Security analysis and improvement of a mutual authentication and key agreement solution for wireless sensor networks using chaotic maps

Abstract

A characteristic of wireless sensor networks (WSNs) different from traditional networks is that WSNs are vulnerable to various types of attacks because of their distinctive features, involving distributed and nomadic attribute, wireless transmission medium, and lack of centralized infrastructure of security protection. Recently, Kumari et al presented a mutual authentication and key agreement scheme for WSNs using chaotic maps. Unfortunately, we find that the scheme of Kumari et al cannot resist sensor node capture attack, session‐specific temporary information attack, sensor node impersonation attack, and man‐in‐the‐middle attack. To overcome the security weaknesses in the solution of Kumari et al, this paper introduces a secure and efficient mutual authentication and key agreement scheme for heterogeneous ad hoc WSNs in fully public channel. Consequently, compared with the solution of Kumari et al, while providing relatively higher level of security and more security features, the proposed solution remains a favorable performance on communication overhead, computation overhead, and storage overhead separately.

An improved user authentication protocol for wireless sensor networks

“…In this section, a review of the scheme of Li et al is presented. All the notations used in this paper are given in Table .…”

“…It issues public parameters { p , q , r } that are calculated as r = q t mod p , where p is a large prime number, q is a generator of Z p , and a random number t ∈ Z p −1 is a private key of GWN. This phase is same as predeployment phase of the scheme of Li et al…”

“…Computational cost is defined as the total execution time taken by user, GWN, and SN in login and authentication phases. Table summarizes the comparison of computational cost of proposed scheme along with schemes . Here, the same approximate time is used as in the works of Li et al and Xu and Wu for running one hash function T h ≈0.0004 ms, time for encryption/decryption T c ≈0.1303 ms, time for modular exponentiation T me ≈1.8269 ms, and T fe ≈63.08 ms time for computing fuzzy extractor .…”

“…It has very large processing and battery power. It acts as a central unit for WSN, which is used to connect user node and SN in WSN, whereas SN has less processing and battery power . User node sends request to GWN for data along with its authenticity, and after verifying the authenticity of user, GWN sends the request to SN.…”

“…In 2019, Jabbari and Mohasefi found user impersonation attack and user anonymity problem in the scheme of Lee et al and proposed a key agreement protocol by overcoming these attacks. Recently, Li et al have showed that the scheme of Kumari et al cannot resist to man‐in‐the‐middle attack, SN impersonation attack, and SN capture attack and designed a mutual authentication protocol for WSN. In this paper, the cryptanalysis of the scheme of Li et al has been performed and pointed out that this scheme suffers from offline password guessing attack, user traceability problem, insider attack, replay attack, and device stolen attack.…”