Cloud Controlled Intrusion Detection and Burglary ... - IEEE Xplore

15 downloads 66617 Views 1MB Size Report
hoc wireless home automation system with en suite intrusion detection and burglary prevention stratagems. Along with an improved infrared camera, each node ...
2012 2nd Baltic Congress on Future Internet Communications

Cloud Controlled Intrusion Detection and Burglary Prevention Stratagems in Home Automation Systems Anindya Maiti and S. Sivanesan School of Computing Science & Engineering VIT University Vellore, India. [email protected], [email protected] In addition to the benefits of cloud computing we used the assistance of location-aware services, where the geolocation of each node of a home automation system is independently detected [3] and stored in the cloud. The location of each node helps the cloud services to stealthily alert the accurate neighbors (who are using our home automation system) or local police, in an event of genuine intrusion. Later in this paper, we discuss how we remotely control and monitor security surveillance system via the cloud using any Internet enabled device as a remote controller, even when we are not in the proximity of the house. To conclude we emphasize the scope of other cloud services that can be integrated in the cloud based home automation system, along with the security surveillance system.

Abstract - Security surveillance partakes in significant number of home automation systems, deploying digital cameras and sensors to monitor and report intrusion events and thereby reducing damages caused by burglary. In this paper, we present the design, implementation and operation of a cloud connected adhoc wireless home automation system with en suite intrusion detection and burglary prevention stratagems. Along with an improved infrared camera, each node of our home automation system has devised intelligent algorithms for intrusion detection and subsequently reports any event to a location-aware cloud service in real-time. In case of an intrusion event, another cloud service alerts the user with a SMS conversation. The user can then monitor the intrusion from anywhere, on any Internet enable device by accessing the cloud’s web interface. If the intrusion is genuine, the user is provided with options to stealthily alert neighbors (who are using our home automation system), play alarm sounds or even report to the police. Using these techniques, burglary can be evaded effectively.

II.

In our earlier works, we designed and implemented our cloud based home power management system (a vital part of most home automation systems). The design of our home automation system makes it low cost, flexible and easy to install. We replaced the traditional electrical switch board of each room with our cloud connected board where each board is a node of the home automation system, as a result creating an ad-hoc wireless network among all the boards in a household, using 802.11n standard. The use of 802.11n empowers us to create an ad-hoc network with adequate signal range to operate across a house and concurrently connect each node of the distributed home automation system directly to the cloud over Internet Protocol, without the requirement of dedicated or specialized gateway as commonly required [4]. In addition to an 802.11n radio, every node can be equipped with a low cost Universal Mobile Telecommunications System (UMTS) Release 5 modem. Using such a UMTS modem, we configured one board to act as the Internet gateway for an entire ad-hoc network, by connecting it to a public High-Speed Downlink Packet Access (HSDPA) network of a national Internet Service Provider (ISP) with down-link speed of 7.2 Megabit/s and uplink speed of 1.8 Megabit/s.

Keywords - burglary; home automation; intrusion detection; cloud; geolocation; fisheye lens; motion detection.

I.

INTRODUCTION

A home automation system delivers state-of-the-art accessibility, comfort, energy efficiency and security, by providing control and monitoring of illumination, heating, ventilation, air conditioning, appliances, security surveillance, multimedia and other systems [1]. Of our primary interest, security surveillance systems are an integral part of numerous home automation systems. We developed our own cloud services which can be used in homes around the world to control and monitor cloud connected home automation systems remotely [2]. Connecting the home automation system to the cloud not only reduces the setup and maintenance cost by eliminating the need of dedicated gateway and web server in each household, but also enables us to deliver additional services to the home automation system, like a cloud controlled power management has been implemented. Our cloud is based on the standard cloud computing model, in which we make our services available to the general public over the Internet as long as they use our web interface. In this paper, once again we call attention to the advantages of cloud based home automation systems, this time from the perspective of a security surveillance system.

Figure 1 shows the internal and external photographs of the board in use. As a part of the power management system, home appliances are plugged into the AC power outputs which are governed by solid state relays controlled by a microcontroller. The microcontroller in turn is controlled by a low cost microprocessor which also controls the network radios of both

978-1-4673-1671-2/12/$31.00 ©2012 IEEE

978-1-4673-1671-2/12/$31.00 ©2012 IEEE

EARLIER WORKS

182

802.11n and UMTS (if configured) over a common bus. The touchscreen on the exterior of the board can be used as a local controller of the power management system, which delivers faster response time than a remote controller. The operating configuration of the 802.11n radio chip (Qualcomm Atheros XSPAN-AR9287) is provided in Table I. The wireless network is secured by Wi-Fi Protected Access II (WPA2) and encrypted with Counter Mode with Cipher Block Chaining Message Authentication Code Protocol (CCMP) encryption protocol. Table I. Operating configuration of the 802.11n radio [5].

