On the Payoff Mechanisms in Peer-Assisted Services With Multiple Content Providers: Rationality and Fairness

Abstract:-

This paper studies an incentive structure for cooperationand its stability in peer-assisted services when there exist multiplecontent providers, using a coalition game-theoretic approach.We first consider a generalized coalition structure consisting ofmultiple providers with many assisting peers, where peers assistproviders to reduce the operational cost in content distribution. Todistribute the profit from cost reduction to players (i.e, providersand peers), we then establish a generalized formula for individualpayoffs when a “Shapley-like” payoff mechanism is adopted. Weshow that the grand coalition is unstable, evenwhen the operationalcost functions are concave, which is in sharp contrast to the recentlystudied case of a single provider where the grand coalition isstable.We also show that irrespective of stability of the grand coalition,there always exist coalition structures that are not convergentto the grand coalition under a dynamic among coalition structures.Our results give us an incontestable fact that a provider does nottend to cooperatewith other providers in peer-assisted services andis separated fromthem. Three facets of the noncooperative (selfish)providers are illustrated: 1) underpaid peers; 2) service monopoly;and 3) oscillatory coalition structure. Lastly, we propose a stablepayoff mechanism that improves fairness of profit sharing by regulatingthe selfishness of the players as well as grants the contentproviders a limited right of realistic bargaining. Our study opensmany new questions such as realistic and efficient incentive structuresand the tradeoffs between fairness and individual providers’competition in peer-assisted services.

Architecture:-

Two coalition structures for a dual-provider peer-assisted service

Existing System:

In Existing system, peer-peer send the packets, the irrespective of stabilityof the grand coalition,there always exist coalition structures that are not convergentto the grand coalition under a dynamic among coalition structures.Our results give us an incontestable fact that a provider does not

tend to cooperatewith other providers in peer-assisted services andis separated fromthem. we adopt a Shapley-likepayoff mechanism, called Aumann–Drèze value, irrespectiveof stability of the grand coalition, there always existinitial states that do not converge to the grand coalition.

Disadvantages:

  1. Limited Access Only.
  2. No Secure because No Encode/Decode Operation of Sharing.
  3. No coalescent process.

Proposed System:

In particular, introduced a coalition game modelfor self-organizing agents (e.g., unmanned aerial vehicles) collectingdata from arbitrarily located tasks in wireless networksand proved the stability of the proposed algorithm by usinghedonic preference. First, even though it is fair to pay peers morebecause they become relatively more useful as the number ofpeer-assisted services increases, the content providers will notadmit that peers should receive their fair shares. The providerstend to persist in single-provider coalitions. In the sense ofthe classical stability notion, called “core,” the cooperationwould have been broken even if we had begun with the grandcoalition as the initial condition. We have proposed an alternate, realisticincentive structure in peer-assisted services, called x value,which reflects a tradeoff between fairness and rationality ofindividuals.

Advantages:

  1. Added only a list of PEER ASSIST currently joined to the P2P network.
  2. More secure because Encode/Decode Operation used.
  3. The charge will calculated by the payoff mechanisms.

Module Description:

  1. SEPARATED:

where there exists a fixed partition of peers for each provider; In theseparated case, a candidate payoff scheme isbased on the Shapley value in each disconnected coalition. Inthe coalescent case, the Shapley value is also a candidate payoffscheme after a worth function of the grand coalition is defined,where a reasonable worth function1 can be the total optimalprofit, maximized over all combinations of peer partitions toeach provider. Consequently, the total payoff for the coalescentcase exceeds that for the separated case, unless the two partitionsof both cases are equivalent. Shapley value is defined by afew agreeable axioms, one of which is efficiency,2 meaning thatthe every cent of coalition worth is distributed to players. Sincesmaller worth is shared out among players in the separated case,at least one individual is underpaid as compared to the coalescentcase. Thus, providers and users are recommended to formthe grand coalition and be paid off based on the Shapley values.

  1. COALESCENT:

where each peer is possible to assist any provider. However, it is still questionable whether peers are willing tostay in the grand coalition and thus the consequent Shapleyvalue-based payoff mechanism is desirable in the multiprovidersetting. In this paper, we anatomize incentive structures in peerassistedservices with multiple content providers and focus onstability issues from two different angles: stability at equilibriumof Shapley value and convergence to the equilibrium. Weshow that the Shapley payoff schememay lead to unstable coalitionstructure and propose a different notion of payoff distributionscheme, value, under which peers and providers stay inthe stable coalition as well as better fairness is guaranteed.

  1. INCENTIVE:

Our study opensmany new questions such as realistic and efficient incentive structuresand the tradeoffs between fairness and individual providers’competition in peer-assisted services. One of nice mathematical tools to study incentive compatibilityof peer-assisted services is the coalition game theory thatcovers how payoffs should be distributed and whether such apayoff scheme can be executed by rational individuals or not.In peer-assisted services, the “symbiosis” between providersand peers are sustained when: 1) the offered payoff schemeguarantees fair assessment of players’ contribution under aprovider–peer coalition; and 2) each individual has no incentiveto exit from the coalition. In the coalition game theory, thenotions of Shapley value and the core have been popularly appliedto address 1) and 2), respectively, when the entire playerscooperate, referred to as the grand coalition. A recent paper byMisra et al. [4] demonstrates that the Shapley-value approachis a promising payoff mechanism to provide right incentivesfor cooperation in a single-provider peer-assisted service.

  1. PAYOFF:

Lastly, we propose a stablepayoff mechanism that improves fairness of profit sharing by regulatingthe selfishness of the players as well as grants the contentproviders a limited right of realistic bargaining. payoff scheme can be executed by rational individuals or not. The grand coalition is expected to be the “best” coalition inthe peer-assisted service with multiple providers in that it providesthe highest aggregate payoff.

Algorithms:

Coalition game-theoretic approach

  • Multiple service Provider

 Payoff Mechanism

  • Calculate the cost

Packet size Based Algorithm

  • Transaction

System Configuration:

Hardware Required:

System:Pentium IV 2.4 GHz

Hard Disk:40 GB

Floppy Drive:1.44 MB

Monitor:15 VGA color

Mouse:Logitech

Keyboard:110 Keys enhanced

RAM:512MB

Software Required:

O/S:Windows XP.

Language : C#.Net