Content uploaded by Murat Ali
Author content
All content in this area was uploaded by Murat Ali on Feb 26, 2015
Content may be subject to copyright.
Technical Development and Socioeconomic
Implications of the Raspberry Pi as a Learning Tool
in Developing Countries
Murat Ali∗, Jozef Hubertus Alfonsus Vlaskamp∗, Nof Nasser Eddin†, Ben Falconer∗and Colin Oram∗
∗School of Engineering, The University of Warwick, Coventry, CV4 7AL, UK, Email: murat.ali@warwick.ac.uk
†Department of Sociology, The University of Warwick, Coventry, CV4 7AL, UK, Email: N.J.Nasser-Eddin@warwick.ac.uk
Abstract—The recent development of the Raspberry Pi mini
computer has provided new opportunities to enhance tools
for education. The low cost means that it could be a viable
option to develop solutions for education sectors in developing
countries. This study describes the design, development and
manufacture of a prototype solution for educational use within
schools in Uganda whilst considering the social implications
of implementing such solutions. This study aims to show the
potential for providing an educational tool capable of teaching
science, engineering and computing in the developing world.
During the design and manufacture of the prototype, software
and hardware were developed as well as testing performed to
define the performance and limitation of the technology. This
study showed that it is possible to develop a viable modular
based computer systems for educational and teaching purposes.
In addition to science, engineering and computing; this study
considers the socioeconomic implications of introducing the EPi
within developing countries. From a sociological perspective, it is
shown that the success of EPi is dependant on understanding the
social context, therefore a next phase implementation strategy is
proposed.
I. INTRODUCTION
The recent development of the Raspberry Pi mini-computer
has unlocked great potential for computing to be applied in
a vast number of areas. Due to the unique advantages of the
Raspberry Pi system, this technology holds great promise for
providing solutions within the developing world. This includes
but is not limited to education tools, especially the use of
GPIO (General Purpose Input/Output) which allows automated
data acquisition and producing simple digital control systems
in a school laboratory setting. To make the best use of the
Raspberry Pi in the developing World, a number of factors
need to be considered. These can be divided into technical,
educational, economic and social factors. While the current
research focuses on providing educational materials, the other
factors are inextricably linked and therefore will also form an
important part of the investigation. A successful implementa-
tion of the proposed solution will rely on understanding the
needs of the local population in providing the right device,
supporting infrastructure and taking into account cultural back-
ground as well as the end user’s expectations of the project.
It is only when the solution is fully accepted by the target
audience that the project can achieve its goals. This research
does not only look at technological advancement and how it
can improve education in developing countries from purely
an engineering perspective, but also takes into consideration
the sociological perspective that tackles the social and the
economic aspects of introducing technologies such as the
Raspberry Pi to disadvantaged communities in The Developing
World.
Following identification of the needs, the objective of this
study is to develop a first generation working prototype
(referred to as ’EPi’) based on the Raspberry Pi which can
supplement the learning of science, engineering and computing
within the developing world. This will be demonstrated by
setting up a temperature sensor as well as an analog-to-digital
converter (adc) through the GPIO using the EPi. Through
hardware design and software development, this will provide
just a few examples of how this solution can provide key
educational tools from secondary education to post-secondary
education levels, including university based projects.
II. CO NT EX T
There is currently a need for developing versatile computer
systems for the education sector within developing countries.
As a relevant example, it has been reported that the primary
cause of low education rates in Africa alone is due to ’unequal
opportunities’ and ’lack of proper schooling facilities’ [1].
Non-specialist flexible computer systems for educational use
minimises the need for technical expertise to develop, maintain
and upgrade complex systems on both a hardware and soft-
ware level. Therefore, the local educational institutes using
the developed computer systems are able to carry out these
tasks independently of external technical specialists and highly
reliable infrastructures such as power sources and networking
solutions.
