Deep Learning-Driven Inverse Design of Doherty Power Amplifiers Using Pixelated Combiners and Dual-State Impedance Synthesis

Collection of recent wireless/EM/ML research papers (arXiv 18 Jun 2026) covering: (1) deep-learning-driven inverse design of three-port pixelated Doherty PA combiners with GA and dual-state impedance synthesis — GaN HEMT prototypes show >44.2 dBm output and high efficiency with DPD; (2) CNN+GA pixelated microwave filter synthesis validated by S-parameters and electro‑optical electric‑field measurements; (3) covert multi‑hop communications for heterogeneous networks monitored by multiple passive wardens, including optimal detectors, DEP expressions, KL‑based bounds and a low‑complexity routing/