PASS-Based Multi-User Communications: Capacity Characterization and Configuration Strategy

The fundamental capacity limits of pinching-antenna systems (PASS)-based multi-user communication network are investigated. Two practical pinching-antenna configuration strategies are considered, namely <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">multiple-time discrete activation</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">one-off continuous sliding</i>. For each strategy, the capacity and rate regions are characterized under the non-orthogonal multiple access (NOMA) and orthogonal multiple access (OMA) schemes, respectively. 1) For NOMA, the capacity region achieved by the discrete activation is first characterized. In particular, the ideal case with the infinite number of activation times is considered, where the optimal PA activation and resource allocation scheme is derived. It is shown that different user groups or decoding orders are served via time-sharing. Inspired by this result, an inner bound of capacity region is obtained for the practical case with a finite number of activation times. Then, for the continuous sliding case, the inner bound of capacity region is characterized by alternately optimizing the resource allocation and PAs’ continuous positions. 2) For OMA, the rate region is first obtained for the discrete activation case. It is unveiled that multiple users are successively served. Then, by alternately optimizing the resource allocation and PAs’ continuous positions, the inner bound of rate region is obtained for the continuous sliding case. Numerical results demonstrate that i) PASS can significantly improve the capacity performance compared with the conventional fixed-antenna systems; ii) the capacity gain can be further enhanced by using the proposed PA activation and sliding and resource allocation schemes; and iii) the continuous sliding has the potential to outperform the discrete activation in NOMA, whereas the latter performs better in OMA.