RESCHEDULED Thesis Defence: Performances of Resource Allocation Schemes for Cable Modems in Data Over Cable Service Interface Specification (DOCSIS)

*The date and time for this defence have been changed to the new date and time noted.

Mirza Mursalin Iqbal, supervised by Dr. Md. Jahangir Hossain, will defend their thesis titled “Performances of Resource Allocation Schemes for Cable Modems in Data Over Cable Service Interface Specification (DOCSIS)” in partial fulfillment of the requirements for the degree of Master of Applied Science in Electrical Engineering.

An abstract for Mirza Mursalin Iqbal’s thesis is included below.

Defences are open to all members of the campus community as well as the general public. Please email jahangir.hossain@ubc.ca to receive the Zoom link for this defence.

---

Abstract

Data Over Cable Service Interface Specification (DOCSIS) is a widely adopted standard enabling high-speed data transmission for Cable Modems (CMs) over Hybrid Fiber-Coaxial (HFC) networks. In this system, the Cable Modem Termination System (CMTS) serves as the central node responsible for allocating resource blocks (RBs) at the CMs, which operate under varying traffic loads. Efficient resource allocation (RA) is essential to achieving high data rate and maintaining low backlog levels. However, since the CMTS does not have prior knowledge of future data arrival rates at the CMs, dynamically allocating resources to match real-time traffic demands becomes a complex challenge. Improper allocation of resources, particularly at high data rates can lead to excessive delays and increased backlogs. To address the challenges of scheduling under diverse traffic demands, two novel resource allocation schemes: hybrid RA and proportional fairness (PF) RA are introduced and evaluated against traditional allocation approaches from wireless communication including random, delay-based and backlog-based RA. The proposed schemes are designed to enhance scheduling adaptability in DOCSIS networks, particularly in balancing delay and backlog performance across both real-time and non-real-time applications. The hybrid RA scheme balances delay and backlog via a tunable parameter, while the PF RA scheme adapts allocations based on real-time traffic conditions. The impact of outlier modems-devices with atypical traffic behaviour is also investigated. Simulation results indicate that both schemes offer measurable improvements in delay reduction and backlog control, contributing to more balanced and fair allocation of resources in DOCSIS networks.