Link reliability analysis in ad hoc networks

Abstract

Mobile Ad-hoc NETworks (MANET) have become more and more popular, during the last ten years. Mobile hosts such as notebook computers are now easily affordable and are becoming quite common in everyday business and personal life. At the same time, network connectivity options for use with mobile hosts have increased dramatically, including support for a growing number of wireless networking products based on radio and infrared. With this type of mobile computing equipment, there is a natural desire and ability to share information between mobile users. MANETs are suitable for situations where infrastructure is either not available, not trusted, or should not be relied on in times of emergency, like in critical mission applications, wherein fault tolerance is of great importance. For wireless (and wire line) networks, the network’s ability to avoid or cope with failure is measured in three ways: reliability, availability and survivability, all of which have long been important areas of research [1]. Because of its importance, in this paper we investigate the link reliability for ad hoc networks, which can be used as important global measure of performances of ad hoc networks. It also can be utilized to improve localized route repairs by the means of connection mean time to failure (MTTF) as a parameter that allows predicting of the link expiration time (LET). Consequently global connection repairs are more seldom necessary, since the route maintenance routine of the routing algorithm could create disjoint bypass routes in advance based on calculated connection mean time between failures. The previous work of ad hoc network reliability includes work on calculating link expiration times (LET) as in [2], where statistical LET in mobile ad hoc networks is reviewed. The authors predict the probability that a link between two nodes exists at time t2, in case the link existed at the starting time t0. They use the random waypoint model as a basic movement model. No forecasts about the duration of uninterrupted link can be made, since the link may cease to exists at time t1 with t0 < t1 < t2. In [3] prediction-based link availability estimation is introduced, while [4] gives a statistical derivation to forecast the average distance when the routing node is within the scope of the two other nodes. With these statistical calculations, the paper investigates the possibility of predictions of the average link expiration times and deviations for different node velocities, independent from the nodes radio transmission ranges and the distances between each other. In [5] a mobility prediction, which allows the creation of new routes in advance for mobile ad hoc networks, is introduced. The routing algorithm determines when a node will shortly leave the proximity of another node and can handoff the connection. Unfortunately this approach requires data from an external global positioning system (GPS) interface to determine the position, the velocity and the moving direction of all nodes plus a synchronized clock in all nodes. Qin and Kunz [6] perform a prediction of link breakage time using the mobile node's signal power strength from the received packets. The source node can perform a pro active route rebuild to avoid disconnection. Sadagopan at al. [7] examine the varying of the statistics of path durations including PDFs with parameters such as mobility model, relative speed, number of hops and radio range. They suggest that at moderate and high velocities the exponential distribution with appropriate parameterizations is a good approximation of the path duration distribution for a range of mobility models. In this paper a link reliability model for two hop ad hoc networks is presented. The main motivation for this work is to use this model for MTTF calculation for a given link, which can be used to forecast the local route repairs. The prediction of link breakage is made on statistical basis and accordingly does not require additional devices like GPS or some special features in the wireless receiver that will enable measurement of the signal strength. The proposed model is validated through series of simulations using NS-2 simulator and application of proposed model is also presented.