Abstract. Peer-to-peer (P2P) file sharing systems have become ubiquitous and at present the BitTorrent (BT) based P2P systems are very popular and successful. It has been argued that this is mostly due to the Tit-For-Tat (TFT) strategy used in BT [1] that discourages free-ride behavior. However, Hale and Patarin [2] identify the weakness of TFT and hypothesize that it is possible to use multiple identities to cheat. To test this hypothesis we modify the official BT source code to allow the creation of multiple processes by one BT client. They use different identities to download the same file cooperatively. We experiment with several piece selection and sharing algorithms and show that BT is fairly robust to the exploitation of multiple identities except for one case. In most cases, the use of multiple identities does not provide siginificant speedup consistently. Interestingly, clients with multiple identities are still punished if they do not maintain a comparable upload rate with other normal clients. We attribute this to the robust way that the TitFor-Tat policy works. From our experiments we observe that the BT protocol is rather resilient to exploits using multiple identities and it encourages self-regulation among BT clients.

No abstract

A novel voice packet scheduling scheme named Modified Adaptive Priority Queuing (MAPQ) is proposed for CDMA downlinks. Closed-loop power control is employed to closely track fast fading, and provides power inputs for MAPQ. Estimation error is considered when measuring received pilots at mobile stations. An adaptive priority profile is defined in the scheme based on queuing delay and required transmission power, which borrows the idea of composite metric from wired systems. This definition ensures system capacity improvement, packet dropping rate reduction, and fairness. Users are allocated resources according to their priorities in a modified PQ fashion constrained by total power budget of base stations. Numerical results show distinct performance gains of the proposed scheme, comparing to the reference scheme.

We propose an adaptive prioritizing soft handoff algorithm for concurrent handoff requests aiming at a same cell. A predicted set, an adaptive priority profile jointly exploiting the impact of required handoff power and call holding time have been developed to realize the proposed algorithm. A link-layer scheduler residing in each base station to ensure the desired operation of the prioritizing procedure is also designed. Numerical results are acquired through comparing the proposed algorithm with no-prioritizing scenario and performance gain is obtained in terms of handoff dropping probability, average guard power utilization, and average guard power efficiency, by no less than 24%, 5%, respectively for the first two criteria, and some amount for the last.

No abstract