The connectivity properties of nanonetworks that employ free diffusion-based molecular communication are analyzed by constructing a model based on Gilbert's disk graph. In free diffusion-based molecular communication, “path loss” has a different functional form compared with wireless propagation and the scenarios under consideration are not generally limited to 2-D environments as in the case of wireless networks. Hence, the constructed model takes into account the peculiarities of diffusion-based communication in 1-D, 2-D, and 3-D scenarios. The model incorporates particle-counting noise and is independent of the signaling mechanism employed. Analytical formulas are derived for the critical values of the signal strength-to-detection threshold ratio and the number of nodes that are required to achieve a connected or an almost-connected network. Monte Carlo simulations are used to validate model predictions and to suggest design guidelines for deploying nanonetworks. An empirical study is also conducted to evaluate the time required to broadcast an alarm in a connected nanonetwork of various sizes. The time to broadcast as a function of the number of nodes in a fixed region is observed to follow a power law, where the exponent depends on the dimensionality of the medium.

Hazards such as fires and floods may destroy and prevent nodes of a wireless sensor network from functioning properly and reporting measurements of interest. Destructions by hazardous events are systematic and hence different from random node and link failures, which are very common in wireless sensor networks due to the fragility of devices and the intermittent degradation in wireless link quality. An information theoretic approach is presented that may be used for detecting systematic node destructions in dense sensor deployments wherein detection by visual inspection of the destroyed sensors on a "sensor map" at a central monitoring station is hard, if not impossible. It is shown that the information theoretic approach combined with the statistical methodology presented can effectively be used for automated triggering of alarms in decision support systems to rapidly contain hazards. It is also shown that as few as approximately 20 close-proximity sensor failures could be considered "unusual" enough to detect a systematic destruction with 90% confidence in dense sensor deployments, and this is observed to be independent of deployment density. Simulations also indicate that connectivity, which is critical in wireless sensor networks for relaying measured data, is with high probability not compromised before detection takes place.

Internet traffic primarily consists of packets from elastic flows, i.e., Web transfers, file transfers (FTP), and e-mail, whose transfers are mediated via the transmission control protocol. We develop a conditional sampling technique to analyze throughput correlations among elastic flow classes based on flow level measurements from current network traffic monitoring tools. The primary contributions of the paper are: (1) a demonstration of throughput correlation among temporally overlapping flows on congested resources by using analytical/simulation models; (2) application of a multivariate statistical method (principal components) to infer network properties, such as the number of resources shared by flows in the network, from non-intrusive, flow level measurements collected at a single site. Our proposal for using flow level measurements to infer network properties differs significantly from previous network tomography research that has employed end-to-end packet level measurements for making inferences.

We develop methods to infer path or bottleneck sharing among TCP flow classes based on flow level measurements available from the current traffic monitoring tools. Our premise is that flows that temporally overlap on the congested resources have correlated throughputs. We propose to use factor analysis to explore the correlation structure of flow class throughputs in order to hypothesize which flow classes might share congested resources. The effectiveness of this "black box" approach is studied using the empirical data. We show that making such inferences based on flow level statistics is viable in practice, and can serve as an effective, novel tool for network design and configuration decisions. Our work on inferring bottleneck sharing differs significantly from the previous work in that we consider flow level instead of packet level statistics, and hence may potentially influence research in that area. Possible applications of this technique include network monitoring and root cause analysis of poor performance.

Video streaming has already been very popular on the Internet through services such as news bulletins from different parts of the world and on-demand music-video clips. With rapidly developing wireless technologies, video streaming to mobile devices will be very common. We investigate the self-similar scaling behavior that is present in variable bit rate (VBR) MPEG-4 video. As the usage of video services over packet-based wireless networks increases, new workload models will be necessary to study the quality of service aspects of video traffic. A key finding of our study is that MPEG-4 video encoder output traffic has fractal behavior and this behavior exists regardless of the compression ratio.