External Research: Digital Inclusion 2005 Awards

Digital Inclusion Through Mobile and Wireless Technologies 2005 Awards

Microsoft Research selected 17 proposals for the Digital Inclusion awards, totaling $1,200,000 (USD) in funding. The objective of the Digital Inclusion award is to encourage and provide support for deep academic research that will strive to achieve breakthroughs in the state of the art of key enabling technologies to address the questions posed above in the domain of Digital Inclusion. With thes awards, Microsoft Research will support academic research in specific technologies with a keen interest in multidisciplinary approaches to research issues.

Digital Inclusion Through Mobile and Wireless Technologies 2005 Award Recipients

Digital Inclusion Kit in Health and Higher Education
Guillermo Marshall
University of Buenos Aires

We propose the construction of a digital inclusion kit to expand the frontiers of computing technology in health and higher education to promote research and development in Latin America. We aim to reach underserved zones, both rural and urban, with mobile technology, which will allow patients data acquisition and post processing in specialized medical centers, and patient treatment. The use of mobile devices (MD) in conjunction with portable acquisition systems, such as ECG or arterial blood pressure, allows the building of a clinic database which could be analyzed and processed to apply preventive medicine in cardiovascular diseases, diabetes, and cancer, among others. To do this, we propose the deployment of a Virtual Cluster (VC) to share all patient data between educational and research communities, and bring computer resources to them. In fact, VC expands the boundaries of distributed computing among scientists and students by notably simplifying the task of performing computational science research.

For more details and current progress, click this link: http://lsc.dc.uba.ar/digital-inclusion.

Mesh Networking: A Testbed for Security Issues and Community Actions
Carlos Osvaldo Rodriguez
Cequinor/Lanais EFO, CONICET-UNLP

Internet access seems natural to have in many countries. This is not the case in poor regions of Argentina. Even though each child in a school could have access to a laptop (for example, as proposed by Nicholas Negroponte at the MIT Media Lab in the $100 laptop project) once that child returns home there are little chances of having access to Internet and benefiting from its possibilities or have access to a Server at school. Access to a computer, the possibility of neighborhood connectivity, and access to Internet can most surely also open the way for job opportunities. Mesh networking offers a step further in simplifying and making less expensive network connectivity. The present project aims to addresses two different aspects of mesh networking which naturally merge into each other:

  • A Fundamental research project focused on: �Mesh Networking security and privacy�
  • An applied research project: �A test-bed for Mesh Networking within a ‘barrio carenciado’ in La Plata, Argentina. Study of educational benefits for primary school children and search opportunities of parents facing unemployment�

Dra. Alicia Jubert, one of the participants of the research group is also one of the founders of an ONG that is working within the Barrio del Plata. The test bed will link 25-30 homes (with children in primary school age and unemployed parents) and have its gateway to the Internet in the local primary School. The whole family would use the access to Internet for different purposes but, the study will concentrate on education issues of primary school children and providing possible means for reverting parent�s situation. The school will also host a Web server with prepared e-learning material for both children�s education and parents training. The project will be carried out by a multidisciplinary research group from the Universidad Nacional de La Plata in conjunction with the ONG Barrios del Plata.

Integrated Healthcare Information Services through Mobile telephony in Botswana
Henry Nyongesa
University of Botswana

In the emerging information society, access to information sources will be vital for African countries to realize their development goals in the areas of healthcare, education, and social development. In healthcare, for example, access to appropriate information can improve the management of chronic conditions, such as asthma, diabetes, hypertension, and HIV/AIDS, and minimize visits to physicians and periods of hospitalization, thus lowering the cost of healthcare. Many African countries are today well placed to join the information society, due an emerging proliferation of information and communications technology devices, especially mobile phones, in the home and work environments. This project aims to develop and test an integrated internet-based healthcare information service, targeting HIV/AIDS patients. At the heart of the system is an Information Server and Intelligent Personal Information Centres designed to derive and present appropriate information for individual users. In this proposal, the unique requirements of providing a healthcare information service in an African country, Botswana are addressed. The expected outcome is an integrated healthcare information portal.

