Example Of Project Overview Dissertation
Intelligent door controlled by Raspberry pi
This paper proposes the use of an embedded system - Rasperry Pi to develop smart home software. The system utilizes the recent phenomenon in computing referred as Internet of Things. The objective of the phenomenon is to extend the positives of regular internet in terms of connected every device in the home environment to create a smart network. A proposed intelligent door control management system is presented. The intelligent system uses 512 MB RAM Raspberry Pi microcontrollers with a Broadcom BCM2835 system chip, a passive infrared sensor and a Wi-Fi dongle and a door bell. The system is designed such that when a user presses the door bell, the Raspberry Pi activates a Wi-Fi camera which consequently takes a snapshot of the visitor and processes it to be send to the owners email address and Twitter handle. The owner upon receiving the picture decides to access the live stream video via a web interface or issue a command via a LCD display located at the door wall to be used to open it. This system is highly effective in detecting visitors and malicious intruders. With malicious users, the doorbell is avoided but a passive infrared reader senses human activity and the Wi-Fi camera records the feed and the pictures. This information is accessed by the owner and if need be, an alarm system is activated.
The internet has brought new means of communication. The underlying technologies and protocols brought about by internet are used to communicate with machines and this has extended the capabilities further. By merging the different technologies, a new intelligent environment is created which can be controlled from remote locations. With respect to residential environments, this is referred as smart home concepts. A smart home concept reflects the current trends in information technology. It facilitates the interaction of the owner with the devices which can be controlled remotely from anywhere in the globe. Such devices comprise of alarm system, keyless access control, water, heat and light management and smoke detectors. The proliferation of smart devices including, Smartphone, iPhones, tablets PDA’s means that they are can be used to remotely control the home environment via internet. In this paper, a proposal is made on a home automation system that can run on an embedded platform such as Rasperry Pi. Rasperry Pi is preferred because of its low resource consumption and cost effectiveness.
Raspberry Pi is an embedded system that has been applied in a number of applications. Raspberry has been utilized as a platform for the KNX/SIP Home Access Gateway. The device is developed by Raspberry Pi Foundation in United Kingdom. The device is an embedded system with powerful capabilities. It can be configured to run on multiple platforms such as Python, Linux, and other platforms. The advantage is that it is portable, and runs on USB devices given the right configurations. It operates at 700MHz with a GNU capable of rendering 1080 pixels. It’s preferred in Smart Home software due to its sufficient capacity and low cost. To gather for future developments in smart home automation, current and future buildings have already equipped featured IP network. The owner upon receiving the picture decides to access the live stream video via a web interface or issue a command via a LCD display located at the door wall to be used to open it. This system is highly effective in detecting visitors and malicious intruders. With malicious users, the doorbell is avoided but a passive infrared reader senses human activity and the Wi-Fi camera records the feed and the pictures. This information is accessed by the owner and if need be, an alarm system is activated.
When everyday things are connected to the internet, the concept is referred as Internet of Things. IoT is used to manage security in a cost effective manner. This paper documents the development of an internet connection to any door so that access control can be managed remotely from any location. In a scenario where homeowner is away and a visitor steps at the door, the homeowner is notified that a visitor is standing at its door. The system is intelligent in the manner that it takes a photo of the visitor and forwards it to the homeowner in a matter of seconds. Through email or social network sites such as twitter, the owner can see the picture of the person and make a decision. The owner can manage interactions by messaging the visitor, ignoring it completely, or initiate more security backup such as locking all other systems at home or calling a police number. A small screen on the frontal part of the door allows communication between the owner and the visitor. A message sent by the owner using a Smartphone or PC is immediately relayed and shown on the screen for the visitor to read. This system presents a real monitoring and control system with evidence of the person accessing a door recorded just in case of an emergency or situation.
