Sökning: "Power processing unit"

Visar resultat 16 - 20 av 77 uppsatser innehållade orden Power processing unit.

  1. 16. Low Power Hardware Accelerator For Gated Recurrent Unit

    Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Malavika Balakumar; [2022]
    Nyckelord :;

    Sammanfattning : Neural Networks are a subset of Machine Learning which are designed to recognize patterns. Recurrent Neural Networks are an important part of AI (Artificial Intelligence) which allows for short term as well as long term dependencies to be captured. LÄS MER

  3. 17. Access Point Selection and Clustering Methods with Minimal Switching for Green Cell-Free Massive MIMO Networks

    Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Qinglong He; [2022]
    Nyckelord :Cell-free massive MIMO; multi-objective optimization; deep reinforcement learning; AP switch ON OFF; energy efficiency; Cellfri massiv MIMO; multiobjektiv optimering; djup förstärkningsinlärning; AP switch ON OFF; energieffektivitet;

    Sammanfattning : As a novel beyond fifth-generation (5G) concept, cell-free massive MIMO (multiple-input multiple-output) recently has become a promising physical-layer technology where an enormous number of distributed access points (APs), coordinated by a central processing unit (CPU), cooperate to coherently serve a large number of user equipments (UEs) in the same time/frequency resource. However, denser AP deployment in cell-free networks as well as an exponentially growing number of mobile UEs lead to higher power consumption. LÄS MER

  4. 18. Acoarse grain reconfigurable memory architecture for linear algebra and deep neural networks

    Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Cyril Armand Koenig; [2022]
    Nyckelord :;

    Sammanfattning : Companies and institutions around the world have been working to develop machines with always more computing power. This race has now found its new objective: hexascale computing (with 1018 flops machines). LÄS MER

  5. 19. On-board processing with AI for more autonomous and capable satellite systems

    Master-uppsats, Luleå tekniska universitet/Institutionen för system- och rymdteknik

    Författare :Tamina Lund; [2022]
    Nyckelord :Satellite telemetry forecasting; battery degradation prediction; LSTM; artificial intelligence; on-board processing;

    Sammanfattning : While the use of Artificial Intelligence (AI) has faced a sharp up-rise in popularity in ground-based industries, such as for autonomous navigation in the automotive industry and predictive maintenance in manufacturing processes, it is yet only rarely used in space industry. Hence, this thesis aims to investigate the possibilities of using AI for processing on-board Earth-orbiting satellites while in orbit. LÄS MER

  7. 20. Configurable, scalable single-ended sense amplifier with additional auxiliary blocks for low-power two-port memories in advanced FinFET technologies

    Master-uppsats, Lunds universitet/Institutionen för elektro- och informationsteknik

    Författare :Limitha Subbaiah Kumar Nangaru; [2022]
    Nyckelord :CMOS Complementary Metal Oxide Semiconductors ; DRC Design Rule Check ; Process Corners; FinFET Fin Field Effect Transistor ; IC Integrated Circuit ; LVS Layout Versus Schematic ; Monte Carlo; Nominal Voltage; PDK Process Design Kit ; Power Delay Product; Read Bit Line; Read Word Line; SoC System on Chip ; SRAM Static Random Access Memory ; Threshold Voltage; Technology and Engineering;

    Sammanfattning : System on Chip (SoC) designs contain a variety of Intellectual Property (IP) cores, including digital signal processing blocks, media and graphics processing units, as well as processing core units that employ multiple-port memories to enhance performance and bandwidth. These memories allow parallel read/write operations from the same memory blocks from different ports. LÄS MER