Robust and Low-Cost Networking for Rural Kiosks
Srinivasan Keshav
University of Waterloo

We propose low-cost and robust networking solutions for rural kiosks using a combination of �mechanical backhaul and long-range directional wireless mesh technologies. These will be used by our NGO partner, eGovServices, to provide connectivity to their rural kiosks, currently 80, expanding to 4200 over the next few years. eGovServices currently used VSAT and dialup connections, which are either costly or unreliable. By eliminating these problems, we hope to reduce their cost of operations and make their affiliated entrepreneurs more productive. Besides helping the specific NGO, our technologies will be placed in the public domain and will enable mesh networking and opportunistic communication technologies in the real world.

Click this link for current information on this supported project http://blizzard.cs.uwaterloo.ca/tetherless/index.php/KioskNet.

Wi-Fi Enabled Phones for Bridging the Cognitive Divide and Transforming the Classroom Experience
Miguel Nussbaum
Universidad Catolica de Chile

Handheld devices with wireless networks (WiFi) allow face-to-face computer-supported collaborative learning. Even quantitative and qualitative results have been shown, its dissemination has been slow due to the technology cost, since to provide equipment to a whole classroom the cost is beyond a poor country or neighborhood budget. However, there are other platforms which are massive even in underdeveloped countries, as mobile phones. Looking therefore to the near future, moving our developed technology to a cheaper platform as Wi-Fi enabled phones, allows to massify the classroom transformation. We also want to show that with technology we can bring the Internet to the classroom without any other hardware. In a recent article, we showed the influence of ICTs, and in particular Chat systems, on the cognitive development of middle-class children in Santiago de Chile. We found a direct correlation between the use of chat and the development of memory and spatial abilities. From our findings, a provocative question arises. Was the finding because of a cognitive development of chatters, or because non chatters have intrinsic less cognitive capacities? We want to show that with Wi-Fi enabled phones the cognitive development of children improve in a one semester experience.

View this link for current status on this project: Part of the Digital Teacher Network - Handheld Learning, Unplugged.

Project of Modern Long-Distance Education for Agricultural and Pastoral Area in Tibet Autonomous Region
Modern Educational Technology Center, Tibet University

Modern Long-Distance Education Center of Tibet University will be based on the Modern Education Technology Center of Tibet University, which consists of the network department and the electro teaching department at present, possessing various large-scale network equipment and the audiovisual teaching equipment. At the same time the network department is also the main node of the China Education and Scientific Research Computer Network (CERNET) in Tibet, which was built and put into practice in 1999. The staffs working in the network department have rich managerial experience of network operation. The main node at present has a SDH with 155 mbps directly connecting with the CERNET center. It has received dozens of education units in Lhasa for the education network. The staffs working in the electrified education department have developed various multimedia coursewares and optimized media courseware. From county to village school, we will use the following methods to connect with network: to connect with wireless since it is good enough to use this in Tibet and the necessities are available here, for example, AWAYA.

  • Setting up a Modern Long-distance Education Center in Tibet University, and offering hardware like teaching resource server.
  • We will select two village schools from one conditioned county around Lhasa as our trial bases.
  • Tibet Telecom has already covered 95% of the counties in Tibet. Modern long-distance center will rent fiber of Tibet Telecom from Tibet University for the county.

Design and Development of a Low-Cost, Low-Power Portable Medical Device Integrated with PC, PDA, and Mobile Platforms for Telemedicine Application
M.B. Srinivas
International Institute of Information Technology

The project proposes to design and develop a low-cost, low-power, and handheld multifunctional medical device for telemedicine application. The device will monitor parameters like Pulse Rate, BP, ECG, and EEG to help people in case of an emergency. It will be particularly useful to rural communities who do not have access to a doctor in case of an emergency such as heart attack or stroke. The device works in conjunction with Pocket PC or Smartphone working on Windows Mobile platform to consult the remote doctor using different types of connectivity, viz., internet, GSM/CDMA, and PSTN. The device is expected to help improve the life expectancy of rural communities across the world.

