In this paper, we propose two RFID identification and authentication schemes based on the previously proposed triggered hash chain scheme by Henrici and Muller. The schemes are designed to mitigate the shortcomings observed in the triggered hash chain scheme and to ensure privacy preserving identification, tag-reader mutual authentication, as well as forward-privacy in the case of RFID tags that have been compromised. The first scheme uses a challenge-response mechanism to defend against an obvious weakness of the triggered hash chain scheme. The second scheme uses an authenticated monotonic counter to defend against a session linking attack that the first scheme is vulnerable to. We compare the level of security offered by our proposed schemes against other previous schemes and find that the schemes perform well, while keeping within reasonable overheads in terms of computational, storage and communication requirements.

In this paper, we introduce a randomized bit encoding scheme that can strengthen the privacy protection on RFID tags. This scheme is used together with the backward channel protection method proposed by Choi and Roh (2006), which serves to protect the unique identifier of an RFID tag from disclosure to close-range eavesdroppers. Choi and Roh's method faces the 'same-bit' problem, in which some bits of the unique identifier could be disclosed, thereby revealing critical information. Our proposed scheme alleviates the 'same-bit' problem to a negligible level. Furthermore, we propose an enhanced system model that can protect the unique tag identifier from disclosure not only against eavesdroppers, but against unauthorized interrogators as well. A metric based on entropy was defined and used to measure the amount of protection offered by the scheme. A method to construct an optimal randomized n-bit encoding scheme was also described. In addition, theoretical analysis and simulations were conducted, which show that the proposed encoding scheme provides significant improvement (achieving almost twice the entropy) over no encoding.

In this paper, we expose a denial of information attack that is possible due to the misuse of the kill password (specified under the EPC Class-1 Gen-2 standard [1]) in a previously proposed RFID tag-reader mutual authentication scheme [2]. We show how a passive eavesdropper can obtain useful information by monitoring the authentication session involving a target tag and correlating the information received. By repeating the process over a few authentication sessions, the eavesdropper can collect enough information about the kill password to launch a successful attack to kill and disable the tag. From our simulation analysis, we find that the attack can be carried out effectively using only three to five eavesdropped sessions in most cases. In addition, we discuss the implications of this attack and describe a few other weaknesses that we have observed in the scheme.

In this paper, we propose a selective identifier masking scheme to prevent eavesdropping or unauthorized reading of the unique identifier of RFID tags. In our system, we differentiate between tags that require privacy protection (private tags) and tags that do not require protection (public tags). Under an environment whereby private and public tags co-exist, the proposed scheme uses an active RFID device to selectively transmit a secret mask when a tag identifier is reported to a reader. This induces bit collisions between the tag identifier and the mask. An eavesdropper or an unauthorized interrogator would not be able to resolve the collided bits without knowledge of the secret mask. Hence, the tag identifier is protected against disclosure. Simulations were conducted to investigate the overhead incurred under the scheme. In addition, theoretical analyses were carried out to examine the 'same-bit' problem and its effects on the scheme. In general, we find that our proposed scheme is simple, practical and secure under a realistic threat model as compared to other similar approaches.

A lightweight radio frequency identification (RFID) tag-reader mutual authentication scheme was recently proposed as an improvement over the original authentication protocol specified under the EPC Class 1 Generation 2 UHF RFID Protocol Standard (otherwise known as the "EPC Gen2" standard in short). The improved scheme seeks to protect the access password of the RFID tag against exposure to adversaries. In this paper, we show the weakness in this scheme by launching an attack that effectively exposes the access password. Thereafter, we propose some possible fixes to the scheme to protect it against the attack. We also present some experiment results, which show that the fixed schemes provide greater resistance against exposure of the access password. Based on insights gained from this work, we find that designing a secure authentication scheme by relying only on the minimal features available on an RFID tag is an extremely challenging task. Furthermore, we also stress on the need to put a proposed scheme through stringent tests to ascertain its effectiveness and resistance against attacks.

In public key infrastructures (PKIs), cryptographically protected responses are used to validate the status of certificates. The online certificate status protocol (OCSP) [1] is a scheme that was proposed for this purpose. However, OCSP is expensive primarily because it requires online trusted servers. Currently, there exists cheaper online validation schemes wherein validating clients download pre-computed responses from untrusted distribution servers. Examples of such schemes include NOVOMODO [3], certificate revocation tree (CRT) [4], authenticated dictionary (AD) [5] and the recently proposed augmented certificate revocation lists (ACRL) [6]. In this paper, we conduct a detailed performance analysis of these schemes with respect to their computational, network and storage costs and compare them against OCSP. Based on the system model that we adopt, our results show that all these schemes perform better than OCSP.

Providing effective certificate revocation status is an important yet burdensome aspect of PKI. It is widely assumed that certificate revocation lists (CRLs) cannot provide bandwidth-efficient online certificate status. Using our reference delta CRL scheme, we show that this assumption is not true. Clients using reference delta CRLs never download complete CRLs times they construct revocation lists locally. Our scheme performs significantly better than any earlier CRL scheme and has comparable bandwidth performance with respect to OCSP (Myers et al., 1999)

In this paper, we present simulation-based studies of the performance and tradeoffs of cooperative communications in wireless ad-hoc networks. While cooperative schemes can potentially improve throughput performance, this improvement often comes at a price in terms of higher energy consumption and the involvement of additional nodes. As such, we contend that it is worthwhile to consider schemes that adapt to different link conditions such that cooperation is only invoked when necessary. We study these issues quantitatively by looking at the throughput and energy consumption of cooperative and non-cooperative schemes in a 3-node network. In particular, two different modes of cooperation, namely multihop routing and cooperative relaying, are considered. We aim to determine which modes of cooperation work well under a variety of link conditions and how their performances compare against direct transmissions. Our results are uniquely presented through vivid graphical representations, which facilitate visualization and easy comparison between the different transmission modes