802.11n Frequency

5 GHz

Bandwidth

40 MHz

Data rate

150 Mbit/s per stream

MIMO streams

2

Modulation

OFDM

Approximate indoor range

60m

200ft

Approximate outdoor range

200m

650ft

III.

Fig. 1. External (top) and internal (bottom) snapshots of an operational cloud connected board (node).

additional wide-angle (also known as fisheye) lens to the camera [6], so as to have a full 180° view of observation (Figure 2). Since the board is installed on a wall, the camera acquires a complete view of a room. And since we replaced the electrical switch board of every room with our board, we get a complete view of the entire house. An application running on each board detects intrusion by the technique of motion detection [7], captured in the infrared camera. In case of an intrusion event, the application communicates with the cloud in order to notify the user.

THE SECURITY SURVEILLANCE SYSTEM

We installed supplementary security surveillance hardware and software to the cloud based home automation system we had implemented earlier. We installed software based location detection system using the following JavaScript example of geolocation application programming interface (API): if(navigator.geolocation) { navigator.geolocation.getCurrentPosition(function(position) { alert('Your lat-long is: ' + position.coords.latitude + ' / ' + position.coords.longitude); alert('You live in ' + position.address.city + ', ' + position.address.region) }); }

Using the above code, we obtain the geolocation of each board (node) of all independent home automation networks in latitude and longitude. This geolocation information is stored in the cloud and is update on a regular basis. On the hardware part, we installed an improved infrared camera in each node. The improvement is made by fitting an Fig. 2. An illustration of how an infrared camera attached to each board would get a full 180° view of a room.

183

Fig. 3. Schematic diagram of how a user (B) is notified about an intrusion event and how he takes action to prevent burglary by alerting neighbor (user A), informing local police or ringing loud alarm bell sound.

IV.

THE CLOUD SERVICES

Cloud computing (or simply cloud) refers to the online services provided over the Internet together with the hardware and software resources of the servers that offer those services [8]. To construct our cloud, we deployed an Intel Xeon and Windows Server 2008 R2 based system. A SMS modem connected to the cloud server is used to notify the users, in case of intrusion events. A schematic diagram of how a user is notified and how he/she controls and monitors the security surveillance system via the cloud services is shown in Figure 3. The figure portrays a direct connection of all boards (nodes) to the cloud over Internet which may or may not be through the ad-hoc wireless network gateway, depending on configuration.

Real-time Monitoring of all rooms.

x

Toggling the security surveillance on or off.

x

Automatic storing and updating geolocation of each board.

SMS notifications and user’s confirmation in case of intrusion detection.

x

In case of genuine intrusion, as confirmed by the user:

V.

Presently, the security services offered in our cloud are:

x

x

ƒ

Stealthily alert neighbors, who are also using our home automation system, using SMS, email or by playing low sound alarms in their home.

ƒ

Inform local police.

ƒ

Play loud alarm sound.

REMOTE CONTROLLING AND MONITORING

The cloud services can be accessed from anywhere in the world on any Internet enable device over an enforced [9] Hypertext Transfer Protocol Secure (HTTPS) connection, encrypted by Secure Sockets Layer (SSL) version 3.0. The web interface of the cloud requires password based user authentication. A user is entitled to add any number of board (nodes) to his account, which can also be removed if necessary.

184

Fig. 5. A photograph of a mobile device displaying a SMS alert to the user in case of an intrusion event.

automation system would require a static public IP address for the web server. B.

No dedicated or specialized gateway required

Our home automation network doesn’t require a dedicated or specialized gateway to connect to an external network because the communication base of the network is Internet Protocol, which can be directly connected to the Internet. Our choice of using 802.11n standard enables us to form this IP network. C.

Resource Sharing

Resources in the cloud are being shared among all users. For example, a single SMS modem is serving to notify all users in case of intrusion event. This type of resource sharing reduces cost and saves energy. D.

The location of each node helps the cloud services to stealthily alert the accurate neighbors (who are using our home automation system) or local police, in an event of genuine intrusion.

Fig. 4. A photograph of the security surveillance page opened on a mobile device, HTC HD7.

After authentication, the user is directed to the security surveillance page where controlling options and monitoring view from individual boards (nodes) are displayed. Figure 4 is a photograph of the security surveillance page opened on a mobile device, displaying view from multiple boards that are added to the user’s account and Figure 5 is a photograph of a SMS alert to the user in case of an intrusion event. VI. A.

