TY - GEN
T1 - Polynomial Time k-Shortest Multi-criteria Prioritized and All-Criteria-Disjoint Paths
T2 - 5th International Symposium on Cyber Security Cryptography and Machine Learning, CSCML 2021
AU - Dinitz, Yefim
AU - Dolev, Shlomi
AU - Kumar, Manish
N1 - Publisher Copyright: © 2021, Springer Nature Switzerland AG.
PY - 2021/1/1
Y1 - 2021/1/1
N2 - The shortest secure path (routing) problem in communication networks has to deal with multiple attack layers e.g., man-in-the-middle, eavesdropping, packet injection, packet insertion, etc. Consider different probabilities for each such attack over an edge, probabilities that can differ across edges. Furthermore, a usage of a single shortest paths (for routing) implies possible traffic bottleneck, which should be avoided if possible, which we term pathneck security avoidance. Finding all Pareto–optimal solutions for the multi-criteria single-source single-destination shortest secure path problem with non-negative edge lengths might yield a solution with an exponential number of paths. In the first part of this paper, we study specific settings of the multi-criteria shortest secure path problem, which are based on prioritized multi-criteria and on k-shortest secure paths. In the second part, we show a polynomial-time algorithm that, given an undirected graph G and a pair of vertices (s, t), finds prioritized multi-criteria 2-disjoint (vertex/edge) shortest secure paths between s and t. In the third part of the paper, we introduce the k-disjoint all-criteria-shortest secure paths problem, which is solved in time O(min (k| E|, | E|3 / 2) ).
AB - The shortest secure path (routing) problem in communication networks has to deal with multiple attack layers e.g., man-in-the-middle, eavesdropping, packet injection, packet insertion, etc. Consider different probabilities for each such attack over an edge, probabilities that can differ across edges. Furthermore, a usage of a single shortest paths (for routing) implies possible traffic bottleneck, which should be avoided if possible, which we term pathneck security avoidance. Finding all Pareto–optimal solutions for the multi-criteria single-source single-destination shortest secure path problem with non-negative edge lengths might yield a solution with an exponential number of paths. In the first part of this paper, we study specific settings of the multi-criteria shortest secure path problem, which are based on prioritized multi-criteria and on k-shortest secure paths. In the second part, we show a polynomial-time algorithm that, given an undirected graph G and a pair of vertices (s, t), finds prioritized multi-criteria 2-disjoint (vertex/edge) shortest secure paths between s and t. In the third part of the paper, we introduce the k-disjoint all-criteria-shortest secure paths problem, which is solved in time O(min (k| E|, | E|3 / 2) ).
KW - Disjoint shortest paths
KW - Multi-criteria
KW - Path selection
KW - Shortest secure path
KW - k-shortest paths
UR - http://www.scopus.com/inward/record.url?scp=85111982841&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-78086-9_20
DO - 10.1007/978-3-030-78086-9_20
M3 - Conference contribution
SN - 9783030780852
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 266
EP - 274
BT - Cyber Security Cryptography and Machine Learning - 5th International Symposium, CSCML 2021, Proceedings
A2 - Dolev, Shlomi
A2 - Margalit, Oded
A2 - Pinkas, Benny
A2 - Schwarzmann, Alexander
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 8 July 2021 through 9 July 2021
ER -