Check this link to view details and progress on this research: http://www.iiit.net/~srinivas/; http://iiit.ac.in/research/vlsi/MSR_DI.htm.

Appropriate Technology Interaction for Facilitating Peer-to-Peer Internet Microfinance in Uganda
Jan Carel Diehl
Delft University of Technology

Positive economic development in emerging countries has been directly linked with those countries gaining increased access to credit. Micro financing has demonstrated significant capacity for facilitating economic and social development in countries around the world. Relevant work in the field has demonstrated that the use of technology can improve the efficiency of the microfinance system. The project�s research area is microfinance facilitation through technology, specifically person to person (P2P) online lending in Uganda. The goal of this multidisciplinary project will be to research which technologies are most appropriate for integration into Uganda�s specific economic, social, and cultural context in order to facilitate entrepreneurs accessing P2P internet micro credit. The research will be developed through the case study of the P2P internet lending system KIVA using the technology Microsoft Windows Mobile Technology Platform, including mobile phone, PDA and smart phone. KIVA is an innovative concept that operates with local NGOs to connect individual entrepreneurs in developing countries, through a direct peer to peer network online, with sponsors in Western countries that fund their business development proposals. Insights generated through research, design, and testing of prototypes will clarify which technology interaction forms can facilitate quick and easy use by eager entrepreneurs with limited access and familiarity with technology. The ultimate goal is to increase the scalability and efficiency of the system.

See this link for current news on this project: http://www.intocontext.org/.

Poor Man�s Broadband: Peer-to-Peer Dialup Networking
Umar Saif
Lahore University of Management Sciences, Pakistan

The scarcity of available bandwidth in the developing world is not a �last-mile� problem; a 56 Kbps modem can typically achieve an average throughput of close to 40 Kbps, more than twice the typical 10-20 Kbps ISP bandwidth allocation over a dialup connection in Pakistan. However, given the high cost of international Internet circuits, end users in the developing world are forced to access the Internet at a fraction of the speed possible over a dialup connection. As a result, the accessibility and utility of Internet in the developing world remains severely limited. We propose a solution to enhance the Internet connectivity in the developing world. The key idea of our proposal is simple: bypass the Internet using peer-to-peer dialup connections, at line-speed, when transferring large content between end hosts. This is achieved by using a network routing layer, dubbed dialup-underlay, which opportunistically interleaves the (low-bandwidth) ISP dialup connection with line-speed peer-to-peer dialup connections to deliver near-broadband throughput to dialup users in the developing world. Our work will build on experience with the MIT�s TEK Web search engine for dialup users in the developing world. Our software will be released as part of the TEK sourceforge distribution.

View current information on this project at: http://www.dritte.org/.

Deployment Wireless Ambient Networks on Heterogeneous Rural Environments
Eduardo Gramp�n
Instituto de Computaci�n. Universidad de la Rep�blica, Uruguay

The aim of this project is to improve connectivity of remote rural communities, which take part in the Link-All initiative, linked together by a strong share of social development principles, sustained over a core of software applications, which rely on remote sites connectivity throughout Latin American countries. Remote communities often depend on unreliable, expensive, and/or low capacity physical links to the networking environment; in particular certain communities in Uruguay are not yet reachable. In this context, a mobile infrastructure, combined with the deployment of long-reach wireless links are foreseen as a suitable solution to bridge the connectivity gap. Applicability of Ambient Network (AN) concepts will be studied in this context, seeking for the development of a framework for providing connectivity to isolated communities at low cost. Wimax, Wi-fi, Bluetooth, and GSM/GPRS technologies are involved in the project, and our aim is to orchestrate network connectivity among these heterogeneous technologies, based on the concept of network composition.

DSH and CAM: Leveraging Low-Cost Technology for Rural India
Thomas Anderson
University of Washington

We propose to build and deploy two demonstration projects in rural India: DSH and CAM. DSH (Digital Study Hall) is a project to address the limited educational opportunities for the rural and urban poor by using video capture, the postal service, DVDs, cable TV, and cheap hand held radios. CAM is a project to reduce the paperwork barrier for rural microfinance and other applications by using intermittently connected camera phones as portable data capture devices. Two common themes run through our work.

