Pengyuan Du

A SDN-controlled underwater MAC and routing testbed

Efficient data communication among autonomous under-water vehicles (AUVs) is difficult. Challenges include the long propagation delays arising with acoustic communication solutions, and line-of-sight requirements for optical transceivers. Existing multi-hop routing approaches are not always appropriate due to node mobility. This work presents a centralized approach to network control, exploiting the observation that AUV networks will have a bounded number of nodes. The paper describes a SDN realization of AUV networking, and documents the implementation of a small-scale replica of the system in our testbed, which can be accessed remotely via a web page and SSH. We then demonstrate the functionality of our implementation by evaluating the performances of two existing MAC protocols namely Slotted FAMA [11] and UW-Aloha [13], in a multi-hop, underwater scenario.

Software Defined naval network for satellite communications (SDN-SAT)

We propose Software Defined Networking (SDN) framework to a fleet of naval ships that relies on multiple satellite communication systems for onboard communication. Our solution addresses practical issues in current shipboard naval networks such as sharing and load balancing of multiple satellite communication links as well as overcoming limited bandwidth constraints. To ameliorate link intermittence and outage, we propose Multi-Path Transmission Control Protocol (MPTCP), which improves end-to-end data delivery by creating several subflows under one TCP session. In our SDN framework, each ship is an SDN switch with multiple SATCOM connections. The management and classification of MPTCP subflows are handled by a remote SDN controller. The cooperation between MPTCP and SDN controller leads to an agile, bandwidth efficient, robust naval network. System analysis and numerical evaluation validate the feasibility and efficacy of our SDN-based solution for such a network.

Multipath TCP over LEO satellite networks

Low earth orbit (LEO) satellite networks like Iridium have played a pivotal role in providing ubiquitous network access services to areas without terrestrial infrastructure because of their potential for global coverage and high bandwidth availability. With low orbit and short range as compared to geostationary satellites, LEO satellites are accessible by mobile devices with limited transmission power and small gain antennas. The drawback, however, is that LEO satellites move fast across the sky with average contact time in the order of 10 minutes, thus requiring frequent handover from one satellite to the next. To achieve smooth handover and efficiently utilize constellation capacity, we propose to use Multipath TCP (MPTCP) in LEO systems and maintain parallel, simultaneous connections between terrestrial handpoints via multiple satellites. In this paper, we discuss the feasibility of using MPTCP over LEO satellite networks and propose a framework of MPTCP-Routing design. Then the performance of this protocol is evaluated through simulation. We show that compared to traditional “single-path” TCP, MPTCP significantly improves throughput performance and prevents the interruption of transmission during handover. Furthermore, we show that our MPTCP-Routing interaction is essential for the end-to-end session to quickly recover from handover.

Multipath TCP in SDN-enabled LEO satellite networks

Satellite systems such as Low Earth Orbiting (LEO) networks play an important role in the next generation 5G networks. To facilitate the integration of satellite and terrestrial networks, software-defined networking (SDN) is embraced which brings flexibility, user-customized services and reduces the cost of network configurations. However, it has been long known that communications via LEO satellite network suffer from long delay and frequent ground-satellite handovers, both are problematic for TCP connections. The emergence of Multipath TCP (MPTCP) provides a new solution to these challenges. In this paper, we study the performance of MPTCP over SDN-enabled LEO satellite networks. MPTCP maintains multiple simultaneous subflows in space to increase throughput. In anticipation of handover, MPTCP creates subflows that run in backup mode and shifts traffic smoothly. To support MPTCP, we design an SDN controller that identifies MPTCP subflows attached to the same MPTCP session and splits them to disjoint paths. The SDN architecture centralizes the routing logic, so the system is more scalable and on-board processing is minimized. Simulations are run to evaluate the proposed MPTCP-SDN framework. It is shown that compared to previous solutions, our strategy significantly improves throughput performance and prevents the interruption of transmission during handover.

An Evolutionary Multi-player Game Model for Two-Hop Routing in Delay Tolerant Networks

Delay-tolerant networks (DTN) are sparse mobile ad hoc networks where contemporaneous end-to-end path is typically not available. Therefore, nodes act as relays for each other to enable data delivery. The cooperation among mobile nodes however can be hindered by selfish users. Incentive schemes are inevitably introduced to regulate the behavior of DTN users. The motivation of this paper is to seek conditions under which cooperators can survive, and even prevail in DTN without incentives. We study the formation of cooperation in DTN routing following an Evolutionary Game Theory (EGT) approach. In particular, the two-hop routing protocol is assumed to be adopted which applies to a large class of DTN routing schemes. We first formulate the two-hop DTN routing as a multi-player game. Using the concept of Evolutionarily Stable Strategy (ESS), we show that defection always dominates cooperation when the population is infinite and well-mixed. Recent developments in evolutionary biology reveal that in finite and structured populations such as graphs, cooperation is nevertheless promoted. We derive a sufficient condition for cooperators to be favored over defectors on K-regular graph. Simulation results validate our theoretical finding, and prove that cooperation indeed can prevail in DTN routing games without incentives.

