Abstract:
This report presents a detailed study of the Internet censorship in India. We consolidated a list
of potentially blocked websites from various public sources and checked their accessibility from
9 popular ISPs within the country. Our results indicate that the censorship mechanism varies
across networks: some are blocked at the DNS level while others at the HTTP level. In this
study, we first demonstrate that existing censorship detection tools like OONI results in high
false positives and negatives. Thus, we developed our own approach for identifying censorship
policy and mechanism, which relies on human intervention for confermation of blocked instances.After identifying the censorship infrastructure | poisoned resolvers for DNS and middleboxes for HTTP filtering, we highlight upon two aspects. First, how well the censorship infrastructure covers the ISP and second, how consistent they are in doing censorship. In the last phase of our research, we device our own novel anti-censorship techniques, that does not depend on any third party tools (like proxy, tor and vpn etc.). We managed to anti-censor all blocked websites in all ISPs under test. Our results indicate that ISPs have placed middleboxes sub-optimally viz., experimental results reveal that middleboxes have extremely low values of coverage, for instance | middleboxes of Idea network it is _ 4:02% (i.e., they intercept only _ 4:02% of total paths under our tests) whereas for Vodafone network it is as low as low as 0:37%. It indicates that, some clients of a particular ISP face near to zero censorship, whereas others experience significant censorship