76 research outputs found
A Survey on Citizens Broadband Radio Service (CBRS)
No full textTo leverage the existing spectrum and mitigate the global spectrum dearth, the Federal Communications Commission of the United States has recently opened the Citizens Broadband Radio Service (CBRS) spectrum, spanning 3550–3700 MHz, for commercial cognitive operations. The CBRS has a three-tier hierarchical architecture, wherein the incumbents, including military radars, occupy the topmost tier. The priority access licenses (PAL) and general authorized access (GAA) are second and third tier, respectively, facilitating licensed and unlicensed access to the spectrum. This combination of licensed and unlicensed access to the spectrum in a three-tier model has opened novel research directions in optimal spectrum sharing as well as privacy preservation, and hence, several schemes have been proposed for the same. This article provides a detailed survey of the existing literature on the CBRS. We provide an overview of the CBRS ecosystem and discuss the regulation and standardization process and industrial developments on the CBRS. The existing schemes for optimal spectrum sharing and resource allocation in CBRS are discussed in detail. Further, an in-depth study of the existing literature on the privacy of incumbents, PAL devices, and GAA devices in CBRS is presented. Finally, we discuss the open issues in CBRS, which demand more attention and effort
The Decisions To Make (In An Advertising Agency)
No full textThis case is about Mr. Pranay Bharadwaj who with his hard word rose to an entrepreneur and initiated Total Advertising after understanding the advertising trade. In the case author has made an effort to understand the need of customers in the changing times. At the time of this problem Mr. Pranay put emphasis on research parameters to use it as a tool of decision making thus leading to customer satisfaction
Efficient learning and planning using spatial side information
No full textThis thesis investigates the following question: how to efficiently integrate side information, available either a priori or online, with existing algorithms for learning and planning in environments with stochastic features? Side information in this context refers to any information that does not directly determine system parameters, but indicates a relationship between them. Such information can often be obtained from existing data, including that collected by onboard sensors. Algorithms that exploit side information are of interest in solving many real-world problems that can be modeled as stochastic control processes with unknown transition probabilities or unknown transition times. Specifically, we consider the problems of reward maximization in grid-world environments with unknown, stochastic dynamics and travel time minimization in urban transit routing problems with deterministic dynamics and stochastic travel times. Exploiting additional information available to solve these problems, when classical algorithms leave much to be desired in terms of performance and accuracy, is the main theme of this thesis.
The first part of the thesis proposes the idea of indirect sampling for accelerated learning in Markov decision processes when additional information is available in the form of bounds on the differences between the transition probabilities at different states. In addition, it proposes a greedy approximation algorithm that utilizes the additional side information to effectively balance exploration and exploitation. It also analyzes the performance of indirect sampling algorithms in different information settings and defines the notion of agent safety, a vital consideration for systems operating in the physical environment, in the context of our problem. Under certain assumptions, it provides guarantees on the safety of an agent exploring with our algorithm that exploits side information.
The second part proposes a methodology and a tool that, given an origin-destination pair, a travel time budget, and a measure of the passenger's tolerance for ambiguity, provide the optimal online route choice in a transit network by balancing the objectives of maximizing on-time arrival probability and minimizing expected travel time. This framework is a significant improvement over existing algorithms where the problem of optimal routing in urban transit networks is usually studied with only the least expected travel time as the performance criteria under the assumption of travel time independence on different road segments. The proposed algorithm utilizes side information, available in the form of historic travel time data and upstream real-time data, to build and update the underlying model online.
