location: Current position: Home >> Scientific Research >> Paper Publications

On the using of discrete wavelet transform for physical layer key generation

Hits:

Indexed by:期刊论文

Date of Publication:2017-09-01

Journal:AD HOC NETWORKS

Included Journals:Scopus、SCIE、EI

Volume:64

Page Number:22-31

ISSN No.:1570-8705

Key Words:Key generation; Wireless channel reciprocity; Discrete wavelet transform; Gray code

Abstract:For key generation between wireless transceivers, key generation leveraging channel reciprocity is a promising alternative to public key cryptography. Several existing schemes have validated its feasibility in real environments. However, in some scenarios, channel measurements collected by the involved transceivers are highly correlated but not identical, i.e., measurement sequences of these transceivers have too many discrepancies, which makes it difficult to extract the shared key from these measurements. In this paper, we propose a scheme to achieve secret key generation from wireless channels. During the proposed scheme, to reduce the amount of the referred discrepancies and further achieve efficient key generation, the involved transceivers separately apply a compressor based on the discrete wavelet transform (DWT) to pre-process their measurements. Then, multi-level quantization is implemented to quantify the output of DWT-based compressor. An encoding scheme based on gray code is employed to establish bit sequence and ensure that the resulting bit mismatch rate can be further reduced so that efficient information reconciliation can be implemented. Accordingly, the shared key between these transceivers can be derived after information reconciliation. Finally, 2-universal hash functions are used to guarantee the randomness of the shared secret key. Several experiments in real environments are conducted to validate the proposed scheme. The results demonstrate that the proposed scheme is available to generate shared secret keys between transceivers even though their measurement sequences have too many discrepancies. (C) 2017 Elsevier B.V. All rights reserved.

Pre One:LIAP: A local identity-based anonymous message authentication protocol in VANETs

Next One:基于文本分类方法识别《史记》的伪作