90
Internet of Things (IoT)
5.5.1.2.5 DoS Attack: Application Layer (WSSN)
Through this layer, a path-based DoS attack is initiated by stimulating the sensor nodes
to create a huge traffic in the route toward the base stations. Karlof and Wagner (2003) and
Chen et al. (2009) discussed the severity of attacks in WSN and classified into five types of
threats.
5.5.2 Security Issues in Social Networks
Privacy of knowledge may be leaked if a social network is released improperly to public.
We have experienced the fact that when an individual, organization, or a social group
innovates successfully, the knowledge on which that progress is based becomes visible, at
least partially, in the immediate neighborhood. As time goes on, such progress is understood and copied. Hence, there is a need for a systematic method to anonymize the social
network data before it is released.
Now we would like to discuss about two basic concepts, label hierarchy and neighborhood, which are basic inputs for anonymization methods in social network.
Consider an example for label hierarchy as shown in Figure 5.7. The items in the label
set L form a hierarchy. If the occupations are used as labels of vertices in a social network, L contains not only the specific occupations such as dentist, general physician,
optometrist, high school teacher, and primary school teacher but also general categories like medical doctor, teacher, and professional. We assume that there exists a meta
symbol * ∈ L which is the most general category generalizing all labels. Similarly, in a
social network G, the neighborhood of u ∈ V (G) is the induced subgraph of the neighbors of u, denoted by Neighbor G(u) = G(Nu), where Nu = {v|(u, v) ∈E(G)}. The components of the neighborhood graph of a vertex are the neighborhood components. The
d-neighborhood graph of a vertex u includes all the vertices that are within the distance
“d” from the vertex u.
5.5.2.1 Neighborhood Attack
The example in Figure 5.8 illustrates how the neighborhood attacks can take place. Let us
consider the graphical representation (see Figure 5.8a) of the social network in which a
group of persons are connected to each other through the relation of friendship. An edge
connecting two nodes represents that they are friends.
In Figure 5.8a, if an attacker has knowledge about the 1-neighborhood of “Fred” (see
Figure 5.8b), the node “Fred” can be easily identified from the basic anonymized graph
of Figure 5.8c, as no other node has similar 1-neighborhoods to that of “Fred.” Thus,
the privacy of the social network is leaked. Once this node is identified, other private
*Profession
Teacher
Doctor
High school
teacher
Primary
teacher
Optometrist
Dentist
FIGURE 5.7
Label hierarchy.
Internet of Things (IoT)
5.5.1.2.5 DoS Attack: Application Layer (WSSN)
Through this layer, a path-based DoS attack is initiated by stimulating the sensor nodes
to create a huge traffic in the route toward the base stations. Karlof and Wagner (2003) and
Chen et al. (2009) discussed the severity of attacks in WSN and classified into five types of
threats.
5.5.2 Security Issues in Social Networks
Privacy of knowledge may be leaked if a social network is released improperly to public.
We have experienced the fact that when an individual, organization, or a social group
innovates successfully, the knowledge on which that progress is based becomes visible, at
least partially, in the immediate neighborhood. As time goes on, such progress is understood and copied. Hence, there is a need for a systematic method to anonymize the social
network data before it is released.
Now we would like to discuss about two basic concepts, label hierarchy and neighborhood, which are basic inputs for anonymization methods in social network.
Consider an example for label hierarchy as shown in Figure 5.7. The items in the label
set L form a hierarchy. If the occupations are used as labels of vertices in a social network, L contains not only the specific occupations such as dentist, general physician,
optometrist, high school teacher, and primary school teacher but also general categories like medical doctor, teacher, and professional. We assume that there exists a meta
symbol * ∈ L which is the most general category generalizing all labels. Similarly, in a
social network G, the neighborhood of u ∈ V (G) is the induced subgraph of the neighbors of u, denoted by Neighbor G(u) = G(Nu), where Nu = {v|(u, v) ∈E(G)}. The components of the neighborhood graph of a vertex are the neighborhood components. The
d-neighborhood graph of a vertex u includes all the vertices that are within the distance
“d” from the vertex u.
5.5.2.1 Neighborhood Attack
The example in Figure 5.8 illustrates how the neighborhood attacks can take place. Let us
consider the graphical representation (see Figure 5.8a) of the social network in which a
group of persons are connected to each other through the relation of friendship. An edge
connecting two nodes represents that they are friends.
In Figure 5.8a, if an attacker has knowledge about the 1-neighborhood of “Fred” (see
Figure 5.8b), the node “Fred” can be easily identified from the basic anonymized graph
of Figure 5.8c, as no other node has similar 1-neighborhoods to that of “Fred.” Thus,
the privacy of the social network is leaked. Once this node is identified, other private
*Profession
Teacher
Doctor
High school
teacher
Primary
teacher
Optometrist
Dentist
FIGURE 5.7
Label hierarchy.