Checkpointing and rollback recovery are well-known techniques for coping with failures in distributed systems. Future generation Supercomputers will be message passing distributed systems consisting of millions of processors. As the number of processors grow, failure rate also grows. Thus, designing efficient checkpointing and recovery algorithms for coping with failures in such large systems is important for these systems to be fully utilized. We presented a novel communication-induced checkpointing algorithm which helps in reducing contention for accessing stable storage to store checkpoints. Under our algorithm, a process involved in a distributed computation can independently initiate consistent global checkpointing by saving its current state, called a tentative checkpoint. Other processes involved in the computation come to know about the consistent global checkpoint initiation through information piggy-backed with the application messages or limited control messages if necessary. When a process comes to know about a new consistent global checkpoint initiation, it takes a tentative checkpoint after processing the message. The tentative checkpoints taken can be flushed to stable storage when there is no contention for accessing stable storage. The tentative checkpoints together with the message logs stored in the stable storage form a consistent global checkpoint. Ad hoc networks consist of a set of nodes that can form a network for communication with each other without the aid of any infrastructure or human intervention. Nodes are energy-constrained and hence routing algorithm designed for these networks should take this into consideration. We proposed two routing protocols for mobile ad hoc networks which prevent nodes from broadcasting route requests unnecessarily during the route discovery phase and hence conserve energy and prevent contention in the network. One is called Triangle Based Routing (TBR) protocol. The other routing protocol we designed is called Routing Protocol with Selective Forwarding (RPSF). Both of the routing protocols greatly reduce the number of control packets which are needed to establish routes between pairs of source nodes and destination nodes. As a result, they reduce the energy consumed for route discovery. Moreover, these protocols reduce congestion and collision of packets due to limited number of nodes retransmitting the route requests.

Location privacy preservation has shifted to the forefront of discussions about next generation wireless networks. While pseudonym-changing schemes have been proposed to preserve an individual's privacy, simulation has shown that new association attack models render these schemes useless. The major contribution of this thesis is the implementation of a tracking network with commodity hardware on the California Polytechnic State University campus which leverages the combination of de-anonymization strategies on captured wireless network data to show the effectiveness of a pseudonym-changing scheme for wireless identification across WiFi and Zigbee protocols.

<p>This thesis presents the lessons learned from building an IEEE 802.11 wireless mesh network (WMN) test-bed. Each network node consists of a Linux processor with multiple IEEE 802.11b/g transceivers operating in the 2.4 GHz band. Each transceiver consists of a medium access control (MAC) and base-band processor (BBP) in addition to a radio. A device driver was modified to control some of the key transceiver functions. The test-bed's Wi-Fi interfaces can be programmed to implement any mesh communication topology. All Wi-Fi interfaces use omni-directional antennas and the IEEE 802.11b operation mode.</p> <p>The test-bed design is easily extendable to incorporate newer Wi-Fi technologies. Measurements of co-channel interference in each Wi-Fi channel including received signal strength (RSS) and signal-to-interference-and-noise ratio (SINR) are presented. The AutoMin algorithm was developed in order to use the captured physical layer (PHY) metrics to avoid Wi-Fi congestion during test-bed operation. A comparison of a software-based spectrum analyzer to a commercial one is described. Key Wi-Fi functions in the Ralink driver source code are explored in depth. The compliance of the Ralink chip-set to the IEEE 802.11b spectral mask was verified. The maximum driver-induced retuning rate for the popular Ralink radio was found experimentally. This data can be used to optimize the performance of IEEE 802.11 WMNs.</p> / Master of Applied Science (MASc)

<p>We examine energy harvesting sensor networks, more specifically, a sensor network using the Geographic Routing with Environmental Energy Supply (GREES) algorithm. We start with a discussion of other sources of energy conservation both in energy harvesting and non-energy harvesting sensor networks. Ideas presented in these works are combined where possible with the GREES algorithm. A sensor network was actually built to test and (if possible) improve the algorithm. There were problems along the way, but they were overcome to produce a functioning energy harvesting sensor network that used solar cells as the energy harvesting unit. Tests were run on the network by giving a consistent light and battery supply, and then changing parameters of the algorithm to see their effect on the lifetime of the network, indicating the network's sensitivity to individual parameters. These results are presented, along with their interpretation, as well as an error analysis detailing the behaviour of the algorithm. We discuss how sensitive the network is to each parameter, indicating which parameters are more important to calibrate or measure correctly.</p> / Master of Applied Science (MASc)

<p>Selfish routing is frequently discussed. The general framework of a system of non-cooperative users can be used to model many different optimization problems such as network routing, traffic or transportation problems.</p> <p>It is well known that the Wardrop user equilibria (i.e. the user optima) generally do not optimize the overall system cost in a traffic routing problem.</p> <p>In order to induce the equilibrium flow to be as close to the optimal flow as possible, the term “toll” is introduced. With the addition of tolls, a traffic system does not show the actual cost to the users but the displayed cost of users, which is the summation of the actual cost and the toll. A common behavioral assumption in traffic network modeling is that every user chooses a path which is perceived as the shortest path, then the whole system achieves the equilibrium of the displayed cost. It is proved that there exists an optimal toll which can induce the equilibrium flow under displayed cost to be the optimal flow in reality.</p> <p>However, this conclusion holds only if the selfish routing executes only once. If the game is played repeatedly, the users will detect the difference between the actual and displayed costs. Then, they will not completely trust the information given by the system and calculate the cost. The purpose of this thesis is to find out the optimal strategy given by the system–how to set tolls in order to maintain the flow as close to the optimal flow as possible.</p> / Master of Science (MSc)

A web app was developed which allows any internet-connected device to remotely monitor a roadway intersection’s state over HTTP. A mapping simulation was enhanced to allow researchers to retroactively track the location and the ad-hoc connectivity of vehicle clusters. A performance analysis was conducted on the utilized network partitioning algorithm. This work was completed under and for the utility of ETSU’s Vehicular Networking Lab. It can serve as a basis for further development in the field of wireless automobile connectivity.

