Important notes:
The details section of each article contains an extremely brief
description and a link to the article itself.
I don't guarantee that the saved article version is the one specified.
The "to be checked" remark indicates that I'm not sure about this
version or that I don't know where this article was published.
The "sensors" section contains articles regarding sensor
networks specifically
The "adhoc" section contains articles regarding ad hoc
networks
The "adhoc/routing" section contains articles
regarding routing protocols in ad hoc networks
The "detectionoffaults
" section contains articles regarding how to detect faults or
misbehaving nodes in both ad hoc and sensor networks,
The "routing_byzantine" section contains
articles which deals with byzantine failures in routing in wired networks.
The "faults" section contains articles
regarding how to overcome faults in a network,
including byzantine faults.
M.Chu,
H.Haussecker,and F.Zhao, "Scalable information-driven sensor querying
and routing for ad hoc heterogeneous sensor networks", International
Journal of High-Performance Computing Applications vol.16,no.3, 2002. – to
be checked, I don't think I have found this version
details
B. Deb, S. Bhatnagar and B. Nath, “A topology
discovery algorithm for sensor networks with applications to network
management (short paper),” in Proceedings of the IEEE CAS Workshop on
Wireless Communications and Networking, Pasadena, USA, Sept. 2002. – to
be checked, I have found only the short paper
details
D. Estrin, R. Govindan, J. Heidemann, and S.
Kumar, “Next century challenges: Scalable coordination in sensor
networks,” in 5th annual
ACM/IEEE International Conference on Mobile Computing and Net-working, 1999,
pp. 263–270.
details
Q.
Fang, J. Gao and L. J. Guibas, "Locating and
Bypassing Routing Holes in Sensor Networks", INFOCOM 2004
details
D. Ganesan, R. Govindan, S. Shenker, and D.
Estrin, “Highly-resilient, energy-efficient multipath routing in wireless
sensor networks,”
Mobile Computing and Communications Review, vol. 4, no. 5, October 2001.
details
T. He, J. A. Stankovic, C. Lu and T. Abdelzaher,
“SPEED: A Stateless Protocol for Real-Time Communication in Sensor
Networks”
– to be checked
details
W. R. Heinzelman, A. Chandrakasan, and H.
Balakrishnan, “Energy-efficient communication protocol for wireless
microsensor networks,”
in 33rd Annual Hawaii International Conference on System Sciences, 2000, pp.
3005–3014.
details
J. Hill, R. Szewczyk, A. Woo, S. Hollar, D.
Culler, and K. Pister, “System architecture directions for networked
sensors,” in Proceedings
of ACM ASPLOS IX, November 2000.
details
L. Hu and D. Evans, "Secure Aggregation
for Wireless Networks", WSNA 2003? – to be
checked!!!!
details
C. Intanagonwiwat, R. Govindan, and
D. Estrin, “Directed diffusion: A scalable and robust communication
paradigm for sensor networks,”
in Proceedings of the Sixth Annual International Conference on Mobile
Computing and Networks (MobiCOM ’00), August 2000.
details
C.
Intanagonwiwat, R. Govindan, D. Estrin, J. Heidemann, and F. Silva,
”Directed Diffusion for Wireless Sensor Networking”– to
be
checked!!!!
details
J. M. Kahn, R. H. Katz, and K. S. J. Pister,
“Emerging Challenges: Mobile Networking for “Smart Dust” – to be
checked!!!!
details
J. M. Kahn, R. H. Katz, and K. S. J. Pister,
“Next Century Challenges: Mobile Networking for “Smart Dust” – to
be checked!!!!
details
C. Karlof and D. Wagner, “ Secure Routing in Wireless Sensor Networks: Attacks and
Countermeasures” – to be checked!!!!
details
J. Kulik,
W. Rabiner, and H. Balakrishnan. “Adaptive Protocols for Information
Dissemination in Wireless Sensor Networks”. In
Proceedings of the Fifth Annual ACM/IEEE International Conference on Mobile
Computing and Networking (MobiCom’99), Seattle,
WA, 1999.
details
J. Kulik, W. R. Heinzelman, and H.
Balakrishnan, “Negotiation-based protocols for disseminating information
in wireless sensor networks,”
Wireless Networks, vol. 8, no. 2-3, pp. 169–185, 2002. – to be
checked!!!!
details
S.S.
Kulkarni and U. Arumugam, “Collision-free Communication in Sensor
Networks” – to be checked!!!!
details
J.
Liu, J. Reich and F. Zhao, "Collaborative In-Network Processing for
Target Tracking", EURASIP,
Journal on Applied Signal
Processing vol.2003,pp.378 391,Mar.2003. – to
be checked, , I don't think I have found this version
details
J.
Liu, F. Zhao and D. Petrovic, "Information-Directed Routing in Ad
Hoc Sensor Networks", WSNA 2003
details
S. Madden, M. Franklin, J. Hellerstein and W.
Hong, “TAG: a Tiny AGgregation Service for Ad-Hoc Sensor Networks,” in
Proceedings of the USENIX Symposium on Operating Systems Design and
Implementation, 2002
details
A. Perrig, R. Szewczyk, V. Wen, D. Culler, and J. Tygar, “SPINS: Security
protocols for sensor networks,” in Proceedings of Mobile
Networking and Computing 2001, 2001. networks,” USC/ISI, Research Report
TR-2000-527, October 2000.
details
G.J.Pottie and W.J.Kaiser, “Wireless
integrated network sensors” – to be checked!!!!
details
D. Tian
and Nicolas D. Georganas, “A Coverage-Preserving Node Scheduling Scheme
for Large Wireless Sensor Networks”, MOBICOM 2002
details
F. Ye, G.
Zhong, S. Lu, L. Zhang, "Energy Efficient Robust Sensing Coverage in
Large Sensor Networks" – to be checked!!!!
details
C.-W.
Yi, P.-J. Wan, X.-Y. Li and O. Frieder, "Fault Tolerant
Sensor Networks with Bernoulli Nodes", WCNC 2003
details
Y.
Yu, R. Govindan, and D. Estrin, “Geographical and energy aware
routing: A recursive data dissemination protocol for wireless sensor
networks,”
details
J. Zhao,
R. Govindan and Deborah Estrin, “Computing Aggregates for Monitoring
Wireless Sensor Networks” – to be checked!!!!
details
S.Basagni,K.Herrin,E.Rosti,and D.Bruschi,“Secure pebblenets,” in ACM
International Symposium on Mobile Ad Hoc Networking and Computing (MobiHoc
2001),October 2001,pp.156–163.
details
L. Buttyan and J. Hubaux, "Stimulating
Cooperation in Self-Organizing Mobile Ad Hoc Networks". Technical Report
DSC/2001/046, EPFL-DI-ICA, August 2001.
details
B.
Chen, K. Jamieson, H. Balakrishnana and R. Morris, "Span: An
Energy-Efficient Coordination Algorithm for Topology
Maintenance in Ad Hoc Wireless Networks," MOBICOM’01, 2001.
details
C.-L.
Hu and M.-S. Chen, "Adaptive
Information Dissemination: An Extended Wireless Data Broadcasting Scheme
with Loan-Based
Feedback Control", IEEE TRANSACTIONS ON MOBILE COMPUTING, VOL. 2, NO. 4,
OCTOBER-DECEMBER 2003,
pp. 322-336
details
Y. C. Hu, A. Perrig and D.B. Johnson, " Packet Leashes: A Defense against Wormhole Attacks in Wireless
Networks", INFOCOM 2003
details
J. Hubaux, L. Buttyan, and S. Capkun, “The
quest for security in mobile ad hoc networks,” in Proceedings of the ACM
Symposium on Mobile Ad Hoc Networking and Computing (MobiHOC 2001), 2001.
details
T.
Jurdzinski, M. Kutyowski, J. Zatopianski, Efficient Algorithms for
Leader Election in Radio
Networks, ACM PODC’2002, 51-57.
details
M.
Kutyowski and W. Rutkowski, "Adversary
Immune Leader Election in Ad Hoc Radio Networks"
details
X.-Y.
Li, P.-J. Wan and Y.
Wang, "Fault Tolerant Deployment and Topology Control in Wireless
Networks", MOBIHOC 2003
details
H. Luo, P. Zefros, J. Kong, S. Lu, and L. Zhang,
“Self-securing ad hoc wireless networks,” in Seventh IEEE Symposium on
Computers
and Communications (ISCC ’02), 2002.
details
E.
Pagani, G. P. Rossi, "Reliable Broadcast in Mobile Multihop Packet
Networks", MobiCom 1997
details
F. Stajano and R. J. Anderson, “The
resurrecting duckling: Security issues for ad-hoc wireless networks,” in
Seventh International Security
Protocols Workshop, 1999, pp. 172–194.
details
I. Stojmenovic and M. Seddigh,
“Broadcasting in Wireless Networks” – to be checked!!!!
details
Y.-C.
Tseng, Y.-F. Li, and Y.-C. Chang, "On
Route Lifetime in Multihop
Mobile Ad Hoc Networks”, IEEE TRANSACTIONS ON MOBILE COMPUTING, VOL. 2, NO. 4,
OCTOBER-DECEMBER 2003, pp. 366-376
details
Y. Zhang and W. Lee. "Intrusion detection in
wireless ad-hoc networks". In Mobile Computing and Networking, pages
275–283, 2000.
details
L. Zhou and Z. Haas, “Securing ad hoc
networks,” IEEE Network Magazine, vol. 13, no. 6, November/December 1999.
– to be checked!!!!
details
M.
