Satadal Sengupta

On the Free Bridge Across the Digital Divide: Assessing the Quality of Facebook's Free Basics Service

Free Basics is an initiative backed by Facebook to provide users in developing countries free mobile Internet access to selected services. Despite its wide-spread deployment and its potential impact on bridging the digital divide, to date, few studies have rigorously measured the quality of the free Internet service offered by Free Basics. In this short paper, we characterize the quality of the Free Basics service offered in Pakistan and South Africa along three dimensions: (i) the selection of accessible Web services, (ii) the functionality of those services, and (iii) the network performance for those services. While preliminary, our findings show that data- driven studies are essential for having more informed public debates on the pros and cons of the current design of the Free Basics service.

A Qualitative Survey on Multicast Routing in Delay Tolerant Networks

Delay Tolerant Networks (DTNs) are a class of networks that make communication in stressed and challenging environments possible. DTN is characterized with a number of unique features by virtue of which a working environment is achieved in situations where traditional networking paradigms fail to deliver satisfactorily or entirely. The utility of multicasting in DTNs extends to numerous potential DTN applications i.e., crisis environments, battlefield situations, deep space communications, dynamic data size management, etc. In this paper, we propose taxonomy for the different multicast routing strategies and thereafter, we present a comprehensive up to date survey of these strategies. Further, we perform a qualitative comparison between the different multicast strategies with respect to important performance issues in DTN. We also highlight some unexplored areas in DTN multicasting that could inspire research in the near future.

SRSnF: A Strategy for Secured Routing in Spray and Focus Routing Protocol for DTN

This paper deals with the aspect of security in Delay Tolerant Networks (DTN). DTNs are characterized with decentralized control. Network performance and trustworthiness of transmitted information in DTNs depend upon the level of co-operation among participating nodes. As a result, DTNs are vulnerable towards untoward activities arising out of node selfishness as well as malicious intentions. In this paper, we limit our focus to the Black Hole Denial-of-Service attack. We develop a table-based strategy to record network history and use this information to detect discrepancies in the behavior of nodes, followed by elimination of those detected as malicious. We explain our detection mechanism considering Spray & Focus routing protocol as the representative routing scheme. The detection mechanism has been described in detail with examples pertaining to various case scenarios. Furthermore, we study the effect of variation of various parameters on detection efficiency and message transmission through simulation results.

Candid with YouTube: Adaptive Streaming Behavior and Implications on Data Consumption

YouTube has emerged as the largest player among video streaming services, serving video content for users using DASH. Research studies on various aspects of YouTube, especially its streaming service, abound in the literature. However, these works study YouTube streaming from the periphery, and report results based on their understanding of general DASH recommendations. In this study, we explore in depth YouTubes implementation of the DASH client. We identify important parameters in YouTubes rate adaptation algorithm, and study their roles. In a departure from existing literature, we observe that YouTube opportunistically adapts segment length, in addition to quality level, in response to bandwidth fluctuations. We report that this scheme results in a much lower average data wastage ratio (0.82x10-6), than reported earlier. We also propose an analytical model, augmented with a machine learning based classifier (with average accuracy of 85.75%), to predict data consumption for a playback session in advance.

An Empirical Analysis of Facebook's Free Basics

ACM Reference format: Siddharth Singh, Vedant Nanda, Rijurekha Sen, Sohaib Ahmad, Satadal Sengupta, Amreesh Phokeer, Zaid Ahmed Farooq, Taslim Arefin Khan, Ponnurangam Kumaraguru, Ihsan Ayyub Qazi, David Choffnes, and Krishna P. Gummadi. 2017. An Empirical Analysis of Facebook’s Free Basics Program. In Proceedings of ACM Sigmetrics conference, Urbana-Champaign,IL, USA, June 05-09, 2017 (SIGMETRICS ’17), 2 pages. https://doi.org/http://dx.doi.org/10.1145/3078505.3078554

Leveraging Facebook's Free Basics Engine for Web Service Deployment in Developing Regions

