THIS ARTICLE HAS BEEN ACCEPTED FOR PUBLICATION IN IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 1 Energy Saving Analysis in Cellular-WLAN Cooperative Scenarios Leonardo Militano, Antonio Iera, Antonella Molinaro, and Francesco Scarcello Abstract—This paper addresses scenarios where cellular mo- bile subscribers implement a cooperative content sharing service by using short-range radio communications in a Wireless Local Area Network (WLAN), and benefit from this situation to reduce the energy consumption. A central entity acts as a coordinator by identifying the data-shares to be downloaded by each cooperating user to guarantee fairness and stability of the cooperative- download process. The intended objective is achieved by carefully modeling the energy consumption on the WLAN radio links and by exploiting the proposed model within the framework of a coalitional game triggered by the coordinator. A performance assessment analysis confirms the expected gains achievable by the cooperative group of users. Index Terms—Wireless Cooperation, Fairness, Energy Con- sumption, WLAN, Game Theory. I. I NTRODUCTION W IRELESS cooperative content sharing [1], [2] is an interesting paradigm gaining momentum in the wire- less communication arena thanks to its easy implementation in modern multi-interface mobile terminals [3]. According to this paradigm, users download portions of data of common interest over long-range cellular links while exchanging the down- loaded portions over short-range radio links. The rewards of cooperation in terms of energy and delay are demonstrated in [4], while the advantages of evolutionary theory and network coding applied to cooperative scenarios are studied in [5]. Expected benefits, which are natural incentives to cooperation, may be in terms of content price, energy consumption, and transfer delay reduction. Notwithstanding, approaching the problem by relying on conventional analytical tools based on linear programming to maximize the performance according to one key parameter only (e.g., energy or delay) could be unsatisfactory or simplistic for some applications. The reason is that such tools distribute service costs disregarding any fairness among cooperating users. On the other hand, even a fair cost distribution could be unstable. The achievement of a fair and stable agreement among cooperating users can be eased by a branch of the Game Theory, called Coalitional Game Theory [6]. In the reference scenario for this paper, nodes have a natural incentive to collaborate, sustained by the inherent benefits guaranteed by the designed cooperative content sharing framework, as also demonstrated in [1]. There, in a cooperative Bluetooth-cellular scenario, nearby users reduce the usage costs of energy-demanding and costly cellular L. Militano, A. Iera, A. Molinaro are with University Mediterranea, DIIES Reggio Calabria, Italy (email: leonardo.militano@unirc.it; anto- nio.iera@unirc.it; antonella.molinaro@unirc.it). F. Scarcello is with University of Calabria, DIMES Cosenza, Italy (email: scarcello@unical.it). links by sharing processing and battery capabilities of their mobile devices. The objective of this paper is to overcome the limitations of the considered scenario (the maximum number of 8 simultaneously active nodes and the impossibility of direct slave-to-slave communications) by considering WLAN as an alternative short-range communication technology for data sharing in the coalition. This requires some research efforts to (i) define a valid energy consumption model in the WLAN as close as possible to a practical implementation while preserving its theoretical value; and (ii) to fully assess cooperation potentialities when exploiting WLAN as the short- range technology. II. THE COOPERATIVE DATA- SHARING FRAMEWORK The proposed framework is applicable to scenarios involv- ing WLANs cooperating nodes in close proximity. Several such reference service scenarios can be identified: groups of students in an aggregation place of a University campus downloading and exchanging lecture notes and other contents; spectators on the stands of a stadium attending a sport or a music event; visitors at tourist attractions receiving additional related information, and so on. In all these cases, users physically close to each other are interested in the same contents and exploit short-range links to cooperatively access the data. For them, we propose a cooperative data-sharing service based on a centralized architecture that relies on a coordinating entity, the Cooperation server. This is responsible of identifying the group(s) of cooperating users, the so-called coalition(s), and orchestrating the cooperation by assigning them complementary fractions of data to download from the cellular network. There are multiple reasons for having a central entity that coordinates the wireless cooperative file-sharing service. The Cooperation Server could be owned by a content provider selling web contents to its subscribers and acting as a promoter of the cooperative service. A service provider gains from this novel data sharing-service because it can widen its subscriber basin, for example by offering group discounts. Subscribers also gain from a controlled cooperation that guarantees impor- tant and fair energy savings. The coordinator also controls that none of the cooperating users is misbehaving (typical of classic P2P systems). Finally, a high-performing server identifies coalitions in a fast and effective way without introducing further overhead and energy costs for the mobile devices (for the server energy consumption is not an issue). In this paper, energy saving is the main objective of each cooperating device; thus, the user utility is defined in terms of the energy amount saved when taking part in the coalition