Peer-To-Peer (P2P) networks are self-organizing, distributed systems, with no centralized authority or infrastructure. Because of the voluntary participation, the availability of resources in a P2P system can be highly variable and unpredictable. In this paper, we use ideas from Game Theory to study the interaction of strategic and rational peers, and propose a differential service-based incentive scheme to improve the system's performance.

In this paper, the effect of spatial diversity on the throughput and reliability of wireless networks is examined. Spatial diversity is realized through multiple independently fading transmit/receive antenna paths in single-user communication and through independently fading links in multiuser communication. Adopting spatial diversity as a central theme, we start by studying its information-theoretic foundations, then we illustrate its benefits across the physical (signal transmission/coding and receiver signal processing) and networking (resource allocation, routing, and applications) layers. Throughout the paper, we discuss engineering intuition and tradeoffs, emphasizing the strong interactions between the various network functionalities.

Historical precedents suggest that the basic issues underlying the debate about network neutrality, dealing with the balance between efficiency and fairness in markets, will never be resolved. Should net neutrality dominate, attention would likely turn to other parts of the economy that might be perceived as choke points for economic activities, such as Net search. Traditionally, the balance between efficiency and fairness that was struck by policy makers depended heavily on cost considerations. When a service was expensive to provide, fairness was deemphasized. In the current discussion of network neutrality, this issue appears to be unduly neglected. 1

Abstract. This paper/presentation aims at introducing the reasons why switching from the current flat-rate Internet pricing to another scheme is required, at briefly classifying the existing propositions, and at highlighting the challenges that still have to be tackled in the area. Pricing has indeed become a hot topic in the networking literature in order to control congestion, differentiate services among users and somehow fairly share the resource, but is still the subject of debate about how, and even if, it should be implemented.

There exists a huge demand for multimedia goods and services in the Internet. Currently available bandwidth speeds can support sale of downloadable content like CDs, e-books etc. as well as services like video-on-demand. The only constraints in the e-content market are availability of resources (server capacity and bandwidth) and consumer demand. In this paper, we develop an analytical framework to price on-demand content based on these constraints. We consider a system where a server handles requests for content on a First-Come-FirstServed (FCFS) basis. We develop a model where customers can refuse the service based on their capacity to pay and their willingness to do so. We formulate the maximum expectation of revenue for such systems as a constrained optimization problem. Since our model is probabilistic and depends on potentially unknown parameter values for customer behavior, we develop an adaptive algorithm that learns these parameters online. We validate our framework using simulations. Our simulations indicate that our algorithm generates revenue close to the maximum expectation. Further, they also indicate that the algorithm is robust to transient customer behavior.

www.jims.cam.ac.uk

We consider the free-rider problem in peer-to-peer file sharing networks such as Napster: that individual users are provided with no incentive for adding value to the network. We examine the design implications of the assumption that users will selfishly act to maximize their own rewards, by constructing a formal game theoretic model of the system and analyzing equilibria of user strategies under several novel payment mechanisms. We support and extend this work with results from experiments with a multi-agent reinforcement learning model.

Mobility and the New Global Economy are accompanied by requirements for dynamism and flexibility, with respect to e-commerce, interorganizational activity, and security.

those of the author(s) and do not necessarily reflect those of the John F. Kennedy

Publication interne n ˚ 1731 — Juillet 2005 — 26 pages Abstract: Third Generation (3G) wireless systems are expected to become more and more popular worldwide, thanks to a higher quality than in current 2G. Direct-sequence code-division multipleaccess (DS-CDMA) is a solution to be implemented in these systems but, due to a limited radio spectrum and more and more demanding applications, it seems likely that congestion will still be a problem. Pricing appears a simple way to tackle this problem. This paper aims at studying the impact of a per-packet static pricing scheme on the use of the reverse-link in a cell where demand (defined through the so-called utility functions) decreases when prices increase or quality of service decreases. The paper also deals with the pricing of multiple classes, since DS-CDMA supports integrated services. In a first step, we determine as a Nash equilibrium the number of customers that will actually apply for service, depending on demand. In a second step, assuming perfect power control, we find the prices and received powers optimizing the service provider’s revenue. We find that in the case where potential demand always exceeds capacity, the base station’s best interest is to favor only one class, but that it is not the case in the more realistic situation when considering potential demand as a random variable over time.