of Development Wireless Stations for Distributed Field Operations
T. Andrew Yang, Sadegh Davari
ISTRIBUTED FIELD OPERA tions involve dispersed mobile units operating in a wide geographical area, such as battlefield operations or exploration on the moon. One or more wireless stations may be deployed for effec tive connectivity among the units. In this project, we investigate the security and performance issues of wireless stations in mobile ad hoc networks (MANET), with a focus on the public key management sys tem using certificates. T. Andrew Yang One of the main issues to consider in a certificate-based scheme is the secure distribution of the public keys to all the nodes in the network. The Public Key Infrastructure (PKI) defines methods to handle public key man agement using X.509 certificates. In a wired network, there exists a centralized certificate server which handles the creation, renew al, and revocation of certificates. This is not feasible in mobile ad hoc networks (MANET), which are composed of mobile nodes that may be constantly moving in the geographical area and do not have a fixed infrastructure or centralized management. Besides, because of the dynamic topology of the network, frequent link fail ures may occur, resulting in issues such as re-authentication and timely communication with the certificate server. To devise a secure and effective certificate-based authentica tion scheme for mobile ad hoc networks, our immediate goal is evaluation of existing approaches by running network simula tions to compare their respective strengths and weaknesses. Our contribution so far includes the following: (1) We analyzed the requirements of a secure distributed authentication system for MANETs. (2) We surveyed some of the existing certificatebased authentication mechanisms by analyzing their features, including pros and cons, in the context of distributed authenti cation. (3) We designed a series of scenario-based simulation experiments and metrics to evaluate these features. In order to study the effectiveness of these mechanisms, we proposed a set of realistic for scenarios simulation. Before defining the scenarios, we first needed to define some parame ters: (1) The mobility model represents the realistic movements of nodes in the network. (2) Node Density also varies according to a particular scenario. (3) Traffic rates vary according to the
D
node linkage failures, congestion, and mobility. Having defined the parameters for the scenarios, we further identified the following metrics, on which basis the authentica tion mechanisms can be evaluated: (a) Successful Certification Ratio () measures the ratio of the number of successful certifi cation services (including issuance, NCISS, and renewal, NCREN, respectively) to the total number of requests for such services (NCTotIss and NCTotRen, respectively). (b) Settling time (st) meas ures the initial time taken for all the nodes in the network to be issued valid certificates. (c) Frequency of Certification (fcert) measures the number of certification services per time interval. (d) Average Certification Delay (acd) is measured as the time delay between the certificate service request (CSReq) and the cer tificate service reply (CSRep) averaged over the simulation time. Publications Sadasivam, K., V. Changrani, and T. A. Yang. Evaluation of Ce...