Traffic optimization in software defined naval network for satellite communications

Naval surface fleets of the United States and its allies rely on multiple satellite communication systems (SATCOM) for onboard communication with other entities such as ships, shore nodes and hosts from external networks. Current practice is for an onboard ship router to select a particular SATCOM link for each outgoing traffic flow based on a mission-specific routing policy. In this paper we propose an alternative solution by viewing the multi-SATCOM link utilization task as a traffic engineering and load balancing problem — in particular, as a Multi-Commodity Flow (MCF) optimization problem. We propose using the Flow Deviation Method (FDM) as a network-wide optimal load-balancing solution that maximizes total throughput and minimizes traffic flow delay and jitter. Our approach is equally valid for both UDP and TCP flows. Network-wide global optimization is carried out via a central controller in a Software Defined Networking (SDN) framework. For TCP flows we propose a novel solution that combines the best attributes of Multi-Path TCP, SDN and FDM. Compared to single-path TCP or MPTCP-SDN without FDM-based traffic optimizer, our proposed combined scheme is more efficient in bandwidth utilization, delay/jitter minimization and also robust against jamming and intermittent link failure. Network performance results are validated via Mininet emulation tests.

Demand-driven Cache Allocation Based on Context-aware Collaborative Filtering

Many recent advances of network caching focus on i) more effectively modeling the preferences of a regional user group to different web contents, and ii) reducing the cost of content delivery by storing the most popular contents in regional caches. However, the context under which the users interact with the network system usually causes tremendous variations in a user groups preferences on the contents. To effectively leverage such contextual information for more efficient network caching, we propose a novel mechanism to incorporate context-aware collaborative filtering into demand-driven caching. By differentiating the characterization of user interests based on a priori contexts, our approach seeks to enhance the cache performance with a more dynamic and fine-grained cache allocation process. In particular, our approach is general and adapts to various types of context information. Our evaluation shows that this new approach significantly outperforms previous non-demand-driven caching strategies by offering much higher cached content rate, especially when utilizing the contextual information.

Towards Opportunistic Resource Sharing in Mobile Social Networks: an Evolutionary Game Theoretic Approach

In mobile social networks, the success of resource sharing depends on a high level of cooperations. The motivation of this work is to seek conditions under which cooperation prevails without additional incentive mechanisms such as credit and reputation-based schemes. We apply the Evolutionary Game Theory framework to investigate the formation of cooperation in opportunistic resource sharing. First, we extend the existing Small World In Motion mobility model to preserve real-world localized mobility patterns. On top of the mobility model, a game theoretic model tailored for resource sharing is developed. Preliminary simulation results show that high user cooperation rate emerges when the cost of resource sharing is sufficiently small, even if the Nash Equilibrium of the resource sharing game is non-cooperation. Moreover, we discovered that heterogeneous user mobility patterns promote cooperation.

A Relay Selection Strategy Based on Power-Law and Exponentially Distributed Contacts in DTNs

Delay Tolerant Networks (DTNs) are sparse mobile ad-hoc networks in which there is typically no complete path between the source and destination. Although many routing algorithms for DTNs have been proposed, prior works generally focus on utilizing the delivery probability of network nodes and the social network structure for data forwarding. In this work, we investigate the use of the inter-contact time (ICT) distribution to derive a new metric that selects the next relay node with the least expected minimum delay (EMD) among all possible routes to the destination. We address the case of exponential and power-law ICTs, which are the most popular assumptions for ICTs that have emerged in recent literature. Extensive simulation results based on the Cabspotting and Cambridge Haggle traces show that our proposed metric can achieve up to 21% higher delivery rate and 23% lower delay than existing schemes.

Throughput and Delay Scaling of Cognitive Radio Networks with Heterogeneous Mobile Users

We study the throughput and delay scaling laws of cognitive radio networks (CRN) with mobile primary and secondary users. They operate at the same time, space and share the spectrum. The primary users (PUs) have higher priority to access the spectrum while the secondary users (SUs) should access opportunistically. Furthermore, we consider a unique situation where both PUs and SUs move according to a General Heterogeneous Speed-restricted Mobility (GHSM) model. In this model, we define (h + 1) heterogeneous moving patterns using a universal set T ={T_i|0 ≥ i ≤ h, A_i = n^-i χ₀/h}, where A_i determines the moving area of each pattern. The set of primary (secondary) moving patterns T^(p) (T^(s)) is a randomly and independently selected subset of T. We assign n (n^β, where β > 1) primary (secondary) nodes to each moving pattern, and their initial positions are subject to a poisson point process. In addition, we have |T^(p)|~|T^(s)| = Θ¹(h) = Θ(log n). By proposing a cooperative routing strategy, we fully utilize the mobility heterogeneity of primary and secondary users and achieve near-optimal throughput and delay performance of order Θ(poly log n) when χ₀ ≥ β. In other cases, our transmission scheme shows advantages over [1] in delay performance of the primary network (PN) and over [2] in capacity of the secondary network (SN).