Work

TDM flexible PON based optical fronthaul networks: fine-grained QoT-aware dynamic bandwidth allocation

Year: 2026
Type: article
Abstract: Time division multiplexing bandwidth-limited flexible passive optical networks (TDM Flex-PON) have emerged as promising optical fronthaul solutions for radio access networks (RANs). Compared with conventional fixed TDM-PON, the TDM Flex-PON provides physical media dependent (PMD)-layer flexible transmission capabilities and quality-of-transmission (QoT) characteristics, introducing what we think is a new degree of freedom for fine-grained QoT-aware dynamic bandwidth allocation (DBA). This motivates the investigation of how such PMD-layer flexibility and QoT characteristics can be effectively exploited by transmission convergence (TC)-layer DBA mechanisms to enhance service performance. Therefore, in this paper, by adopting vehicle-to-infrastructure (V2I) communication as a service type, a TDM Flex-PON fronthaul architecture is first proposed by exploiting the PMD-layer flexibility of Flex-PON for supporting V2I communications. A TDM Flex-PON fronthaul resource allocation problem is then formulated as a mixed-integer nonlinear programming (MINLP) problem, which can be solved to obtain a theoretically optimal solution. To efficiently solve the MINLP problem, a Flex-PON-based QoT-aware cooperative excess DBA algorithm is proposed as a practical heuristic solution. Results show that the proposed DBA algorithm achieves near-optimal performance compared with the MINLP-based solution and outperforms the quarter frame based DBA algorithm, improving network throughput by 4.11× and channel utilization by 15.82% while maintaining a maximum upstream latency of ≤ 234.64 µs under the same Flex-PON PMD-layer conditions. Furthermore, compared with the case without exploiting PMD-layer flexibility, the proposed DBA algorithm achieves up to 60.09% higher network throughput and reduced upstream latency.
Language: English
Open Access status: gold
Estimated APC paid: $2,270
OpenAlex ID: W7223023742
FWCI:
Cites: 16
Cited by:
Related to: 10