The size of the developing world is vast, therefore to narrow
down the scope of the project one country was focused on,
and this country was Uganda. This country within Africa
was selected through academic links at The University of
Warwick. Uganda is considered one of the poorest countries
in the world with 37.7% of the population living on 1.25$ a
day [2], [3]. Poverty levels are mainly concentrated in rural
areas where almost 80% of Ugandans live. A number of
health, economic and social factors contribute to Uganda’s
high poverty levels, and those include a lack of security, large
size of households, gender inequality, unemployment, lack of
2013 5th Computer Science and Electronic Engineering Conference (CEEC) University of Essex, UK
978-1-4799-0383-2/13/$31.00 ©2013 IEEE 103
health services and most importantly low education levels [4],
[5]. Education is considered one of the most important factors
that can eradicate poverty in developing countries, as it opens
different opportunities for people. Uganda has a literacy rate
of 66.8% (76.8% male and 57.7% female). To increase the
level of education in developing countries new methods can
be introduced to school pupils. For example, the ’one laptop
per child’ (OLPC) program which was originally funded by
a number of large organisations including AMD, Google and
Red Hat, targeted children in elementary schools. This project
and initiative aimed to improve education of children in the
poorest areas of the world. The rationale behind it was to
provide low cost and power rugged connected laptops to
children in order to create a collaborative and empowered
learning environment.
The OLPC project had an optimistic vision of improving
children’s education through technological interventions, how-
ever the project was not one without its challenges. Despite
the ambitious and philanthropic goals of the program, there
are certainly areas which can be improved [6], [7]. The
implementation of new computers into the education system
such as OLPC and EPi require that teachers are comfortable
with and understand these technologies in order to maximise
the learning experience for the students. There also needs to be
a clear strategy for dealing with faulty or redundant systems.
By gaining an overall understanding of the outcomes, strengths
and weaknesses of the OLPC project, a strategy for EPi is
proposed in section VII.
III. DESIGN PHILOSOPHY
A. Concept
In order to meet the objectives of the project the concept
and vision of the EPi must be defined. Along with using the
Raspberry Pi hardware, an open source operating system and
software are utilised and modified to suit the requirements of
the project. A modular assembly system allows for flexibility
and changes to be made specific to the needs of the end user.
Overall, the proposed prototype offers a flexible learning tool
to be utilised within the education sector in Uganda, as well as
provide a range of scientific learning options and manageable
complexity to cater for the specific users and environments.
B. Parts selection and Prototype Manufacture
The parts of the EPi were selected based on both the
capabilities of the Raspberry Pi as well as meeting the product
specification and objectives. To reduce the size and cost of the
solar panel, and to improve the operating life of the EPi under
battery power, components with low power consumptions were
carefully selected. The power consumptions were measured to
confirm the specification, which is discussed further in this
study. The final selection of parts and peripherals for the
prototype model is listed in Table I with the associated cost,
however for volume production these costs are expected to
be significantly lower. The common input and output devices
that any modern computer system would be supplied with have
all been included. From a cost perspective the solar panel was
TABLE I
EPIPART AND PERIPHERAL COS TS (EX C. VAT)
Item Cost (£)
Raspberry Pi 26.88
3.5” LCD screen 12.60
Mini keyboard/mouse 16.31
4-port usb hub 7.16
mini webcam/microphone 7.58
bluetooth dongle 0.80
Wifi dongle 2.41
Battery Pack 30.39
Solar Panel 173.34
Total Cost (exc. solar panel) 104.13
Total Cost (inc. solar panel) 277.47
the main concern, therefore the system has been designed with
the option of providing the end product without the need for
a solar panel. The EPi system as a whole can be powered
by a mains power supply, which will also charge the battery.
This still allows the EPi to be used as a mobile system. The
modular and flexible assembly philosophy means that other
input/output devices can be attached as per the limitations of
the EPi. The most important connectivity is the GPIO which
will be further discussed in sections V and VI.
All components are assembled to allow for interchangeabil-
ity of parts as well as providing the option for alternatives
to be used. The arrangement of components and an assembly
schematic is shown in Fig. 1 and Fig. 2. This prototype has
been designed to be accessible and viewable by the teachers
and students within the school laboratory environment, there-
fore no fixed enclosures have been provided. This is an area
that will be assessed in more detail during the development of
a production model.
