A performance analysis of wireless ATM multiplexing system

by Zhang, Xiaomei

Performance Analysis of Wireless ATM Multiplexing System Xiaomei Zhang Wireless ATM is a direct result of the success of ATM on wired networks. Applying ATM over wireless channels is challenging since ATM was originally designed for band-width rich and nearly error-free medium, while in wireless environrnent, the radio channel is both bandwidth limited and the bit-error-rate is timevarying. It is therefore important to obtain good understanding of the statistical ATM multiplexing of the aggregate traffic on wireless channels. ln this thesis, we study the effects of channel error characteristics on a wireless ATM rnultipiexing system. The main characteristic of such a channel is that its error behavior is time-varying, with periods of low error rate (Good) transmission alternating with periods of high error rates (Bad) transmission. This variability is modeled as a two-state Markov chain with two states corresponding to Bad and Good states, respectively. The channel is modeled as being synchronous with the basic time unit being the slot. The transitions from Bad to Good states, and vice versa occur at the slot boundaries. We assume that during a Good state, the channel transmits a packet, while in a Bad state no packet is transmitted. The arriva1process is modeled as the superposition of independent binary Markov sources, each source alternates between On and Off periods. In the source Off state, no data is ernitted while, in the On state, the source generates a random number of packets in a dot each of which fits into a channel slot. We present a discrete-time queuing analysis of such a system. Based on imbedded Markov chah analysis, a functional equation relating the joint probability generating function (PGF) of the system behiveen two consecutive dots under the assumption of an infinite buffer is presented. The functional equation is transformed into a suitable form which makes it possible to derive the steady-state expressions for the marginal PGFs of the queue length, the state of the channel and the number of On sources. Then some performance measures such as mean queue length and mean packet delay are calculated. The relationshipbetween the autocovariance of the channel state and the performance is also demonstrated. Finally, some numerical results are presented to show the effects of the wireless channel error characteristics 0.n the behavior of the ATM multiplexing system.