A Balanced Capacity Implementation of DSM for DSL Systems

This paper investigates a fair resource allocation method in crosstalk limited DSL systems. As a measure of fairness, the concept of balanced capacity (BC), introduced in T. Sartenaer et al. (2005), is used. BC corresponds to a point of the boundary of the capacity region where each user experiences the same relative loss with respect to his maximum capacity. Our goal is to propose algorithms to compute the corresponding point in DSL (digital subscriber line) systems where some users suffer from a near-far situation and for constraints on individual power spectrum densities. Results are reported for typical crosstalk limited cases and are compared to a resource allocation optimized for the sum rate

[1]  Marc Moonen,et al.  Iterative spectrum balancing for digital subscriber lines , 2005, IEEE International Conference on Communications, 2005. ICC 2005. 2005.

[2]  John M. Cioffi,et al.  Dynamic spectrum management for next-generation DSL systems , 2002 .

[3]  Wei Yu,et al.  Dual methods for nonconvex spectrum optimization of multicarrier systems , 2006, IEEE Transactions on Communications.

[4]  Luc Vandendorpe,et al.  Balanced capacity of wireline multiuser channels , 2005, IEEE Transactions on Communications.

[5]  Wei Yu,et al.  Optimal multiuser spectrum management for digital subscriber lines , 2004, 2004 IEEE International Conference on Communications (IEEE Cat. No.04CH37577).

[6]  Wei Yu,et al.  Distributed multiuser power control for digital subscriber lines , 2002, IEEE J. Sel. Areas Commun..

[7]  Luc Vandendorpe,et al.  Balanced capacity of wireline multiple access channels with individual power constraints , 2005, SPAWC 2005.

[8]  Marc Moonen,et al.  Optimal Spectrum Management , 2004 .

[9]  Thomas Starr,et al.  DSL Advances , 2003 .