Zorzi and R.R. Rao, "Geographic
Random Forwarding (GeRaF) for
Ad Hoc and Sensor Networks: Energy and Latency
Performance", IEEE TRANSACTIONS ON MOBILE COMPUTING, VOL. 2, NO. 4,
OCTOBER-DECEMBER 2003,
pp. 349-365
details
M.
Zorzi and R.R. Rao, "Geographic
Random Forwarding (GeRaF) for
Ad Hoc and Sensor Networks: Multihop Performance",
IEEE TRANSACTIONS ON MOBILE COMPUTING, VOL. 2, NO. 4, OCTOBER-DECEMBER 2003, pp. 337-348
details
B. Awerbuch, D. Holmer, C. Nita-Rotaru
and H. Rubens, "An On-demand Secure Routing Protocol Resilient to Byzantine
Failures", Wise 2003
details
J. Broch, D. A. Maltz, D. B. Johnson,
Y.-C. Hu and J. Jetcheva, “A performance Comparison of Multi-hop Wireless Ad
Hoc Network Routing Protocols”.Mobicom'98,Dallas Texas,25–30 October,1998.
details
Y.
Ganjali and A. Keshavarzian, "Load
Balancing in Ad Hoc Networks: Single-path
Routing vs. Multi-path Routing", INFOCOM 2004
details
J.
Gao and L. Zhang, "Load Balanced Short Path Routing in Wireless
Networks", INFOCOM 2004
details
J.J. Garcia-Luna-Aceves, M. Mosko
and C. E. Perkins, “A New Approach to On-Demand Loop-Free Routing in Ad
Hoc Networks”
– to be checked
details
Y.-C. Hu, A. Perrig, and D. B. Johnson,
“Ariadne: A secure on-demand routing protocol for ad hoc networks,”
Department of
Computer Science, Rice University, Tech. Rep. TR01-383, December 2001.
details
Y. C. Hu, D. B. Johnson and
A. Perrig, "SEAD: secure Efficient Distance Vector Routing for Mobile Wireless
Ad Hoc Networks", in Proceedings of the 4th IEEE Workshop on Mobile Computing Systems and
Applications (WMCSA 2002),June 2002,pp.3–13.
details
P.Jacquet,
P.muhlethaler, T.Clausen, A.Laouiti,A.Qayyum, L.Viennot, "Optimized
Link State Routing Protocol. for Ad Hoc Networks" – to be checked!!!!
details
D. B. Johnson and D. A. Maltz, “Dynamic
source routing in ad hoc wireless networks,” in Mobile Computing,
Imielinski and Korth, Eds.
Kluwer Academic Publishers, 1996, vol. 353. – to be checked!!!!
details
B.
Karp and H. T. Kung, "GPSR: Greedy Perimeter Stateless Routing for
Wireless Networks", in
Proc. MobiCom, 2000, pp.243 - 254
details
Y.-B. Ko and N. Vaidya, Location-Aided Routing (LAR)
in mobile ad hoc networks, in: International Conference on Mobile
Computing
and Networking (MobiComÕ98) (1998). for ad-hoc wireless networks, in:
International Workshop on Security Protocols (1999).
details
S.-J. Lee
and M. Gerla, "Split Multipath
Routing with Maximally Disjoint Paths in Ad hoc Networks", ICC 2001
details
P. Papadimitratos, Z. Haas and
E. G. Sirer, "Path Set Selection in Mobile Ad Hoc Networks", MobiHoc 2002
details
P. Papadimitratos and Z. Haas,
“Secure routing for mobile ad hoc networks,” in SCS Communication
Networks and Distributed Systems
Modeling and Simulation Conference (CNDS 2002), January 2002.
details
P. Papadimitratos and Z. Haas,
“Secure message transmission in mobile ad hoc networks", Ad Hoc Networks 1
(2003) 193 - 209
details
M. R. Pearlman, Z. J. Haas, P. Sholander and
S. S. Tabrizi, “On the Impact of Alternate Path Routing for Load Balancing
in Mobile
Ad Hoc Networks,”
Proceedings of the first
workshop on mobile and ad hoc networking and computing (MobiHoc 2000), Boston,
MA, Aug. 2000.
details
C. Perkins and E. Royer, “Ad-hoc on-demand
distance vector routing,” in MILCOM ’97 panel on Ad Hoc Networks, 1997.
details
M.
Spohn, J.J. Garcia-Luna-Aceves, "Neighborhood Aware Source
Routing", MobiHOC 2001
details
A. Srinivas and E. Modiano, "Minimum
Energy disjoint Path Routing in Wireless Ad-hoc Networks," Mobicom 2003
details
K. Wu and J. Harms, “On-Demand Multipath Routing
for Mobile Ad Hoc Networks” – to be checked
details
Y. Xu, J,
Heidemann and D. Estrin, "Geography-informed Energy Conservation for Ad Hoc
Routing", MobiCom 2001
details
Z. Ye, S.V. Krishnamurthy and S. K.
Tripathi, "A Framework for Reliable Routing in Mobile Ad Hoc Networks,"
Infocom 2003
details
S. Yi, P. Naldurg, and R. Kravets.
“Security-aware ad-hoc routing for wireless networks”. MobiHOC Poster
Session, 2001.
details
M. G. Zapata and N. Asokan, "Securing Ad hoc
Routing Protocols", WiSe 2002
details
S. Buchegger and J.-Y. L. Boudec, “Nodes
bearing grudges: Towards routing security, fairness, and robustness in
mobile ad hoc
networks,” in Proceedings of the Tenth Euromicro Workshop on Parallel,
Distributed and Network-based Processing. Canary
Islands, Spain: IEEE Computer Society, January 2002, pp. 403–410. Infocom
2002, 2002.
details
S. Buchegger and J.-Y. L. Boudec, “Performance Analysis of the CONFIDANT Protocol: Cooperation Of
Nodes - Fairness In
Dynamic Ad-hoc NeTworks. In Proceedings
of IEEE/ACM Symposium on Mobile Ad Hoc Networking and Computing (MobiHOC), Lausanne, CH, June 2002. IEEE.
details
S. Buchegger and J.-Y. L. Boudec,
“Coping with False Accusations in Misbehavior Reputation Systems for
Mobile Ad-hoc Networks.”,
EPFL Technical Report IC/2003/31, EPFL-IC-LCA, CH-1015 Lausanne, Switzerland
details
S. Chessa and P. Santi, “Comparison-Based System-Level Fault Diagnosis in Ad-Hoc
Networks,” in Proceedings of the 20th IEEE Symposium on Reliable
Distributed Systems, Oct. 2001.
details
P.
Kyasanur and N. H. Vaidya, “Detection
and Handling of MAC Layer Misbehavior in Wireless Networks” – to be
checked
details
S. Marti, T. J. Giuli, K. Lai, and M. Baker,
“Mitigating routing misbehavior in mobile ad hoc networks,” in Sixth
annual ACM/IEEE
Internation Conference on Mobile Computing and Networking, 2000, pp.
255–265.
details
P. Michiardi
and R. Molva, “CORE: A collaborative reputation mechanism to enforce node
cooperation in mobile ad hoc networks.”
Sixth IFIP conference on security communications, and multimedia (CMS 2002),
Portoroz, Slovenia., 2002.
details
J. Staddon, D. Balfanz and G. Durfee,
“Efficient Tracing of Failed Nodes in Sensor Networks,” in Proceedings
of the First ACM
International Workshop on Wireless Sensor Networks and Applications,
Atlanta, USA, Sept. 2002, pp. 122–130.
details
I. C. Avramopoulos, H.
Kobayashi and R. Y. Wang, "A Routing Protocol with Byzantine Robustness", IEEE
Sarnoff symposium, 2003
details
I. C. Avramopoulos, H.
Kobayashi, R. Wang and A. Krishnamurthy, "Highly Secure and Efficient Routing",
INFOCOM 2004
details
K. A. Bradley, S. Cheung, N.
Puketza, B. Mukherjee, and R. A. Olsson, "Detecting disruptive routers: A
distributed network monitoring approach," in IEEE Symposium on Security and
Privacy, 1998.
details
Q. Huang, H. Kobayashi and B. Liu,
"Energy/Security scalable Mobile Cryptosystem"
details
B. R. Smith, S. Murthy, and J. Garcia-Luna-Aceves.
"Securing distance-vector routing protocols." In Proceedings of
Internet
Society Symposium on Network and Distributed System Security, San Diego, CA,
pages 85 – 92, February 1997.
details
I. Gupta, R. van Renesse and K. P. Birman,
“Scalable Fault-Tolerant Aggregation in Large Process Groups,” in Proc.
Conf. on
Dependable Systems and Networks, 2001.
details
L. Lamport, R. Shostak, and M. Pease.
"The Byzantine Generals Problem." IEEE Transactions on Programming
Languages and Application Systems, 4(3):382–401, July 1982.
details
Details of sensors
M.Chu,
H.Haussecker,and F.Zhao, "Scalable information-driven sensor querying
and routing for ad hoc heterogeneous sensor networks", International
Journal of High-Performance Computing Applications vol.16,no.3, 2002. – to
be checked, I don't think I have found this version
Article:
chu_2001scalableinformationdriven.pdf
Brief: IDSQ and CADR are presented. Greedy routing algorithm. The next node will be the
most informative one. In IDSQ - there is a leader which at each step chooses
the next node and receives its measurement. In CADR - there is no leader,
but the current is acting as the leader, it calculates the current belief
based of his and previous measurements and passes it on to the next most
informative neighbor. Holes are not handled.
return to article name
B. Deb, S. Bhatnagar and B. Nath, “A
topology discovery algorithm for sensor networks with applications to
network
management (short paper),” in Proceedings of the IEEE CAS Workshop on
Wireless Communications and Networking, Pasadena, USA, Sept. 2002. – to
be checked, I have found only the short paper
Article: deb_2002topology.pdf
Brief: A tree is constructed by flooding. Each node learns
about its neighbor by listening to the traffic. On the reply messages the
lists
of neighbors are sent and thus the full network topology can be
constructed.
