Llorenç Cerdà-Alabern

Topology patterns of a community network: Guifi.net

This paper presents a measurement study of the topology and its effect on usage of Guifi.net, a large-scale community network. It focuses on the main issues faced by community network and lessons to consider for its future growth in order to preserve its scalability, stability and openness. The results show the network topology as an a typical high density Scale-Free network with critical points of failure and poor gateway selection or placement. In addition we have found paths with a large number of hops i.e. large diameter of the graph, and specifically long paths between leaf nodes and web proxies. The usage analysis using a widespread web proxy service confirms that these topological properties have an impact on the user experience.

Wireless Systems and Mobility in Next Generation Internet

WLAN Characterization.- Performance Analysis of Wireless Multihop Data Networks.- On the Shaping Introduced by IEEE 802.11 Nodes in Long-Range Dependent Traffic.- Vehicular Networks.- Measurements of IEEE 802.11g-Based Ad-Hoc Networks in Motion.- TrafficNet: A L2 Network Architecture for Road-to-Vehicle Communication.- WLAN and Sensor Networks Protocols.- A Protocol Stack for Cooperative Wireless Networks.- Cross Layer Routing and Medium Access Control with Channel Dependant Forwarding in Wireless Ad-Hoc Networks.- An Energy-Efficient Low-Latency Multi-sink MAC Protocol for Alarm-Driven Wireless Sensor Networks.- QoS and Routing in Ad-Hoc Networks.- Adaptive QoS Reservation Scheme for Ad-Hoc Networks.- DiffServ in Ad Hoc Networks.- Analytical Evaluation of the Overhead Generated by a Routing Scheme with Subnets for MANETs.- Heterogeneous Networks.- Framework for Resource Allocation in Heterogeneous Wireless Networks Using Game Theory.- On the Performance of Mobile IP in Wireless LAN Environments.- Network Selection Box: An Implementation of Seamless Communication.- Resource Management in Cellular Networks.- Joint Connection and Packet Level Analysis in W-CDMA Radio Interface.- Trunk Reservation in Multi-service Networks with BPP Traffic.- Feasible Load Regions for Different RRM Strategies for the Enhanced Uplink in UMTS Networks.- TCP in Wireless.- On the Interactions Between TCP Westwood and the AODV Routing Protocol.- Stability and Dynamics of TCP-NCR(DCR) Protocol in Presence of UDP Flows.- Mobility Agents.- Optimization Models for Application Migration to Support Mobile Thin Clients.

Modeling and comparison of candidate selection algorithms in opportunistic routing

Opportunistic Routing (OR) has been investigated in recent years as a way to increase the performance of multihop wireless networks by exploiting its broadcast nature. In contrast to traditional routing, where traffic is sent along pre-determined paths, in OR an ordered set of candidates is selected for each next-hop. Upon each transmission, the candidates coordinate such that the most priority one receiving the packet actually forwards it. Most of the research in OR has been addressed to investigate candidate selection algorithms. In this paper we propose a discrete time Markov chain to assess the improvement that may be achieved using opportunistic routing. We use our model to compare a selected group of candidate selection algorithms that have been proposed in the literature. Our main conclusion is that optimality is obtained at a high computational cost, with a performance gain very similar to that of much simpler but non-optimal algorithms. Therefore, we conclude that fast and simple OR candidate selection algorithms may be preferable in dynamic networks, where the candidates sets are likely to be updated frequently.

On the topology characterization of Guifi.net

In this paper it is analyzed the topology of Guifi.net Wireless Community Network (WCN). WCN have emerged in recent years organized and deployed by the cooperation of its own customers. WCN keep parallelisms with the construction of the Internet, and seems logical to look for common characteristics among them. More specifically, it is investigated whether there are topology patterns of Guifi.net, and to what extend they share the power-law properties that have been found to obey many topology parameters of the Internet. The paper also proposes a topology generator algorithm that it is able to reproduce specific graph properties that have been obtained for Guifi.net.

