| 1. |
- Yang, Yu
(författare)
-
High-Level Synthesis for SiLago : Advances in Optimization of High-Level Synthesis Tool and Neural Network Algorithms
- 2022
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Embedded hardware designs and their automation improve energy and engineering efficiency. However, these two goals are often contradictory. The attempts to improve energy efficiency often come at the cost of engineering efficiency and vice-versa. High-level synthesis (HLS) is a good example of this challenge. It has been researched for more than three decades. Nevertheless, it has not become a mainstream design flow component concerning custom hardware synthesis due to the big efficiency gap between the HLS-generated hardware design and the manual RTL design.This thesis attempts to address the HLS challenge. We divide the research challenge of improving state-of-the-art HLS into three components: 1) the hardware architecture and its underlying VLSI design style, 2) the design automation algorithms and data structures, and 3) the optimization of the algorithm to be mapped.The SiLago hardware platform has been reported as a prominent hardware architecture that can deliver ASIC-like efficiency and could be an ideal HLS hardware platform. It has the following features: 1) SiLago embodies parallel distributed two-level control. 2) SiLago blocks are hardened blocks that can create valid VLSI designs by abutment without involving logic or physical synthesis.Consequently, when targeting the SiLago hardware platform, the SiLago HLS tool generates not a single controller but multiple collaborative controllers, each of which is a hierarchy of two levels. The distributed two-level control scheme poses unique challenges in synchronization and scheduling tasks. Unique data structures and instruction scheduling models are developed for the SiLago HLS tool to support the distributed two-level control scheme. The SiLago HLS tool also generates a valid GDSII macro whose average energy, area, and performance are not estimated but known with post-layout accuracy thanks to the predictable SiLago hardware blocks. Moreover, the SiLago HLS tool is not intended for the end-user. It is designed to develop a library of algorithm implementations used by the application-level synthesis (ALS) tool in the SiLago framework. The application is defined as a hierarchy of algorithms. This library would include algorithms that vary in their function, dimension, and degree of parallelism. The ALS tool explores the design space in terms of number and type of algorithm implementation, rather than arithmetic resources, as HLS tools do.Algorithms are often developed by domain experts. For efficient implementation in hardware, such algorithms often need to be optimized with the hardware platform in mind. Two algorithm instances have been chosen for demonstration purposes. The first instance is a self-organizing map (SOM) based genome recognition algorithm. The second example concerns a complex model of cortex called Bayesian confidence propagation neural network (BCPNN). As developed by computational neuroscientists, the original model demands too much memory storage and memory access.This thesis addresses the latter two components because the first component has been addressed in the literature. We will first demonstrate the design of the SiLago HLS tool to support the hardware features like the distributed two-level control system. Moreover, we will use the two complex algorithm instances -- SOM and BCPNN, to demonstrate both general-purpose and algorithm-specific hardware-oriented algorithm optimization techniques. With the research carried out in this thesis, the SiLago HLS framework is greatly improved.
|
|
| 2. |
- Yu, Yang
(författare)
-
Alzheimer disease : a super-resolved picture of the amyloid β-peptide producing machinery
- 2021
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Alzheimer disease (AD) is the most common neurodegenerative disease, influencing 40 million patients in the world. Pathologically, AD is hallmarked by intracellular tau tangles and extracellular Amyloid-β peptide (Aβ) plaques. Aβ is generated from the amyloid precursor protein (APP) and its C-terminal fragment (APP-CTF). Aβ is involved in the pathology of AD and has been studied for over 30 years. However, detailed subcellular information of Aβ and the APP processing pathway is still lacking. In this thesis, we use super-resolution microscopy, which allows approximately ten times higher resolution than traditional confocal microscopy and therefore can be used to study the subcellular localization of full-length APP (FL-APP), APP-CTF, and Aβ in mature primary cultured mouse hippocampal neurons from both wild type and a disease model (AppNL-F knock in mice). In Paper I, we showed super-resolution microscopy is a powerful technique to study the subcellular localization of Aβ42 in neurons, especially in synapses. With super-resolution microscopy, we can observe lots of details that cannot get by confocal microscopy, it is possible to distinguish between the pre- and the postsynapse and resolve the synaptic cleft. We found that Aβ42 was enriched in the presynaptic side, but not in the postsynaptic side. Aβ42 was only partly colocalized with synaptophysin, which means that some Aβ42 is localized to other vesicles besides the synaptic vesicles. In Paper II, we continued to illustrate the subcellular localization of Aβ42, FL-APP, APP-CTF and N-terminal fragment (APP-NTF) in hippocampal neurons and showed that Aβ42 was present in early endosomes in immature presynapse. In mature synapses, partly Aβ42 was localized in clathrin coated vesicles. Interestingly, in the soma region, Aβ42 was not present in early endosomes, but in late endosome/lysosome and autophagosome. We also found that some FL-APP, and a high percentage of APP-CTF/-NTF were present in early endosomes in the soma region. In the synapse, only APP-CTF, but not FL-APP, was located in presynapse, in line with the presynaptic localization of Aβ42. In Paper III, we investigated the subcellular localization of APP-CTF and Aβ42 in AppNL-F knock in neurons. We found that mature and immature APP levels were altered in AppNL-F neurons. In Paper IV, we used iPALM microscopy to further characterize the Aβ42 containing vesicles in the presynaptic regions of hippocampal neurons. With this technique, the resolution along the z-axis was improved, and therefore, we can get more details in three dimensions. Overall, our studies use different types of super-resolution microscopy techniques to reveal the subcellular localization of FL-APP, APP-CTF, and Aβ. Subcellular details of APP amyloidogenic pathway in hippocampal neurons were illustrated, such as APP-CTF and Aβ42 are only enriched in presynapse, but not postsynapse; mostly Aβ42 in presynapse has different pool rather synaptic vesicles; No Aβ42, but FL-APP located in early endosome. These finds will be helpful to find a high targeting to specific site, disease modifying drug for AD.