The algorithm uses three colors, black (can be thought of as cluster head),
gray (a normal node) and white (an undiscovered node).
Black nodes wait for an answer from their sons and return the aggregated
information. The protocol is based of timing which is based
on the distance from the route. A black node will wait longer for its sons
if it is closer to the root.
return to article name
D. Estrin, R. Govindan, J. Heidemann,
and S. Kumar, “Next century challenges: Scalable coordination in sensor
networks,” in 5th annual
ACM/IEEE International Conference on Mobile Computing and Net-working, 1999,
pp. 263–270.
Article: estrin_99nextcenturychallenges.pdf
Brief:
Using clusters and possibly a cluster hierarchy. Cluster heads
aggregate data of their children.
The sensors with the higher remaining energy have a better chance to become
a cluster head. The process of electing the cluster head is performed
periodically in order to support network changes and energy status. This
process requires sending and receiving messages
(and thus the bi-directional communication link is guaranteed).
Not all the cluster heads may participate in a computation. The decision is
task specific.
This article also gives a short description of directed diffusion.
return to
article name
Q.
Fang, J. Gao and L. J.
Guibas, "Locating and
Bypassing Routing Holes in Sensor Networks", INFOCOM 2004
Article: fang_2004locatingandbypassing.pdf
Brief: Using greedy forwarding. Weak/strong stuck node are defined
and identified by the TENT rule. Stuck nodes are on the borders of an hole.
Holes are defined by the BOUNDHOLE algorithm which uses the right hand rule
with some restrictions (forbidden region) and handling of crossed edges.
Each hole has a leader and an hole can be bypassed by either side depending
on the destination's location.
return to
article name
D. Ganesan, R. Govindan, S. Shenker,
and D. Estrin, “Highly-resilient, energy-efficient multipath routing in
wireless sensor networks,”
Mobile Computing and Communications Review, vol. 4, no. 5, October 2001.
Article: ganesan_2001highlyresilient.pdf and technical report in
ganesan_2001highlyresilient_technicalreport.pdf
Brief: Based on the article “Directed diffusion: A scalable and robust
communication paradigm for sensor networks”. This article changes
the protocol described in Directed Diffusion to support multipath routing.
Two approaches are described, disjoint multipaths and
braided multipath. Each approach has its characteristics regarding
maintenance overhead and resilience to failures.
return to
article name
T. He, J. A. Stankovic, C. Lu and T.
Abdelzaher, “SPEED: A Stateless Protocol for Real-Time Communication in
Sensor Networks”
– to be checked
Article: he_2003speedastatelessprotocol.pdf
Brief: Each node knows its location. A periodic beacon is sent to the
neighbors in order to know the neighbors' locations and delay.
Each node stores the location and delay information about its neighbors.
Packets are sent according to the distance of the neighbor
to the destination and according to the delay. In case the node got stuck,
i.e. it has no neighbor to forward the message to, it sends
backwards a back pressure beacon.
Unicast, area multicast and area anycast are defined.
return to
article name
W. R. Heinzelman, A. Chandrakasan,
and H. Balakrishnan, “Energy-efficient communication protocol for wireless
microsensor networks,”
in 33rd Annual Hawaii International Conference on System Sciences, 2000, pp.
3005–3014.
Article: heinzelman_00energyefficient.pdf
Brief:
LEACH – low energy adaptive clustering hierarchy.
Assumption - the cost of transmission equals the cost of receiving. The
algorithm is a combination of directed transmission (transmission
from a mote to the base station directly), minimum transmission energy (MTE)
(each message must go through n transmits and receives)
and clustering. The number of cluster heads is determined a priori by the
system.
The cluster head aggregates the data, and use lossy compression. There is a
tradeoff between the quality of the data and the amount of compression.
Comparisons to direct transmission, MTE and static clustering.
The operation is broken into rounds. Each round lasts a fixed amount of
time. At each round there is the set up phase (electing the cluster head)
and the steady state (data is transferred to the base station).
A node uses the number of cluster heads, the number of times it has been a
cluster head so far and the current round number to calculate a threshold,
then is chooses a random number between 0 and 1. If the number is less than
the threshold, the node becomes a cluster head.
The decision to become a cluster head is done at each node independently of
the other nodes.
The cluster heads use CSMA MAC protocol to advertise their election. A non
cluster heads node chooses its cluster head according to the signal strength
and use CSMA MAC to inform the cluster head that it will be a member of the
cluster.
Each cluster head creates a TDMA schedule telling each node when it can
transmit.
The radio of each node is turned on only in its allocated transmission time.
In order to prevent collision between the clusters, each cluster uses a
different CDMA code chosen randomly by the cluster head.
The cluster head aggregates and compresses the data, and sends it to the
base station.
Hierarchical clustering can be used as well.
return
to article name
J. Hill, R. Szewczyk, A. Woo, S. Hollar,
D. Culler, and K. Pister, “System architecture directions for networked
sensors,” in Proceedings
of ACM ASPLOS IX, November 2000.
Article: hill_2000systemarchitecture.pdf
Brief: Characteristics and architecture of sensor networks are
described.
return to
article name
L. Hu and D. Evans, "Secure
Aggregation for Wireless Networks", WSNA 2003? – to be
checked!!!!
Article: hu_2003secureaggregation.pdf
Brief:
the
article addresses the problem of securely aggregating results from individual
sensors and sending the aggregated data to the base station (BS). A tree like
routing hierarchy is assumed where the BS is the root of the tree. Adversaries
can perform Byzantine attacks, but two adversaries cannot be on consecutive
levels of the tree (i.e. father and son). The BS can broadcast messages to all
nodes directly. Each node uses a MAC to authenticate its message. The keys
will be disclosed later by the BS (the idea is taken from the micro TESLA
technique). The BS can only authenticate its children and grandchildren
messages, all the other messages are authenticated by intermediate nodes. The
temporary keys that are used to generate the MACs are based on the shared
symmetric key between each node and the BS. These temporary keys will be
computed by encrypting a counter value using the shared key. Nodes retransmit
the messages received from their immediate children and aggregate the date
received from their grandchildren (via their children) and transmit the MAC of
the aggregation value. An adversary that will change the messages will be
caught by its parent because when the keys are revealed there will be a
mismatch between the messages and their MACs. In this solution the BS must use
wide area broadcast in order to reveal all the temporary keys. This solution
doesn’t scale well, so another solution is presented. Micro TESLE is used to
authenticate keys locally. The K0 key for each node will be
transmitted by the BS and authenticated using the BS’s micro TESLA. The key
used for calculating the MAC for the current message will be revealed with the
next message and authenticated using the TESLA technique.
Once a parent detects an inconsistency MAC from a child or grandchild, it
sends an alarm to the BS along with a MAC computed using the node’s temporary
key.
Replay attacks are ineffective since the keys are replaced every reading. If
the adversary changes the reading and sends an alarm, its parent will of
course send an alarm as well. In this case it is unclear who the adversary is.
Selective forwarding is handled by nodes snooping on their parents and sending
the message in a way that does not depend on the parent.
return to
article name
C. Intanagonwiwat, R.
Govindan, and D. Estrin, “Directed diffusion: A scalable and robust
communication paradigm for sensor networks,”
in Proceedings of the Sixth Annual International Conference on Mobile
Computing and Networks (MobiCOM ’00), August 2000.
Article:
intanagonwiwat_00directeddiffusionascalable.pdf
Brief:
Using RF. No routers, but each mote has a cache and the
ability to interpret interests and decide locally. No global identifiers.
Using naming conventions (decided upon in advance), attribute-value
pair – data centric. Aggregation – a mote may aggregate
data received from several motes, for example, more accurately pinpointing a
location of an object. The sink node sends interests.
The gradient is the direction of the neighbor from which the interest was
received. Each mote may use flooding, constrained flooding
based on location and directional propagation based on previously cached
data.
The return path is the reverse of the propagation path. Each interest has a
data rate. Reinforcement – when the data is detected and
sent at lower rate (at the beginning) the sink might sent a new interest
with smaller interval (better data rate) to one of its neighbors.
Negative enforcement – lowering the date rate.
return
to article name
C.
Intanagonwiwat, R. Govindan, D. Estrin, J. Heidemann, and F. Silva,
”Directed Diffusion for Wireless Sensor Networking”
Article:
intanagonwiwat_03directeddiffusionforwireless.pdf
Brief:
A revision of “Directed diffusion:
A scalable and robust communication paradigm for sensor networks”. A
section describing how
loops can be removed by negative reinforcement was added. An analytic
evaluation and a section describing a generic diffusion substrate
were added as well.
return
to article name
J. M. Kahn, R. H. Katz, and K. S. J.
Pister, “Emerging Challenges: Mobile Networking for “Smart Dust” – to
be checked!!!!
Article: kahn_00emergingchallenges.pdf
Brief:
Optical transmitters and receivers.
There should be a line of sight between source and destination. Avoiding
collision is problematic.
Advantage – motes can passively pass information to the BTS (base station
transceiver) by reflecting it’s beam.
The BTS can receive several reflections simultaneously. Using several kinds of motes is optional.
SNR – signal to noise ratio.
return to
article name
J. M. Kahn, R. H. Katz, and K. S. J.
Pister, “Next Century Challenges: Mobile Networking for “Smart Dust”
– to be checked!!!!
Article: kahn_00nextcenturychallenges.pdf
Brief: The same article as “Emerging Challenges: Mobile Networking for “Smart Dust”.
return to
article name
C. Karlof and D. Wagner, “ Secure Routing in Wireless Sensor Networks: Attacks and
Countermeasures” – to be checked!!!!
