Heejun Roh

Cooperative Bridges: Topology Control in Cooperative Wireless Ad Hoc Networks

Cooperative Communication (CC) is a technology that allows multiple nodes to simultaneously transmit the same data. It can save power and extend transmission coverage. However, prior research work on topology control considers CC only in the aspect of energy saving, not that of coverage extension. We identify the challenges in the development of a centralized topology control scheme, named Cooperative Bridges, which reduces transmission power of nodes as well as increases network connectivity. We observe that CC can bridge (link) disconnected networks. We propose two algorithms that select the most energy efficient neighbor nodes, which assist a source to communicate with a destination node; an optimal method and a greedy heuristic. In addition we consider a distributed version of the proposed topology control scheme. Our findings are substantiated by an extensive simulation study, through which we show that the Cooperative Bridges scheme substantially increases the connectivity while consuming a similar amount of transmission power compared to other existing topology control schemes.

Resource pricing game in geo-distributed clouds

Cloud computing enables larger classes of application service providers to distribute their services to world-wide users in multiple regions without their own private data centers. Heterogeneity and resource limitation of geo-graphically distributed cloud data centers impose application service providers to have incentives to optimize their computing resource usage while guaranteeing some level of quality of service. Recent studies proposed various techniques for optimization of computing resource usage from cloud users (or application service providers) perspective with little consideration of competition. In addition, optimization efforts of application service providers motivate cloud service providers owning multiple geo-distributed clouds to decide their computing resource prices considering their efforts. In this context, we formulate this problem for cloud service providers as a game of resource pricing in geo-distributed clouds. One of the main challenges in this problem is how to model the best responses of application service providers, given resource price information of clouds in non-overlapped regions. We propose a novel concave game to describe the quantity competition among application service providers reducing payment while guaranteeing fair service delay to end users. Furthermore, we optimize the prices of computing resources to converge to the equilibrium. In addition, we show several characteristics of the equilibrium point and discuss their implications to design computing resource markets for geo-distributed clouds.

Topology Control in Cooperative Wireless Ad-Hoc Networks

Topology control is to determine the transmission power of each node so as to maintain network connectivity and consume the minimum transmission power. Cooperative Communication (CC) is a new technology that allows multiple nodes to simultaneously transmit the same data. It can save transmission power and extend transmission coverage. However, prior research work on topology control considers CC only in the aspect of energy saving, not that of coverage extension. We observe that CC can bridge (link) disconnected networks and therefore identify the challenges in the development of a centralized topology control scheme, named shape Cooperative Bridges, which reduces transmission power of nodes as well as increases network connectivity. We propose three algorithms that select energy efficient neighbor nodes, which assist a source node to communicate with a destination node: an optimal method and two greedy heuristics. In addition, we consider a distributed version of the proposed topology control scheme. Our findings are substantiated by an extensive simulation study, through which we show that the shape Cooperative Bridges scheme substantially increases the connectivity with tolerable increase of transmission power compared to other existing topology control schemes, which means that it outperforms in terms of a connectivity-to-power ratio.

Joint flow and virtual machine placement in hybrid cloud data centers

With the advance of virtualization technology, the current generation of cloud data centers contains diverse applications which generate massive inter-rack traffic in a distributed and unpredicted manner. However, since existing network architectures are not suitable to supply enough network capacity, there have been several research trials to improve the network capacity with augmented wireless links. Especially, architectural design and link scheduling of wireless-cum-wired hybrid data center networks are of their main interests. However, the existing approaches for hybrid data center networks with direct wireless links have limited performance improvements, since virtual machines are typically placed with less consideration of traffic locality.To this end, in this paper, we conduct a novel approach to flow and virtual machine placement problems in hybrid data center networks. We first design a threshold-based, wireless link-aware flow placement algorithm with low complexity. To enhance traffic locality, we also suggest a set of virtual machine placement algorithms under the flow placement algorithm. To fully exploit the extra capacity of the wireless links, we propose a new clustering metric for the algorithms. Extensive simulation results in hybrid data center networks with 60 gigahertz wireless links shows that combination of the proposed algorithms achieves better performance compared to baseline algorithms in flow completion time, aggregate throughput, and computational complexity.

Multisource wireless energy harvesting-based medium access control for rechargeable sensors

By collecting data from sensor devices, wireless sensor networks enable consumer product management in indoor environments. However, most off-the-shelf sensor devices are battery-powered and hence hampered by the limitation of battery life. In this context, wireless rechargeable sensor networks (WRSNs) which adopt wireless power transfer (WPT) technique - leveraging electromagnetic waves for sensor energy replenishment - have emerged as a promising scenario for the architecture of self-sustainable and resilient sensor networks. Nevertheless, while the rapid proliferation of studies on discussion WPT and data collection in WRSNs has been witnessed in recent years, their unilateral investigation makes them insufficient to construct high-performance WRSNs. Therefore, this paper firstly explores joint WPT and data collection in a WRSN and accordingly presents a medium access control protocol called FarMac. The WRSN employs multiple sink nodes which can either conduct WPT to or data collection from numerous deployed sensor devices. The sensor devices, on the other hand, fall into two categories: lethargic sensor device which needs energy replenishment from the sink nodes before conveying its data and energetic sensor device which can disseminate its data directly. In this context, FarMac leverages a centralized algorithm to achieve multisource WPT for maximizing the transferred power to a lethargic sensor device. In addition, each lethargic sensor device executes a distributed algorithm to compute its necessary energy harvesting time. Furthermore, FarMac achieves concurrent WPT and data collection through interference cancellation technique. Simulation results demonstrate that FarMac improves network throughput by up to 41% compared with a benchmark approach and guarantees network resilience 1.

