Shravan Mohan

Krishna Chaitanya Kosaraju, Shravan Mohan, Ramkrishna Pasumarthy

The aim of this paper is to study the convergence of the primal-dual dynamics pertaining to Support Vector Machines (SVM). The optimization routine, used for determining an SVM for classification, is first formulated as a dynamical system. The dynamical system is constructed such that its equilibrium point is the solution to the SVM optimization problem. It is then shown, using passivity theory, that the dynamical system is global asymptotically stable. In other words, the dynamical system converges onto the optimal solution asymptotically, irrespective of the initial condition. Simulations and computations are provided for corroboration.

Shravan Mohan, Bharath Bhikkaji

This paper considers the problem of designing a multilevel pulse width modulated waveform (PWM) with a prescribed harmonic content. Multilevel PWM design plays a major role in many diverse engineering disciplines. In power electronics, multilevel PWM design corresponds to determining the inverter switching times and levels for selective harmonic elimination and harmonic compensation. In mechatronics, the same design corresponds to shaping input signals to damp residual vibrations in flexible structures. More generally, in most applications, the aim of PWM design is to minimize the total harmonic distortion while adhering to a prescribed harmonic content. The solution approach presented in this paper is based on linear programming with the objective of minimizing the total harmonic distortion. This objective is achieved within an arbitrarily small bound of the optimal solution. In addition, the linear programming formulation makes the design of such switching waveforms computationally tractable and efficient. Simulations are provided for corroboration.

Shravan Mohan

This preliminary note presents a heuristic for determining rank constrained solutions to linear matrix equations (LME). The method proposed here is based on minimizing a non-convex quadratic functional, which will hence-forth be termed as the \textit{Low-Rank-Functional} (LRF). Although this method lacks a formal proof/comprehensive analysis, for example in terms of a probabilistic guarantee for converging to a solution, the proposed idea is intuitive and has been seen to perform well in simulations. To that end, many numerical examples are provided to corroborate the idea.

Shravan Mohan, Mithun Im, Bharath Bhikkaji

The aim of this paper is to design a band-limited optimal input with power constraints for identifying a linear multi-input multi-output system. It is assumed that the nominal system parameters are specified. The key idea is to use the spectral decomposition theorem and write the power spectrum as $ϕ_{u}(jω)=\frac{1}{2}H(jω)H^*(jω)$. The matrix $H(jω)$ is expressed in terms of a truncated basis for $\mathcal{L}^2\left(\left[-ω_{\mbox{cut-off}},ω_{\mbox{cut-off}}\right]\right)$. With this parameterization, the elements of the Fisher Information Matrix and the power constraints turn out to be homogeneous quadratics in the basis coefficients. The optimality criterion used are the well-known $\mathcal{D}-$optimality, $\mathcal{A}-$optimality, $\mathcal{T}-$optimality and $\mathcal{E}-$optimality. The resulting optimization problem is non-convex in general. A lower bound on the optimum is obtained through a bi-linear formulation of the problem, while an upper bound is obtained through a convex relaxation. These bounds can be computed efficiently as the associated problems are convex. The lower bound is used as a sub-optimal solution, the sub-optimality of which is determined by the difference in the bounds. Interestingly, the bounds match in many instances and thus, the global optimum is achieved. A discussion on the non-convexity of the optimization problem is also presented. Simulations are provided for corroboration.