Article:
karlof_2002secureroutinginwireless.pdf
Brief: Description of possible attacks and analyzing the major sensor network routing protocols regarding these attacks.
Basically all the articles are very vulnerable and exposed to these attacks.
Countermeasures are suggested against these attacks. Attacks are divided to
outsider/insider attacks, or laptop class/ mote calls attacks.
Examples for countermeasures: the usage of authentication and encryption
implemented by symmetric keys, or the usage of geographical information in order to prevent wormhole attacks.
Knowing the topology of the network can help identifying suspected/extreme changes.
Note that compromising the nodes which are closer to the base station can cause more damage since these nodes relay more traffic.
return to article
name
J.
Kulik, W. Rabiner, and H. Balakrishnan. “Adaptive Protocols for
Information Dissemination in Wireless Sensor Networks”. In
Proceedings of the Fifth Annual ACM/IEEE International Conference on Mobile
Computing and Networking (MobiCom’99), Seattle,
WA, 1999.
Article: kulik_99adaptiveprotocols.pdf
Brief: SPIN – sensor protocols for information via
negotiation.
Problem of flooding – implosion (receiving duplicates data from several
motes), overlap (motes has overlapping data, can be good for
authentication), resource blindness (a mote has single behavior regardless
of the energy left)
Sensors communicate with each other (by sending meta data) about the data
they already have and the data they still need to obtain and by this
eliminate implosion and overlap. There are costs related to the management
of the meta data.
A node can send an advertisement message (ADV), a request message (REQ)(both
meta data messages), and the data itself (DATA).
The protocol consists of three stages, first advertising the data, second
requesting the wanted data and third sending the requested data.
(The article directed diffusion can be thought of as a two stage protocol,
only REQ and DATA)
A node will only participate in the three stage protocol if it has enough
energy.
SPIN-1 and SPIN22 (SPIN-1 with low energy threshold) are described. The
proposed protocols assume unicast and not multicast.
Use of aggregation. Loss can be compensated by resending.
A node with a higher degree will need more energy. Comparison to gossiping, flooding and ideal
dissemination.
return to article
name
J. Kulik, W. R. Heinzelman, and H.
Balakrishnan, “Negotiation-based protocols for disseminating information
in wireless sensor networks,”
Wireless Networks, vol. 8, no. 2-3, pp. 169–185, 2002. – to be
checked!!!!
Article: kulik_99negotiationbased.pdf
Brief: A revision of
“Adaptive Protocols for
Information Dissemination in Wireless Sensor Networks”.
The unicast protocols are called SPIN-PP (equivalent to SPIN-1) and SPIN-EC
(equivalent to SPIN-2).
In addition broadcast protocols are described, SPIN-BC and SPIN-RL. The last
one achieves reliability by using retransmissions.
return to
article name
S.S.
Kulkarni and U. Arumugam, “Collision-free Communication in Sensor
Networks” – to be checked!!!!
Article: kulkarni_2003collisionfree.dvi
Brief: The algorithm is based on a two-dimensional
mesh. Each node has a communication range and an interference range.
MCG – maximum collision group, this is the maximum number of sensors that
can affect the communication of any sensor.
A node can send diffusion message only in specific intervals according to
the MCG, i.e. in its previously determined time slot (TDM).
This message is used both to diffuse data and to synchronize the clocks of
the nodes. The assumption is that based on to the time
of the message and the path that the diffusion took to reach the node, a
node can calculate the current time.
Based on the direction from which the diffusion message was received the
node decides when to transmit it (In order to avoid collisions). Stabilization –
if a node doesn’t receive a diffusion message for certain consecutive
number of times, it will not transmit any message
until it receives a diffusion message.
return
to article name
J.
Liu, J. Reich and F. Zhao, "Collaborative In-Network Processing for
Target Tracking", EURASIP,
Journal on Applied Signal
Processing vol.2003,pp.378 391,Mar.2003. – to
be checked, , I don't think I have found this version
Article: liu_2002collaborativeinnetwork.pdf
Brief: Greedy routing algorithm. The next node will be the
most informative one. Defined mutual information (U:V), how information can
V conveys about U. The idea is to choose the next node without knowing its
measurement.
return
to article name
J.
Liu, F. Zhao and D. Petrovic, "Information-Directed Routing in Ad
Hoc Sensor Networks", WSNA 2003
Article: liu_2003informationdirected.pdf
Brief: Two routing scenarios are described: Routing from an
arbitrary node to the high activity region and back. Routing from an
arbitrary node to an exit node, maximizing information gain along the path.
The algorithm is greedy. The next node will be the
most informative one. The first routing scheme is based on CADR. Holes are
handled by choosing the next node by calculating M-hop paths instead of only
checking the next node. The next node will be the first node of the most
informative M-hop path. In the second scheme, there is a limited
communication cost, the most informative path which doesn't exceed this cost
is chosen.
This article is based on "Collaborative In-Network Processing for
Target Tracking" and "Scalable
information-driven sensor querying and routing for ad hoc heterogeneous
sensor networks".
return
to article name
S. Madden, M. Franklin, J. Hellerstein
and W. Hong, “TAG: a Tiny AGgregation Service for Ad-Hoc Sensor
Networks,” in Proceedings
of the USENIX Symposium on Operating Systems
Design and Implementation, 2002
Article: madden_2002tagatinyaggregation.pdf
Brief: A tree is constructed. The queries are SQL like.
Timing is used. A parent expects to hear from its children in a certain
bounded
interval.
return to article
name
A.
Perrig, R. Szewczyk, V. Wen, D. Culler, and J. Tygar, “SPINS: Security
protocols for sensor networks,” in Proceedings of Mobile
Networking and Computing 2001, 2001. networks,” USC/ISI, Research Report
TR-2000-527, October 2000.
Article: perrig_2001spinssecurity.pdf and
perrig_2002spinssecurity.pdf
This article is based on an article with the same name from 2001.
Brief: Symmetric encryption and authentication (MAC) are used in order to send messages between the nodes and the base station.
The assumption is that there is a symmetric key between each node and the base station.
An authenticated broadcast by the base station (µTESLA) is described.
The network topology is a forest, and each tree is rooted at the base
station. The construction of the forest is by periodic beacons.
The routing from the base station to the nodes is by source routing.
The
SNEP protocol is
described.
Counters are used in order to create different encryption to the same message and in order to prevent replay.
A protocol to synchronize counter between two nodes is described.
Authenticated routing and node to node key agreement are also described.
return to
article name
G.J.Pottie and W.J.Kaiser, “Wireless
integrated network sensors” – to be checked!!!!
Article: pottie_00wireless-sensor.pdf
Brief: Physical principles and limitations of sensors are described.
The WINS network and the WINS node's architecture are shown.
return to
article name
D.
Tian and Nicolas D. Georganas, “A Coverage-Preserving Node Scheduling
Scheme for Large Wireless Sensor Networks”, MOBICOM 2002
Article: tian_2002acoveragepreserving.pdf
Brief: The basic idea is that a node can go to sleep if its
neighbors cover its sensing area.
The eligibility rule is defined and a back off mechanism.
The scheme is implemented as an extension to LEACH.
return to article
name
F. Ye,
G. Zhong, S. Lu, L. Zhang, "Energy Efficient Robust Sensing Coverage in
Large Sensor Networks" – to be checked!!!!
Article:
ye_energyefficientrobustsensingcoverage.dvi
Brief: Describes the usage of probing in order to know
whether a node should sleep or not. A sleeping node wakes up after sleeping
for an exponentially distributed period of time specified by wakeup rate λ.
return to article
name
C.-W.
Yi, P.-J. Wan, X.-Y. Li and O. Frieder, "Fault Tolerant
Sensor Networks with Bernoulli Nodes", WCNC 2003
Article: yi_2003faulttolerantsensor.pdf
Brief: Determining the radius of transmition as a function
of the number of nodes, the probability that a node is functioning and the
required probability for connectivity.
return to article
name
Y. Yu,
R. Govindan, and D. Estrin, “Geographical and energy aware routing: A
recursive data dissemination protocol for wireless sensor networks,”
Article: yu_2001GEAR.pdf
Brief: The GEAR protocol is described. The next node will
be chosen from the nodes which are closer to the destination according to
its distance from the destination and its remaining energy. In case of an
hole, it will the node with the lowest learned cost. Recursive geographic
forwarding is described. Restricted flooding is used in not dense
networks.
return to article
name
J.
Zhao, R. Govindan and Deborah Estrin, “Computing Aggregates for
Monitoring Wireless Sensor Networks” – to be checked!!!!
Article: zhao_2003computingaggregates.pdf
Brief: Digested Diffusion is defined (digests are aggregates
of network properties). By Digested Diffusion a tree is constructed.
This tree is constructed by calculating specific digests such as maximum.
For example, these digests are not influenced by duplicates.
Each node on the way calculates part of the result and the root of the tree
knows the final result.
Once this tree has been constructed, other non exemplary digests can be
computed. For example, digests which are sensitive to duplicates, such as
average.
return to article
name
Details of adhoc
S.Basagni,K.Herrin,E.Rosti,and
D.Bruschi,“Secure pebblenets,” in ACM International Symposium on Mobile Ad Hoc
Networking and Computing (MobiHoc 2001),October 2001,pp.156–163.