The concept of intelligent home networks is well established in IT literature. Many institutions and companies have developed their own software to automate processes in their operation areas. Most of the systems developed are proprietary in nature and cannot operate with other home automation systems. KNX provides a framework to develop an automation system that can be controlled remotely from already established intelligence. The works of YanboZhao saw the development of a GSM/GPRS based wireless home monitoring and security system. However, this system was too expensive given the GSM fixed charges on calls and SMS. Mokhtar, Bakar and Hassan invented face recognition technology based on auto-switching magnetic door lock controller. However, due to complex algorithms required for face recognition, and accuracy, the system lacked usability. The other type of security system is sensor based controlled system which has extensively been studied.
With the capabilities of internet, a new system is proposed that is least in complexity and convenient. It is a plug and play system with proven performance. The homeowner can see who is at the door and keeps the picture of the person as evidence if anything undesirable was to occur. The user controls the door with a single command via internet using his handheld device. This system has a lot of applications ranging from domestic door controls to industrial automation. In case of an incidence such as fire or lost keys, the door is opened using simple and single command.
The new integrated system is scalable, re-configurable and based on internet. Home users are more likely to add devices such as more doors, windows and garage openings to the intelligent system. This is possible through a configuration interface at the LCD screen. The device is also re-configurable such that two devices can be made to act differently. Raspberry pi can be set up to perform a number of functions in a home including physical firewall or a secure anonymous gateway to the outside world.
This project involves the use of Raspberry Pi microcontroller. The system is made up of two components; an input and output component. An input component comprises of a calling bell, PIR, wireless camera and a sensor. The processing section is conducted by raspberry microcontroller which is outfitted with a Wi-Fi dongle. The output components of the system include an LCD screen, magnetic door lock, reporting platforms such as an email account. A door bell is installed at the door. The door bell is pressed by a visitor upon visiting and a signal is generated to the raspberry pi signifying the presence of a person. With malicious visitors trying to break into the house, an alternative form of detecting them is available. A passive infra red human motion detector is used. During a break in, a PIR sensor detects motion and transmits an alarm signal to the raspberry pi. The wireless camera, which is strategically placed installed in the door is activated a records a video and a snapshot of the person, which is then forwarded to the owners email address using the raspberry. The raspberry pi processes the input and transmits it to the control email address. In a matter of seconds, an email notification or a tweet feed alerts the user of some activity at home. The user controls the proceeding activities using a custom application on the Smartphone or computer. The options including opening the door using the magnetic lock or rejecting access by posting a message on the LCD screen.
A typical model B Raspberry Pi has 512MB RAM, Ethernet port and two USB ports. It has a Broadcom BCM2835 system chip and an ARM 700MHz processor. Others include and SD card and a Video Core IV GPU. The fast 3D core is accessed via Open VG AND OpenGL ES2.0 libraries. Other functionalities in the Raspberry include HDMI support. Programming languages supported by the pi is Python with support for BBC, C, Perl and Debian and Arch Linux ARM distributions.
The usage of Rasperry Pi is founded on its cost and power consumption. Raspberry is selected as the preferred hardware platform because it consumes less power and is low cost. This way there is a lot of space to be saved in comparison with normal PCs.
Power consumption: The Raspberry Pi is rated 750mA at 5V and this translates to 3.75W per hour. At 100% work rate, the consumption of 1A gives 5W which is equivalent to 3 W an hour. Thus a total consumption of 17 kW is realized in a year.
Cost: This translates to a cost of 5 EUR a year and when compared against the cost of running a desktop computer, this is significantly low. Consumers normally consume 100W in an hour. Another cost factor is that or purchasing the device. Raspberry Pi only cost about EUR 50 as a complete set with all the peripheral equipment. Thus, an initial investment of EUR 50 and a permanent cost of EUR 5 yearly are anticipated. Where the device needs to be replaced, Raspberry Pi and SD-card will incur minimal costs.