We demonstrate the utility and the performance of the proposed algorithms with the help of realistic numerical experiments conducted (i) on a fixed-route bus system that serves the residents of the Champaign-Urbana metropolitan area and, (ii) in the setting of a Mars rover navigating on unknown or partially known terrain. In both of these problems, data from onboard sensors and external sources acts as the side information.Submission published under a 24 month embargo labeled 'U of I Access', the embargo will last until 2022-12-01The student, Pranay Thangeda, accepted the attached license on 2020-12-07 at 20:36.The student, Pranay Thangeda, submitted this Thesis for approval on 2020-12-07 at 21:47.This Thesis was approved for publication on 2020-12-09 at 15:15.DSpace SAF Submission Ingestion Package generated from Vireo submission #16086 on 2021-03-04 at 16:20:43Made available in DSpace on 2021-03-05T21:42:51Z (GMT). No. of bitstreams: 2
THANGEDA-THESIS-2020.pdf: 2804943 bytes, checksum: 7ceb6a1733dd8f8ba27f08e2bb31d0a0 (MD5)
LICENSE.txt: 4212 bytes, checksum: 1604efe3ab15ae4102972a9c2dbfde93 (MD5)
Previous issue date: 2020-12-09Embargo set by: Seth Robbins for item 117236
Lift date: 2023-03-05T21:43:00Z
Reason: Author requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemAuthor requested U of Illinois access only (OA after 2yrs) in Vireo ETD systemU of I Onl
Numerical investigation of anisotropic and time-dependent behaviours of foliated rock mass
No full textSqueezing ground conditions have become a major challenge faced by underground hard rock
mines exploiting reserves at greater depth and in a high-stress environment. This thesis
investigates the influence of various parameters on the severity of squeezing ground
conditions in foliated rock mass utilizing continuum numerical modelling in FLAC3D finite
difference code. A series of numerical simulations were carried out for selected cases of two
mines subject to squeezing ground conditions to investigate the rock mass failure
mechanisms, the influence of various geological, mining and stress parameters, time
dependence and the role of rock reinforcements. The numerical models were calibrated using
field data and underground observations. The numerical models were successful in capturing the observed failure mechanisms at the
two mines. The calibrated numerical models were used to investigate the influence of varying
interception angle, excavation shape, excavation over-break, mining depth and parallel
excavations on the severity of squeezing ground conditions utilizing ubiquitous joint model.
The time dependence in the numerical models was simulated by using the power ubiquitous
model and the role of various rock reinforcement elements was simulated by using structural
elements. The results of the numerical simulations were found to be in good agreement with
the underground convergence data and observations. The modelling methodology used in the
thesis can be used to improve the understanding of the anisotropic and time-dependent
behaviour of the foliated rock mass to various mining conditions. The methodology can also
be used to evaluate performance of various rock reinforcement strategies and elements that
are used in squeezing ground conditions
Performance and security challenges in next-gen networks: SDN using FFNN as a DDoS mitigation solution
No full textClick on the DOI link to access this article at the publishers website (may not be free).In recent years, there has been a proliferation in the development of next-generation networks, specifically SDN (software defined networks) and IoT (the Internet of Things). However, this proliferation has also led to detrimental growth in cyber terrorism, particularly in the form of various types of DDoS attacks. This research paper evaluates the efficacy of FFNN based back propagation algorithm and Gradient Descent (GD by comparing their performance. We performed a deep analysis of this algorithm and proposed the performance evolution of DDoS attack detection in data set, such as CICIDS2019. We analyze the training, testing, and validation processes of these algorithms using Matrix Laboratory (MATLAB) R2020B. We compare these methods based on various performance parameters like accuracy, precision, recall, F-1 score, and time complexity analysis. We use the big O notation to analyze the time complexity of the optimization algorithm. The GD method provides a linear big O notation, n (o), which is superior to that of other algorithms. A detailed analysis of performance analysis is discussed in the result analysis of the proposed work. A detailed performance analysis is discussed in result section. The proposed assistance aims to identify an improved optimization algorithm for distinguish and mitigating DDoS attacks in the SDN network. © 2025 IEEE
The future of software testing automation: Innovations, challenges, and emerging alternatives
No full textClick on the DOI link to access this article at the publishers website (may not be free).Software testing automation is seeing fast evolution, propelled by innovative developments in artificial intelligence (AI), machine learning (ML), and cloud computing technologies. These advances are transforming the software development environment, providing new opportunities to improve the effectiveness, precision, and adaptability of testing operation. This study investigates the future of software testing automation by analyzing the key advancements that are set to transform testing methodologies in the next years. Prominent advancements include AI-driven test generation methods that utilize machine learning algorithms to automatically produce test cases based on application behavior, as well as self-healing test scripts capable of autonomously identifying and adjusting to alterations in the application interface, thereby substantially minimizing maintenance burdens. Important developments include AI-driven test generation methods that utilize machine learning algorithms to automatically produce test cases based on application behavior, as well as self-healing test scripts that can independently identify and adjust to modifications in the application interface, thereby substantially decreasing maintenance burdens. The incorporation of continuous testing into DevOps pipelines is enhancing the agility and reliability of software delivery, enabling real-time feedback and expedited release cycles. Automation has many benefits, but implementing it is difficult. Maintaining automated test scripts may be complicated and resource-intensive, particularly for rapidly evolving codebases. Due to these constraints, low-code and no-code testing frameworks are becoming popular, democratizing automation by enabling testers without coding skills to build and execute automated tests. Combining these strategies with traditional automated techniques is improving software testing by offering a more complete and effective assessment framework. In research work focus on recent development in the area software testing also compare the different method which is involved in automated software testing. © 2025 IEEE
- …
