We consider the problem of information-theoretic secrecy in identification schemes rather than transmission schemes. In identification, large identities are encoded into small challenges sent with the sole goal of allowing at the receiver reliable verification of whether the challenge could have been generated by a (possibly different) identity of his choice. One of the reasons to consider identification is that it trades decoding for an exponentially larger rate, however this may come with such encoding complexity and latency that it can render this advantage unusable. Identification still bears one unique advantage over transmission in that practical implementation of information-theoretic secrecy becomes possible, even considering that the information-theoretic secrecy definition needed in identification is that of semantic secrecy. Here, we implement a family of encryption schemes, recently shown to achieve semantic-secrecy capacity, and apply it to a recently-studied family of identification codes, confirming that, indeed, adding secrecy to identification comes at essentially no cost. While this is still within the one-way communication scenario, it is a necessary step into implementing semantic secrecy with two-way communication, where the information-theoretic assumptions are more realistic.

翻译：我们考虑识别方案而非传输方案中的信息论安全问题。在识别过程中，大量身份被编码成简短挑战信号，其唯一目标是让接收方可靠验证该挑战是否可能由其选定的（可能不同的）身份生成。采用识别方案的理由之一在于，它通过解码换取指数级增长的速率，但这可能带来过高的编码复杂度和延迟，导致该优势无法实际应用。然而，识别在传输方案中仍具有独特优势：即便需要采用语义保密这一信息论安全定义，识别仍可实现信息论安全性的实际部署。本文实现了一组近期被证明能达到语义保密容量的加密方案，并将其应用于一类新近研究的识别码，结果证实：为识别添加安全性确实几乎不增加额外代价。虽然这仍属于单向通信场景，但它是实现双向通信语义保密的必要步骤，因为后者所依据的信息论假设更为现实。