Blockage-Aware Pinching-Antenna-Enabled Mobile Edge Computing for Indoor IoT: Joint Antenna Placement, Serving Order, Task Offloading, and Computing Resource Allocation

Indoor millimeter-wave Internet of Things (IoT) terminals are constrained by limited computing capability and energy, while blockage can severely degrade the uplinks required for computation offloading. This paper investigates a pinching-antenna system (PASS)-enabled uplink mobile edge computing (MEC) framework, where a single pinching antenna (PA) is sequentially repositioned along a dielectric waveguide to serve multiple terminals through time-division multiple access (TDMA). Accounting for three-dimensional blockage geometry and PA movement, the PA serving positions, serving order, partial-offloading ratios, and edge computing resources are jointly optimized to minimize a weighted cost of the normalized task latency and the normalized terminal energy consumption under frame-duration, task-deadline, energy-budget, and computing-capability constraints. A two-layer algorithm is developed by combining structured position and order searches with biconvex resource allocation. Simulations show that the proposed design achieves a constraint feasibility rate of 0.96, compared with 0.65 and 0.28 for the fixed- and random-position baselines, and that on jointly feasible realizations it reduces the weighted cost by 13.0% over the multi-resolution grid-search baseline and by 36.5%–45.9% over the two position baselines. The results further show that task-oriented PA positioning outperforms rate-oriented placement and that this gain cannot be compensated for by simply increasing the terminal transmit power.