Emily Kumpel, University of California, Berkeley; Anurag Sridharan, Thejo Kote, and Ari Olmos, Next Drop; Tapan S. Parikh, University of California, Berkeley
Households around the world access water through an intermittent piped water supply; however, delivery of water in these systems is often unpredictable, creating a burden for households waiting to collect water and utilities managing its distribution. We present NextDrop, a system that allows water operators to report information through existing networks of mobile phones and use that information to relay water delivery times to households. NextDrop has been deployed for six months in Hubli, a mid-sized city in India. We describe the NextDrop system, evaluate the accuracy of data collection, and present data collected through NextDrop. The experience of NextDrop shows that crowd-sourced observations can be used to monitor the operation of a complex infrastructure system in a low-resource setting.
Brian G. Omwenga and Pauline W. Githinji, Nokia Research—Africa
We describe the Buzzenger project, a mobile phone based missed-call-duration interpretation project that makes use of a time-sliced messaging communication protocol useful for transmitting small data packets that can form complete messages at no cost. This project takes the basic beeping habits (missed calling/flashing) and extends them into a formalized messaging protocol. This can be particularly useful for sending small structured messages at no cost where useful applications can be developed for emergency situations, feedback, etc. where the user doesn’t have sufficient funds (phone credit) to place a call or send an SMS.
Only radically lower cost service delivery models can close the growing “broadband gap that characterizes much of the developing world. Inveneo’s Haiti Rural Broadband Initiative seeks to demonstrate one such model by combining low-cost wireless technologies, shared infrastructure and local IT capacity building to extend broadband to 22 un-served population centers in Haiti. While progress is possible, scaling such efforts involves a complex set of factors. This paper describes the program's main elements, key goals and current status as well as preliminary lessons learned.
Mohit Jain and Ajeet Pal Singh, JIIT, Noida, India; Soshant Bali and Sanjit Kaul, IIIT-D, New Delhi, India
There is evidence to support the claim that speed-breakers can cause accidents and injury. When a vehicle approaches a speed-breaker at a speed greater than some threshold velocity, the risk of accident or injury is substantial. Speed-breakers are inconspicuous in low visibility conditions, like at night, or when there is fog, rain or snow. This problem is particularly acute in developing countries where speed-breakers don’t always accompany warning signs. We propose an early warning system that uses a smartphone based application to alert the driver in advance when the vehicle is approaching a speed-breaker. In addition, the application constantly monitors the smartphone accelerometer to detect previously unknown speed-breakers. The proposed detection algorithm is easy to implement because it does not require accelerometer reorientation. This is one of the main contributions of our work since previous approaches have used expensive computations to reorient the accelerometer. The algorithm was evaluated using 678 Km of drive data, which involved 22 different drivers, 5 different types of vehicles (bus, auto rickshaw, cycle rickshaw, motorcycle, and car), and 4 smartphones. The results are very promising and can be further improved by aggregating detection reports from multiple smartphones.
Rijurekha Sen and Bhaskaran Raman, Indian Institute of Technology, Bombay
Road traffic congestion is a recurring problem worldwide. In India, a fast growing economy, the problem is acutely felt in almost all major cities. This is primarily because infrastructure growth is slow compared to growth in number of vehicles, due to space and cost constraints. Secondly, Indian traffic being non-lane based and chaotic, is largely different from the western traffic. The difference can be understood fully only through experience, but some example scenarios can be seen at . Thus, Intelligent Transport Systems (ITS), used for efficient traffic management in developed countries, cannot be used as it is in India. ITS techniques have to undergo adaptation and innovation to suit the contrasting traffic characteristics of Indian roads.
In this position paper, we present a comprehensive study of all available ITS systems,including both research prototypes and deployed systems. We next pose a set of interesting open research problems in the context of Indian ITS. Finally, we list a set of public and private organizations, that play a role in Indian traffic management and research, as meaningful collaboration between field practitioners and researchers is needed for efficient transfer of relevant technology. Though our paper focuses on the Indian traffic scenario because of our hands-on experience of working with it [2, 3, 4], many of the problems and solutions outlined in this paper, are relevant for other developing countries as well.
Azarias Reda and Brian Noble, University of Michigan
As available bandwidth improves, the role of latency as the system bottleneck becomes even more pronounced. This is particularly concerning as cellular networks become the primary avenue of connectivity for many users in developing regions. This paper proposes the moving cloud, an architecture for the proactive delivery of data. The moving cloud leverages route fingerprints in individual mobility with users' contextualized behavior of data access. The moving cloud trades bandwidth and storage for latency by predictively placing content where it needs to be in the near future. This is enabled by a novel mechanism for augmenting location prediction algorithms with time bounds, allowing for a timely delivery of data. This paradigm enables a number of networking scenarios ranging from bulk data access and mobile resource augmentation to personal content distribution.
Aditya Vashistha and William Thies, Microsoft Research India
Interactive Voice Response (IVR) systems play an important role in collecting and disseminating information in developing regions. Recently, researchers have used IVR technology to build voice forums, in which callers leave messages that can be heard over the Internet and over the phone. However, despite their appeal, voice forums remain difficult to set up, and difficult to scale due to the overhead of moderating content and the cost of phone calls.
This paper discusses the challenges and opportunities in creating scalable voice forums. We also present a new open-source system, IVR Junction, that leverages existing free services and commercial tools to simplify the process of creating a voice forum. IVR Junction utilizes familiar cloud-based services to provide free content hosting and moderation, as well as a novel mechanism for automatically synchronizing content across geographically-dispersed offices, thereby enabling local access points with decreased calling costs.
Hao Zhuang, Royal Institute of Technology, Sweden, and Aalto University, Finland; Hervé Ntareme, Royal Institute of Technology, Sweden; Zhonghong Ou, Aalto University, Finland; Björn Pehrson, Royal Institute of Technology, Sweden
An increasing number of web-based Delay Tolerant Network (DTN) applications are being developed by researchers, ranging from monitoring the environment to supplementing other network infrastructure located or installed in harsh environments like wireless sensor networks. These applications need to communicate with the DTN daemon to facilitate data transmission. Hence, it is necessary to provide a middleware layer that allows the communication between applications based on different platforms and DTN service daemons. In this paper, a DTN Service Adaptation Middleware (DSAM) is proposed to provide a communication layer between DTN service daemons and different applications based on different platforms (e.g., Java, Python, C/C++, PHP). Furthermore, we delineate the architecture of DSAM and describe two supporting applications based on our middleware, namely DTN2 network management tool and WSN environmental monitoring application. Finally, we present our findings grounded on performance evaluations in terms of throughput and power consumption.
Amal Fahad, Zhuan Chen, Kai Shen, and Jeffrey Bigham, University of Rochester, U.S.A.; Assmaa Fahad, Baghdad University, Iraq
Web clients in the developing world experience long response time due to poor network connectivity. This paper evaluates existing and new web acceleration techniques using access traces from Iraq, Zambia, and Cambodia. We validate the effectiveness of proxy caching and observe around 40% cache hit ratio. But our evaluation finds limited benefit of web prefetching (<1% hit rate by the history-based prefetching that previously reported around 10% hit rate ). This is because proxy caching diminishes the benefits of prefetching at individual clients and further many (40–60%) cache misses are accesses to dynamic web applications that are not prefetchable. To accelerate dynamic web applications for the developing world, we explore the feasibility and performance of mirroring dynamic applications on near-client proxy servers. Preliminary evaluation on a collaborative office application shows an order of magnitude performance improvement by the local mirroring.