Anonymizing Networks Applications
Internet in the present days has become important in the daily life of everybody in doing many things. With internets impact on people and society, they become more sensitive especially on the issues of privacy. However, the design of the internet did not give consideration t anonymity. The IP packets contains many finger prints. The paper will analyze security, usability and deployability features of one anonymity application called I2p.
I2P is a peer to per bases, message oriented anonymous network of communication of low latency. The I2P network was designed largely to enhance anonymous communication entirely between two parties within a network . Variety of applications inside the network of I2P exists such as anonymous file sharing, web browsing, web hosting, and email among others.
Conrad & Shirazi , further elaborated that I2P is a network that is overlaying that permits the users to exchange within the network anonymously . The network technically is applies framework of java of many applications that is designed to offer P2P networking anonymously. Every user runs the I2P router that is the central part of the I2P software. Moreover, relaying of the messages is done through the built tunnels using other I2P peers by each I2P router. Conrad & Shirazi  pointed out that the tunnels cancan be used only in one direction hence the need for building incoming and outgoing traffic (outbound and inbound tunnels). The messages relayed through this network are encrypted from one end to the other using garlic encryption.
I2P network feature different encryption layers, beginning with the transport encryption layer provided by the TLS connection that is maintained by peers of I2P. Also I2P displays another supplementary tunnel encryption. The relayed messages within the I2P network is also garlic encrypted which implies that the connection to the tunnel from the user is encrypted always. As along as the interaction is within the network, the information or messages sent in I2P are also encrypted from one end to the other .
In terms of attacks targeted to I2P, the known key attacks are categorised as intersection, the partitioning and DoS attacks. Conrad & Shirazi  highlighted some of the proposed attacks on I2P by other experts in the field. For instance, the proposed attack by Herrmann et al that uses a kind of attack that are selective on DoS and exploits bias in the selection of node towards the good performance nodes so as to de-anonymize the peers that hosts the Eepsites. Similarly Crenshaw  also examined the attacks on de-anonymization on connections of I2P by conducting an analysis on the leaked data by the applications that are operated using the I2P network . Lastly, Conrad & Shirazi  also pointed out the more recent proposal by Egger of some attacks on I2P that are practical, where the hacker or attacker uses Sybil and DoS attacks to break the users’ anonymity.
In terms performance, I2P has the capability of achieving better results when simple HTTP-GET-Requests are issued. This could be explained by good peers of I2P distribution in European region and hence good latencies once simple HTTP-GET-Requests are issued. Conrad & Shirazi  indicated that in accessing web pages as a whole and also downloading files, 50% of the I2P requests used up to 103.19 seconds. On the other hand, in downloading, the download speed of I2P was averagely12.91kB/s
In terms of scalability, Conrad & Shirazi  found out that I2P is influenced by increasing the clients’ numbers participating in the network. Despite the fact that the anonymity set becomes bigger and thus there could be stronger anonymity. Moreover, increase in the network traffic may result into problems such as congestion.
For I2P, new peers that join the network may be used in building tunnels, assuming that they offer enough bandwidth and capacity. Therefore, there is no likelihood of congestion appearing. However, if the number of clients seeking to access the services are sufficient and are outside the I2P network, there would be need of providing more out proxies. Other than that, Conrad & Shirazi  asserted that more customers’ connecting to the I2P network offers numerous benefits:
- Most likely the quantity of potential peers that are fast layer will increase and hence tunnels that has more bandwidth and less latency could be the result
- The quantity of the networks cover traffic shall increase with a higher likelihood and hence provide anonymity that is sturdier
- With additional customers in the network and using the network, there is a higher likelihood that a wide range of services will be offered.
In terms of avoiding congestion I2P uses packet switching and this results to some implicit balancing of the load and assists in avoiding service interruptions and congestion. This is particularly essential for transfer of large files.
Deployability is the centralization or how the network is distributed. In I2P, the network is fully distributed with no centralization like in Tor. Each participating in I2P network relay locally and also maintains the list of all relays that are known . Moreover, it is formed by many volunteer run routers .
B. Conrad and F. Shirazi, ‘A Survey on Tor and I2P’, geti2p, 2015. [Online]. Available: https://geti2p.net/_static/pdf/icimp_2014_1_40_30015.pdf. [Accessed: 19- Nov- 2015].
A. Crenshaw, ‘Darknets and hidden servers: Identifying the true IP/network identity of I2P service hosts’, 2015. .
P. Liu, L. Wang, Q. Tan, Q. Li, X. Wang and J. Shi, ‘Empirical Measurement and Analysis of I2P Routers’, Journal of Networks, vol. 9, no. 9, 2014.