Hardware Memory Management for Future Mobile Hybrid Memory Systems

Wen, Fei; Qin, Mian; Gratz, Paul V.; Reddy, A. L. Narasimha

doi:10.1109/tcad.2020.3012213

Cited by 20 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The CTR-DRBG algorithm, widely used for secure random number generation, faces several inherent issues [38]- [40]. Firstly, in the derivation function, generating secure random numbers using the counter value leads to increased computational time as the counter starts at 0 and increments, causing a significant rise in computation time with larger counter values [41].…”

Section: Problemsmentioning

confidence: 99%

Enhanced ARIA-based counter mode deterministic random bit generator random number generator implemented in verilog

Rhee,

Lee

2024

IJEECS

View full text Add to dashboard Cite

This paper presents a study aimed at effectively implementing a deterministic random bit generator (DRBG) IP in verilog language, based on the standard encryption algorithm. By controlling the existing round generation and key generation blocks, the internal modules of the counter mode deterministic random bit generator (CTR-DRBG) were successfully implemented and operated, ensuring the secure and efficient generation of random bit sequences. The research focused on parallel operation of modules and optimized module placement to achieve improved clock frequencies. By concurrently operating two modules in the derivation and internal update modules of CTR-DRBG, the processing speed was enhanced compared to the conventional algorithm. Additionally, integrating the reseeding and initialization modules of CTR-DRBG into a single module successfully reduced size. Furthermore, this IP supports the special function register (SFR) interface. The safety of the CTR-DRBG was validated through known answer test (KAT) verification utilizing test vectors from certification. Future research should explore additional studies on CTR-DRBG operating on real FPGA or ASIC, not only using normal algorithm but also employing other block cipher algorithms.

show abstract

Section: Problemsmentioning

confidence: 99%

Enhanced ARIA-based counter mode deterministic random bit generator random number generator implemented in verilog

Rhee,

Lee

2024

IJEECS

View full text Add to dashboard Cite

show abstract

“…5G allows higher network speeds, improved reliability, and less latency while consuming a fifth of energy for a single bit compared to 4G [39]. Wen et al [40] proposed a hardware-accelerated memory manager with data placement data migration policies for use in future mobile hybrid systems achieving a 39% reduction in energy usage and only a 12% loss in performance. Samsung developed the next-generation mobile device RAM, LPDDR4X, with 17% less power usage, 15% performance increase, and up to 12GB of storage in a compact form.…”

Section: B Energy Saving Designs In Hardwarementioning

confidence: 99%

Energy Efficiency on Edge Computing: Challenges and Vision

Holmes

McLarty

Shi

et al. 2022

2022 IEEE International Performance, Computing, and Communications Conference (IPCCC)

View full text Add to dashboard Cite

The Internet of Things (IoT) has been the key to many advancements in next-generation technologies for the past few years. With a conceptual grouping of ecosystem elements such as sensors, actuators, and smart objects connected together, complex operations like environmental monitoring, intelligent transport systems, smart buildings, and smart cities are able to be performed. Edge computing technology extends the reach/scope of IoT ecosystems, offering robust and powerful computational capabilities by connecting multiple devices through the Internet. Unfortunately, this form of computation comes with a significant drawback with strict energy constraints and low power efficiency, which highly limits its potential and usage. In this paper, we present some of the challenges in the planning of energy-efficient IoT edge devices and discuss some of the recent research efforts that proposed promising solutions that address these challenges. Specifically, we first analyze the challenges and reasons for improving the energy consumption of edge platforms and IoT devices. Next, we perform case studies that outline the energysaving techniques in smart grids, smart cities, electric vehicles (EV), smart home devices, and Virtual Reality and Augmented Reality (VR/AR). We further discuss different approaches such as computation offloading, edge devices hardware and software designs, and a number of algorithms that help reduce energy consumption. Finally, we outline possible future directions and our vision of improving energy efficiency on edge platforms.

show abstract

“…Short term monitoring cannot guarantee capturing participants' symptoms and their associated behavior changes when they are unwell. Meanwhile, if we retrain models using all existing data, the computation and memory burden would be high and prohibitive for mobile devices [11][12][13]. If we retrain models using only the newly incoming data, a classic problem referred to as catastrophic forgetting will significantly downgrade model performance.…”

Section: Introductionmentioning

confidence: 99%

Semi-supervised Graph Instance Transformer for Mental Health Inference

Dong

Tang

Cai

et al. 2021

2021 20th IEEE International Conference on Machine Learning and Applications (ICMLA)

View full text Add to dashboard Cite

Mobile sensing appears as a promising solution for health inference problem (e.g., influenza-like symptom recognition) by leveraging diverse smart sensors to capture fine-grained information about human behaviors and ambient contexts. Centralized training of machine learning models can place mobile users' sensitive information under privacy risks due to data breach and misexploitation. Federated Learning (FL) enables mobile devices to collaboratively learn global models without the exposure of local private data. However, there are challenges of on-device FL deployment using mobile sensing: 1) long-term and continuously collected mobile sensing data may exhibit domain shifts as sensing objects (e.g. humans) have varying behaviors as a result of internal and/or external stimulus; 2) model retraining using all available data may increase computation and memory burden; and 3) the sparsity of annotated crowd-sourced data causes supervised FL to lack robustness. In this work, we propose FedMobile, an incremental semi-supervised federated learning algorithm, to train models semi-supervisedly and incrementally in a decentralized online fashion. We evaluate FedMobile using a real-world mobile sensing dataset for influenza-like symptom recognition. Our empirical results show that FedMobile-trained models achieve the best results in comparison to the selected baseline methods.

show abstract

Hardware Memory Management for Future Mobile Hybrid Memory Systems

Cited by 20 publications

References 25 publications

Enhanced ARIA-based counter mode deterministic random bit generator random number generator implemented in verilog

Enhanced ARIA-based counter mode deterministic random bit generator random number generator implemented in verilog

Energy Efficiency on Edge Computing: Challenges and Vision

Semi-supervised Graph Instance Transformer for Mental Health Inference

Contact Info

Product

Resources

About