Article: basagani_2001securepebblenets.pdf
Brief: this article
addresses the problem of replacing a global symmetric key common to all the
nodes. Only outside attacks are assumed. The idea is to use as much as
possible temporary keys. The reason is that the probability that a key will be
broken is increased as the key is used more often. These keys are generated by
applying a known hash function on the constant global key common to all the
nodes. Each temporary key is valid for one round. Each node has a weight which
is determined by application specific parameters such as battery power. During
this round, HELLO messages are sent and used to create clusters. The cluster
heads (CH) are the nodes with the highest weight. Clusters are at most two
hops wide. These CHs form a backbone. The CH with the highest weight will
generate the next global key (TEK) which will be used to encrypt the routing
and data messages.
return to
article name
L. Buttyan and J. Hubaux. "Stimulating
Cooperation in Self-Organizing Mobile Ad Hoc Networks". Technical Report
DSC/2001/046, EPFL-DI-ICA, August 2001.
Article: buttyan_2001stimulatingcooperation.pdf
Brief:
Asymmetric cryptography is used (there is a certificate for each public
key).
Security module which is tamper resistant is assumed. This module is
responsible on forcing the algorithm and thus malicious
node cannot change the algorithm. Specifically, the security module manages
the counter and produces the security header.
The algorithm relies on a counter. A message can be sent if the counter is
equal or bigger to the estimated number of hops of the path.
The counter is incremented by one if relaying a message for other nodes and
decremented in path length if sending a massage.
The result of this protocol is that it is best for a node to cooperate in
order to send the greatest amount of messages.
Neighbors establish a Security Association in order to create symmetric key.
return to
article name
B.
Chen, K. Jamieson, H. Balakrishnana and R. Morris, "Span: An
Energy-Efficient Coordination Algorithm for Topology
Maintenance in Ad Hoc Wireless Networks," MOBICOM’01, 2001.
Article: chen_2001mobicom_span.pdf
Brief:
This paper presents the SPAN technique that reduces energy consumption. In
this technique, some of the nodes are asleep and some are coordinators.
Coordinators take part in the routing protocol and relay messages as needed.
A node will be a coordinator depending on its energy level and the number of
nodes it connects.
return to
article name
C.-L.
Hu and M.-S. Chen, "Adaptive
Information Dissemination: An Extended Wireless Data Broadcasting Scheme
with Loan-Based
Feedback Control", IEEE TRANSACTIONS ON MOBILE COMPUTING, VOL. 2, NO. 4,
OCTOBER-DECEMBER 2003,
pp. 322-336
Article:
hu_2003adaptiveinformationdissemination.pdf
Brief: N/A
return to
article name
Y. C. Hu, A. Perrig and D.B.
Johnson, " Packet Leashes: A Defense against Wormhole Attacks in Wireless
Networks", INFOCOM 2003
Article: hu_2003packetleashes.pdf
Brief:
–
geographical and temporal leashes are presented. Time synchronization is
assumed. Based on the time the message was sent and the received time, it can
be checked if the distance the message has passed is possible. The time bounds
the distance. The transmission distance is known. Both hash chain and hash
tree are presented. In hash tree the root is used for authentication. The
leaves are the keys. Given a key and a path to the root, the key can be
authenticated. The problem is the high cost of storing the tree or the high
cost of computing the tree. The TIK protocol which is based on TESLA and hash
tree is presented. The TIK protocol is not suitable for sensor networks, it is
too expensive
return to
article name
J. Hubaux, L. Buttyan, and S. Capkun,
“The quest for security in mobile ad hoc networks,” in Proceedings of
the ACM Symposium on Mobile Ad Hoc Networking and Computing (MobiHOC 2001),
2001.
Article: hubaux_2001thequestforsecurity.pdf
Brief: Attacks on ad hoc networks and countermeasure are described.
Asymmetric cryptography is used. An algorithm to implement certificate authority (CA) by constructing a CA
chain is described.
Each node authenticates another node and by constructing a chain of
certificates. By this chain the last node is authenticated by
the first node in the chain. The suggested algorithm is called shortcut hunter.
return to
article name
T.
Jurdzinski, M. Kutyowski, J. Zatopianski, Efficient Algorithms for
Leader Election in Radio
Networks, ACM PODC’2002, 51-57.
Article:
jurdzinski_2002efficientalgorithmsforleader.pdf
Brief: The idea of masters and slaves is presented. Each master has
several slaves which executes its part in the protocol of the leader
election in order to save energy from the master. It is assumed that there
is no collision detection.
return to
article name
M.
Kutyowski and W. Rutkowski, "Adversary
Immune Leader Election in Ad Hoc Radio Networks"
Article:
kutylowski_2003adversdaryimmuneleader.pdf
Brief: The idea is to build chains based of the nodes IDs and then to
merge the chains. All the station knows a secret s which the adversary
doesn't know. At each chain a leader is elected according to the secret s
and the IDs in the chain. It is assumed that there is no collision
detection. All the adversary can do is to scramble the communication.
return to
article name
X.-Y.
Li, P.-J. Wan and Y.
Wang, "Fault Tolerant Deployment and Topology Control in Wireless
Networks", MOBIHOC 2003
Article: li_2003faulttolerantdeployment.pdf
Brief: Given the number of nodes, the required connectivity level of
the graph (k) and alpha, the radius is chosen such that the graph is k+1
connect. The topology control scheme is based on Yao graph - the use of
cones and k closest neighbors.
return to
article name
H. Luo, P. Zefros, J. Kong, S. Lu, and L.
Zhang, “Self-securing ad hoc wireless networks,” in Seventh IEEE
Symposium on Computers
and Communications (ISCC ’02), 2002.
Article: luo_2002selfsecuring.pdf
Brief: Asymmetric cryptography is used. A CA is divided by any set of fixed number of nodes.
This is done by using the polynom's shares (Adi Shamir). All the nodes in the network have shares and participate in producing
certificates, refreshing shares and refreshing certificates. In order for this to work all the nodes shares must be updated.
return to
article name
E.
Pagani, G. P. Rossi, "Reliable Broadcast in Mobile Multihop Packet
Networks", MobiCom 1997
Article: pagani_1997reliablebroadcast.pdf
Brief: The basic idea is to use clusters and ACKs
return to
article name
F. Stajano and R. J. Anderson, “The
resurrecting duckling: Security issues for ad-hoc wireless networks,” in
Seventh International Security
Protocols Workshop, 1999, pp. 172–194.
Article: stajano_1999resurrectingduckling.pdf
Brief: Availability, Integrity, Confidentiality and Authenticity are
defined. Authenticity can be implemented by imprinting.
The thermometer example is given in order to explain these issues.
return to
article name
I. Stojmenovic and M. Seddigh,
“Broadcasting in Wireless Networks” – to be checked!!!!
Article:
stojmenovic_2000broadcasting-in-wireless-networks.pdf
Brief: Internal nodes are used. Internal nodes are the nodes that
belong to the domination set.
The location of all the neighbors or the lists of neighbors of the neighbors
must be known in order to choose the internal nodes.
It is suggested that the maintenance of internal nodes is smaller than the
maintenance of clustering.
return
to article name
Y.-C.
Tseng, Y.-F. Li, and Y.-C. Chang, "On
Route Lifetime in Multihop
Mobile Ad Hoc Networks”, IEEE TRANSACTIONS ON MOBILE COMPUTING, VOL. 2, NO. 4,
OCTOBER-DECEMBER 2003, pp. 366-376
Article: tseng_2003onroutelifetime.pdf
Brief: N/A
return to
article name
Y. Zhang and W. Lee. "Intrusion
detection in wireless ad-hoc networks". In Mobile Computing and Networking,
pages 275–283, 2000.
Article: zhang_2000intrusiondetection.pdf
Brief: A generic model to for intrusion detection is described. Each
node monitors local activities and it can cooperate with its neighbors in
order to detect intrusions. Normal behavior patterns and deviations from the
normal should be defined.
Intrusions detection should be performed at each layer (MAC layer, network
later, application later etc.). An example is given
regarding route updates and the routing table change is measured by
percentage of change route (PCR) and percentage in sum of
hops of all the routes (PCH).
return to
article name
L. Zhou and Z. Haas, “Securing ad hoc
networks,” IEEE Network Magazine, vol. 13, no. 6, November/December 1999.
– to be checked!!!!
Article: zhou_1999securingadhoc.pdf
Brief: Asymmetric cryptography is used. Digital signature are used.
Since there are multiple paths a malicious node can be avoided.
A CA is divided by a specific set of nodes called servers. These
servers have shares and participate in producing certificates,
refreshing shares and refreshing certificates. The idea of threshold
cryptography is presented.
The certificate proves that a
this public key belongs to that node.
Another point mentioned is asynchrony (there is no bound on message delivery
or processing time). Asynchrony can be handled if there
are enough correct servers who can compute all the subshares.
return to
article name
M.
Zorzi and R.R. Rao, "Geographic
Random Forwarding (GeRaF) for
Ad Hoc and Sensor Networks: Energy and Latency
Performance", IEEE TRANSACTIONS ON MOBILE COMPUTING, VOL. 2, NO. 4,
OCTOBER-DECEMBER 2003,
pp. 349-365
Article: zorzi_2003GeRAFenergyandlatency.pdf
Brief: N/A
return to
article name
M.
Zorzi and R.R. Rao, "Geographic
Random Forwarding (GeRaF) for
Ad Hoc and Sensor Networks: Multihop Performance",
IEEE TRANSACTIONS ON MOBILE COMPUTING, VOL. 2, NO. 4, OCTOBER-DECEMBER 2003, pp. 337-348
Article: zorzi_2003GeRAFmultihop.pdf
Brief: N/A
return to
article name
Details of adhoc/routing
B. Awerbuch, D. Holmer, C.
Nita-Rotaru and H. Rubens, "An On-demand Secure Routing Protocol Resilient to
Byzantine Failures", Wise 2003
Article:
awerbuch_2002anondemantsecurerouting.pdf
Brief:
both
the request and the reply are flooded. The request is signed by the source.
