Abstract Wiretap coding, evolving in parallel with cryptography for nearly 50 years, focuses on secure transmission under the assumption that the wiretap channel is no less noisy than the main channel. Most provably secure schemes rely on information-theoretic security but often achieve limited practical rates. This paper proposes a framework for computationally secure modular wiretap coding. We integrate error correction encoding with the channel transition process to define the wiretap channel function, which consists of an invertible function and a lossy function. Secure encoding is then modeled as a computational entropy extractor. A detailed analysis of the lossy function for symmetric wiretap channels is presented. To leverage this lossiness, we design two computational extractors: the invertible fooling extractor (IFE) and the compressed randomness extractor (CRE). For practical implementation, we demonstrate that a 4-round optimal asymmetric encryption padding serves as an IFE in the random oracle model. Experimental comparisons show that our scheme achieves approximately 3 times and 2.7 times the code rates of the Invert-then-Encode and code-based schemes—classical information-theoretic schemes—under equivalent channel conditions. By instantiating IFE and CRE with hash algorithms such as SHAKE-128/256, we develop a practical wiretap coding scheme that achieves high rates with reasonable computational overhead.

In physical-layer key generation (PLKG), researchers often assess key security using randomness tests or secret key capacity analysis. However, randomness tests can be misleading, as low-entropy inputs may still pass when obscured by hash functions or extractors. While key capacity captures the theoretical upper limit of secret bits from raw channels, it overlooks the impact of practical preprocessing. Recent studies have explored min-entropy as a more practical security metric. However, commonly used min-entropy estimators, such as those in NIST SP 800-90B (90B), may yield inaccurate results for complex sources, even overestimating entropy and thereby undermining security. Their applicability to PLKG also remains unverified. To address the above research gap, this paper focuses on modeling the received signal envelope, a common feature extracted for key generation. Leveraging typical wireless fading scenarios considered in key capacity analysis, we adopt the more general Weibull distribution to characterize the envelope statistics, which generalizes the Rayleigh model and better captures diverse propagation conditions. We first derive an explicit expression for the min-entropy in the absence of preprocessing, and then systematically analyze the impact of typical preprocessing operations on the statistical properties of the observed channel. Building on this foundation, we further establish theoretical upper and lower bounds on the min-entropy under representative preprocessing strategies. We further evaluate all 90B entropy estimators using both simulations and real-world channel measurements, comparing their estimates with our analytical results. While some estimators work well on raw data, they often fail or overestimate entropy after preprocessing, compromising security. In contrast, our conservative lower bound remains robust across all tested scenarios and preprocessing conditions. It offers a reliable fallback when standard estimators break down, providing a stronger theoretical foundation for secure key extraction in PLKG systems.