Price-based tethering for cooperative networking

Along with the speedy increase of cellular traffic due to new generation mobile devices, cellular network operators have been looking forward to offloading cellular traffic through WiFi networks. However, it is practically hard to satisfy both the operators and users in considering the poor coverage of current WiFi networks. Instead, WiFi tethering, which is a different way to provide WiFi networks, can solve this problem sufficiently and achieve mobile data offloading. This paper proposes a mobile tethering-based cooperative network which can make heterogeneous users gain certain profits after they participated in it. In the first stage of our system, each users cellular traffic demand is allocated to others as a way which can minimize the total cellular traffic cost. According to this optimal resource allocation, the tethering price can be determined in the direction of maximizing total utility of all the participated users. Through the simulation results, we observed the reduction of total cellular traffic cost and found various optimal tethering prices in terms of different traffic demands. Although we made some assumptions in order to simplify our system model, it can be verified that it may also show an excellent performance in practice.

Secondary user games with spectrum leasing market in cooperative cognitive radio networks

Recently various cognitive radio network models are suggested and analyzed. In particular, cooperative cognitive radio network model considers cooperative relaying of secondary users in cognitive radio networks, which utilize the spectrum occupancy efficiently. However, such cooperation may not occur when secondary users have a better alternative, spectrum leasing. Therefore, in this paper, we study the impact of leasing in a network model with game-theoretic analysis. We first show the importance of the problem with a simple two-secondary-user game. Finally we extend our formulation and analysis to explicitly incorporate the effects of lease strength for a n-secondary-user game given the occupancy model of primary users.

A stackelberg game for cooperative cognitive radio network with active SUs

Recently a new class of methods called Cooperative Cognitive Radio Networks (CCRNs) has been proposed with the advance of Cooperative Communication (CC) and Cognitive Radio (CR) technologies. In a CCRN, secondary users can operate as cooperative relays for a primary user in order to obtain a temporal right to dynamic spectrum access. Many recent CR studies focus on the spectrum market including pricing and leasing in economic contexts. However, prior research efforts on CCRN models typically assume passive operations of secondary users, which ignores that active secondary users may not cooperate with the primary user due to the existence of spectrum market. Therefore, in this paper, a CCRN model named YACCRN is formulated as a Stackelberg game and analyzed. We show several interesting characteristics of our CCRN model due to the rational choice of secondary users between cooperation and band leasing. Our analysis shows that the consideration of CC in the future spectrum market design is inevitable.

The Useful Impact of Carrier Aggregation: A Measurement Study in South Korea for Commercial LTE-Advanced Networks

Carrier aggregation (CA) is one of the main features of the long-term evolution (LTE)-Advanced network that was introduced in Third-Generation Partnership Project (3GPP) Release 10 (Rel-10). CA was applied to commercial cellular networks in South Korea in the middle of 2013; however, the performance of CA in commercial networks has not yet been well studied. This article describes how CA technology is applied to a commercial network and how it performs. An extensive field drive test was conducted to compare CA technology performance with that of non-CA technology by using commercial evolved node B (eNB) and user equipment (UE) in a dense urban area and a suburban area. Downlink (DL) CA with two component carriers (CCs) of 30-MHz aggregated bandwidth (BW) was used in the network with one CC of 20-MHz BW at Band 7 and the other with 10-MHz BW at Band 5. The measurement results verified that the maximum DL data rate of CA reached 203 Mb/s, close to the theoretical peak bit rate of 225 Mb/s, and the average DL data rate was 76 Mb/s during the suburban drive test. As a comparison, the maximum DL data rate of single-carrier Band 7 was 141 Mb/s, and the average DL data rate was 51 Mb/s in the same area.

PND: a p-persistent neighbor discovery protocol in wireless networks

In wireless communications research, a number of literature assume that every node knows all of its neighbor nodes. To this end, neighbor discovery research has been conducted, but it still has room for improvement in terms of discovery delay. Furthermore, prior work has overlooked energy efficiency, which is considered as the critical factor in wireless devices or appliances. For better performance with respect to the discovery delay and energy efficiency, we proposed a novel p-persistent-based neighbor discovery protocol and devised a simple and light algorithm estimating the number of neighbor nodes to support the proposed protocol. Our protocol requires a lower delay and a smaller number of messages for the discovery process than the existing protocols. For extensive performance evaluation, we adopted extra comparison targets from other research areas within the same context. Copyright