Qingsong Cai

Predict and spread: An efficient routing algorithm for opportunistic networking

With their proliferation and increasing capabilities, mobile devices with local wireless interfaces can be organized into opportunistic networks that exploit communication opportunities arising out of the movement of their users. Because the nodes are carried by people, these opportunistic networks can also be viewed as social networks. Unfortunately, existing routing algorithms for opportunistic networks rely on relatively simple mobility models that rarely consider these social network characteristics. In this paper, we propose PreS (Predict and Spread), an efficient routing algorithm for opportunistic networking that employs an adapted Markov chain to model a nodes mobility pattern, and capture its social characteristics. A comparison with state-of-the-art algorithms suggests that PreS can yield better performance in terms of delivery ratio and delivery latency, and approaches the performance of the Epidemic algorithm with lower resource consumption.

Message delivery delay analysis in VANETs with a bidirectional traffic model

A VANET consists of vehicles equipped with on board units (OBUs) that can communicate with each other and the road side base stations. Due to the mobility and sparse distribution of vehicles, the delivery delay of messages in the VANET is mainly caused by the message transmissions between vehicles. The message delivery delay directly impacts the deployment of applications in VANET, and hence, an in-depth study of message delivery delay in the VANET is significant. In this paper, we focused on the investigation of the message delivery delay in the V2V stage with a bidirectional setting. The bidirectional traffic was modeled as a combination of two Poisson point processes. Based on the sub-additive ergodic theory, we found theoretically that the message delivery delay has a linear relationship with the message forwarding distance. Further, the upper bound of the coefficient of the linear relationship has an exponential polynomial relation with the density of vehicles on the road and decreases with the increment of the velocity of the traffic.

Disseminating real-time messages in opportunistic mobile social networks: A ranking perspective

There has been a significant body of work on evaluating node criticality in information networks. However, most of the existing works are developed for static networks and are not applicable to dynamic settings where connectivities among nodes change frequently over time. In this paper, we treat an opportunistic mobile social network as a time-evolving, dynamic graph, and propose a scheme to ascertain the information dissemination capability for each node based on its contact history. In particular, we analyze the node importance in spreading or forwarding real-time messages which are assumed to become less important or even stale over time. To this end, we take a dynamic walk counting approach to calculate all possible temporal-spatial routes associated with each node, by using the down-weighting method. Since the age of a message increases with time, the old walks are discounted to represent the fading influence on the target node. Extensive experiments are conducted based on 4 real-world trace datasets, and the results show that, our analytical result is effective at ranking the node criticality in disseminating or acquiring real-time messages in opportunistic mobile social networks.

Properties of Message Delivery Path in Opportunistic Networks

One of the main challenges in opportunistic networks is how to deliver messages effectively. Mobile nodes have to rely on encounter opportunities to exchange data due to no complete end-to-end path existing in such networks. In this paper, based on in-depth analysis of encounter occurrence process and contact frequency, we find that both of them exhibit unique power-law distributions. The great majority of contacts occurred in short period of time shows that mobile nodes cluster into communities during moving, which indicates the spatial dependency existing among them. The fact that most node pairs only encountered a few times implies that the network connectivity greatly depends on those rare contacts. Using Time Evolving Graph (TEG) theory we analyze the Minimum Delay Path (MDP) for each node pair and find that although there are large number of nodes in networks, the average length of MDP is relative small, which indicates that communities are inherently organized into a hierarchy structure as human society is, and some rare encounters have a significant influence on the average length of MDP as well as the message delivery delay. Our results suggest that decentralized community detection algorithms will achieve optimal message delivery performance with the help of node encounter history information about inter-community.

Segment-Based Adaptive Caching for Streaming Media Delivery on the Internet

Delivering streaming media objects on the Internet consume a significant amount of network resources. Segment-based proxy caching schemes have been an effective solution. In this paper, we investigate the bandwidth consumption of streaming multiple objects from a remote server through a proxy to multiple clients. We propose a caching strategy to realize the suffix segment data adaptively caching at proxy, and integrate it with multicast-based batched patching transmission strategy to develop a proxy—assisted scheme that can efficiently support to deliver media objects in the Internet-like environment. We then introduce a simple transmission cost model to quantify the overall usage of network bandwidth of our framework. Simulation results show that our scheme can reduce the consumption of aggregate transmission cost significantly under the small amount cache size at proxy.

