IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, VOL. 56, NO. 1, JANUARY 2007 271 Impact of Mobility on the BER Performance of Ad Hoc Wireless Networks Gianluigi Ferrari, Member, IEEE, and Ozan K. Tonguz, Member, IEEE Abstract—In this paper, the authors quantify the impact of mo- bility on the bit error rate (BER) performance of ad hoc wireless networks. Analytical expressions, relating the BER at the end of a multihop route with the mobility characteristics of the nodes and the switching strategy, are derived on the basis of a rigorous detection-theoretic approach. In particular, two network switching scenarios are considered: 1) opportunistic nonreservation-based switching (ONRBS), where a message flows from source to des- tination by opportunistically choosing the available shortest con- secutive links and 2) reservation-based switching (RBS), where, after the creation of a multihop route from source to destination, the message is “forced” to flow over the reserved links, regardless of their actual lengths. The network performance is evaluated for both an ideal case (without interference) and a realistic case (with interference). The improved robustness against mobility offered by ONRBS, with respect to RBS, is analyzed and quantified. In particular, two node mobility models, known as direction- persistent (DP) and direction-non-persistent (DNP), are consid- ered, and it is shown that DP mobility causes a much more profound degradation in the end-to-end route BER than DNP mobility. This conclusion is more pronounced in ad hoc wireless networks employing RBS. Overall, the results show that if the medium access control (MAC) protocol is not efficient in canceling or mitigating the interference, then the role of the switching/ routing strategy in network performance is quite minor. Index Terms—Ad hoc wireless networks, internode interference (INI), mobility, nonreservation-based switching (NRBS), RBS. I. I NTRODUCTION M ULTIHOP wireless ad hoc and sensor networks have re- cently attracted a lot of attention due to their potential to provide ubiquitous connectivity [1], [2]. In particular, in future generations of ad hoc wireless networks, nodes are likely to be mobile (e.g., car-based ad hoc wireless networks) [1], [3]–[5]. However, maintaining multihop communication routes is a challenge, especially in the case of mobile nodes: the topology is time-varying and, once a route has been established, local route maintenance is necessary in order for that route to con- tinue to work when a link is broken [6]–[9]. In [10], it is shown that designed mobility might be helpful in surveillance sensor networks. Manuscript received March 29, 2005; revised January 3, 2006 and February 17, 2006. This work was supported in part by an ARO Grant to Cylab of Carnegie Mellon University. The review of this paper was coordinated by Dr. W. Zhuang. G. Ferrari is with the Department of Information Engineering, University of Parma, 43100 Parma, Italy (e-mail: gianluigi.ferrari@unipr.it). O. K. Tonguz is with the Electrical and Computer Engineering Depart- ment, Carnegie Mellon University, Pittsburgh, PA 15213-3890 USA (e-mail: tonguz@ece.cmu.edu). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TVT.2006.883790 Recently, the concept of transport capacity has been intro- duced and quantified, via an approach that is information- theoretically inspired, in [11]. The concept of transport capacity, which simultaneously takes into account the amount of information transferred in the network and the distance over which the information is transferred, represents a very useful measure of the maximum sustainable information flow in multihop ad hoc wireless networks. While mobility can, the- oretically speaking, increase the transport capacity (as shown in [12]), the effect of mobility on the performance of practical ad hoc wireless networks is deleterious [13]. In [14], Neely and Modiano consider a network with cell-partitioned structure, where nodes move according to a simplified uniform mobil- ity model, and redundant packets are sent by each source to the corresponding destination through multiple paths. For this scenario, fundamental tradeoffs between network capacity and queuing delay are derived. In [15], the maximum transmission rate and an upper bound on the transmission delay are evaluated in an ad hoc wireless networking scenario with mobile nodes, taking into account the presence of fading. A novel communication-theoretic framework for ad hoc wireless networks has been proposed in [16]–[19], on the basis of a “bottom-up” approach. In particular, the impact of the physical-layer characteristics on the network performance, jointly with the medium-access-control (MAC) protocol em- ployed and the specific switching strategy (either reservation- based switching (RBS) [16], [17] or non-RBS (NRBS) with disjoint multihop routes [18]), is evaluated. While in [16]–[18] a network-communication scenario with static nodes placed at the vertices of a regular square grid is considered, in this paper we extend the proposed frame- work by incorporating the effects of node mobility on the performance of ad hoc wireless networks. Rather than relying heavily on computer simulations, we propose a novel semi- analytical approach for quantifying the impact of mobility. We consider both an ideal network-communication scenario, without internode interference (INI), and a realistic network- communication scenario, where communication is affected by INI. In the latter case, reserve-and-go (RESGO) MAC protocol, originally proposed in [16] and characterized by multihop route reservation and absence of collision-based retransmission in intermediate links, 1 is used. Two possible switching strategies 1 This MAC protocol was incorrectly referred to in [16] and [17] as the Aloha MAC protocol, for its resemblance, in terms of route activation being independent from the activity of other nodes in the network, with the classic Aloha MAC protocol [20]. However, there are significant differences that make the proposed protocol different from the classic Aloha MAC protocol: 1) multihop route reservation and 2) no use of retransmission techniques. 0018-9545/$25.00 © 2007 IEEE