Location Awareness

E.

Remote Control & Monitoring

The security surveillance system can be controlled and monitored in real-time, from anywhere, via the Internet. F.

Social Interaction

Since all users are using a central cloud service, we easily locate and contact the neighbors who are also using our cloud based home automation system.

ADVANTAGES OF CLOUD CONTROLLED HOME SECURITY SYSTEMS

VII. QUALITY OF SERVICE

No dedicated web server required

As with any real-time process, the process of security surveillance should comply with good operability and integrity [11]. So, we did a few benchmark tests of the network gateway and found the result to be encouraging (Figure 6). On an average, less than 10% of the gateway bandwidth is in use, while carrying out security surveillance in 5 rooms

Our cloud connected home automation network doesn’t require a dedicated web server, to be remotely controlled and monitored over the Internet, whereas ordinary Internet enabled home automation network require a 24/7 running web server [10]. Furthermore, every mere Internet enabled home

185

REFERENCES

simultaneously. This leaves out enough bandwidth to set up Resource Reservation Protocol - Traffic Engineering (RSVP-TE) across the IP network [12], which is anticipated to give real-time controlling and monitoring authority to the users.

[1]

J. L. Ryan, “Home automation”, Electronics & Communication Engineering Journal, Volume: 1, Issue: 4, August 1989 pp. 185 – 192. [2] Anindya Maiti and S. Sivanesan, “Controlling and Monitoring of Wireless Home Power Management Systems through Public Cloud Services”, in press. [3] Xinrong Li, Kaveh Pahlavan, Matti Latva-aho and Mika Ylianttila, “Indoor geolocation using OFDM signals in HIPERLAN/2 wireless LANs”, The 11th IEEE International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC) 2000, Volume: 2, September 2000 pp. 1449 – 1453. [4] Guangming Song, Yaoxin Zhou, Weijuan Zhang and Aiguo Song, “A multi-interface gateway architecture for home automation networks”, IEEE Transactions on Comsumer Electronics, Volume: 54, Issue: 3, August 2008 pp. 1110 – 1113. [5] Qualcomm Atheros XSPAN-AR9287 – Product Bulletin, October 2010. [6] Wonjun Kim and Changick Kim, “An efficient correction method of wide-angle lens distortion for surveillance systems”, IEEE International Symposium on Circuits and Systems (ISCAS) 2009, May 2009 pp. 3206 – 3209. [7] Metkar Shilpa P. and Talbar Sanjay N., “Dynamic Motion Detection technique for fast and efficient video coding”, IEEE Region 10 Conference (TENCON) 2008, November 2008 pp. 1 – 5. [8] Michael Armbrust, Armando Fox, Rean Griffith, Anthony D. Joseph, Randy Katz, Andy Konwinski, Gunho Lee, David Patterson, Ariel Rabkin, Ion Stoica and Matei Zaharia, “A view of cloud computing”, Communications of the ACM, Volume: 53, Issue: 4, April 2010 pp. 50 – 58. [9] Adonis P.H. Fung and K.W. Cheung, “HTTPSLock: Enforcing HTTPS in Unmodified Browsers with Cached Javascript”, 4th International Conference on Network and System Security (NSS), 2010, September 2010 pp. 269 – 274. [10] Ali Ziya Alkar and Umit Buhur, “An Internet based wireless home automation system for multifunctional devices”, IEEE Transactions on Comsumer Electronics, Volume: 51, Issue: 4, November 2005 pp. 1169 – 1174. [11] E. Topalis, L. Mandalos, S. Koubias and G. Papadopoulos, “QoS support for real-time home automation networks management via highspeed Internet connection”, 10th IEEE International Conference on Networks (ICON) 2002, August 2002 pp. 130– 135. [12] Raymond Peterkin and Dan Ionescu, “A Hardware/Software Co-Design for RSVP-TE MPLS”, Canadian Conference on Electrical and Computer Engineering (CCECE) 2006, May 2006 pp. 1409 – 1412.

Fig. 6. A graph showing gateway bandwidth usage for duration of 2 hours. The maximum down-link network bandwidth is 7.2 Megabit/s and the maximum up-link network bandwidth is 1.8 Megabit/s.

VIII. CONCLUSION AND FUTURE WORK The use of cloud services in home automation derives many benefits extending from cost reduction to value added services. For further work on the cloud based home automation network, we plan to add a few more services like a cloud based power scheduling system, a cloud based music player and many more. On improving the security surveillance system, we plan to add more social integration through social networking sites like Facebook and Google+. With the help of these online social networks, we can easily contact and notify a user’s friends in case of an intrusion event and thus make burglary prevention more effective.

186