LPNet: A DNN based latency prediction technique for application mapping in Network-on-Chip design

Analytical models used for latency estimation of Network-on-Chip (NoC) are not producing reliable accuracy. This makes these analytical models difficult to use in optimization of design space exploration. In this paper, we propose a learning based model using deep neural network (DNN) for latency predictions. Input features for DNN model are collected from analytical model as well as from Booksim simulator. Then this DNN model has been adopted in mapping optimization loop for predicting the best mapping of given application and NoC parameters combination. Our simulations show that using the proposed DNN model, prediction error is less than 12% for both synthetic and application specific traffic. More than 108 times speedup could be achieved using DPSO with DNN model compared to DPSO using Booksim simulator.     

Software-defined GPS receiver on USRP-platform

A non-real-time GPS software-defined radio was designed, built, and tested using the Universal Software Radio Peripheral (USRP) platform. A hardware front-end to receive the GPS signal and downconvert it to an intermediate frequency was also implemented. Subsequently, the system was used to acquire and process real GPS data to demonstrate the capability of the design.     

Simulation Study of Dielectric Modulated Dual Material Gate TFET Based Biosensor by Considering Ambipolar Conduction

Silicon - In this paper, a dielectric modulated dual material gate TFET (DM-DMG_TFET)based biosensor is proposed. In order to detect various biomolecules, a nanogap cavity is formed by the...     

Radiation-absorption,chemical reaction,Hall and ion slip impacts on magnetohydrodynamic free convective flow over semi-infinite moving absorbent surface

The investigation of radiation-absorption,chemical reaction,Hall and ion-slip impacts on unsteady MHD free convective laminar flow of an incompressible viscous,electrically conducting and heat generation/absorbing fluid enclosed with a semi-infinite porous plate within a rotating frame has been premeditated.The plate is assumed to be moving with a constant velocity in the direction of fluid movement.A uniform transverse magnetic field is applied at right angles to the porous surface,which is absorbing the fluid with a suction velocity changing with time.The non-dimensional governing equations for present inves-tigation are solved analytically making use of two term harmonic and non-harmonic functions.The graphical results of velocity,temperature and concentration distributions on the analytical solutions are displayed and discussed with reference to pertinent parameters.It is found that the velocity profiles decreased with an increasing in Hartmann number,rotation parameter,the Schmidt number,heat source parameter,while it increased due to an increase in permeability parameter,radiation-absorption param-eter,Hall and ion slip parameters.However,the temperature profile is an increasing function of radiation-absorption parameter,whereas an increase in chemical reaction parameter,the Schmidt num-ber Sc or frequency of oscillations decrease the temperature profile on cooling.Also,it is found that the concentration profile is decreased with an escalating in the Schmidt number or the chemical reaction parameter.     

The performance of a hybrid routing intelligent algorithm in a mobile ad hoc network

End-to-end delay, power consumption, and communication cost are some of the most important metrics in a mobile ad hoc network (MANET) when routing from a source to a destination. Recent approaches using the swarm intelligence (SI) technique proved that the local interaction of several simple agents to meet a global goal has a significant impact on MANET routing. In this work, a hybrid routing intelligent algorithm that has an ant colony optimisation (ACO) algorithm and particle swarm optimisation (PSO) is used to improve the various metrics in MANET routing. The ACO algorithm uses mobile agents as ants to identify the most feasible and best path in a network. Additionally, the ACO algorithm helps to locate paths between two nodes in a network and provides input to the PSO technique, which is a metaheuristic approach in SI. The PSO finds the best solution for a particle’s position and velocity and minimises cost, power, and end-to-end delay. This hybrid routing intelligent algorithm has an improved performance when compared with the simple ACO algorithm in terms of delay, power consumption, and communication cost.     

GMM based language identification system using robust features

In this work, we have proposed new feature vectors for spoken language identification (LID) system. The Mel frequency cepstral coefficients (MFCC) and formant frequencies derived using short-time window speech signal. Formant frequencies are extracted from linear prediction (LP) analysis of speech signal. Using these two kind of features of speech signal, new feature vectors are derived using cluster based computation. A GMM based classifier has been designed using these new feature vectors. The language specific apriori knowledge is applied on the recognition output. The experiments are carried out on OGI database and LID recognition performance is improved.     

Dynamic surface control of a piezoelectric fuel injector during rate shaping

Fuel injection rate shaping is a strategy to improve fuel efficiency and reduce harmful emissions in diesel engines. Due to their fast response, piezoelectric fuel injectors are capable of rate shaping operation. This paper describes a model-based closed-loop controller of injection flow rate for a piezoelectric fuel injector. This within-an-engine-cycle control strategy utilizes a dynamic surface control scheme and shows an injection flow rate tracking capability. Practical issues with LabVIEW FPGA control implementation are also addressed. The performance of the controller is verified with simulation and experimental results at different rail pressures and desired injection rates. The experiments show a maximum error of total fuel per one injection event of 2.5%. A stability analysis is also included.     

Low intensity visual data improvement using DWT and LWT based regularizer

Image Completion plays a vital role in compressed sensing, machine learning, and computer vision applications. The Rank Minimization algorithms are used to perform the image completion. The major problem with rank minimization algorithms is the loss of information in the recovered image at high corruption ratios. To overcome this problem Lifting wavelet transform based Rank Minimization (LwRM), and Discrete wavelet transform based Rank Minimization (DwRM) methods are proposed, which can recover the image, if the corrupted observations are more than 80%. The evaluation of the proposed methods are accomplished by Full Reference Image Quality Assessment (FRIQA) and No Reference Image Quality Assessment (NR-IQA) metrics. The simulation results of proposed methods are superior to state-of-the-art methods.

     

Joint underwater target detection and tracking with the Bernoulli filter using an acoustic vector sensor

In this paper, we study the problem of joint underwater target detection and tracking using an acoustic vector sensor (AVS). For this challenging problem, first a realistic frequency domain simulation is set up. The outputs of this simulation generate the two dimensional FRequency–AZimuth (FRAZ) image. On this image, the random finite set (RFS) framework is employed to characterize the target state and sensor measurements. We propose to use the Bernoulli filter, which is the optimal Bayes filter emerged from the RFS framework for randomly on/off switching single dynamic systems. Moreover, to increase the performance of detection and azimuth tracking in low signal-to-noise ratio (SNR) scenarios, a track-before-detect (TBD) measurement model for AVS is proposed to be used with the Bernoulli filter. Sequential Monte Carlo (SMC) implementation is preferred for the Bernoulli filter recursions. Extensive simulation results prove the performance gain obtained by the proposed approach both in estimation accuracy and detection range of the system.