  • Cost realism: many of the villages where we work lack basics, like clean water and blackboards. This drives us to avoid expensive �wire the world� solutions in favor of the cheapest workable approach; both DSH and CAM use asynchronous �sneaker net� solutions as a key building block. Being inexpensive means we can scale to additional villages more quickly.
  • Total system approach: instead of developing technology for technology�s sake, both DSH and CAM have long-standing relationships with NGOs on the ground in India helping us focus our technical abilities on building and deploying real solutions to the problems of the rural third world poor.

For current progress on this project go to the following link: http://dsh.cs.washington.edu/.

To Download Recent Papers go to the following links:

  • http://cam.cs.washington.edu/papers/mobeav-schwartzman.pdf
  • http://cam.cs.washington.edu/papers/www2006-parikh.pdf
  • http://cam.cs.washington.edu/papers/ER178-schwartzman.pdf

Wireless Grids: A Technology for Rural Networking
Suman Banerjee
University of Wisconsin-Madison

This project considers the problem of providing low-cost Internet connectivity and services to rural areas of developing and under-developed countries. In most rural communities of countries such as India and China, lack of infrastructure presents a significant barrier to the deployment of such connectivity and services. The goal in this work will be to construct a low-cost, yet resilient, self-managing wireless mesh network that provides high-bandwidth connectivity from an urban hub to neighboring villages up to a 40-mile radius. To meet this goal, the project will address various research challenges that occur both in the management and data planes. While we will design a whole gamut of mechanisms as necessary in the management plane, some of the most interesting research questions arise in the contexts of resilient topology planning and spectrum management. The research in the data plane will focus on demonstrating the utility of this infrastructure through customization of a multi-party video-conferencing application. Beyond this research agenda, a key focus of this project will be in realizing a pilot deployment of this system in a specific rural environment, near Hyderabad, India, spanning 32 villages affecting its 500,000 inhabitants. This will be achieved through our partnership with an NGO, Byrraju Foundation in India. The Foundation plans to deploy our solution to provide an array of electronic services, such as distance education for children, vocational and agricultural training, remote healthcare, e-governance, and e-banking. The technology to be used in this project will be low-cost multi-radio mesh nodes, some of which will be equipped with directional (parabolic grid and sector) antenna. The Microsoft Research Mesh Networking Toolkit, already available to the PI, will be a starting point of this project. The entire system and software developed in this project will be made publicly available.

Check the status on rural networking at: www.cs.wisc.edu/~suman/projects/rural.wifi.

AIR: Advancement Through Interactive Radio
John Bennett
University of Colorado at Boulder

Community radio is the dominant mass media in rural developing regions, but the potential for community radio to serve as an agent of social and economic advancement is limited by its inherent unidirectionality. We propose to develop and evaluate a mechanism for asynchronous listener feedback that addresses this deficiency. We will create and deploy simple, rugged, portable, and inexpensive computing devices capable of recording, storing, and forwarding voice feedback from rural listeners. We will initially deploy approximately 100 prototype devices, and will carefully monitor their use in order to evaluate the effectiveness of bidirectional community radio as a means to achieve development goals. In addition to its potential to serve as an agent of social empowerment and as a vehicle for building technical capacity in developing regions, AIR (Advancement through Interactive Radio) poses interesting research questions in the areas of mobile computing, ad-hoc networking, human-computer interaction, power management, encryption, and VOIP low-power radio. Our prototype deployment will target rural, disadvantaged women in five rural communities. The proposed approach has been met with enthusiasm by governmental, NGO, and technology partners. The proposed funding level will support the project through a two-year cycle of prototype development, deployment, and evaluation.