Experimental evaluation of a wireless community mesh network

Nowadays there are inexpensive WiFi devices that have fostered the deployment of wireless communities. Well known routing protocols used in the Internet do not fit well to time varying characteristic of wireless networks. This has motivated an intensive research on routing protocols for wireless mesh networks. At this time there are a number of mature and stable implementations that are being deployed in production networks. In this paper we focus on the experimental evaluation of a production Wireless Mesh network being deployed in a testbed at Universitat Politècnica de Catalunya (UPC) and a quarter of the city of Barcelona, Spain. To our best knowledge, this is the first paper where a production community wireless network using 802.11an is analyzed.

Candidate selection algorithms in opportunistic routing

Opportunistic Routing (OR) has been investigated in recent years as a way to increase the performance of multi-hop wireless networks by exploiting its broadcast nature. In contrast to traditional routing, where traffic is sent along pre-determined paths, in OR an ordered set of candidates is selected for each next-hop. Upon each transmission, the candidates coordinate such that the most priority one receiving the packet actually forward it. Most of the research in OR has been addressed to investigate candidate selection algorithms. In this paper we compare a selected group of algorithms that have been proposed in the literature. Our main conclusion is that optimality is obtained at a high computational cost, with a performance gain very similar to that of much simpler but non optimal algorithms. Therefore, we conclude that fast and simple OR candidate selection algorithms may be preferable in dynamic networks, where the candidate sets are likely to be updated frequently.

On the performance modeling of opportunistic routing

In this paper we propose a discrete time Markov model to assess the improvement that may be achieved using opportunistic routing. We show that our model can be used to compute the probability distribution and moments of the number of transmissions needed to send a packet from the source to the destination in a variety of scenarios. We use our model to compute some opportunistic routing performance measures in a network with a shadowing radio propagation model with different network topologies, and varying the maximum number of candidates used by the opportunistic routing algorithm. Our model can be easily extended to evaluate other aspects of opportunistic routing not covered in this paper. Therefore, we believe that our model yields a simple, yet flexible and powerful tool to evaluate opportunistic routing protocols.

Adaptive Channel Assignment for Wireless Mesh Networks using Game Theory

Channel assignment has been extensively researched for multi-radio wireless mesh networks, but it is still very challenging when it comes to its implementation. In this paper we propose a semi-dynamic and distributed channel assignment mechanism called SICA that uses game theory. To the best of our knowledge this is the first game formulation of channel assignment which takes the co-channel interference into account. SICA is an interference aware, distributed channel assignment which preserves the network connectivity without relying on a common channel nor central node for coordination between mesh routers. SICA applies an on-line learner algorithm which assumes that nodes do not have perfect information. We have simulated SICA and compared against another interference-aware channel assignment mechanism proposed in the literature called Urban-X. Simulation results show that SICA outperforms Urban-X, even using fewer radio interfaces per node.

Energy Efficient Cloud Service Provisioning: Keeping Data Center Granularity in Perspective

The cost of power and its associated delivery are becoming significant factors in the total expenditure of large-scale data centers. Numerous techniques have been proposed to address the energy efficiency issue in cloud systems. Recently, some efforts have been made to decentralize the cloud via distributing data centers in diverse geographical positions, at different scales. In this paper, we elaborate on the energy effectiveness of service provisioning on different cloud architectures, from a mega-data center to a nano data center, which provides the extreme decentralization in terms of cloud architecture, as well as P2P-clouds or community network clouds. We study the energy consumption through an analytical and simulation framework for video streaming and MapReduce applications.

Utility based channel assignment mechanism for multi radio mesh networks

We address the channel assignment problem in a multi-radio mesh network that involves assigning channels to radio interfaces for eliminating the effect of wireless interference. Due to the insufficient number of frequency channels and available radios per node, interference is still present which limits the available bandwidth on wireless links and eventually decrease the achievable throughput. In this paper we investigate the effect of considering the diverse delivery probability of the wireless links on the channel assignment solutions. We show that it is possible to classify the wireless links and omit some of them to benefit from a more diverse-channel solution. We propose a new channel assignment aiming to minimize the interference over high performance links. Finally a performance study is carried to assess the effectiveness of our proposed algorithm. Evaluations show that the multi-channel network obtained from our proposed algorithm achieves significant improvement in terms of reducing the interference and increasing the network capacity.