Despite the world-changing success of the Internet, shortcomings in its routing and forwarding system have become increasingly apparent. One symptom is an escalating tension between users and providers over the control of routing and forwarding of packets: providers understandably want to control use of their infrastructure, and users understandably want paths with sufficient quality-of-service (QoS) to improve the performance of their applications. As a result, users resort to various “hacks” such as sending traffic through intermediate end-systems, and the providers fight back with mechanisms to inspect and block such traffic. To enable users and providers to jointly control routing and forwarding policies, recent research has considered various architectural approaches in which provider- level route determination occurs separately from forwarding. With this separation, provider-level path computation and selection can be provided as a centralized service: users (or their applications) send path queries to a path service to obtain provider- level paths that meet their application-specific QoS requirements. At the same time, providers can control the use of their infrastructure by dictating how packets are forwarded across their network. The separation of routing and forwarding offers many advantages, but also brings a number of challenges such as scalability. In particular, the path service must respond to path queries in a timely manner and periodically collect topology information containing load-dependent (i.e., performance) routing information. We present a new design for a path service that makes use of expensive pre- computations, parallel on-demand computations on performance information, and caching of recently computed paths to achieve scalability. We demonstrate that, us- ing commodity hardware with a modest amount of resources, the path service can respond to path queries with acceptable latency under a realistic workload. The ser- vice can scale to arbitrarily large topologies through parallelism. Finally, we describe how to utilize the path service in the current Internet with existing Internet applica- tions.

A common feature of emerging future Internet architectures is the ability for applications to select the path, or paths, their packets take between a source and destination. Unlike the current Internet architecture where routing protocols find a single (best) path between a source and destination, future Internet routing protocols will present applications with a set of paths and allow them to select the most appropriate path. Although this enables applications to be actively involved in the selection of the paths their packets travel, the huge number of potential paths and the need to know the current network conditions of each of the proposed paths will make it virtually impossible for applications to select the best set of paths, or just the best path. To tackle this problem, we introduce a new Network Path Advising Service (NPAS) that helps future applications choose network paths. Given a set of possible paths, the NPAS service helps applications select appropriate paths based on both recent path measurements and end-to-end feedback collected from other applications. We describe the NPAS service abstraction, API calls, and a distributed architecture that achieves scalability by determining the most important things to monitor based on actual usage. By analyzing existing traffic patterns, we will demonstrate it is feasible for NPAS to monitor only a few nodes and links and yet be able to offer advice about the most important paths used by a high percentage of traffic. Finally, we describe a prototype implementation of the NPAS components as well as a simulation model used to evaluate the NPAS architecture.

Campus Area Networks (CANs) are a subset of enterprise networks, comprised of a network core connecting multiple Local Area Networks (LANs) across a college campus. Traditionally, hosts connect to the CAN via a single point of attachment; however, the past decade has seen the employment of mobile computing rise dramatically. Mobile devices must obtain new Internet Protocol (IP) addresses at each LAN as they migrate, wasting address space and disrupting host services. To prevent these issues, modern CANs should support IP mobility: allowing devices to keep a single IP address as they migrate between LANs with low-latency handoffs. Traditional approaches to mobility may be difficult to deploy and often lead to inefficient routing, but Software-Defined Networking (SDN) provides an intriguing alternative. This thesis identifies necessary requirements for a software-defined IP mobility system and then proposes one such system, the Software-Defined Mobile Campus Area Network (SD-MCAN) architecture. SD-MCAN employs an OpenFlow-based hybrid, label-switched routing scheme to efficiently route traffic flows between mobile hosts on the CAN. The proposed architecture is then implemented as an application on the existing POX controller and evaluated on virtual and hardware testbeds. Experimental results show that SD-MCAN can process handoffs with less than 90 ms latency, suggesting that the system can support data-intensive services on mobile host devices. Finally, the POX prototype is open-sourced to aid in future research.

The General Direction Routing Protocol (GDRP) is a Wireless Sensor Network (WSN) multi-path routing protocol which abstracts localization information (commonly GPS coordinates) into relative direction information in order to perform routing decisions. By generating relative direction information GDRP is able to operate with fewer precision requirements than other protocols. This abstraction also allows the integration of other emerging hardware-based localization techniques, for example, Beamforming Sensor Arrays. GDRP does not specifically address the next hop a packet should take, but instead specifies a direction it should travel. This direction abstraction allows for multiple paths to be taken through the network thus enhancing network robustness to node mobility and failures. This indirect addressing scheme also provides a solution to sensor node unique identification. GDRP is simulated in a custom simulator written in Java. This simulator supports interfaces for multiple protocols for layers 1, 2, 3, and 7 of the OSI model. For performance comparisons, GDRP is compared against multiple WSN routing protocols. GDRP operates with a significantly lower setup cost in terms of bytes transmitted and a lower setup latency for networks of varying sizes. It also demonstrates an exponentially lower routing cost when compared to another multi- path routing protocol due to a more efficient packet propagation in the network.