Estimation of line properties in the copper access network

The copper access-network operators face the challenge of developing and maintaining cost-effective digital subscriber line (DSL) services that are competitive to other broadband access technologies. The way forward is dictated by the demand of ever increasing data rates on the twisted-pair copper lines. This forces a deployment of the DSL transceivers in cabinets closer to the customers with a joint expansion of the accompanying optical-fiber backhaul network. The equipment of the next generation copper network is therefore becoming more scattered and geographically distributed, which often leads to increased maintenance for the operator. Another challenge for the operators is to maintain high quality of service even when the signal-to-noise-ratio margin is reduced in favor of higher bit-rates, required by new services such as IP-TV and high-definition TV (HD-TV). Moreover, the emerging multi-user techniques for increasing the total throughput in the copper network utilize spectrum management, which relies on accurate information of the usually unknown crosstalk channels in the cable binders. What these challenges have in common is the need of an efficient tool for estimating the properties and the states of the transmission lines. A viable solution is to use the already installed DSL transceivers for line qualification and monitoring, combined with rapid fault detection and localization, in order to prevent unnecessary and costly manual intervention. This thesis addresses estimation of various kinds of line properties based on one-port or two-port measurements. For the purpose of line qualification and fault detection, the focus is on: low-order and causal cable modeling (Paper I), estimation of the characteristic impedance (Paper II), detection and localization of load coils (Paper III), and estimation of line topology (Paper IV). For the application of dynamic spectrum management (DSM), the focus is on: estimation of the crosstalk channels (Paper V) and the impact of estimation errors on the DSM performance (Paper VI). Finally, estimation of the echo transfer function for echo cancellation is addressed (Paper VII). (Less)