This paper presents the design, implementation, and characterization of a hardware platform applicable to self-powered wireless sensor network (WSN) node. Its primary design objective is devise hybrid energy harvesting system extend operational lifetime WSN node after they are deployed in field environment. Besides implementation optimal components (microcontroller, sensor, radio frequency (RF) transceiver, others) achieve lowest power consumption, it also necessary consider sources instead...

The number of vias generated during the global routing stage is a critical factor for yield final circuits. However, most routers only approach problem by charging cost in maze function. In this paper, we present router that addresses via optimization throughout entire flow. We introduce aware Steiner tree generation, 3-bend and layer assignment with careful ordering to reduce count. integrate these three techniques into FastRoute 3.0 achieve significant reduction both count runtime.

The number of vias generated during the global routing stage is a critical factor for yield final circuits. However, most routers only approach problem by charging cost in maze function. In this paper, we present router that addresses via optimization throughout entire flow. We introduce aware Steiner tree generation, 3-bend and layer assignment with careful ordering to reduce count. integrate these three techniques into FastRoute 3.0 achieve significant reduction both count runtime.

Double Patterning Lithography (DPL) is one of the few hopeful candidate solutions for lithography CMOS process beyond 45nm. DPL assigns patterns less than a certain distance from each other on layer onto two masks instead mask in traditional lithography. In this paper, we prove that conflict graph used to model conflicts layout planar graph. Based planarity graph, propose new face merging based framework which formulates decomposition as problem pairing odd faces simultaneously inimize...

Double patterning technology (DPT) has emerged as the most hopeful candidate for next node of ITRS roadmap [1]. The goal a DPT decomposer is to decompose entire layout on each layer onto two masks. It assigns features different masks if their spacing less than predefined threshold. Besides, some must be sliced and put so that there would feasible solution mask assignment. Such slicing will cause stitches affect yield. So needs minimize number.

Purpose Offline reinforcement learning (RL) acquires effective policies by using prior collected large-scale data, while, in some scenarios, collecting data may be hard because it is time-consuming, expensive and dangerous, i.e. health care, autonomous driving, seeking a more efficient offline RL method. The purpose of the study to introduce an algorithm, which attempts sample high-value transitions prioritized buffer, uniformly from normal experience improving efficiency learning, as well...

In this paper, an asynchronous multi-user deep reinforcement learning scheme is developed to control the handover (HO) processes across multiple user equipments (UEs), in goal of lowering HO rate while ensuring certain system throughput. scheme, we use a neural network (DNN) as controller learned by each UE via collaborative fashion. Moreover, supervised initializing DNN before execution exploit what already know with traditional schemes and mitigate negative effects random exploration at...

As an easily implemented approach, ripup and reroute has been employed by most of todaypsilas global routers, which iteratively applies maze routing to refine solution quality. But traditional is susceptible get stuck at local optimal results. In this work, we will present a fast high quality router FastRoute3.0, with the new technique named virtual capacity. Virtual capacity proposed guide stage achieve higher results in terms overflow runtime. During stage, works as substitute for real...

As an easily implemented approach, ripup and reroute has been employed by most of todaypsilas global routers, which iteratively applies maze routing to refine solution quality. But traditional is susceptible get stuck at local optimal results. In this work, we will present a fast high quality router FastRoute3.0, with the new technique named virtual capacity. Virtual capacity proposed guide stage achieve higher results in terms overflow runtime. During stage, works as substitute for real...

Global routing faces an increasing problem size and urgent demand on improvement in solution quality. Despite of the recent developments global routers, there exist only two types choices: slow 3D routers with good quality or efficient 2D relatively poor We propose a multi-level router called MGR to fill gap. resorts framework reroute nets congested region grid graph. Routing coarsened graph speeds up while introduces less vias. The powerful rerouting wraps three innovative techniques...

Global routing faces an increasing problem size and urgent demand on improvement in solution quality. Despite of the recent developments global routers, there exist only two types choices: slow 3D routers with good quality or efficient 2D relatively poor We propose a multi-level router called MGR to fill gap. resorts framework reroute nets congested region grid graph. Routing coarsened graph speeds up while introduces less vias. The powerful rerouting wraps three innovative techniques...

Energy harvesting technology has been popularly adopted in embedded systems. However, unstable energy source results unsteady operation. In this paper, we devise a long-term efficient task scheduling targeting for solar-powered sensor nodes. The proposed method exploits reinforcement learning with solar prediction to maximize the efficiency, which finally enhances quality of services (QoS) Experimental show that improves efficiency by 6.0%, on average and achieves better QoS level 54.0%,...

Multi-agent deep reinforcement learning (MDRL) is an emerging research hotspot and application direction in the field of machine artificial intelligence. MDRL covers many algorithms, rules frameworks, it currently researched swarm system, energy allocation optimization, stocking analysis, sequential social dilemma, with extremely bright future. In this paper, a parallel-critic method based on classic algorithm MADDPG proposed to alleviate training instability problem cooperative-competitive...

Solar powered sensor nodes have been adopted in many applications, but unstable energy source and high loss are hindrances to their wide spreading. Storage-less converter-less solar non-volatile reduce the a great extent. However, without buffers, become more sensitive variations. Making full use of harvested provide better quality services (QoS) guarantee stable operations under this circumstance is crucial. In paper, we devise an efficient task mapping strategy for storage-less nodes. The...

Accurately estimating concrete mechanical parameters using artificial intelligence-based methods can save time and energy. Existing nonlinear relationships between components have entered uncertainty in the estimation of hardness properties slump compressive strength as one most important design. Employing regular approaches to use AI models individually dependent variables has been adopted many studies. Therefore, current study aimed develop predictive two categories ensemble hybrid...

With the increasing power of machine learning-based reasoning, use meta information (e.g., digital signal modulation parameters, channel conditions, etc.) to predict performance various processing techniques has become feasible. One such problem practical interest is choosing a proper interference mitigation method based on received signal. Since heuristic table-based methods suffer from limited prediction capability for unseen cases, we propose recommendation system Random Forests (RF)....

Global Routing has been a traditional EDA problem. It congestion elimination as the first and foremost priority. Despite of recent development for popular rip-up reroute framework, process remains arbitrary requires significant tunings. In order to achieve more consistent elimination, we propose new preprocessing framework global routing. identify most congested routing locations by an interval overflow lower bound technique. Then use auction based detour algorithm compute which nets where...

Recent advances in the fields of neuroscience, computer science, and biomedical engineering now allow for analysis large-scale neurophysiological data sets to be carried out on-line real time. Here, we described an on-going effort our research laboratory build a system that will on-line, real-time analyses response properties ensembles neurons (as many as 256) recorded brains awake animals perform behavioral tasks. A cluster workstations allows us carry sequential simultaneous neuronal...

In the context of China's economic growth, situation national new areas is particularly important, and it great significance to study its expansion factors. this paper, regression random forest algorithm (RF) used panel data Shanghai, China from 2011-2022 selected for learning, which obtained with low error high accuracy. After was learned, OOB coefficients influencing factors were analyzed their impact on regional inflation. The results show that per capita disposable income ratio added...

Commonly used UAV emergency inspection methods are executed by the instructions of ground command center. The response rate depends on stability communication network and rapid ability commander. critical time window is fleeting, which likely to cause unnecessary loss. Crisis capability has become key measuring system capabilities. In order improve system’s capability, a method deploying decision-making agents airborne computers early warning systems proposed. This uses technologies such as...