In this paper we analyze Facebooks Free Basics program, which provides free Internet access to a restricted set of web services. As the program grows to 60+ developing countries, an independent and data-driven audit of its scope and outreach is highly relevant to the ICTD community. We provide the first large scale empirical observations on how content providers are using the Free Basics platform and what kind of user traffic is expected once a Free Basics service goes live. Implementing an Android app for data collection and recruiting participants from 15 countries, we analyze the current set of Free Basics services and their growth over time. We also deploy our own Free Basics services to gather first hand experience about Facebooks gate-keeping procedure in the program. One of our services Bugle News, an RSS news feed aggregator offered in English, Spanish and French, attracted 95.6K unique visitors from 55+ countries since Sep 2016. This enables us to characterize the nationality, demographics and interests of this Free Basics user population. We specifically deploy an ICTD related Free Basics service called Awaaz: My Voice. Awaaz is a web-service, where citizens can report local issues with location and images. This citizen journalism portal has attracted several hundred users during its short two months deployment in ten cities across South Africa. Visitors have reported concrete issues in categories of road, electricity, water, health and sanitation, school and education, crime and others. Overall our experimental observations allow the ICTD community to understand how Free Basics works and our deployment experiences pave the way for other applications to be launched in future, geared towards important use cases the ICTD community cares about.

Understanding data traffic behaviour for smartphone video and audio apps

This poster is the first known attempt towards traffic engineering for smart-phone audio and video apps - it seeks to report network traffic characteristics, like packet size distribution, traffic burstiness and self-similarity in data traffic. We consider different candidate apps from three different groups - interactive apps, buffered apps and streaming apps, collect packet traces for three months with four customized Android smart-phones, and then analyze their internal patterns. We observe significant differences in traffic characteristics among various application groups, which can be explored in the future for the development of network level service provisioning and traffic management mechanisms.

Prediction of quality degradation for mobile video streaming apps: A case study using YouTube

The growing popularity for developing streaming media applications over HTTP triggers new challenges for managing video quality over mobile devices. Quality of online videos gets significantly affected due to the capacity fluctuations of underlying communication channel, which is very much common for cellular mobile networks. Such fluctuations lead to re-buffering and sudden drops in video quality, adversely affecting video watching experience. In this poster, we propose a light-weight method for early detection of network capacity degradation. We explore the traffic characteristics of mobile streaming video apps, by considering YouTube Android app as a use case. We show that by observing the traffic pattern, we can predict possible video quality degradation and video re-buffering events. We develop a methodology for early prediction of possible re-buffering. The experimental results reveal that our proposed scheme works with very high accuracy.

Predicting social dynamics based on network traffic analysis for CCN/ICN management

Proliferation of online social networks (OSNs) has resulted in an unprecedented surge in the volume of multimedia content consumed by users on a daily basis. Popular OSNs such as Facebook enable users to view and share embedded videos and images on their feeds, which increases visibility, prompting repeated requests for the same piece of content. Maintaining desirable quality of service for all users becomes challenging in such a scenario, especially when low-bandwidth cellular network is being used for data download. Such problems have prompted the research community to focus heavily on the emerging paradigm of Information-or Content-Centric Networking (ICN/CCN), where in-network content management (e.g., content distribution, caching, etc.) forms the crux of an enhanced user experience. In this abstract, we argue that social dynamics among OSN users can provide concrete hints regarding future popularity of content. We propose a strategy to identify viewing and sharing patterns of Facebook users served by a cellular base station, by analyzing network traffic. We utilize these patterns to infer social dynamics among cellular users (mapped to cellphone numbers). We validate our strategy with proof-of-concept experiments on real data, and extensive simulations on a simulation framework proposed by us.

CommBox: Utilizing sensors for real-time cricket shot identification and commentary generation

Online cricket commentary has become very popular as the internet provides access to a large number of sports websites. A key challenge for them is to offer their readers an insightful and fast paced live commentary. In this paper, we propose a framework to automate cricket shot identification and commentary generation using sensor data as features for machine learning models.