The route is accumulated and stored in the reply. Each intermediate node
checks the signature on the reply, signs and sends the reply. Each link has a
weight and the weights are accumulated in the reply. The routes with the
lowest weights are used. Intermediate nodes re-send only lower cost routes.
The weight list of the source is included in the request and used by the other
nodes. Sequence numbers are used to uniquely identify requests and replies.
Detection of faults is based on ACKs and timeouts. If an ACK was not received
till the timeout, then there is a problem. Normally only the destination sends
an ACK. Upon error, the detector of the error (for example a MAC layer
notification.If such exists!!!) signs and sends an error message. If there is
no such detector, upon timeouts which exceed the threshold, the source adds a
list of probes which will also send ACKs. This list is signed and thus the
adversary cannot alter it. The probes are chosen according to a binary search.
And thus after logn faults the faulty link is detected. The list is onion
encrypted, and uses both encryption and HMAC. The symmetric keys between the
source and the probes and used for both encryption and HMAC. The onion
encryption forces the packet to traverse the probes in order. Each probe waits
before sending its ACK. The idea is to accumulate ACKs before sending them to
the source. The symmetric keys and the onion technique are used in the ACKs. A
route with a fault is still used in case the fault was temporary. And thus a
mechanism to handle probes and change the links weights is described. Probes
will be removed and links weights will be reduced if there are no errors.
Only after a route is established the detection mechanism can operate.
The ACKs can only be verified by the source since they are authenticated by
the symmetric key between the sender of the ACK and the source.
return to
article name
J. Broch, D. A. Maltz, D. B.
Johnson, Y.-C. Hu and J. Jetcheva, “A performance Comparison of Multi-hop
Wireless Ad Hoc Network Routing Protocols”.Mobicom'98,Dallas Texas,25–30
October,1998.
Article:
broch_1998aperformancecomparison.pdf
Brief: Comparing the protocols DSDV, TORA, DSR and AODV. AODV was
implemented without the hello messages. The neighbors are discovered on
demand. DSR uses caching to reduce the number of messages and thus the number
of messages is less than in AODV but each message size is bigger than the
messages in AODV. When the nodes are static the performance of DSR and AODV is
almost the same.
The state at each node is not taken into account.
return to
article name
Y.
Ganjali and A. Keshavarzian, "Load
Balancing in Ad Hoc Networks: Single-path Routing vs. Multi-path
Routing", INFOCOM 2004
Article:
ganjali_2004loadbalancinginadhocnetworks.pdf
Brief: It is shown that in a circle the load is not balanced between
the nodes even if multi-path routing is used, unless a very large number of
paths are used. The nodes in the center of the circle relays most of the
messages. Given two points A and F, they have calculated all the points B,
such that the path between A and B goes through F.
return to
article name
J.
Gao and L. Zhang, "Load Balanced Short Path Routing in Wireless
Networks", INFOCOM 2004
Article:
gao_2004loadbalancedshortpathrouting.pdf
Brief: Several greedy algorithms are described (GREEDY1 - GREEDY4).
GREEDY1/2 handle a line and GREEDY3/4 handle a strip.
The basic idea is to choose the node with the lowest load or the bridge with
the lowest load.
return to
article name
J.J. Garcia-Luna-Aceves,
M. Mosko and C. E. Perkins, “A New Approach to On-Demand Loop-Free Routing
in Ad Hoc Networks”
– to be checked
Article:
garcialunaaceves_2003anewapproachtoondemand.pdf
Brief: The LDR protocol is described.
The protocol is based on AODV.
Sequence numbers and feasible distance are used.
LDR removes the requirement of AODV for nodes to independently increase
other node's sequence numbers, placing firm control
of a sequence number with the owning node.
Conditions NDC, FDC and SDC avoid the creation of routing table loops. If
the start distance of an advertisement is less than a node's feasible
distance, there can be no loop, even if the measured distance increases.
return
to article name
Y.-C. Hu, A. Perrig, and D. B. Johnson,
“Ariadne: A secure on-demand routing protocol for ad hoc networks,”
Department of
Computer Science, Rice University, Tech. Rep. TR01-383, December 2001.
Article: hu_2001ariadne_techreport.ps and the article is
hu_2002ariadne.pdf
Brief: Various attacks and countermeasure in ad hoc networks are
described.
The protocol is based on DSR. The source and the destination have symmetric key used for encryption and authentication (MAC).
Each node on the route authenticated itself by using TESLA. TESLA is authenticated broadcast which doesn't need a key in advance.
Hash chain is used in order to prevent an adversary from omitting nodes from
the message.
Blackmail attack is defined (an adversary says a good node is bad).
The attacker model is active n-m model, n is the number of compromised nodes and m is the number of nodes an adversary
owns.
A mechanism to limit the number of route discoveries initiated by the same node is described.
This is done by authentication.
return to
article name
Y. C. Hu, D. B.
Johnson and A. Perrig, "SEAD: secure Efficient Distance Vector Routing for
Mobile Wireless Ad Hoc Networks", in Proceedings
of the 4th IEEE Workshop on Mobile Computing Systems and Applications (WMCSA
2002),June 2002,pp.3–13.
Article: hu_2002SEAD.pdf
Brief:
proactive protocol based on DSDV. The same idea as in "Securing Ad hoc Routing
Protocols" but the same seed is used for several messages. Max hop count (m)
is assumed. The correct hash is calculated according to the sequence number.
The same seed is used for many messages. The advantage is that only once in a
while an authenticated value should be broadcast. The disadvantage is that the
authentication is more expensive since the hash should be applied more times.
return to
article name
P.Jacquet,
P.muhlethaler, T.Clausen, A.Laouiti,A.Qayyum, L.Viennot, "Optimized
Link State Routing Protocol. for Ad Hoc Networks"
Article: jacquet_2001olsr.ps
Brief: OLSR is a proactive link state protocol. Each node has a
routing table to all the rest of the nodes. Messages are relayed through
multipoint relays (MPRs).
return to
article name
D. B. Johnson and D. A. Maltz,
“Dynamic source routing in ad hoc wireless networks,” in Mobile
Computing, Imielinski and Korth, Eds.
Kluwer Academic Publishers, 1996, vol. 353. – to be checked!!!!
Article:
johnson_1996_dynamicsourcerouting.pdf
Brief: The DSR protocol is described. This is one of the mostly used routing protocols.
The protocol is based on requests and replies. Each node has an ID/address.
The source node initiates the request. In the request the
the unique request id and destination are specified. The nodes addresses are
accumulated in the request as it is relayed towards the destination. Only
the first request for a specific request id is handled. The route reply
contains all the nodes addresses it has relayed
through as well. The reply's route can be the reverse of the request route
or another route.
Route error messages are used in order to report on errors to the source.
Nodes cache is used for optimizations (a request will not reach the
destination if a node on the way has a path to the destination).
return to
article name
B.
Karp and H. T. Kung, "GPSR: Greedy Perimeter Stateless Routing for
Wireless Networks", in
Proc. MobiCom, 2000, pp.243 - 254
Article:
karp_2000GPSR.pdf
Brief:
The GPSR protocol is described. The greedy approach is used, the next node
will be the neighbor which is closer to the destination and it is the
closest among all the neighbors. If not such node exist, i.e. the current
node is a local minimum, perimeter mode is used. In this mode a planar graph
is used and the next node will be chosen according to the right hand rule.
Once a node which is closer to the destination is reached the greedy mode is
used again.
return to
article name
Y.-B. Ko and N. Vaidya, Location-Aided
Routing (LAR) in mobile ad hoc networks, in: International Conference on
Mobile Computing
and Networking (MobiComÕ98) (1998). for ad-hoc wireless networks, in:
International Workshop on Security Protocols (1999).
Article:
ko_1998locationaidedrouting.pdf
Brief: Each node knows its location and has an estimate on
the location of the destination.
The message will be relayed by intermediate nodes only if they are closer to
the destination.
Two methods are described. The first one uses request zone and the other one
uses distances.
return to
article name
S.-J.
Lee and M. Gerla, "Split
Multipath Routing with Maximally Disjoint Paths in Ad hoc Networks", ICC 2001
Article:
lee_2001splitmultipath.pdf
Brief: The SMR protocol is presented. RREQs are flooded
through out the network and accumulate the route they have traversed through
(source routing). Duplicate RREQ are not dropped in certain conditions - the
duplicate packed traversed through different incoming link than the link from
which the first RREQ is received, and whose hop count is not larger than that
of the first received RREQ. The destination chooses the shortest path and the
path which is maximally disjoint from it. Only the destination replys to RREQs,
no cache is used by intermediate nodes. RREPs include all the route's nodes'
IDs and are sent back to the source. Upon error, RERR is sent back to the
source of the message. The RERR contains a route to the source and the
problematic edge which is then removed from all the routes of the source. The
disjoin paths are used for load balancing.
return to
article name
P. Papadimitratos, Z. Haas and
E. G. Sirer, "Path Set Selection in Mobile Ad Hoc Networks", MobiHoc 2002
Article:
papadimitratos_2002pathsetselection.pdf
Brief: The DSDP protocol is presented. This protocol selects
edge disjoint paths. The DSDP protocol relies on an underlying routing
protocol to produce a set of candidate paths. Each edge has a probability to
fail. In an iterative way the paths are chosen and their reliability improves.
Given a new path which interlaces with already chosen paths, the interlacing
is tried to be removed. If the result of adding the path is better reliability
the path is added to the chosen paths, otherwise it is removed.
This protocol can be converted to vertex disjoint paths.
return
to article name
P. Papadimitratos and Z. Haas,
“Secure routing for mobile ad hoc networks,” in SCS Communication
Networks and Distributed Systems
Modeling and Simulation Conference (CNDS 2002), January 2002.