Fig. 1. EPi Prototype Assembly
IV. HAR DWAR E DESIGN
The modular prototype has been designed using as many
off-the-shelf components as possible. This minimises the de-
sign time and costs required to build the first generation
prototype, while still retaining high performance. This modular
approach also simplifies maintenance, a faulty component can
be easily replaced without advanced tools.
With portability and off-grid capability being essential de-
sign goals, a suitable solar panel and battery combination is
the most important additional hardware to be selected. As the
2013 5th Computer Science and Electronic Engineering Conference (CEEC) University of Essex, UK
978-1-4799-0383-2/13/$31.00 ©2013 IEEE 104
Fig. 2. EPi Assembly Schematic
Fig. 3. Solar Simulator - The University of Warwick, School of Engineering
solar panel is by far the most expensive component, this needs
to be carefully selected as overcapacity will increase the bill of
materials very significantly. Experiments have been performed
to determine the actual power output of the solar panel, and
the power consumption of the EPi and all peripherals.
The solar panel selected for the prototype is the Powerfilm
R14, a rollable panel with dimensions of 107x37cm and a rated
output of 14W. To verify the power output under various levels
of solar radiation, experiments have been performed on the
Solar Simulator at the University of Warwick (Fig. 3). The
Solar Simulator contains a large array of halogen and LED
light sources, which allow the panel to be subjected to solar
radiation levels between 0.6-1.2 kW/m2. A summary of the
maximum power output under various conditions is shown in
table II.
TABLE II
OUT PUT O F TH E SOL AR PAN EL AT DI FFER EN T RAD IATI ON LE VE LS
Radiation (W/m2) 469 608 772 884 1001 1009 1174
Pout (W) 2 3 7 9 11 11 13
TABLE III
MON THLY AVER AGE D INSO LATI ON INCIDENT ONA HORIZONTAL
SUR FACE INUGA NDA I N (kWh/m2/day)
Jan Feb Mar Apr May Jun
6.07 6.36 6.09 5.66 5.39 4.97
Jul Aug Sep Oct Nov Dec
4.96 5.25 5.70 5.44 5.48 5.83
The climate in Uganda is ideal for using solar power,
with sunshine for most of the year. Average solar radiation
for Uganda is shown in table III [8]. For design purposes,
two cases have been assumed: best case of the equivalent
of 6 hours of 1 kW/m2or at optimal placing of the panel,
corresponding to 6 kWh/m2/day, and worst case of 4 hours
at only 0.6 kW/m2, corresponding to 2.4 kWh/m2/day. While
in practice in many geographical regions, including the UK,
worse conditions exist for a large part of the year than the
worst case assumed above [9], in these cases a solar panel is
not considered a practical solution within the context of this
project.
The assumptions above give a total energy budget of 14-60
Wh/day. To determine the total working time of the prototype
under various conditions, the power consumption of each
individual component is to be identified. A small circuit board
was built to measure the current consumption of the USB
devices while in operation, consisting out of one USB-A and
one USB-B connector. An ampere-meter is placed between the
power pins of the sockets, with the ground and signalling pins
connected directly. The results are shown in table IV and V.
Assuming 5.5W total power consumption, the solar power
allows 2.5-11 hours of use per day. When the solar radiation is
less than 0.7kW/m2, the battery will be required to provide a
backup. For the prototype, the Anker Astro3 has been chosen,
with a 50Wh capacity, providing 9 hours of backup when fully
charged.
TABLE IV
CURRENT CONSUMPTION OF THE PI(5V) AN D SCR EEN (12V) UNDER
VARIO US L OAD S
Idle 100% CPU Playing Screen
(Ethernet Off/On) Video (Idle/Active)
I(mA) 335/392 393 350 80/220
P(W) 1.68/1.96 1.98 1.75 0.96/2.64
TABLE V
CURRENT CONSUMPTION OF THE PERIPHERALS
Wifi Usb Usb Keyboard Mouse
Hub Storage (Idle)/(Active) (Idle)/(Active)
I(mA) 93 70 72 4.7/4.7 4.5/13
P(mW) 465 350 360 24/24 23/65
2013 5th Computer Science and Electronic Engineering Conference (CEEC) University of Essex, UK
978-1-4799-0383-2/13/$31.00 ©2013 IEEE 105
V. SO FT WARE
All the software included is open source; this reduces
cost substantially compared to commercial licenses. It is also
”free” in the sense that it can be adapted to individual needs.
