Optimal jammer placement in a wireless communication network by clustering

Date

2023-5-30

Author

Özcan, Mahmut Anıl

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

122
views

174
downloads

Cite This

Wireless communication system is a communication technology that enables the wireless transmission of information such as data, sound, and image using wireless signals. The system uses radio frequencies to transmit wireless signals. Today, it is one of the most common communication tools due to its flexibility of movement, easy installation, and wide coverage. However, these systems are vulnerable to jamming attacks. The functionality of wireless communication networks can be disrupted by devices that produce unwanted signals, called jammers. However, targeting a single connection on a wireless network with the jammer may not be enough to break the communication. Since the communication traffic in the network can continue over alternative connections, the communication network should be considered as a whole, and signal jamming strategies should be developed accordingly. This thesis deals with the problem of jammer placement to jam a wireless communication network by dividing it into disconnected clusters. There are limited number of studies on this problem in the literature. As one of the studies, Feng et al. (2015) formulate this problem as an optimal clustering problem on the graph and find the optimal solution with binary integer programming. Their proposed mathematical model finds the minimum number of homogeneous jammers required to divide the wireless network into the desired number of clusters. Jammers can only be positioned at the locations of the radios whose signals are desired to be jammed in the network. The maximum number of radios in subnets is given as input to the model. In this thesis, the mathematical model proposed by Feng et al. (2015) is considered as the base model. New models with different approaches are proposed to improve the jamming performance by eliminating the lacks of the base model. With the first proposed model, it is aimed to minimize the maximum number of radios a cluster can have by giving the number of jammers as input. The second model hierarchically minimizes the number of radios for all clusters, not just the cluster with the maximum number of radios. The optimal solutions are obtained with integer programming formulation in both models. With another proposed model in the thesis, the restriction on the position of the jammers is removed, and placement is provided in the real plane. In the final model, the variable of antenna is added to the jammers, and the problem is addressed with heterogeneous jammers for the first time in the literature. The last two proposed models obtain optimal solutions by mixed integer second-order cone programming formulation due to the real plane placement of the jammers. In a different article, Feng et al. (2019) remove the placement restriction on the jammers and work on the real plane instead of the discretized plane. For this reason, the base model is adapted to the real plane by keeping the approach same to compare the developed models with the base model on the real plane in addition to the discretized plane. In this way, the developed models and the base model are compared with a comprehensive experimental design both in the discretized and real plane. In addition, the final model is compared within itself to examine the effect of using omnidirectional and directional antennas at different angles on the results.

Subject Keywords

Jammer, Radio, Communication link, Network, Jammer placement, Partitioning, Graph clustering

URI

https://hdl.handle.net/11511/104132

Collections

Graduate School of Natural and Applied Sciences, Thesis