Article: papadimitratos_2002securerouting.pdf
Brief: The Secure Routing Protocol (SRP) is presented.
The protocol is based on DSR. There must be a symmetric key between the source and the destination.
Both source and destination use MAC (message authentication code) and thus preventing from intermediate nodes to do almost any
harm. Each messages has 2 identifiers in order to prevent replay and injection of new messages.
Each node has an IP address and the addresses are accumulated in the message.
The assumption is that there are no collusions.
return
to article name
P. Papadimitratos and Z.
Haas, “Secure message transmission in mobile ad hoc networks", Ad Hoc Networks
1 (2003) 193 - 209
Article: papadimitratos_2003securemessage.pdf
Brief:
The SMT
protocol is presented, a complementary protocol to the SRP protocol. Assumes
that there are several node disjoint routes between the source and the
destination. Only the source and the destination have symmetric key which is
used for calculating MAC for the request and the reply. In addition the same
matrix is known by both ends for calculating the dispersion of the message.
The message is disperse along several routes. Upon receiving only several
parts of the original message, the whole message can be calculated. Each route
as a rating according to its reliability. ACKs and timers are used for
feedback
return
to article name
M. R. Pearlman, Z. J. Haas, P.
Sholander and S. S. Tabrizi, “On the Impact of Alternate Path Routing for
Load Balancing in Mobile
Ad Hoc Networks,” Proceedings of the first workshop on mobile and ad hoc networking
and computing (MobiHoc 2000), Boston, MA, Aug. 2000.
Article:
pearlman_2000ontheimpactofalternate.pdf
Brief: Advantages and disadvantages of alternate routes are
described.
The idea is that the source calculates the node disjoint routes from
the source to the destination. The idea is base on Diversity Injection
(“Improving the Performance of Query-Based Routing Protocols Through
‘Diversity Injection’”).
When a node receives a route query it caches the previous hop node ID and
the hop count from the query source. During the reply phase
the cached path information is used to redirect replies along more diverse
paths back to the source. The chosen hop is the one used
least often and closest to the source. The reply accumulates the IDs of the
replying nodes. The source node extracts the route and decomposes it to a
set of links. The set of all the links is used to calculate node disjoint
routes. The next messages from the source
to the destination will contain the specific route.
Partially based on DSR.
return
to article name
C. Perkins and E. Royer, “Ad-hoc
on-demand distance vector routing,” in MILCOM ’97 panel on Ad Hoc
Networks, 1997.
Article: perkins_1997AODV.pdf
Brief: The AODV protocol is described. This is one of the mostly used routing protocols.
The protocol is based on requests and replies. Each node has an ID/address.
The source node initiates the request. In the request there
are the sequence numbers of the source and destination. Each node stores the
next and previous hop. This information is used to
forward the message and for the return path. The reply's path is the reverse
of the request's path. The sequence numbers enables the
sender to make sure that the information is up to date and that the route
doesn't contain loops. A mechanism to handle broken links is
presented. The nodes that identify the broken links prevent others from
using this route by incrementing the destination sequence number
and by setting the hop count to infinity in the route reply.
Hello messages are
used in order to learn about a node's neighbors and to ensure
bi-directionality communication.
Timeouts are used to expire old information at each node's routing table.
return to
article name
M.
Spohn, J.J. Garcia-Luna-Aceves, "Neighborhood Aware Source
Routing", MobiHOC 2001
Article:
spohn_2001neighborhoodaware.pdf
Brief: The NSR protocol is described. The ides is that each node
holds information on it's 2-hop neighbors. Source routing is used. The idea
is to use local information in order to fix broken routes.
return to
article name
A. Srinivas and E. Modiano,
"Minimum Energy disjoint Path Routing in Wireless Ad-hoc Networks," Mobicom
2003
Article:
srinivas_2003minimumenergydisjointpath.pdf
Brief: A centralized algorithm is presented which finds the minimal
energy disjoint paths. Algorithm STPS - changing the transmission power of the
source, for each transmission power, running an algorithm for calculating k
node disjoint paths. After calculating all the options choosing the one with
the minimal energy. Complexity O(kN^3). OCND - calculating 2 link disjoint
paths. For each possible pair of nodes, calculate the minimal energy 2 node
disjoint paths, then run , for example. Dijkstra's algorithm for minimal path
on the resulting graph. The minimal path will be the minimal 2 link disjoint
path. Note that 2 link disjoint path is actually a set of 2 node disjoint
paths. Complexity O(kN^5). STPS and OCND are optimal. In addition 3
sub-optimal algorithms which are more efficient (O(kN^2)) are presented.
return to
article name
K. Wu and J. Harms, “On-Demand Multipath
Routing for Mobile Ad Hoc Networks” – to be checked
Article:
wu_2001ondemandmultipathrouting.pdf
Brief:
The idea is to use DSR but the criteria include: node disjoint, small
length difference between the routes (important in case of node balancing
when the routes are used simultaneously) and small correlation factor
between any two routes, i.e. minimal number of connecting links between any
two routes.
Route queries include a parameter d, which indicates the permitted maximal
difference of the length between the primary (shortest) route
and the alternate routes. When a node receives a route query message it
caches and broadcast it, if it is the first time to receive this query
message or the route included in this query message is node-disjoint with
the routes included in previously cached same route query
messages. When a destination receives a query message, a reply message is
sent back to the source if the query message carries a source route that is
node-disjoint from the existing routes and the difference of the length
between this route and the primary route is less than d.
The reply message contains the whole route information between the source
and the destination.
Neighborhood information of each node on the route is
also piggybacked in the reply message. The source node will calculate
the
correlation factor using the neighborhood information. The source will
calculate the node disjoint paths according to the required
maximal correlation factor.
return to
article name
Y. Xu, J,
Heidemann and D. Estrin, "Geography-informed Energy Conservation for Ad Hoc
Routing", MobiCom 2001
Article:
xu_2001geographyinformedenergy.pdf
Brief: The algorithm GAF is presented. Each GAF node uses local
location information to associate itself with a "virtual grid", where all
nodes in a particular grid square are equivalent with respect to forwarding
packets. Nodes in the same grid then coordinate with each other to determine
who will sleep and how long.
return to
article name
Z. Ye, S.V. Krishnamurthy and S.
K. Tripathi, "A Framework for Reliable Routing in Mobile Ad Hoc Networks,"
Infocom 2003
Article:
ye_2003aframeworkforreliablerouting.pdf
Brief: The AODV Multipath (AODVM) algorithm is presented. In AODVM
multiple RREQ are forwarded by intermediate nodes (as opposed to AODV, in
which only the first RREQ is forwarded). Only the destination sends RREP along
the reverse path the RREPs were sent. Note that a node can participate in only
one path. According to their simulations very few node disjoint paths are
discovered and thus there is a need to use more reliable nodes called R-nodes.
These nodes should be placed in problematic places such as bottlenecks and
thus enable reliable paths. Each node will know its k-hop neighborhood and
calculate the min-cut of its local topology without itself. The idea is to
check how much a node is important. The R-nodes will move to the areas with
the lowest min-cut. The path from the source to the destination will be
comprised of reliable segments. A segment is reliable if there are k node
disjoint paths between its end nodes or if the segment consists of only
R-nodes.
The fault model is that nodes can stop functioning (fail stop fault model).
return to
article name
S. Yi, P. Naldurg, and R. Kravets.
“Security-aware ad-hoc routing for wireless networks”. MobiHOC Poster
Session, 2001.
Article: yi_2001securityaware.pdf
Brief: The SAR protocol is presented.
The nodes are divided into trust levels. Each level may have a unique key.
In the query, the security level is specified and only the nodes with the
appropriate security can participate in the routing protocol.
return to
article name
M. G. Zapata and N. Asokan, "Securing Ad
hoc Routing Protocols", WiSe 2002
Article: zapta_2002securingadhocrouting.pdf
Brief:
SAODV
is presented. The hop count is authenticated by hash chains. Assuming max hop
count. The top hash, computed by applying the hash max hop count on the seed,
is sent with the message and used to verify the message. Each intermediate
node applies the hash before rebroadcasting the message again. The hop count
is part of the AODV protocol. Top hash should be equal to applying the hash on
the given hash max_hop_count - hop_count times. Each message uses a different
seed.
A digital signature is used to authenticate the non-mutable parts of the AODV
and SAODV message including the top hash, max hop count, addresses and
sequence numbers.
return to
article name
Details of detectionoffaults
S. Buchegger and J.-Y. L. Boudec,
“Nodes bearing grudges: Towards routing security, fairness, and robustness
in mobile ad hoc
networks,” in Proceedings of the Tenth Euromicro Workshop on Parallel,
Distributed and Network-based Processing. Canary
Islands, Spain: IEEE Computer Society, January 2002, pp. 403–410. Infocom
2002, 2002.
Article:
buchegger_2002nodesbearinggrudges.pdf
Brief: Analogy from nature - sucker, cheat, grudger (birds).
Identifying malicious neighbors by listening to their traffic. Once a bad
behavior occurs an ALARM is sent to the node's friends.
Each node manages information about the other nodes. Actions are taken in
case of malicious nodes, a request from a malicious node
may be ignored, a path is deleted if it contains a malicious node,
etc.
Authentication is prerequisite in order for the protocol to work. There is
no handling of false accusations.
The protocol is based on DSR. Only negative information is kept.
The protocol assumes there are no collusions.
return
to article name
S. Buchegger and J.-Y. L. Boudec,
“Performance Analysis of the CONFIDANT Protocol: Cooperation Of
Nodes - Fairness In
Dynamic Ad-hoc NeTworks. In Proceedings
of IEEE/ACM Symposium on Mobile Ad Hoc Networking and Computing (MobiHOC), Lausanne, CH, June 2002. IEEE.