Immersive Language Learning Using Smart Phones
John Canny
University of California at Berkeley

Computer literacy is often seen as a fast track out of poverty for developing regions. However, literacy in a widely-spoken language is an even deeper pre-requisite for economic success. Most developing country governments strive to do this through public education, but face many obstacles. We propose to develop widely deployable, immersive language learning systems using cell phones. Our project will use Microsoft Smartphones with a large-vocabulary speech recognition system being developed by us. We will also use design patterns gleaned from the most effective language courses to create guidelines for new speech-based language learning modules. We will develop several modules using these patterns, using templating and XML to support localization to various first languages. We will develop delay-tolerant networking in support of aggregation of student data in support of collaborative games, and for assessment. The intended outcomes include a high-accuracy LVCSR speech recognizer which is easily retargetable to new languages, design patterns, and course modules for English learning (and their evaluation), and at least one recognizer for a local language (such as Tamil).

See current progress on this project at: http://www.cs.berkeley.edu/~mattkam/millee/index.html.

Smartphone Health Information and Epidemiology Network (SHIEN) — Vietnam
Joseph Rosen, MD
Thayer School of Engineering, Dartmouth College

As Vietnam�s health care system evolves it faces increasing health care disparities and emerging infectious diseases. We propose to build a wireless, electronic medical information system that leverages existing infrastructures to face these growing challenges. Because the local internet and PC infrastructure is still limited, we are designing a system based primarily on Windows Mobile Smartphones. In doing so we will tackle the engineering hurdles inherent in making an intuitive, capable system based on low-bandwidth networks and small user-interfaces. Vietnam has shown remarkable ingenuity in response to challenges; without sophisticated medical technology, the health care system in Vietnam was able to tackle SARS more effectively than any other country in the region. In that same spirit, our priorities are simplicity, flexibility, and reliability. Our system — designed to connect doctors to vital medical information and each other — will be instrumental in ensuring greater access to quality care.

See new details on the progress of this research at: http://www.dartmouth.edu/~engs13/courseinfo/JRosen.html.

Speech Interfaces for Health Information-Access in Underserved Communities
Roni Rosenfeld
Carnegie Mellon University

We are investigating speech interfaces for health workers in underprivileged communities in Karachi, Pakistan. We will be using the Microsoft Speech Server, as well as Windows Mobile Smartphones, to enable access to vast unstructured health information resources to semi- and non-literate health workers in an underprivileged community in Pakistan in collaboration with a community-based NGO. We are aiming to create a novel speech user interface for accessing proven and relevant resources to strengthen existing, proven networks of health care provision. Through the evaluation of this system, we hope to prove that speech-based information access interfaces are compelling for semi-literate users, and enable them to access large information resources that would otherwise be closed to them. Furthermore, we would potentially prove this technology as a means for improving healthcare service provision in the developing world.

See the developing progress of this novel project at: http://www.cs.cmu.edu/~healthline/.

Early Warning Flood Systems Based on Sensor Networks
Daniela Rus
Massachusetts Institute of Technology

Flooding caused by natural disasters crosses cultural boundaries, devastating lives in developed countries (e.g., USA, Germany) as well as developing countries (e.g., Honduras, India, Sri Lanka). The effects of these floods are significantly worse and the recovery efforts are substantially more challenging in developing countries due in part to the lack of early warning systems and infrastructure. We propose a research program in designing and building early warning systems using sensor networks to predict floods. Traditional flood warning systems usually requires complex centralized computing that can be difficult without adequate infrastructure. To address this problem, this project proposes to explore new techniques for distributing the computation of flood detection within a wireless sensor network, along with developing algorithms to adapt to the changing topologies of the network and the calculation. We propose to design and installation an early warning system for flooding and do a first field deployment in the developing country of Honduras. This location typifies the flooding scenario as well as utilizing an existing connection between MIT and a local NGO. This NGO, the Honduran government, and MIT have all committed to this project. In the future we plan to field the system in other parts of the world as well. Overall, this proposal addresses three key research directions: analysis of design requirements for early warning systems within developing countries to further development appropriate work in this area, prototyping and fielding an early warning system prototype installed on the Rio Agu�n in Honduras, and developing algorithms for distributing the flood detection computation within a sensor network and adapting to the morphing topologies this network experiences.

Read what progress has been made on this project at the following link: http://groups.csail.mit.edu/drl/wiki/index.php/floodews.

 > Collaboration > Opportunities