Mehdi Mohseni

Optimal Resource Allocation for OFDMA Downlink Systems

This paper proposes efficient rate and power allocation algorithms for OFDMA downlink systems where each tone is taken by at most one user. Weighted sum rate maximization (WSRmax) and weighted sum power minimization (WSPmin) problems are considered. Since these resource allocation problems are non-convex, complexity of finding the optimal solutions is prohibitively high. This paper employs the Lagrange dual decomposition method to efficiently solve both optimization problems. Because of their non-convex nature, there is no guarantee for the solution obtained by the dual decomposition method to be optimal. However, it is shown that with practical number of tones, the duality gap is virtually zero and the optimal solutions can be efficiently obtained

Optimized transmission for fading multiple-access and broadcast channels with multiple antennas

In mobile wireless networks, dynamic allocation of resources such as transmit powers, bit-rates, and antenna beams based on the channel state information of mobile users is known to be the general strategy to explore the time-varying nature of the mobile environment. This paper looks at the problem of optimal resource allocation in wireless networks from different information-theoretic points of view and under the assumption that the channel state is completely known at the transmitter and the receiver. In particular, the fading multiple-access channel (MAC) and the fading broadcast channel (BC) with additive Gaussian noise and multiple transmit and receive antennas are focused. The fading MAC is considered first and a complete characterization of its capacity region and power region are provided under various power and rate constraints. The derived results can be considered as nontrivial extensions of the work done by Tse and Hanly from the case of single transmit and receive antenna to the more general scenario with multiple transmit and receive antennas. Efficient numerical algorithms are proposed, which demonstrate the usefulness of the convex optimization techniques in characterizing the capacity and power regions. Analogous results are also obtained for the fading BC thanks to the duality theory between the Gaussian MAC and the Gaussian BC

The ITU-T's new g.vector standard proliferates 100 mb/s dsl

This article explores the recently issued ITUT G.vector (G.993.5) that allows expanded use of 100 Mb/s DSL. A tutorial description on G.vectors crosstalk noise reduction methods leads to specific projections and measurements of expanded DSL 100 Mb/s reach. A discussion on dynamic maintenance to enhance G.vectors practical application then concludes this article.

On the capacity of AWGN relay channels with linear relaying functions

This paper describes the capacity of frequency-division AWGN relay channels with linear relaying functions. A sequence of nonconvex optimization problems solving are also described in this paper

A Proof of the Converse for the Capacity of Gaussian MIMO Broadcast Channels

The paper provides a proof of the converse for the capacity region of the Gaussian MIMO broadcast channel under total average transmit power constraint. The proof uses several ideas from earlier works on the problem including the recent converse proof by Weingarten, Steinberg and Shamai. First the duality between Gaussian multiple access and broadcast channels is employed to show that every point on the boundary of the dirty paper coding region can be represented as the optimal solution to a convex optimization problem. Using the optimality conditions for this convex problem, a degraded broadcast channel is constructed for each point. It is then shown that the capacity region for this degraded broadcast channel contains the capacity region of the original channel. Moreover, the same point lies on the boundary of the dirty paper coding region for this degraded channel. Finally, the standard entropy power inequality is used to show that this point lies on the boundary of the capacity region of the degraded channel as well and consequently it is on the boundary of the capacity region of the original channel

Dynamic spectrum management for mixtures of vectored and non-vectored DSL systems

A number of Vectored DSL system prototypes have confirmed the substantial performance gains from Far-End Crosstalk (FEXT) cancellation. With field trials and initial deployments expected to materialize in the next two years, a very important question is how such Vectored DSL systems can coexist with non-vectored DSL systems sharing the same cable. This paper shows that the performance gains for Vectored DSL systems can be maintained in a mixture of vectored and non-vectored lines if a Spectrum Management Center (SMC) is assigned to control the impact of crosstalk from the non-vectored to the vectored lines. Simulation results are presented for certain scenarios, first, to illustrate that an Iterative Waterfilling strategy can achieve performance near the optimal, and second, to explore the trade-off between limiting the rates of the non-vectored lines and boosting the rates of the Vectored DSL systems.

Distributed Crosstalk Management for Upstream VDSL Using Dynamic Power Control

The severe interference from neighbor copper lines, commonly known as crosstalk, is a well-known limitation that can reduce the upstream rate of a victim user by 50% or more in dense VDSL (Very high bit-rate Digital Subscriber Lines) deployment. This problem has been partially addressed by the use of UPBO (Upstream Power BackOff). UPBO blindly controls transmit PSD (Power Spectral Densities) based on the pre-defined parameters and channel measurements without considering each users target data rate. Although UPBO can provide fair protection against crosstalk, it still leaves room for practical improvement. This paper proposes a distributed algorithm that dynamically controls transmit power according to each lines need and capacity. The algorithm effectively improves data rates and line reaches in real VDSL deployment.

Compatibility of vectored and non-vectored VDSL2

Vectoring cancels the crosstalk between multiple VDSL2 lines and can greatly improve performance, particularly on short loops. Crosstalk cancellation can only be performed on lines within a vector group and vectored lines may experience uncancelled crosstalk from nearby lines that are non-vectored or in a separate vector group. This paper shows that a substantial consensus exists on the possibility of mitigating the impact of this uncancelled crosstalk on the vectored lines. Among the possible mitigation techniques, we show here that very good levels of compatibility between vectored and non-vectored VDSL2 can be achieved with Dynamic Spectrum Management (DSM]).

Sub-optimal throughput maximization schemes for Gaussian vector broadcast channels

This paper considers two sub-optimal transmission schemes for a family of parallel Gaussian vector broadcast channels. One of the schemes is based on the QR precoding of Ginis et al. (2000). In QR precoding, the maximum achievable throughput depends on the order in which users are encoded. This scheme is a new algorithm for obtaining the best user ordering and channel-input covariance matrix that maximizes the total channel throughput. The proposed algorithm has linear complexity in the number of multi-carrier frequencies. The simplicity of a linear transmitter-and-receiver architecture is attractive for a transmission scheme. The design of a linear transmitter and multiple linear receivers that maximize the total throughput is studied. It is shown that under discrete bit loading assumption for each user, this problem reduces to solving a series of second order cone programming problems.

Signal processing for G.fast

The first version of G.fast occupies frequencies up to 106 MHz and uses relatively simple linear channel-inverting matrix precoders to cancel the crosstalk in a multi-pair telephone cable. However, to truly deliver Gbps+ data rates to most customers, even higher frequencies should be used; where the crosstalk can be stronger than the signal, and so more advanced non-linear signal processing is envisioned. This paper derives optimum signal processing structures for G.fast crosstalk cancelling precoders, and compares approximate and optimum linear and non-linear structures for G.fast running up to 212 MHz. The optimum linear and non-linear precoders are derived. Simulations compare the performance of optimum and QR-decomposition based non-linear precoders, and find that the QR-decomposition with Tomlinson-Harishima precoding is practically optimal. An algorithm is shown for maximizing the minimum bit rate for line ordering of nonlinear precoders; this algorithm is shown to have max-min performance near the optimum.