No abstract

IEEE 802.16j is defining standards for Multihop Relay (MR) networks. With many low cost Relay Stations (RSs), a Base Station (BS) can diminish the radio range but extend the coverage by RSs. Besides, frequency can be reused in different RSs. When using real-time services in an 802.16j MR network, fixed bandwidth is allocated to each intermediate hop along the path between the MS and the MR-BS. The MS does not need to request bandwidth for each packet in each hop. Therefore overhead is eliminated and packet latency is reduced. However, if an MS suffers from bad channel and cannot transmit packets to its access RS correctly, all bandwidths allocated from the access RS to the MR-BS along the multihop path are wasted. This will decrease the system performance. In this paper, we propose a Cooperative Adaptive Partner Selection (CAPS) algorithm to select a relaying partner to help the MSs experiencing bad channel condition. Simulation results show that the proposed technique can improve the system performance significantly.

No abstract

Abstract-It is essential to model channel holding time (CHT), cell residence time (CRT), and handoff rate for performance analysis and algorithm evaluation in mobile cellular networks. The problem has been extensively studied in the past for circuitswitched (CS) cellular networks. However, little research has been done on packet-switched (PS) cellular networks. Unlike that a call occupies a dedicated channel during its whole lifetime in CS networks, an active session in PS networks occupies and releases channels iteratively due to discontinuous reception (DRX) mechanism. In this paper, we investigate the key quantities in PS cellular networks. We present a set of comprehensive new models to characterize the quantities and their relationship in PS networks. The models shed light on the relationship between CHT and CRT and handoff rate. The analytical results enable wide applicability in various scenarios and therefore have important theoretical significance. Moreover, the analytical results provide a quick way to evaluate traffic performance and system design in PS cellular networks without wide deployment, which can save cost and time.Index Terms-Cell residence time (CRT), channel holding time (CHT), handoff rate, mobility, modeling and analysis, packet session.

Abstract-In this paper, we analyze the packet session channel holding time for packet-switched cellular networks. Channel holding time depends on mobility which is characterized by cell residence time in this paper. We apply HyperErlang distribution model to investigate the packet session channel holding time in which a session will experience active and idle periods. In terms of new session, handoff session in busy mode, handoff session with new arrival packets in idle mode and handoff session without arrival packet in idle mode, we demonstrate that the effective channel holding time is exponentially distributed if and only if the cell residence time is exponentially distributed. The analytical results provide a new approach to evaluate traffic performance and system design in packet-switched cellular networks.