32

Outline

Fog Computing

Fog in Wireless Sensor Networks

Applications

Open Issues

17/32

Forest Fire Alarm

• High real-time:

−Fog nodes can respond response to fire detection in real-time

−Fog nodes will turn on extinguishers (actuators) automatically

18/32

Airplane Operation Data

• Local data storage and process:

−10TB data per half hour by sensors during flight

−The fog process data and provide emergency measures services

19/32

Structural Health Moniting

• High dependability:

−Fog devices can detect corrupt results by fault sensors

−The fog will recover those wrong diagnosis

20/32

Train Self-Maintain

• Self-maintaining system:

−The fog stores data from sensor which are on train's ball-

bearing and detect heat levels

−The fog alert the train operator to stop the train for emergency

21/32

Outline

Fog Computing

Fog in Wireless Sensor Networks

Applications

Open Issues

22/32

Researches and Open Issues

• Data transmission methods

• Fog devices cooperation

• Dependability awareness

• Security insurance

23/32

Data transmission methods

• The communication ability of one single sink in

WSNs is limited

• When many data come into one sink at the same

time, they will suffer from upload delay

• A mobile fog node can serve as a mobile sink where

data can be fused and compressed before uploading

• Several mobile fog nodes can build up a multi-input

multi-output (MIMO) network for load balance

24/32

Data Transmission Maximizing

• Mobile MIMO networks:

−Dynamic adjustment

A

B C

D

Mobile MIMO

Network

The Cloud

The Fog

25/32

Fog Devices Cooperation

• Several fog devices should cooperate together to

guarantee the quality of service (QoS) and improve

network performance

• Swarm intelligence method can be used to optimize

services and even to a modify physical motion of

mobile fog devices

26/32

Fog Devices Dispatch

• Mobile fog nodes:

−Cooperation by swarm intelligence

27/32

Dependability Awareness

• Sensors are inclined to faults, such as energy

outage, connection break, program halt and wrong

decision

• Fault sensors can corrupt results of a abnormal

event and make it without being detected

• Possible solutions in the fog:

−Fault detection methods

−Simple prediction and recovery of data

−Duty cycle optimization for underlying sensors

28/32

Security Insurance

• Convenient accessibility of the cloud raises security

risks at the same time

• Sensors fall short in strong encryption methods for

communication with the cloud

• The fog storage can distribute data among the fog

and the cloud in order to improve security

• The fog can establish strong encrypted connection

with the cloud

29/32

Conclusions

• Internet of things and cloud computing can be

mutually beneficial

• Fog computing is a significant extension for cloud

computing and wireless sensor networks

• The fog framework can be used in many scenarios

and applications

• There are also some open issues for future research

30/32

Thanks

Tian Wang

国立华侨大学Email: wangtian@hqu.edu.cn

2015/12/26

References

1. Evans D. The internet of things: How the next evolution of the internet is

changing everything [J]. CISCO white paper, 2011, 1: 1-14.

2. Gantz J, Reinsel D. The digital universe in 2020: Big data, bigger digital

shadows, and biggest growth in the far east [J]. IDC iView: IDC Analyze

the Future, 2012, 2007: 1-16.

3. Luis M Vaquero, Luis Rodero-Merino. Finding your way in the fog:

Towards a comprehensive definition of fog computing. ACM SIGCOMM

Computer Communication Review, 2014, 44(5): 27-32.

4. Shanhe Yi, Cheng Li, Qun Li. A Survey of Fog Computing: Concepts,

Applications and Issues. William and Mary University, 2015: 1-6.

5. Ivan Stojmenovic, Sheng Wen. The Fog computing paradigm: Scenarios

and security issues. Proceedings of the IEEE Computer Science and

Information Systems (FedCSIS), 2014 Federated Conference on, 2014:

1-8.

6. Flavio Bonomi, Rodolfo Milito, Jiang Zhu, et al. Fog computing and its role

in the internet of things. Proceedings of the ACM Proceedings of the first

edition of the MCC workshop on Mobile cloud computing, 2012: 13-16.

References

7. Md Bhuiyan, Guojun Wang, Jie Wu, et al. Dependable structural health

monitoring using wireless sensor networks. IEEE Transactions on

Dependable and Secure Computing, 2015, to be published: 1-14.

8. Gubbi J, Buyya R, Marusic S, et al. Internet of Things (IoT): A vision,

architectural elements, and future directions[J]. Future Generation

Computer Systems, 2013, 29(7): 1645-1660.

9. Alamri A, Ansari W S, Hassan M M, et al. A survey on sensor-cloud:

architecture, applications, and approaches[J]. International Journal of

Distributed Sensor Networks, 2013, 2013: 1-18.