mmKey : Channel-Aware Beam Shaping for Reliable Key Generation in mmWave Wireless Networks

Poorya Mollahosseini, Yasaman Ghasempour · 2025

Read mmKey : Channel-Aware Beam Shaping for Reliable Key Generation in mmWave Wireless Networks by Poorya Mollahosseini, Yasaman Ghasempour (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-19010.txt.

Introduction

Wireless links can mint secret keys from the channel itself, since two partners see the same fading pattern while an eavesdropper sees another. At millimeter-wave frequencies this trick falters: channels are sparse, noisy in phase, and lossy. Mollahosseini and Ghasempour present mmKey, which uses antenna arrays to stir randomness into a quiet channel by shaping the beam.

A genetic algorithm evolves the beam weights step by step. The winning shapes suppress the direct line-of-sight path while respecting the channel's sparseness and signal strength, balancing secrecy against reliability. In simulations, mmKey widened the secrecy gap by about thirty-nine percent over random beamforming and thirty-four percent over null beamforming. The authors pitch the method for securing next-generation wireless networks where keys must be born from the air itself.

Catalog record

CREATORPoorya Mollahosseini, Yasaman Ghasempour
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE32 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-19010

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

mmKey : Channel-Aware Beam Shaping for Reliable Key Generation in mmWave Wireless Networks Thanks: This work was supported by the Center for Ubiquitous Connectivity (CUbiC), sponsored by Semiconductor Research Corporation (SRC) and the Defense Advanced Research Projects Agency (DARPA) under the JUMP 2.0 program, award number HR0011-23-3-0002. Poorya Mollahosseini, Yasaman Ghasempour Abstract Physical-layer key generation (PLKG) has emerged as a promising technique to secure next-generation wireless networks by exploiting the inherent properties of the wireless channel. However, PLKG faces fundamental challenges in the millimeter wave (mmWave) regime due to channel sparsity, higher phase noise, and higher path loss, which undermine both the randomness and reciprocity required for secure key generation. In this paper, we present mmKey , a novel PLKG framework that capitalizes on the availability of multiple antennas at mmWave wireless nodes to inject randomness into an otherwise quasi-static wireless channel. Different from prior works that sacrifice either the secrecy of the key generation or the robustness, mmKey balances these two requirements. In particular, mmKey leverages a genetic algorithm to gradually evolve the initial weight vector population toward configurations that suppress the LOS component while taking into account the channel conditions, specifically, the sparsity and the signal-to-noise ratio (SNR). Extensive simulations show that mmKey improves the secrecy gap by an average of 39.4% over random beamforming and 34.0% over null beamforming, outperforming conventional schemes. I Introduction In recent years, there has been growing interest in adopting cross-layer approaches to secure wireless networks. One promising strategy is to complement traditional symmetric encryption schemes with physical-layer key generation (PLKG) [ 1 ] . PLKG leverages two fundamental properties of the wireless channel that are critical for secret key generation: randomness and reciprocity . Randomness relies on a rich multipath environment to ensure unpredictability, while reciprocity requires accurate channel measurements at the two legitimate parties, Alice and Bob, to guarantee that both observe highly correlated channels and, hence, keys. Unfortunately, PLKG faces fundamentally new challenges in the millimeter wave (mmWave) regime compared to the sub-6 GHz band, which undermine both the randomness and reciprocity principles. mmWave channels are inherently sparse [ 2 ] due to increased reflection loss and reduced diffraction [ 3 ] , making the channel largely a deterministic function of node positions rather than the environment. This line-of-sight (LOS) dominance has two key consequences: (i) for stationary Alice and Bob, the channel becomes static, leading to repetitive and non-random key sequences, and (ii) an adversary, Eve, can accurately estimate the channel simply by observing the LOS path. Furthermore, while the uplink and downlink channels are theoretically reciprocal, impairments such as noise—whether additive, like additive white Gaussian noise (AWGN), or multiplicative, like phase noise—can degrade reciprocity in practice. To mitigate these effects, a high signal-to-noise ratio (SNR) is required, which is challenging to achieve due to the increased path loss at mmWave frequencies. Fig. 1: System Overview. mmKey generates a series of carefully-designed random-like