Premium
Evaluation of channel access protocol performance for a mobile multimedia terminal in a multihop network
Author(s) -
PomalazaRaez Carlos A.,
Alam Mohammad S.
Publication year - 2000
Publication title -
international journal of communication systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.344
H-Index - 49
eISSN - 1099-1131
pISSN - 1074-5351
DOI - 10.1002/1099-1131(200009)13:6<489::aid-dac462>3.0.co;2-t
Subject(s) - computer science , computer network , time division multiple access , channel (broadcasting) , network packet , asynchronous communication , protocol (science) , quality of service , medicine , alternative medicine , pathology
The performance of the channel access protocol is a main factor when designing and evaluating the overall performance of a multimedia and multihop radio network. A search in the current open literature shows almost no results or investigation on this subject. The purpose of this research is to design and evaluate several candidates for the channel access protocol. The main objective is to find one, or more, protocol that can meet or exceed the performance requirements for all informed voice service to be provided by a hand‐held mobile multimedia terminal (MMT) in a multihop network. Extensive simulation models of MMT have been developed using both asynchronous and synchronous protocol modes. The first one accesses the channel using a suitable random policy. The second one reserves time slots or periods in a fashion similar to conventional time division multiple access (TDMA) techniques. Within each mode there are several options involving various levels of implementation complexity. Based on the simulation results, the following recommendations and conclusions are made: (i) use an asynchronous policy that meets the requirements for all informed voice. At least one of the options studied and simulated meets those requirements; (ii) design a convenient mechanism to reserve channel capacity to ensure promised quality of service. This is particularly important for restricting the channel use for data streams on those nodes which are also handling voice packet streams; (iii) design practical methods to estimate channel capacity; (iv) the level of complexity for implementing any synchronous mode protocol greatly outweighs any potential network performance gain. No performance gain was observed with the synchronous mode procedure implemented in this research work; (v) refine the model to closely match the current design specification for the MMT system at the link layer level. Copyright © 2000 John Wiley & Sons, Ltd.