The operating system and user interface is built on top of
Raspbian, the most common and (currently) best maintained
of the Raspberry Pi targeted operating systems; this provides
the option to flexibly install any of a multitude of packages
that are available on the Raspberry Pi as well as developing
software specific to the needs of the EPi. As less than 10% of
Ugandans have access to the internet [10], [11], the distribution
has been modified for the EPi to include as many useful pieces
of software as possible, as downloading further packages may
be difficult or impossible for the end user.
The EPi software and user interface must enable users
who have never used computers before, to develop skills and
knowledge from the basics of computer usage to creating
useful applications with the EPi. To aid in the first stage of
this, some basic user interface adjustments have been made
to accustom the user to keyboard and mouse usage and the
basics of a windowed desktop environment. This will allow
the inexperienced computer user to easily get started with an
intuitive system before using the command line interface.
Several knowledge bases, such as Wikipedia [12] and the
Khan Academy [13], are packaged as webpages and stored on
the EPi for general education. Wikipedia provides a general
encyclopaedia while Khan Academy offers educational videos
on a wide range of academic subjects. These resources could
be used in a school environment, especially Khan Academy
which has, for example, a full course on mathematics from
basic addition to university level content. It is understood that
live updates would not be available through offline versions,
however the opportunities to provide planned updates could
be considered as an option based on the internet capabilities
of particular locations. Although these online resources are
regularly being updated, the main curriculum subject matters
required for education are well established and are not ex-
pected to change greatly within a short space of time.
For computer development specific education, Scratch pro-
vides a visual flowchart-like programming interface to teach
the basics of coding without the difficulties of syntax and
command lines. Following on from learning the basics of
programming, other programming languages can be used and
understood more easily. Python was chosen as the main pro-
gramming language, as it is generally accepted to be both easy
to learn and a fully fledged programming language suitable
for real world applications. With the addition of NumPy,
SciPy, Matplotlib, IPython, and PyLab, Python can be used
for computational mathematics as well as for the analysis of
experimental data or control systems. Python provides access
to the GPIO facilities on the EPi, and a number of examples
are included ranging from some simple I/O programs to a dig-
ital storage oscilloscope. The examples are written to be easily
reusable in different student projects, where students can go on
to develop their own solutions. All example programs include
Fig. 4. Digital Thermometer Connected to EPi with Typical Output
the full source code, which means that students and teachers
can investigate the code and make their own improvements
and adaptations.
VI. EL EC TRO NI CS
The most distinctive feature of the Raspberry Pi when used
for educational purposes is the GPIO module, which allows
interfacing with general purpose electronics. This allows stu-
dents to gain experience with data acquisition, instrumentation
and control systems, as well as using the EPi as a general tool
during science education. As the intended users are students in
secondary education and early tertiary education, the emphasis
in the curriculum will be on using ready made modules which
plug into the EPi, rather than on electronics design.
The use of ready made modules allows the curriculum
to be focused on understanding the fundamental principles
of scientific experimentation and engineering design, while
still offering the possibility of a more in-depth electronics
education for more advanced students. The modular approach
offers the additional advantage of remaining relatively low-
cost, as components can be reused for various projects.
The prototype developed in the current project comes with
two modules, an ADC module to be used as a digital oscillo-
scope and datalogger and a digital thermometer.
The digital thermometer, based on the DS1631 IC, is
connected to the Raspberry Pi through the I2C bus (Fig. 4).
The current version is not waterproof, which limits is appli-
cation to monitoring dry environments. It has been used in
a simple demonstration of the capabilities of the EPi as a
control system, where a small fan is turned on if the ambient
temperature is above a certain level.
The MAX1270 which is used on the ADC module provides
8 channels (4 channels implemented on the prototype), with
a resolution of 12-bits and a maximum throughput of 110
ksamples/s. Depending on the software, this can either be used
as an oscilloscope or a datalogger. The high storage capacity
provided by the SD-card in the EPi makes it ideal for long-
term experiments. An example lab project investigating the
frequency components of the human voice has been created,
using a simple microphone and opamp. The FFT libraries in
2013 5th Computer Science and Electronic Engineering Conference (CEEC) University of Essex, UK
978-1-4799-0383-2/13/$31.00 ©2013 IEEE 106
Python have been used to convert the signal to the frequency
domain. The project can be extended to include musical
instruments.
