서브메뉴
검색
Towards Artificially Intelligent Communications
Towards Artificially Intelligent Communications
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260209102906
- ISBN
- 9798265406309
- DDC
- 530
- 서명/저자
- Towards Artificially Intelligent Communications
- 발행사항
- [Sl] : Georgia Institute of Technology, 2023
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2023
- 형태사항
- 171 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-05, Section: B.
- 주기사항
- Advisor: Ralph, Stephen E.
- 학위논문주기
- Thesis (Ph.D.)--Georgia Institute of Technology, 2023.
- 초록/해제
- 요약In modern society, communication systems play an essential role in everyday life. Up to this point, digital signal processing (DSP) has supported communications systems (radio, cellular, Internet, etc.) as well as the connected devices. With ever increasing demands on the amount, speed, reliability, and security of transmitted information, the underlying communications technologies need constant improvement. Artificial Intelligence (AI) has been tasked to replace DSP in order to meet forecasted performance demands, expand capabilities of current systems, as well as inspire future communications technologies. However, there are shortcomings of modern AI when trying to integrate it into communications systems: interpretability ("Can we understand why modern AI methods succeed and fail?"), deployability ("Can modern AI operate on everyday electronics, without the need for the cloud?"), and generalizability ("Can these systems perform well in unforeseen circumstances?"). This thesis addresses known technical problems withholding the future of communications with respect to these three shortcomings: (1) enabling complex-valued computations in real-valued algorithms (interpretability & generalizability), (2) neural network pruning methods enabling edge-ready AI (interpretability & deployability), and (3) a gray-box, physics-based AI simulation tool for fiber optic networks (interpretability & generalizability). This thesis demonstrates advancements in AI towards realizing the ultimate goal of artificially intelligent communications.
- 일반주제명
- Physics
- 일반주제명
- Deep learning
- 일반주제명
- Fourier transforms
- 일반주제명
- Signal to noise ratio
- 일반주제명
- Fiber optic networks
- 일반주제명
- Neural networks
- 일반주제명
- Signal processing
- 일반주제명
- Communications networks
- 일반주제명
- Communications systems
- 일반주제명
- Computer science
- 일반주제명
- Electrical engineering
- 일반주제명
- Information technology
- 일반주제명
- Mathematics
- 일반주제명
- Medical imaging
- 기본자료저록
- Dissertations Abstracts International. 87-05B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260203s2023 us c eng d■001000017365974
■00520260209102906
■006m o d
■007cr#unu||||||||
■020 ▼a9798265406309
■035 ▼a(MiAaPQ)AAI32315711
■035 ▼a(MiAaPQ)GeorgiaTech73206
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a530
■1001 ▼aKrzyston, Jakob Anthony.
■24510▼aTowards Artificially Intelligent Communications
■260 ▼a[Sl]▼bGeorgia Institute of Technology▼c2023
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2023
■300 ▼a171 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-05, Section: B.
■500 ▼aAdvisor: Ralph, Stephen E.
■5021 ▼aThesis (Ph.D.)--Georgia Institute of Technology, 2023.
■520 ▼aIn modern society, communication systems play an essential role in everyday life. Up to this point, digital signal processing (DSP) has supported communications systems (radio, cellular, Internet, etc.) as well as the connected devices. With ever increasing demands on the amount, speed, reliability, and security of transmitted information, the underlying communications technologies need constant improvement. Artificial Intelligence (AI) has been tasked to replace DSP in order to meet forecasted performance demands, expand capabilities of current systems, as well as inspire future communications technologies. However, there are shortcomings of modern AI when trying to integrate it into communications systems: interpretability ("Can we understand why modern AI methods succeed and fail?"), deployability ("Can modern AI operate on everyday electronics, without the need for the cloud?"), and generalizability ("Can these systems perform well in unforeseen circumstances?"). This thesis addresses known technical problems withholding the future of communications with respect to these three shortcomings: (1) enabling complex-valued computations in real-valued algorithms (interpretability & generalizability), (2) neural network pruning methods enabling edge-ready AI (interpretability & deployability), and (3) a gray-box, physics-based AI simulation tool for fiber optic networks (interpretability & generalizability). This thesis demonstrates advancements in AI towards realizing the ultimate goal of artificially intelligent communications.
■590 ▼aSchool code: 0078.
■650 4▼aPhysics
■650 4▼aPartial differential equations
■650 4▼aDeep learning
■650 4▼aFourier transforms
■650 4▼aSignal to noise ratio
■650 4▼aFiber optic networks
■650 4▼aMagnetic resonance imaging
■650 4▼aNeural networks
■650 4▼aSignal processing
■650 4▼aCommunications networks
■650 4▼aCommunications systems
■650 4▼aComputer science
■650 4▼aElectrical engineering
■650 4▼aInformation technology
■650 4▼aMathematics
■650 4▼aMedical imaging
■690 ▼a0800
■690 ▼a0605
■690 ▼a0984
■690 ▼a0544
■690 ▼a0489
■690 ▼a0405
■690 ▼a0574
■71020▼aGeorgia Institute of Technology.
■7730 ▼tDissertations Abstracts International▼g87-05B.
■790 ▼a0078
■791 ▼aPh.D.
■792 ▼a2023
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17365974▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


