In this paper, we analyze the spectral efficiency and outage probability of random unitary beamforming with adaptive modulation(RUBAM) in multi-user MIMO systems. While achievable sum-rate is meaningful from the view point of information theory, the spectral efficiency is more preferable in terms of practical implementation since practical systems do not allow the continuous level modulations. With the help of the exact statistics, we derive the analytical expression of the spectral efficiency and outage probability of RUBAM. Based on your analytical results our proposed RUBAM can be used in designing the practical RUB based MU-MIMO systems.
In order to improve performance, image preprocessing for enhancement is essential in a fingerprint recognition system. Most prominent fingerprint enhancement methods perform well for relatively high-quality images but less effectively for low-quality images, especially where the ridge patterns are very noisy and corrupt. To enhance such images, this paper proposes an effective three-step algorithm, a method that locally normalizes input images, computes the local ridge orientation, and then applies a local ridge compensation filter with a rotated window in order to enhance the ridges by matching the local ridge orientation. Experimental results indicate that the proposed method performs better when compared to some noted methods of enhancing low-quality images.
As the technology scales down to the deep submicron domain, the leakage energy in memory devices could contribute to a significant portion of the total energy consumption. Therefore, evaluation of energy consumption including the leakage energy is necessary. In this paper, we investigate the effectiveness of scratch-pad memory on energy reduction considering both the dynamic and leakage energy. The experiments are performed for 65nm, 45nm, and 32nm technologies. The results demonstrate the effectiveness of scratch-pad memory in deep submicron technology. It is also observed that the leakage energy becomes less significant along with the technology scaling.
This paper proposes a high performance, rail-to-rail, Differential Voltage Current Conveyor (DVCC), using Quasi-Floating Gate (QFG) transistors. With a compact structure, the circuit operates at very low supply voltage (±0.5V), dissipates very low standby power (90µW), has a high bandwidth (f-3dB of 81MHz and 82MHz for Ai and Av, respectively), and a very low input resistance (0.2Ω). Using the proposed DVCC a novel fully differential Switched Capacitor (SC) integrator is also introduced. HSPICE simulation results confirm the functionality of proposed DVCC and integrator.
We demonstrate an all-optical single to multi-wavelength converter using gain modulation in a Fabry-Perot laser diode (FP-LD at 10Gb/s. It can simultaneously provide 1 to 4 output channels and support both up and down conversion. We observed over 14dB extinction ration (ER) and power penalty of around 1.5dBm at a BER of 10-9. The results ensure to increase the number of output channels. The proposed scheme can be applied to multicasting function as well as 1xN wavelength conversion in wavelength division multiplexed optical networks.
This paper proposes a solution for inconsistency pruning of neurons within a sequential learning Radial Basis Function (RBF) Network. This paper adopts the concept that a specific RBF neuron which continuously exhibits low output in a sequence of training patterns does not justify the proposition that the neuron is insignificant to the whole function to be approximated. We establish additional criterions to provide protection from error in pruning RBF neurons within the hidden layer, which we prove is able to improve consistency and stability of neuron evolution. With such stability within the sequential learning process, we also show how the convergence speed of the network can be improved by reducing the number of consecutive observations required to prune a neuron in the hidden layer.