Quality of Service in Mobile Ad Hoc Networks

Abstract

Mobile ad hoc networking is a challenging task due to thelack of resources residing in the network as well as the fre-quentchangesinnetworktopology.Althoughmuchresearchhas been directed to supporting quality of service (QoS) inthe Internet and traditional wireless networks, present re-sults are not suitable for mobile ad hoc network (MANET).QoS support for mobile ad hoc networks remains an openproblem, drawing interest from both academia and indus-try under military and commercial sponsorship. MANETshave certain unique characteristics that pose several diffi-culties in provisioning QoS, such as dynamically varyingnetwork topology, lack of precise state information, lack ofcentral control, error-prone shared radio channels, limitedresource availability, hidden terminal problems, and inse-cure media, and little consensus yet exists on which ap-proaches may be optimal. Future MANETs are likely to be“multimode” or heterogeneous in nature. Thus, the routerscomprising a MANET will employ multiple, physical-layerwireless technologies, with each new technology requiring amultiple access (MAC) protocol for supporting QoS. Abovethe MAC layer, forwarding, routing, signaling, and admis-sion control policies are required, and the best combinationofthesepolicieswillchangeastheunderlyinghardwaretech-nology evolves.In response to the above demand for mobile ad hocnetworks, this special issue aims at providing a timely andconcise reference of the current activities and findings in therelevant technical fields, and focuses as well on the state-of-the-art and up-to-date efforts in design, performance anal-ysis, implementation and experimental results for variousQoS issues in MANETs.Webelievethatallofthesepapersnotonlyprovidenovelideas, new analytical models, simulation and experimentalresults, and handful experience in this field, but also simu-late the future research activities in the area of the quality ofserviceformobileadhocnetworks.Abriefsummaryofeachpaper is listed as follows.The first paper by Qi He et al. first identifies two criti-cal issues leading to the TCP performance degradation: (1)unreliable broadcast, since broadcast frames are transmittedwithout the request-to-send and clear-to-send (RTS/CTS)dialog and Data/ACK handshake, so they are vulnerable tothehiddenterminalproblem;and(2)falselinkfailurewhichoccurs when a node cannot successfully transmit data tem-porarily due to medium contention. Secondly, the authorspropose a scheme to use a narrow-bandwidth, out-of-bandbusy-tonechanneltomakereservationforbroadcastandlinkerror detection frames only. The proposed scheme is sim-ple and power efficient, because only the sender needs totransmit two short messages in the busy tone channel beforesending broadcast or link error detection frames in the datachannel. Analytical results show that the proposed schemecan dramatically reduce the collision probability of broad-cast and link error detection frames. Extensive simulationswith different network topologies further demonstrate thatthe proposed scheme can improve TCP throughput by 23%to150%,dependingonusermobility,andeffectivelyenhanceboth short-term and long-term fairness among coexistingTCP flows in multi-hop wireless ad hoc networks.The second paper by Deying Li et al. discusses the en-ergy efficient QoS topology control problem for nonhomo-geneous ad hoc wireless networks. Given a set of nodes withdifferent energy and bandwidth capacities in a plane, andgiven the end-to-end traffic demands and delay bounds be-tween node pairs, the problem is to find a network topol-ogy that can meet the QoS requirements, and the maximumenergy utilization of nodes is minimized. Achieving this ob-jective is vital to the increase of network lifetime. We con-sider two cases of the problem: (1) the traffic demands arenot splittable, and (2) the traffic demands are splittable. For