Electrical Discharge Machining (EDM) is an indis-pensable non-contact electro-thermal subtractive man-ufacturing process deployed for micro-scale shaping of extreme-hardness, high-strength aerospace and biomed-ical alloys.  In the Industry 4.0 paradigm, integrat-ing Internet-of-Things (IoT) sensing clusters—spanning microsecond-sampled spark gap voltages, discharge cur-rents, high-frequency acoustic emissions, and dielectric fluid dynamics—enables predictive maintenance, closed-loop gap impedance regulation, and workpiece qual-ity verification. However, streaming sensitive teleme-try across cyber-physical edge networks exposes ma-chining workcells to severe attack vectors, including telemetry spoofing, unauthorized tool retraction manip-ulation, proprietary alloy recipe extraction, and mali-cious servo feed tampering. Securing this infrastruc-ture requires cryptographic integrity and confidential-ity under unprecedented constraints: ultra-low microsec-ond latency deadlines to avert catastrophic arcing dam-age (< 50 µs), microcontrollers with severely constrained SRAM and Flash budgets, and an electromagnetically violent environment where high-power spark transients (dv/dt > 10 kV/µs, di/dt > 10⁹ A/s) induce severe ra-diated electromagnetic interference (EMI). In this pa-per, we present an exhaustive empirical performance and security evaluation of five premier National Insti-tute of Standards and Technology (NIST) Lightweight Cryptography (LWC) authenticated encryption with as-sociated data (AEAD) ciphers—Ascon-128, Ascon-128a, TinyJAMBU-128, Photon-Beetle-AEAD, and Xoodyak—alongside GIFT-COFB and the legacy enterprise standard AES-128-GCM. We evaluate these ciphers across physical embedded hardware platforms, in-cluding ARM Cortex-M4 (STM32F446RE @ 180 MHz), ESP32-S3 (Xtensa LX7 @ 240 MHz), Raspberry Pi RP2040 (Cortex-M0+ @ 133 MHz), and an industrial edge gateway (NXP i.MX8M Plus Quad Cortex-A53 @ 1.6 GHz). Our physical testbed couples a CNC sinker EDM apparatus machining Inconel 718 with galvanic iso-lation front-ends, Keysight 1 GHz oscilloscopes, and pre-cision DC power analyzers. We systematically quantify execution latency across payload sizes (16–1024 bytes), cycles-per-byte (cpb) efficiency, dynamic SRAM/Flash footprints, energy dissipation per authenticated byte (nJ/byte), packet error resilience under severe radiated spark EMI, and side-channel vulnerability via Welch’s t-test Test Vector Leakage Assessment (TVLA). Our empirical findings demonstrate that Ascon-128a and Xoodyak achieve outstanding execution latencies of 21.8 µs and 21.2 µs for 64-byte telemetry payloads on Cortex-M4, comfortably satisfying the critical 50 µs anti-arcing deadline with sustained throughputs exceeding 4.25 MB/s.  Furthermore, Ascon-128a dissipates merely 31.9 nJ/byte, reflecting a 73.4% energy reduction com-pared to software AES-128-GCM (120.2 nJ/byte). We demonstrate that AEAD 128-bit authentication tags pro-vide a 100% deterministic rejection guarantee against EMI-corrupted packets under spark currents up to 65 A. Finally, a 2-share Boolean masked Ascon-128 implemen-tation eliminates first-order side-channel leakage (|t| < 4.5 across 100,000 traces), demonstrating that lightweight authenticated encryption provides rigorous, determinis- tic cyber-physical security for mission-critical industrial manufacturing.

 

Keywords: Electrical Discharge Machining (EDM), In-dustrial Internet of Things (IIoT), Lightweight Authen-ticated Encryption (AEAD), NIST LWC, Ascon-128, Cyber-Physical Systems (CPS), Real-Time Deadlines, Electromagnetic Interference (EMI), Side-Channel Anal-ysis.