Article:
buchegger_2002performanceofCONFIDANT.pdf
Brief: Based of the article,
“Nodes bearing grudges: Towards routing security, fairness, and robustness
in mobile ad hoc
networks,”. In addition, simulation and performance analysis are
described.
return
to article name
S. Buchegger and J.-Y. L. Boudec,
“Coping with False Accusations in Misbehavior Reputation Systems for
Mobile Ad-hoc Networks.”,
EPFL Technical Report IC/2003/31, EPFL-IC-LCA, CH-1015 Lausanne, Switzerland
Article:
buchegger_2003copingwithfalseaccusations.pdf
Brief: Both negative and positive information is kept on each node
(alpha - bad behavior, beta - good behavior). Each node has a rating
by using the Bayesian approach. Second hand information may be accepted. A
rating that deviates too much, i.e. above a certain
threshold will not be accepted. Ratings are exchanged periodically between
the nodes. A nodes observations will get a higher weight
in the rating calculation.
Authentication is prerequisite in order for the protocol to work.
return
to article name
S. Chessa and P. Santi, “Comparison-Based System-Level Fault Diagnosis in Ad-Hoc
Networks,” in Proceedings of the 20th IEEE Symposium on Reliable
Distributed Systems, Oct. 2001.
Article: chessa_2001comparisonbased.pdf
Brief: Nodes are checked for faultiness by sending tests. (test
request and test response)
Soft faults - a limited version of byzantine faults.
Hard faults - a node is dead.
The assumption is that two different misbehaving nodes will return different
answers.
return to
article name
P.
Kyasanur and N. H. Vaidya, “Detection
and Handling of MAC Layer Misbehavior in Wireless Networks” – to be
checked
Article:
kyasanur_2002detectionandhandling.pdf
Brief: The MAC protocol is changed in order to detect
misbehaving nodes. The receiver of the message determines the random backoff
of the sender. If the sender sends its message too early, the receiver
punishes it by increasing the backoff (this protocol assumes the
receiver is good).
In case the receiver may be bad, the sender sends a function to the
receiver, from which it can derive the backoff.
return
to article name
S. Marti, T. J. Giuli, K. Lai, and M.
Baker, “Mitigating routing misbehavior in mobile ad hoc networks,” in
Sixth annual ACM/IEEE
Internationl Conference on Mobile Computing and Networking, 2000, pp.
255–265.
Article: marti_2000mitigatingrouting.pdf
Brief: The watchdog and pathrater mechanism are described. The
watchdog identifies misbehaving nodes by listening to the traffic
send and received from the node's neighbors. The pathrater helps the routing
protocol avoid these nodes.
In the description only negative information is used, in the implementation
also positive information is used.
In case of misbehaving node, a message is sent to the source. The protocol
is base on DSR.
Misbehaving is not punished, only avoided. (the nodes will forward packets
from misbehaving nodes).
Each node manages a table of misbehaving nodes.
return to
article name
P. Michiardi and R. Molva,
“CORE: A collaborative reputation mechanism to enforce node cooperation in
mobile ad hoc networks.”
Sixth IFIP conference on security communications, and multimedia (CMS 2002),
Portoroz, Slovenia., 2002.
Article:
michiardi_2002core.ps
Brief: The watchdog mechanism is used. Several functions may
be used (for routing, for relaying messages etc.) in order to calculate
a node's reputation. Each function has a weight in the global rating
of a node. A generic pattern is described.
A negative reputation is received by second hand information. A positive
information is received by direct observation. New
observations have smaller weight, reliable nodes have higher weight.
Assumes selfish nodes and not malicious nodes (that is why authentication is
not needed).
Each node manages a rating table on the other nodes. Good nodes don't
cooperate with bad nodes.
return
to article name
J. Staddon, D. Balfanz and G.
Durfee, “Efficient Tracing of Failed Nodes in Sensor Networks,” in
Proceedings of the First ACM
International Workshop on Wireless Sensor Networks and Applications,
Atlanta, USA, Sept. 2002, pp. 122–130.
Article:
staddon_2002efficienttracing.pdf
Brief: Each node sends its list of neighbors to the base
station and thus the base station knows the network topology.
A tree is constructed. Each node has an ID. Once messages are not received
from a certain part of the tree, the base station changes the
routes. By this mechanism dead nodes can be identified.
return to
article name
Details of routing_byzantine
I. C. Avramopoulos,
H. Kobayashi and R. Y. Wang, "A Routing Protocol with Byzantine Robustness",
IEEE Sarnoff symposium, 2003
Article:
avramopoulos_2003aroutingprotocol.pdf
Brief: see the article Highly Secure and Efficient
Routing
return to
article name
I. C. Avramopoulos,
H. Kobayashi, R. Wang and A. Krishnamurthy, "Highly Secure and Efficient
Routing", INFOCOM 2004
Article: avramopoulos_2004highlysecure.pdf
Brief:
wired network. The source knows
the topology of the whole network and thus can compute a route to the
destination. Source routing is used. Every pair of routers has a symmetric
key. MACs are used to authenticate the source by each intermediate router. The
MACs are computed from the destination to the first intermediate node in an
onion like style using the symmetric keys. First the MAC for the destination
is calculated, then the MAC for the node before the destination is calculated
on top of it and so on till the MAC of the first node is calculated. This
technique is also used for ACKs and FAs (fault announcements). A window of
sequence numbers is used to prevent replay attacks and accommodate out of
order arrival of legitimate packets. The source dictates the end points of the
window. Timeouts are used to identify and pinpoint failures. Timeouts are set
according to the worst round trip time. Reserved buffers are used to prevent
congestion and DoS attacks. The source runs an algorithm similar to bellman
ford in order to calculate the best path. The notion of prefix spans is
introduced. FAs are only used by the source since faulty sources can cause non
faulty routers to generate FAs. A router can only report in a FA on a
downstream link and thus a false FA can only incriminate a link of a faulty
router. When a source of the route that received the FA wants to advertise a
faulty link there is an uncertainty whether the source is faulty or the link
is faulty and thus the best that can be done is not to include the lead link
(the first link in the route) and the alleged faulty link on the same route. A
mechanism to block faulty routers is described. If a router knows that one of
two links is faulty but both of them are used in a route then it may drop or
generate a FA so that the source will know that one of the links is faulty.
return to
article name
K. A. Bradley, S. Cheung,
N. Puketza, B. Mukherjee, and R. A. Olsson, "Detecting disruptive routers: A
distributed network monitoring approach," in IEEE Symposium on Security and
Privacy, 1998
Article:
bradley_1998detectingdisruptive.pdf
Brief:
wired network. The WATCHERS
protocol is presented. Each router counts the numbers of bytes that has passed
through it. At each round, the counters are passes through out the network and
inconsistencies are detected. Each router holds the routing table of its
neighbors and thus it can tell if its neighbor relay message according to the
table. Router that decided that its neighbor is faulty, doesn’t pass message
to it but does not inform the others about it
return to
article name
Q. Huang, H. Kobayashi and B.
Liu, "Energy/Security scalable Mobile Cryptosystem"
Article:
huang_energysecurity.pdf
Brief:
wired network?!. The elliptic
curve cryptography (ECC) is used to establish certificates between the CA and
the end user. These public/private keys are used to generate a symmetric key.
A security manager which is a stronger entity is assumed. A routing protocol
is described next. The first message is signed by the source and authenticated
by intermediate nodes. The following messages are authenticated by using hash.
This is done by generating a dependency of hashes where the previous message
contains a hash which can only be calculated by an intermediate node when
receiving the current message and thus the message is authenticated. ACKs and
FAs are authenticated by the same way. A link that is reported in a FA must be
the first link downstream from the generator of the FA. FAs are generated on
timeout which is calculated according to the path length. This mechanism also
defends against replay. All the intermediate nodes must receive all the
messages in order to authenticate the message. This is guaranteed if the
source doesn’t send a new message before receiving an ACK or FA on the current
message. Each new route requires a signature
return to
article name
B. R. Smith, S. Murthy, and J.
Garcia-Luna-Aceves. "Securing distance-vector routing protocols."
In Proceedings of Internet
Society Symposium on Network and Distributed System Security, San Diego, CA,
pages 85 – 92, February 1997.
Article:
smith_97securingdistance.pdf
Brief: Byzantine attacks are described on distance vector
routing. The suggested solution is adding sequence number and predecessor
link
in the messages. This information will enable a node to verify the given
information iteratively using information reported by the routers
directly adjacent to the destination and routers immediately adjacent to
each intermediate hop in the path.
return to
article name
Details of faults
I. Gupta, R. van Renesse and K. P.
Birman, “Scalable Fault-Tolerant Aggregation in Large Process Groups,”
in Proc. Conf. on
Dependable Systems and Networks, 2001.
Article:
gupta_2001scalablefaulttolerand.pdf
Brief: Each node calculates a global function. The nodes are
divided into grid boxes. Each group (grid box) has a number according to the
number of nodes. At each stage a leader is chosen for the current subtree.
At the end of the protocol each node knows all the information.
return to
article name
L. Lamport, R. Shostak, and M.
Pease. "The Byzantine Generals Problem." IEEE Transactions on
Programming Languages and Application Systems, 4(3):382–401, July 1982.
Article:
lamport_82byzantine.pdf
Brief: The Byzantine generals problem is described and two
algorithms are shown to solve the problem. The two algorithms differ in
their
assumptions. One of them relies on signatures and the other one does not.
return to
article name
Comments can be sent to Yoav
Reshef
Last updated on 09/03/2004 05:27 PM