This paper deals with the assessment of induced specific absorption rate (SAR) in various human models under different exposure scenarios, including both laboratory measurements and simulations. Firstly, SAR values were measured in a standardized SAR laboratory using a phantom for two radiofrequency electromagnetic field (RF-EMF) sources at 900 MHz and 1800 MHz. These laboratory measurements served as a reference for SAR calculations conducted on a specific anthropomorphic mannequin (SAM) using a computer simulation technology (CST) program, thus enabling the determination of antenna location and excitation signal levels for further evaluation. Subsequently, simulations were carried out with CST to evaluate average SAR for the head and for specific head tissues such as the brain, muscles, and fat. Realistic computational human models were also used alongside SAM in CST to explore the influence of gender, age, and tissue type on SAR. Various power levels representing low, moderate, and high RF-EMF exposure were applied to the human models to compare against basic restrictions and reference levels. The simulation results indicate significantly higher SAR values calculated for 1800 MHz compared with 900 MHz. The ratio of the highest SAR values at 1800 MHz to 900 MHz is approximately 1.70 for a baby, 2.59 for a child, and 2.84 for both adult female and adult male. While the SAR values for the brain, fat, muscle, and head are comparable at 900 MHz for the baby, the brain’s SAR value at 1800 MHz stands out significantly from the other tissues. In contrast with the baby, the difference in SAR values between 900 MHz and 1800 MHz is more pronounced for the child, adult female and adult male. The lowest SAR values at 900 MHz and 1800 MHz were obtained for brain tissue in all human models, while the head has the highest SAR value. The maximum SAR change ratio between the brain and the head is calculated to be 4.44 for the male at 1800 MHz. The results reveal that, although the applied electromagnetic field levels were below reference levels for general public local exposure, some local SAR values exceeded the International Commission of Non-Ionizing Radiation Protection’s basic restriction for the general public at certain power levels, particularly at 1800 MHz. The SAR analysis derived from this study is significant in understanding the impact of wireless technologies on health, establishing safety standards, guiding technology advancement, conducting risk assessments, and increasing public awareness.

The rapid expansion of sub-6 GHz 5G and Internet of Things (IoT) networks demands highly miniaturized Multiple-Input Multiple-Output (MIMO) antennas. However, balancing extreme physical compactness with rigorous inter-port isolation introduces severe computational bottlenecks for conventional optimization algorithms. To overcome these multidimensional challenges, this paper proposes a novel Deep Surrogate Active Learning framework for the autonomous design and empirical validation of an ultra-compact dual-band MIMO antenna. By using a surrogate-assisted closed-loop strategy to reduce reliance on repeated full-wave evaluations, the methodology combined a custom-penalized Deep Neural Network with dynamic boundary reduction. After training the initial surrogate model with 440 valid full-wave responses obtained from the offline design-of-experiments (DOE) stage, the best CST-validated candidate was identified at the 83rd active learning cycle. The optimized nested-loop geometry, incorporating a partial defected ground structure (DGS), occupies an extremely confined footprint of only 1634 mm² on a Rogers RO4350B substrate (Rogers Corporation, Chandler, AZ, USA). The selected geometry provided simulated −10 dB impedance bands of 3.35–3.88 GHz and 4.34–5.05 GHz, while the complete two-port model maintained inter-port isolation better than 13.8 dB and 14.9 dB across the lower and upper target passbands, respectively. Measurements of the fabricated prototype showed the intended dual-band behavior, a maximum measured gain of 4.54 dBi, and total radiation efficiencies of approximately 51–63% across both ports at the evaluated frequencies. The simulated Envelope Correlation Coefficient (ECC) remained below 0.035 across the target passbands, supporting the suitability of the compact geometry for the investigated sub-6 GHz MIMO bands.