No abstract

No abstract

OverviewBy monitoring Internet traffic in the vicinity of a given host-which can be a single computer or a gateway device with multiple computers behind it-an attacker can observe communications originating from and destined to this host. Even if the message contents are encrypted, the IP addresses must remain visible for the packets to be properly routed throughout the open Internet. So the attacker can eventually discover the identity of all of the hosts with which this host communicates. In particular, if this host belongs to a collaborating set, such an attacker would be able to identify the remaining members of the set from simply observing the traffic in and out of this host. Moreover, by analyzing the traffic patterns the attacker may be able to glean useful information and also identify "high-value" hosts within the collaborative against which to mount a denial-of-service (DOS) attack.The CONTRA system camouflages traffic among a set of collaborating hosts, and camouflages critical hosts by spreading the identity of each across multiple IP addresses. One realization of this system comprises a virtual network topology and supporting protocols that operate on top of the network transport layer. The protocol employs a synergistic combination of multipath relay transmissions, K-out-of N message encoding, packet encryption, heteromorphic packet relay and dynamically assignable IP addresses. The characteristics of the virtual network topology and protocols together impede the attacker's ability to analyze traffic patterns, limit the visibility of real IP addresses to those cooperating hosts that are topologically adjacent to a host whose traffic is being monitored, and allow hosts to spread their IP identities and to modify the IPs associated with a host. These system characteristics will reduce the ability of a hostile entity to mount a successful denial-of-service attack against the operations among the set of hosts. ProblemIn the Internet environment it is relatively easy to mount a distributed denial-of-service attack against a specific host, often associated with a specific IP address. There is currently no good, easily-deployable mechanism for preventing such targeted attacks, and no good way to overcome the effect of such an attack in the short run, i.e., a few hours.Because real IP addresses are used to route traffic they are necessarily visible to anyone capable of monitoring traffic over a link on the public Internet. By looking up network assignments at a domain registry, an attacker could identify the set of hosts assigned to a particular organizations; and by analyzing traffic patterns among specific set of IP addresses, an attacker could identify "high-value" hosts within the set that would make good targets for a DOS attack.Furthermore, even if the contents of messages are encrypted, the attacker may observe the traffic pattern variations in response to known network activities, which could be initiated by the attacker, among the set of IP addresses and deduce the relevant information with regard...

The application of remote sensing to the social sciences is an emerging research area. People's behavior and values shape the environment in which they live. Similarly, values and behaviors are influenced by the environment. This study explores the relationship between features observable in overhead imagery and direct measurements of attitudes obtained through surveys. We focus on three topic areas:• Income and Economic Development • Centrality and Decision Authority • Social CapitalUsing commercial satellite imagery data from rural Afghanistan, we present an exploration of the direct and indirect indicators derived from the imagery. We demonstrate a methodology for extracting relevant measures from the imagery, using a combination of human-guided and automated methods. These imagery observables indicate characteristics of the villages which will be compared to survey data in future modeling work. Preliminary survey modeling, based on data from sub-Saharan Africa, suggests that modeling of the Afghan data will also demonstrate a relationship between remote sensing data and survey-based measures of economic and social phenomena. We conclude with a discussion of the next steps, which include extensions to new regions of the world.

No abstract