The long-term evolution will encompass a much larger
array of modules, including sensors an actuators, together
with teaching materials. This will supply schools in The
Developing World with a ”lab-in-a-box”, with a number of
standard experiments that can be performed. The emphasis,
like in the UK, is on providing the students with both the skills
and a platform that allows them to design and implement their
own ideas.
VII. SOCIOECONOMIC IMPLICATIONS OF THE PROJECT:
IMP LE ME NTATION AND THE WAY FORWAR D
As discussed above the ’one laptop per child’ program was
one of the largest ambitious initiatives that aimed to improve
and create a better learning environment for impoverished
children in developing countries. This project has shown
that the process of introducing new technologies is not as
straightforward as some might assume, and there are many
factors and challenges to be faced [7]. For the EPi to achieve
its goals and be a success in The Developing World, the
proposed action plan aims to ensure that ’development’ and
’suitability’ are accurately defined. Therefore, implementation
of the EPi solution will be based on ’people’s’ own perspective
and needs in relation to these technologies [6]. Willoughby
argues that there are factors that should be taken into consider-
ation when determining whether or not such technologies are
suitable for the targeted nation. He states: ”the Appropriate
Technology notion points to the need for knowledge of a
diversity of technical options for given purposes, careful anal-
ysis of the local human and natural environment, normative
evaluation of alternative options, and the exercise of political
and technological choice” [14]. Therefore a thorough analysis
of socioeconomic and political factors is to take place before
introducing EPi into a specific area.
In Uganda there is a large number of communities and
languages spoken, thus different cultural values, folklore,
traditions and customs exist [15]. Therefore, it is very im-
portant to integrate EPi in a way that is tailored to suit
the context and the custom of the targeted school, students
and area. The philosophy of EPi means that this necessity is
considered. A relevant example of this is shown through the
availability of a specific GNU/Linux distribution which has
been customised for particular users [16]. A similar strategy of
software development has been adopted for the design of EPi,
therefore supporting the technology to be accepted by the end
user and their communities. It is not accurate to adopt the same
implementation strategy of EPi in all schools and communities,
the delivery will take into consideration differences within
the targeted areas. One of the reasons EPi is developed with
such a focused goal of improving science education within
schools is that. Even for one developing country, it is not
expected that such a project would succeed in all areas of
improving the well being of people and communities. In some
cases, what might work for one community might not have the
Fig. 5. Socioeconomic EPi Implementation
same impact for the other. Many factors such as class, gender,
religion and others are to be taken into consideration by uti-
lizing the intersectionality sociological framework [17]. This
shows the merit of this project’s interdisciplinary approach to
project implementation and delivery in general, contributing
to establishing a foundation to a successful delivery of EPi in
the future.
Before implementing the EPi to schools within the de-
veloping world, a pilot study is to be conducted to further
assess the specific needs in the target area, thus helping to
establish the context, and thus tailoring programs that are
suited for the target area. The main purpose of the pilot
study is to explore people and children’s perspectives ’from
below’ rather than have a preconceived idea of what people’s
necessities are ’from above’. In other words, this social
research informs the process of introducing this technology.
The EPi implementation pilot study will be in the form of
interviews, questionnaires and focus groups with local teachers
and pupils within schools in the UK. This pilot study will
then be extended to the target areas in Uganda to firmly
establish their needs as well as confirm what they see as
fit and suitable for their school environment. To avoid the
situations of the EPi being delivered to schools without taking
into consideration that pupils and teachers would need training
along with instructions on how to best use the technology for
education purposes [18], documentation and training courses
for school teachers and children must be provided. An EPi
2013 5th Computer Science and Electronic Engineering Conference (CEEC) University of Essex, UK
978-1-4799-0383-2/13/$31.00 ©2013 IEEE 107
implementation strategy has provided in Fig. 5.
