The premise of physical-layer security is to exploit the physical properties of the communication channel to enhance communication security through appropriate coding and signal processing. While the pioneering work on the fundamental limits of physical-layer security can be traced back to the 1970s, there has been a resurgence of interest in the topic over the last 10 – 15 years, motivated in large part by the enormous opportunities created by wireless communications. Researchers have brought in their expertise in multi-antenna signal processing, channel-aware adaptive coding and signaling, as well as networking design, to engineer asymmetries in channel quality between legitimate users and adversaries, which could in turn be exploited to guarantee secrecy at the physical layer. Furthermore, location-specific or device-specific properties of the wireless channels have been exploited to enable secret-key generation and authentication at the physical layer. While much of the early research was essentially focused on analyzing fundamental performance limits of physical-layer security, it is only recently that viable pathways for implementing physical-layer security technologies in practice have been identified. Nevertheless, significant challenges have already become apparent in moving towards practical implementations. These challenges include accurate modeling of the adversary at the physical layer, channel knowledge acquisition, as well as constraints from regulations and standardization.
Full acceptance of physical-layer security by the wider security research community will still require much effort, but physical-layer security has gained some recognition as a technology complementary to cryptography and certainly not as a replacement. More specifically, cryptography guarantees security regardless of the signal reception capability of adversary but under assumptions constraining the adversary’s computational power. On the other hand, the security achieved at the physical layer holds irrespective no matter how large the adversary’s computational power is but holds assuming constraints on the adversary’s capability in receiving and processing the communication signal. Since realistic adversaries may suffer from both finite computational power and limited received signal strength, it is natural to consider a cross-layer integrated approach to security taking advantage of both technologies. This is an exciting research direction with many open problems waiting to be discovered and solved.
In this Best Readings, we highlight overview articles, archival technical papers, as well as special issues on physical-layer security that are representative of physical-layer security today. While the list is far from exhaustive, the selected papers have been influential by identifying new problems, providing innovative and elegant solutions, pointing out important future directions, as well as reflecting on and rethinking the approaches undertaken to study physical-layer security.
Issued April 2018
Xiangyun (Sean) Zhou
The Australian National University
Professor and Canada Research Chair
Georgia Institute of Technology