Passive Infrared Sensor
Passive Infrared sensors are used to detect the presence of humans in the observed area. Humans generate heat and infrared radiation with the strongest magnitude at 9.4um. Infrared radiation exists in electromagnetic spectrum with its wavelength being more than that of visible light. A PIR sensor can detect a change in radiation as a result of human presence. PIR senses by generating a sine wave with magnitude of +5v and -5v when human subjects pass in front of it. The signal is detected by a breakout board and converted to a longer digital signal which can be felt for approximately 1 second.
WiFi dongle is connected via a USB port. A speed of 300Mbps can be achieved and with MIMO technolog,y interoperability with existing IEEE 802.11b/g networks is possible. This creates a wireless Access Point which is protected by WPA2 secure connection.
Raspberry pi minicontroller is the heart of this system. Wi-Fi dingle is integrated into the system using USB port. The Wi-Fi module scans for available networks and connects to them with a password. The door and PIR sensor connects to GPIO of the raspberry pi. When the calling bell is pressed, it transmits sound. A resistor connected to the bell generates a signal of +5V. This causes an interruption in the Raspberry pi. The pin is lowered and the sensor generates a high voltage which detects the presents of infra red radiation emanating from a human. The Raspberry pi continuously detects inputs from PIR or door bell ringer. Two alerts can be generated at the same time when a visitor presses on the door bell. In the algorithm, this indicates a normal condition that signals the arrival of a visitor. But in case of a single signal detected, especially that of a PIR sensor, an extra ordinary visit is imminent. When the two signals are detected, raspberry pi snaps the person through the Wi-Fi camera. The camera is integrated in the system through Wi-Fi. Raspberry processes the input and transmits it to indicated gmail account. Multiple accounts may be used, one social such as Twitter and the other email. If PIR sensor is generated alone, an alarm system is activated that notify security personnel of an imminent security breaking. However, there are situations when the intention was not a break-in. For instance, a domestic worker cleaning the areas of the house near the door might be detected by the PIR. A feed is immediately established and send to the owner. This might create a false positive about an incident.
In order to rule this scenario out, the order of activities in cases where a PIR signal is generated without the door bell is as follows:
PIR sensor generates and interrupts in the Raspberry pi
The Wi-Fi camera is activated and takes snapshots
Snapshots are processed and forwarded to the owners email address
Several feeds and notifications are established in social networks
Action from the owners is awaited before an alert is issued
A 1.8 inch screen is connected to the system using serial peripheral interface and the magnetic lock is controlled by the GPIO.
This system works in a number of steps. They include image acquisition process where the raspberry activates the Wi-Fi camera to take a snapshot of the human upon detection of signals. The second process involves image processing and mailing. Finally, there is the door opening section where the user provides the door lock command to open or close it. Diagrammatically, the process is represented as follow:
Figure 2 A chart of processes in the intelligent door control system
A standard USB webcam is used to take pictures and videos on the Rasperry Pi. An installation of the fswebcam package is the first step to setting up the camera. A command fswebcam is entered followed by the filename and a picture will be taken and saved to the filename specified. Instances when the camera takes a short are when there is an interrupt. For instance, if the interrupt from PIR sensor is recorded, the cam is programmed to take a shot.
Figure 3. PIR interrupt session code
The system need a username and password to automatically login and generates a mail. The system connects to the server and demands the email address of the user. The address and password allow automatic login and the email address of the user is provided. Ann image is attached identifying the visitor and mailed to the owner of the house.
Connecting to social networks to generate a fed requires a module interfacing with the account. A Twython application is used to interface Raspberry with twitter and is installed on Raspberry Pi. A OAuth access token is required for accessing users twitter account. By default, the application is read only meaning that users cannot change the feed. However, any alteration may be done from the settings tab. On the twitter app, users token information are required to update the feed. The user can decide to have the image of the visitor posted on twitter directly. A number of steps are required to post the picture from raspberry pi to twitter account. In this case, pygame libraries are used to take a picture of the camera.