VIII. CONCLUSION AND FURT HE R WOR K
The EPi prototype demonstrated the design objectives of
developing a modular system based on the Raspberry Pi for
scientific and computing educational purposes. Future work
would involve the implementation of the prototype within
an educational environment in Uganda. Upon feedback from
potential end-users from local and developing countries, the
system can be improved upon further to better comply with
their needs of specific communities. When series production
is considered, a more integrated design can be achieved by
working with the manufacturers of the individual components,
reducing cost and size. Following on from studying the out-
comes of other technological education based projects; for
the EPi to be a success within the developing world, the
solution must be accompanied with an understanding of the
socioeconomic aspects involved. This led to the a strategy
proposed in this study to fulfil this requirement.
Further work includes design and development of the next
generation EPi based on serial production. Along with the
further development of EPi; local and target location based
pilot studies are to be conducted directly with the current EPi
prototype.
ACKNOWLEDGMENT
The research has been funded by the EPSRC (Engineering
and Physical Sciences Research Council). We would also
like to acknowledge the support from Dr. Roger Thorpe
and Dr. Stan Shire from the The University of Warwick,
School of Engineering, Sustainable Energy Engineering and
Design (SEED) group for providing the solar simulator testing
equipment and guidance. Our appreciation also goes to Ian
Griffith for his help and support in the printed circuit board
manufacture. We would like to thank Tracey Moyle, Dean
Boni and Steven Jones for their continued support throughout
the project.
REFERENCES
[1] (2011) Investing in empowerment international -
sustaining lives in africa. [Online]. Available:
http://www.ieinternational.ca/pages/educ.html
[2] R. McGee, “Meeting the international poverty targets in uganda: Halving
poverty and achieving universal primary education,” Development Policy
Review, vol. 18, no. 1, pp. 85–106, 2000.
[3] K. Deininger, “Does cost of schooling affect enrollment by the poor?
universal primary education in uganda,” Economics of Education Review,
vol. 22, no. 3, pp. 291–305, June 2003.
[4] T. A. Kasule, “The causes of rural poverty in uganda: a case study of
wakiso sub county, wakiso district,” Master’s thesis, Makerere Univer-
sity, 2005.
[5] A. O. K. Lawson, D. McKay, “Poverty persistence and transitions
in uganda: a combined qualitative and quantitative analysis,” Global
Poverty Research Group, Tech. Rep., 2005.
[6] M. Warschauer and A. Morgan, “Can one laptop per child save the
worlds poor?” Journal of International Affair, vol. 64, no. 1, pp. 33–51,
2003.
[7] J. S. P. Kenneth, K. Dedrick, “One laptop per child: Vision vs. reality,”
Communications of the ACM, 2009.
[8] (2012) Surface meteorology and solar energy. [Online]. Available:
https://eosweb.larc.nasa.gov/sse/r
[9] (2013) The metoffice - solar energy. [Online]. Available:
http://www.metoffice.gov.uk/renewables/solar
[10] (2013) Internet world stats - uganda. [Online]. Available:
http://www.internetworldstats.com/af/ug.htm
[11] “The world in 2013: Ict facts and figures,” International Telecommuni-
cations Union, Tech. Rep., 2013.
[12] (2013) Wikipedia. [Online]. Available: http://en.wikipedia.org/
[13] (2013) Khan academy. [Online]. Available:
http://www.khanacademy.org/
[14] K. Willoughby, Technology choice: a critique of the appropriate tech-
nology movement. Westview, 1990.
[15] S. Bbumba, “The uganda national culture policy: A culturally vibrant,
cohesive, progressive nation,” 2006.
[16] (2013) Sabily linux operating system. [Online]. Available:
http://www.sabily.org
[17] L. McCall, “The complexity of intersectionality,” Signs, vol. 30, no. 3,
pp. 1771–1800, 2005.
[18] D. Nugroho and L. M, “Evaluation of olpc programs globally: a literature
review,” Australian Council for Educational Research, Tech. Rep., 2010.
2013 5th Computer Science and Electronic Engineering Conference (CEEC) University of Essex, UK
978-1-4799-0383-2/13/$31.00 ©2013 IEEE 108
