My research addresses problems in the areas of distributed computing, networks, and security. Generally, I'm interested in whether networks can be made robust to failures using protocols that consume minimal resources. and still yield high performance. be protected from powerful attacks . Addressing this topic often requires making use of probabilistic techniques, graph theory, and cryptography.
Wireless Sensor Networks One application domain for the algorithms I design is wireless sensor networks (WSNs). These networks hold the promise of large-scale data collection in challenging environments without the need for a physical infrastructure. However, these networks are also vulnerable. For instance, WSN devices are typically battery powered and, as a result, they are subject to strict energy-usage constraints which are often at odds with achieving a high degree of fault tolerance. Furthermore, the wireless communication medium can be easily disrupted by an attacker.
Peer-to-Peer Networks In the wired domain, the peer-to-peer (P2P) paradigm remains a popular framework for providing large-scale distributed services. Despite issues involving copyright violation, the area of peer-to-peer (P2P) continues to mature and today there exist systems tht facilitate legitimate content exchange, play a pivotal role in anonymous communication, and offer certain distributed cloud services. Consequently, there has been significant interest in designing algorithms that allow us to tolerate badly-behaved peers.
Resource-Competitive Algorithms A strong current interest is the design of algorithms whose cost is measured as a function of the adversary's resource expenditures. For example, in WSNs, energy is a scarce resource for both correct and adversarial devices, and the latter must expend energy to carry out attacks on the network (such as jamming the wireless channel). By designing algorithms that force the attacking devices to drain their respective batteries more rapidly than the correct devices, we can achieve an advantage over an adversary. In particular, attacking the system will quickly bankrupt the adversarial devices and the correct devices will succeed in their operations. Resource-competitiveness represents a novel approach to designing attack-resistant distributed algorithms that my collaborators and I have been developing, and much remains to be discovered.