beam patterns for robust and secure PLKG in sparse mmWave channels. To address the challenges of secure PLKG in the high-frequency mmWave regime, we propose mmKey , a novel channel-aware beam-shaping framework that leverages the multi-antenna capabilities of mmWave hardware to intelligently inject randomness into the channel while enabling robust key extraction under high path loss and phase noise. An overview of the system is shown in Fig. 1 . At a high level, dynamically varying the beamforming vector at Alice introduces fluctuations in Bob’s observed channel, which translate into variations in the generated bit sequence. The goal of mmKey is to induce fluctuations in the observed channel at the legitimate receiver by distributing power and phase across different sparse paths in the environment. This is accomplished by designing a sequence of time-varying weight vectors at the transmitter that jointly satisfy two objectives: (i) Robustness —ensuring highly correlated keys between Alice and Bob with minimal mismatches, and (ii) Secrecy —ensuring that the eavesdropper’s key is statistically independent. We consider a strong passive adversary with perfect knowledge of Alice and Bob’s locations (and hence their LOS channel), and extremely high receive sensitivity, enabling detection of arbitrarily weak signals without being exposed. Due to the unique channel characteristics at mmWave frequencies, there exists an inherent tradeoff between robustness and secrecy. The LOS path carries most of the signal power, so steering a highly directive beam toward it enhances robustness by improving key agreement between Alice and Bob. However, this also enables Eve to accurately estimate the effective channel by monitoring LOS variations, compromising secrecy. Alternatively, allocating minimal power to the LOS—e.g., through nulling—improves secrecy by reducing Eve’s channel correlation but severely degrades Bob’s SNR, increasing the key mismatch rate. Furthermore, precise nulling may not be feasible in practice due to the limited phase and amplitude resolution at the transmitter. In contrast, mmKey balances secrecy and robustness by leveraging prior channel knowledge and accounting for limited phase and amplitude resolution at the transmitter. It employs a genetic algorithm (GA) to efficiently explore the large, discrete beam space and maximize the secrecy gap, defined as the difference between Alice–Bob and Bob–Eve key disagreement rates. The GA evolves pseudo-random beams toward LOS-suppressing solutions while maintaining population diversity. A subset of weight vectors is then selected based on the observed channel conditions (e.g., sparsity and SNR), ensuring that the final beam sequence achieves a desirable tradeoff between secrecy and robustness. We perform extensive simulations to evaluate the performance of mmKey against two baseline methods—–random beamforming and null beamforming—–across a wide range of channels. Overall, our results indicate that mmKey offers a flexible and effective solution for enabling secure PLKG in static mmWave environments, achieving an average secrecy gap improvement of 39.4% and 34.0% compared to random beamforming and null beamforming, respectively. II System Model II-A System Architecture and Eavesdropper Model We consider a mmWave wireless system consisting of two legitimate nodes, Alice and Bob, aiming to generate a shared secret key from their reciprocal wireless channel, while an eavesdropper, Eve, attempts to infer the key from overheard signals. Alice is equipped with an analog uniform linear array (ULA) of N antennas with inter-element spacing d , while both Bob and Eve are equipped with a single antenna. mmKey ’s framework can be easily extended to support a multi-antenna Bob. 1 1 1 A multi-antenna eavesdropper may employ more sophisticated key reconstruction strategies depending on her hardware. Extending mmKey to account for such advanced adversarial models is left for future work. We assume that the environment, the nodes, and thus the wireless channel remain static during the key generation process; in other words, the channel coherence time exceeds the duration required to complete key extraction. Consequently, Alice must inject randomness into the channel via her beamforming capabilities—otherwise, the resulting bit sequences would become repetitive and fail to meet randomness requirements. We focus on the downlink scenario, where Alice transmits and Bob receives, though the same principles apply to the uplink. Prior to key generation, Alice and Bob perform an initial beam training phase in which Alice transmits a sequence of directional beams and Bob records the corresponding received powers. From this process, Alice acquires key channel information: the angular location of Bob (i.e., the LOS direction), the highest observed SNR at Bob, and

Read the complete text