===================================================================================
A Performance Evaluation of Intrusion-Tolerant Routing on Wireless Sensor Networks
===================================================================================
Advantages:
- Builds a secure routing protocol, rather than placing security layer on top of exising routing protocols
- INSENS prevents DoS-style attacks by not allowing individual nodes to broadcast to the entire
  network,
- The resource rich base station is chosen as the central point for computation rather than resource
  poor network nodes
- redundant multipath routing is used to achieve secure routing
- The one-way cryptographic hash function used to generate the sequence is helping hiding
  attacker from guessing the next sequence number to spoof the network
- It is not constrained by time synchronization or delayed release schedule.

Disadvantages:
- Base stations is given too much responsibility and thus the prime target for hackers to bring the entire network down
- If an alternate path is not avialable then the network is susceptible to partitioning under attack
- do not mention the advantages of building a bottom up secure routing proocol (i.e. do not numerically compare their approach with other approaches)  

Improvements:
- For multipath routing table dissemination, meshed multipath routing algorithm can be used.
- Further route failure detection via flow monitoring and overlay routing for route reconfiguration can
  be added to ensure fault tolerance in WSN.
- Better algorithm to find disjoint multi paths with minimum number of common nodes between node and base station.

===================================================================================
Secure Routing in Wireless Sensor Networks: Attacks and Countermeasures
===================================================================================
Advantages:
- The authors outline a number of attacks that are possible on a sensor network. They introduce
  two new kinds of attacks that are specific to sensor networks
- The authors present the drawbacks of existing protocols to overcome these threats
- report that the majority of outsider attacks against sensor network routing protocols can be prevented 
  by simple link layer encryption and authentication using a globally shared key.
- The analysis of various possible attacks on WSN give insight in to the sort of effective
  countermeasures required for security in WSN.
  
Disadvantages:
- Energy requirements and overheads of implementing the countermeasures are not presented.
- The authors have not simulated or provided any platform to show that the countermeasure actually
works.
- The use of geographical information for security carries a lot of overhead.

Improvements:
- Multipath routing to multiple destination base stations can be as a strategy to provide tolerance against
individual base station attacks and/or compromise.
- Relocation of the base station in the network topology can be studied as a means of enhancing resiliency
  and mitigating the scope of damage.
- To come up with application speicifc security schemes and counter measures for given attacks

===================================================================================
SPINS: Security Protocols for Sensor Networks.
===================================================================================
Advantages:
- SNEP provides data confidentiality, two party data authentication and data freshness, which are
  important baseline security primitives.
- µTESLA provides authenticated broadcast for severely resource-constrained environments.
- The counter used in SNEP is incremented after each message, hence the same message is encrypted
differently each time and also prevents the replaying of old messages.
- µTESLA discloses key once per epoch, hence doing away with the need of disclosing a key with every
packet, which requires a lot of energy.

Disadvantages:
- Information leakage due to covert channels problem not addressed.
- The protocol assumes that the receivers know the key disclosure schedule.
- Fails to address DoS type attacks in general
- µTESLA requires that base station and the sensor nodes are loosely time synchronized
- for broadcasting data there should be a node-node key agreement.
Improvements:
- when there is a necessity for a node to broadcast it could send the data packet to the base station for broadcasting
- simulating SPIN under various kinds of attacks (DoS, sinkhole, hello, spoofing, sniffing etc) and comparison with existing security protocols