Deep reinforcement learning (deep RL) has the potential to replace classic robotic controllers. State-of-the-art Reinforcement algorithms such as Proximal Policy Optimization, Twin Delayed Deterministic Gradient and Soft Actor-Critic Algorithms, mention a few, have been investigated for training robots walk. However, conflicting performance results of these reported in literature. In this work, we present analysis above three state-of-the-art constant velocity walking task on quadruped. The...

Fibre channel based storage area networks have been a common implementation for data in enterprise centers. As the size of network tends to grow along with geographical distribution leading distributed network, also known as large fabric, seems exhibit scaling and stability issues. The issues seem stem from construction, route discovery, management process which are fibre channel-2 level traffic more than actual I/O transmission process. To determine an adequate tool is required. usually...

In music classification tasks, Convolutional Recurrent Neural Network (CRNN) has achieved state-of-the-art performance on several data sets. However, the current CRNN technique only uses RNN to extract spatial dependency of signal in its time dimension but not frequency dimension. We hypothesize latter can be additionally exploited improve performance. this paper, we propose an improved called Time and Frequency dimensions (CRNN-TF), which captures dependencies both multiple directions....

The problem in modeling large systems by artificial neural networks (ANN) is that the size of input vector can become excessively large. This condition potentially increase likelihood convergence problems for training algorithm adopted. Besides, memory requirement and processing time also increase. paper addresses issue ANN dimension reduction. Two different methods are discussed compared efficiency accuracy when applied to transient stability assessment.

The performance of the state-of-the-art Deep Reinforcement algorithms such as Proximal Policy Optimization, Twin Delayed Deterministic Gradient, and Soft Actor-Critic for generating a quadruped walking gait in virtual environment was presented previous research work titled "A Comparison PPO, TD3, SAC Algorithms Quadruped Walking Gait Generation". We demonstrated that algorithm had best certain instances sensor configurations environment. In this work, we present analysis above generation...

In an effort to reduce vehicle collisions with snowplows in poor weather conditions, this paper details the development of a real time thermal image based machine learning approach early collision avoidance system for snowplows, which intends detect and estimate distance trailing vehicles. Due operational conditions include heavy-blowing snow, traditional optical sensors like LiDAR visible spectrum cameras have reduced effectiveness detecting objects such environments. Thus, we propose using...

Scale Invariant Feature Transform (SIFT) algorithm is a widely used computer vision that detects and extracts local feature descriptors from images. SIFT computationally intensive, making it infeasible for single threaded im-plementation to extract high-resolution images in real time. In this paper, an approach parallelization of the demonstrated using NVIDIA’s Graphics Processing Unit (GPU). The parallel-ization design on GPUs divided into two stages, a) Algorithm de-sign-generic strategies...

A mobile ad-hoc network (MANET) of small robots (sensor nodes) adjusting their positions to establish connectivity would be able provide a communication infrastructure in an urban battlefield environment. sensor node capable moving particular position connectivity, provided it knows its current position, other nodes and the radio propagation characteristics area. Typically, MANET determine using localization algorithm. Most algorithms use either free space or "two ray" models, which are only...

IEEE 802.11e standard has defined EDCA to guarantee prioritized QoS by service differentiation. However, it induces significant performance deterioration in both delay and collision probability at heavy load due inherent drawbacks of binary exponential backoff (BEB) scheme. In this paper we proposed a novel scheme, channel based adaptive (CCBAB)1 scheme which compels each station react collisions irrespective whether or not is involved in, combat the problem. CCBAB selectively forces some...

A deep learning algorithm for Gaussian noise removal from both grayscale and color images is developed. As opposed to most existing discriminative methods that train a specific model each level, the proposed method can handle wide range of levels using only two trained models, one low other high levels. In algorithm, training process consists three successive steps. first step, classifier classify noisy clean images. second denoiser network aims remove in image features are extracted by...

Directional antennas shape transmission patterns to provide greater coverage distance and reduced angle. When used in an ad-hoc network, this reduces interference among transmitting nodes thereby increases throughput. A problem that has not been addressed is how compute individual beam maximize some measure of global network performance. Historically, the focus on finding node antenna give locally optimal In paper we investigate a low hardware complexity beamforming approach aimed at...

The server virtualization architecture, encompassing the sharing of data storage subsystems among virtual servers or operating systems on a single host using I/O channel capabilities in fibre fabrics, was pioneered by IBM through their system z <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">9</sup> mainframe and its predecessors. implementations were based connectivity standard. With advent small computer interface devices area networks...

This paper considers anomaly detection with multi-view data. Unlike traditional on single-view data which identifies anomalies based inconsistency between instances, view within each instance. Current approaches are mostly unsupervised and transductive. may have limited performance in many applications, labeled normal prefer efficient new In this paper, we propose an inductive semi-supervised approach. We design a probabilistic generative model for data, assumes different views of instance...

Small antenna arrays are advantageous for ad-hoc networks. Typically array geometry is designed using simplified objective measures to make analysis tractable or maximize search efficiency. This paper investigates design fitness functions more tightly related network performance. Particle swarm optimization (PSO) used planar with minimum mean-square error noise-to-signal ratio. Fitness averaged across look angle include effects of random signal and interference directions, resulting in a...

Abstract This paper describes, a technique of the transient stability assessment, known as Extended Equal Area Criterion (EEAC) applied to multimachine power system. The EEAC is implemented using MATLAB program provide software tool with GUI. developed incorporate analysis, into an undergraduate curriculum in engineering. Another criterion for developing this reduce effort normally associated instruction complex theory and use classroom environment. results obtained from on two systems are...

Directional antennas shape transmission patterns to provide greater coverage distance and reduced angle. Use of adaptive directional antenna arrays can minimize interference while also being more energy efficient. When used in an ad-hoc network, this reduces among transmitting nodes thereby increases throughput. Such “smart antennas” use digital beamforming based on signal processing algorithms compute the appropriate weights form effective patterns. Smart require knowledge received at each...

The efficient construction of contours a radio propagation map is crucial in using maps number real-time communication and network applications. In this research work, we first propose an adaptive region (ARC) technique capable constructing different resolutions map. Next, the process implementing ARC for execution on GPU presented. drawbacks implementation only global memory are discussed, optimization techniques to improve performance discussed implemented. Simulations performed with...

Artificial Intelligence (AI) and, more specifically, Machine Learning (ML) methodologies have successfully tailored commercial applications for decades. However, the recent profound success of large language models like ChatGPT and enormous subsequent funding from governments investors positioned ML to emerge as a paradigm-shifting technology across numerous domains in coming years. To cultivate competent workforce prepare students this new AI-focused evolving world, integration is proposed...

Abstract NOTE: The first page of text has been automatically extracted and included below in lieu an abstract Embedded Systems Design: Responding to the Challenge A recent IEEE-USA Today’s Engineer Online article indicated that U.S. institutions higher learning need provide embedded systems design programs. cited “only a few colleges universities have good programs place.” further mentioned “many engineers development are getting close retirement age.” In this paper we will outline our...

Antenna array gain is a relative measure of performance defined differently in various literature. Most definitions are not power consistent, and thus cannot be used directly link budget analysis. In this short paper, we present correction factor for common antenna arrays that allows them to standard calculations.