Figure 4. Implementation process
As illustrated in the figure below, the implemented system set up comprises of interconnected hardware. All the hardware are connected as per the circuitry setup.
Figure 5 Implemented system
When the PIR sensor is interrupted, the Wi-Fi camera is activated to take a shot and email the attachment to the owners account. A tweet is also generated to notify the authorized party that a visitor is waiting at the door. The owner can access a live video stream of the visitor using a web or acts based on the picture alone. If the owner desire to communicate with the visitor, the owner can just type and tweet and post which is the posted on the screen located at the door. Among the message is a command prompt which can be used to open the door.
This technology is effective in the control of door systems as long as the rights of visitors are not violated. An extension to the system is the use of social networking accounts such as twitter to provide feedback. In some cases, it will not be appropriate to be posting peoples pictures in the public domain. It is a violation of people’s rights.
The security of the system is also a concern. Given that the system is compromised and the fact that no encryption measures are in place, people’s privacy may be violated. For instance, a blackhat can deploy their own devices to become the new network controllers. The blackhat devices uses the global encryption key – which in many cases is only one and is publicly known with a signal for new network configuration to cut legitimate users out of the picture. Another way of compromising the system is using custom applications. A number of applications such as Killer Bee has been developed to sniff keys in Raspberry Pi and render it easy to penetrate in spite of the encryption.
This project initiates a series of developments and studied which should be pursued to make the intelligent door control system using Raspberry Pi effective. They include; Development of Smartphone apps to be used together with the system. It is more effective to use Smartphone apps a rather than social sites and email accounts. Given the magnitude of spread of these devices, it is mandatory that apps be integrated with Raspberry for an effective control system
Improvement in security issues is highly recommended. Security is an important aspect in the development and deployment of this application. They have the potential of making the home environment much insecure because even though they use AES-128 encryption, these encryptions has been proved to be easily bypassed. Examples of control systems that have failed due to security considerations include using ZigBee and Z-Wave technologies.
The system has been shown to be effective in automated control of door systems. It application varies but the most practical ones include residential premises, banking halls, hospital facilities and other commercial entities. However, customization is required to be used commercially. The system is effective in reducing the hazards of unauthorized entry. This system is effective in secluded locations where human traffic is low. For instance, it can be applied in cottages, vacation homes and hotels which are used seasonally. The deployment of Raspberry to act as the brain and gateway of operations is not only effective but also economical. Raspberry pi is an effective small, powerful and lightweight microcomputer. The device with a range of accessories offers unlimited choice of uses. This system has been build with simple devices that are economical and readily available. The advantage with the system is that it does not depend on any network communication protocol. Rather, it is a plug and play system that is highly customizable to fit the needs of the user. A challenge that is likely to be encountered in the implementation of the system is with respect to security. A redesign is proposed to make sure that the system is secure in s much ways as possible. A recommendation for future consideration is the development of a Smartphone app on Android or iOS platform to be used in conjunction with the system. Overall, the system is highly effective in managing access control in domestic and commercial doors.
Assaf, M. et al., 2012. Sensor based home automation and security system ” Instrument a-tion and Measurement Technology Conference. s.l., s.n.
Coleman, E. G., 2013. Coding Freedom: The Ethics and Aesthetics of Hacking. s.l.:s.n.
Hassan, H., Bakar, R. & Mokhtar, A., 2012. Face recognition based on auto -switching magnetic door lock system using microcontroller. s.l., ICSET.
Ibrahim, R. & Zin, Z., 2011 . Study of automated face recognition system for office door access control application. s.l., ICCSN.
Lee, J. H. et al., 2007. Security Door System Using Human Tracking Method with Laser Range Finders” ,. s.l., ICMA .
YanboZhao & ZhaohuiYe, 2008. A low cost GSM/GPRS based wireless home security system. IEEE Transactions on Consumer Electronics , pp. Voume:54 , Issue: 2.