Identifying high dissemination capability nodes in opportunistic social networks

Although social-aware opportunistic networking paradigms are considered to have broad potential applications, so far very little is known about which nodes are more important in both sustaining the network topology and forwarding or disseminating messages. To address this issue, this paper redefines the concept of walk and extends traditional Katz Centrality measurement to dynamic opportunistic social networks. Based on the Time Evolving Graph model, we derive a convenient formula to identify each nodes information dissemination capability through computing the product of the adjacent matrix of each snapshot along the direction of time. The resulting matrix, in which the spatial and temporal dependency of the network nodes are fully captured, can conveniently be used to evaluate each nodes relative dissemination capability. We apply our method to two real experiment trace datasets and the results show that, several mobile nodes with highest communicability identified by our method are more efficient in message dissemination than the others in the whole network. Those nodes can be chosen as good candidates when some interventions, such as accelerating or suppressing the speed of information spreading in network, are required to be made on network.

Quantifying individual communication capability in opportunistic mobile social networks

Conventional methods with measuring information propagation in static networks mainly rely on paths or the shortest path connecting nodes, whereas in opportunistic mobile social networks the existence of a path or the shortest path between nodes cannot be assumed due to the dynamic topological partition nature of the networks. This paper extends the concept of walk to dynamic settings and combines it with the Greens function that originates from statistical physics to quantify how much information flows through each node in the networks changing over time. By means of the time-evolving graph model and the calculation of weighted combinatorial dynamic walks on the graph, a concise theoretic result is derived to account for the relative information propagation capability of each mobile node based on the historical contacts. In addition, the iteration-form result can be conveniently computed at any time point and therefore can be used for predicting the future network behavior when the time interval is appropriately chosen in specific scenarios. Extensive experiments are conducted based on four real trace datasets and the results show that, the formula derived in this paper is very effective at quantifying the information that flows through each mobile node.

Properties of message forwarding paths in social-aware disconnected mobile networks

One of the main challenges in social-aware disconnected mobile networks is how to effectively forward messages in the highly dynamic evolving topology. Most of the previous work rarely considers real traces of the mobile users, and consequently their proposed forwarding schemes cannot work efficiently in real scenarios. In this paper, we systematically analyze the node contact pattern based on the datasets collected from real experiments to study how the messages are delivered from end to end. We find that both the global encounter occurrence and the contact frequency exhibit unique power-law distributions, which implies apparent spatial dependencies among nodes, and the network connectivity greatly depends on some rarely occurring contacts. Using Time Evolving Graph, we analyze the Minimum Delay Paths (MDPs) for each node pair and find that the average length of a MDP is relatively small, even with a large number of nodes in the networks, which indicates that node cliques are inherently organized into a hierarchy structure as our human society is, and some rare encounters have a significant impact on the average length of the MDP and the message delivery delay.

Capturing the Evolving Properties of Disconnected Mobile P2P Networks

The design of message routing algorithms and its performance evaluation in disconnected mobile P2P networks are challenging issues due to its high dynamic evolving topology. To capture the evolution of the connectivity properties of such networks, a Time-Evolving Graph (TEG) model is proposed in this paper. Our intuitive observation is that the connectivity graph of a dynamic network can be obtained from the union of every snapshots observed over a sequential of past time steps. To avoid losing the network connectivity information in all past evolution, each possible edge is weighted by a set of the 2-tuples which consist of the start time and the duration of each contact. We then present the TEG model through using discrete time Markov process to deal with the time dependencies of consecutive time-step indexed network snapshots. As a further simplification, the dynamic of each possible edge is seen as an independent birth-death process. In addition, given the observed sequence of time-step indexed data, the birth and death probability of each possible edge are estimated using Laplaces rule of succession. We note that a TEG eventually converges to an un-uniform random graph in which the birth probability of all possible edge follows a unique power law distribution. The TEG model is validated through the computation of all possible time-evolved end-to-end fastest paths existing in real experimental datasets.

The Minimal Delay Path and Its Evolving Properties in Intermittently Connected Mobile Networks

We study the Minimal Delay Path (MDP) of message delivery and its evolving properties in social opportunistic networks where the connectivity is intermittent and evolving over time. Through in-depth analysis of the public released trace dataset from CRAWDAD community, our results show that the network connectivity is highly depended on some rare nodal contacts that occurred only few times rather than those frequently occurred nodal contacts in whole trace dataset. By constructing the Time Evolving Graph (TEG) and computing its MDPs using our modified version of the Shortest Path algorithm we illustrate how those rare contacts impact on the delay of message delivery in opportunistic network. Our results is in sharp contrast to previous work that choose the most frequent contact node as next hop forwarder (e.g. PROPHET), and we imply that the algorithms identifying those occur less frequently but important nodes as next hop will achieve better message delivery performance.