3 results
Search Results
Now showing 1 - 3 of 3
Item Power-Efficient Adaptive Memory Design and Optimization for Video and Deep Learning(North Dakota State University, 2020) Das, HritomMemory devices such as Static Random-Access Memory (SRAM) and Dynamic Random-Access Memory (DRAM) are dominating members of today’s semiconductor industry. Most of the silicon area in a digital system is occupied by memory devices. The video decoder and deep learning are especially constrained by memory devices to process a large amount of data. For example, memory devices are consuming lots of power for video processing. Nowadays, all mobile electronics, such as mobile phones and laptops, are using video data a lot. Due to that, the battery life of mobile devices is highly dependent on power consumption of memory devices. To enhance the battery life of mobile devices, supply voltage can be scaled down. However, memory devices are error prone at low supply voltages. To obtain high quality video, a functionally stable memory design is needed, which means we must provide a higher VDD or use a larger memory cell. As a result, there will be a tradeoff between quality, and silicon area or power consumption. For mobile devices, memory needs to be designed to operate in the sub-threshold region to maximize battery life; however, reducing the supply voltage slows down memory devices, resulting in poor video quality. Hence, memory design is very complicated and time consuming. So, a smart way to design memory devices for a specific application is needed. Mathematical models can be developed to design memory devices based on specific requirements such as silicon area, while optimizing video quality for a target supply voltage. Similarly, optimized memory is needed to better support differentially private deep learning algorithms in local devices. This dissertation first develops a mathematical model for designing optimal memory devices for videos, then develops an optimized memory for differentially private deep learning systems in edge computing devices, and finally develops a run-time adaptable Error Correction Code (ECC) video storage scheme, with minimal area overhead and negligible video quality degradation, in order to significantly reduce power.Item Embracing Visual Experience and Data Knowledge: Efficient Embedded Memory Design for Big Videos and Deep Learning(North Dakota State University, 2019) Edstrom, JonathonEnergy efficient memory designs are becoming increasingly important, especially for applications related to mobile video technology and machine learning. The growing popularity of smart phones, tablets and other mobile devices has created an exponential demand for video applications in today’s society. When mobile devices display video, the embedded video memory within the device consumes a large amount of the total system power. This issue has created the need to introduce power-quality tradeoff techniques for enabling good quality video output, while simultaneously enabling power consumption reduction. Similarly, power efficiency issues have arisen within the area of machine learning, especially with applications requiring large and fast computation, such as neural networks. Using the accumulated data knowledge from various machine learning applications, there is now the potential to create more intelligent memory with the capability for optimized trade-off between energy efficiency, area overhead, and classification accuracy on the learning systems. In this dissertation, a review of recently completed works involving video and machine learning memories will be covered. Based on the collected results from a variety of different methods, including: subjective trials, discovered data-mining patterns, software simulations, and hardware power and performance tests, the presented memories provide novel ways to significantly enhance power efficiency for future memory devices. An overview of related works, especially the relevant state-of-the-art research, will be referenced for comparison in order to produce memory design methodologies that exhibit optimal quality, low implementation overhead, and maximum power efficiency.Item Intelligent Energy-Efficient Storage System for Big-Data Applications(North Dakota State University, 2020) Gong, YifuStatic Random Access Memory (SRAM) is a critical component in mobile video processing systems. Because of the large video data size, the memory is frequently accessed, which dominates the power consumption and limits battery life. In energy-efficient SRAM design, a substantial amount of research is presented to discuss the mechanisms of approximate storage, but the content and environment adaptations were never a part of the consideration in memory design. This dissertation focuses on optimization methods for the SRAM system, specifically addressing three areas of Intelligent Energy-Efficient Storage system design. First, the SRAM stability is discussed. The relationships among supply voltage, SRAM transistor sizes, and SRAM failure rate are derived in this section. The result of this study is applied to all of the later work. Second, intelligent voltage scaling techniques are detailed. This method utilizes the conventional voltage scaling technique by integrating self-correction and sizing techniques. Third, intelligent bit-truncation techniques are developed. Viewing environment and video content characteristics are considered in the memory design. The performance of all designed SRAMs are compared to published literature and are proven to have improvement.