|
|
| 3. |
- Yu, Yang, 1991-
(författare)
-
Design and Security Analysis of TRNGs and PUFs
- 2022
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- True Random Number Generators (TRNGs) and Physical Unclonable Functions (PUFs) are two important types of cryptographic primitives. TRNGs create a hardware-based, non-deterministic noise that is often used for generating keys, initialization vectors, and nonces for various applications that require cryptographic protection. PUFs have been proposed as a tamper-resistant alternative to the traditional secret key generation and challenge-response authentication methods. A compromised TRNG or PUF can lead to a system-wide loss of security.The conventional TRNG or PUF designs are challenged by new attack vectors such as deep learning-based side-channel analysis. In this dissertation, we propose several new PUF and TRNG designs and evaluations of their performance and security.The first PUF we introduce is called threshold PUF. We show that, in principle, any n-input threshold logic gate can be used as a base for building an n-input PUF. We implement and evaluate a threshold PUF based on recently proposed threshold logic flip-flops using SPICE simulation as a proof of concept. Threshold PUFs open up the possibility of using the rich body of knowledge on threshold logic implementations for designing PUFs. The second proposed design is a lightweight PUF construction called CRC-PUF, which focuses on protecting PUFs against machine learning-based modeling attacks. In CRC-PUF, input challenges are de-synchronized from output responses to make the PUF model difficult to learn. The input transformation which does the de-synchronization is based on a Cyclic Redundancy Check (CRC), thus the name CRC-PUF. By changing the CRC generator polynomial for each new response, we assure that recovering the transforming challenge has a success probability of at most 2-86 for 128-bit challenge-response pairs.The first TRNG design we introduce is based on a Non-Linear Feedback Ring Oscillator (NLFRO). The proposed NLFRO-TRNG structure harvests randomness from noise and unpredictable variations in delay cells and bi-stable elements, which is further amplified by the formation of non-linear feedback loops. The NLFRO outputs have chaotic behavior, allowing the construction of TRNGs with high entropy and speed. We implement three NLFRO-TRNGs on FPGA and evaluate the properties of the implementations with the NIST 800-90B entropy estimation and NIST 800-22 statistical test suits. The second proposed TRNG design is based on a strong PUF. The PUF based TRNG exploits the inherent determinism of PUF to enable in-field testing of the entropy sources by known answer tests. We present a prototype FPGA implementation of the proposed TRNG based on an arbiter PUF that passes all NIST 800-22 statistical tests and has the minimal entropy of 0.918 estimated according to NIST 800-90B recommendations.Apart from TRNG and PUF designs, it is crucial to consider potential attack vectors that can be created leveraging recently emerged technologies. To that end, in the second part of this dissertation, we introduce a novel attack on FPGA-based PUF and TRNG implementations that combines bitstream modification along with deep learning-based side-channel analysis. We evaluate this new attack vector on the design of an arbiter PUF and a ring oscillator-based TRNG implemented on Xilinx Artix-7 28nm FPGAs. In both cases, we are able to achieve close to 100% classification accuracy to recover the output or response. In the case of the arbiter PUF, the attack can even overcome countermeasures that are based on encrypting the challenges or responses.With such potent attack vectors readily available, the construction of strong countermeasures is necessary. Unfortunately, many of the state-of-the-art countermeasures are one-sided. In the final part of the dissertation, we use a countermeasure proposed for the protection of the Advanced Encryption Standard as an example. We conduct experiments and conclude that it can assist another type of side-channel attack that is not considered by the countermeasure.
|
|
| 4. |
- Yu, Yang, 1970-
(författare)
-
Molecular Mechanisms Underlying Abnormal Placentation in the Mouse
- 2007
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Placental development can be disturbed by various factors, such as mutation of specific genes or maternal diabetes. Our previous work on interspecies hybrid placental dysplasia (IHPD) and two additional models of placental hyperplasia, cloned mice and Esx1 mutants, showed that many genes are deregulated in placental dysplasia. Two of these candidate placentation genes, Cpe and Lhx3, were further studied. We performed in situ hybridization to determine their spatio-temporal expression in the placentas and placental phenotypes were analyzed in mutant mice. Our results showed that the placental phenotype in Cpe mutant mice mimics some IHPD phenotypes. Deregulated expression of Cpe and Cpd, a functionally equivalent gene, prior to the manifestation of the IHPD phenotype, indicated that Cpe and Cpd are potentially causative genes in IHPD. Lhx3 mutants lacked any placental phenotype. Deletion of Lhx3 and Lhx4 together caused an inconsistent placental phenotype which did not affect placental lipid transport function or expression of Lhx3/Lhx4 target genes. Down regulation of Lhx3/Lhx4 did not rescue the placental phenotype of AT24 mice and hence could be excluded as causative genes in IHPD. Analysis of placental development in diabetic mice showed that severe maternal diabetes resulted in fetal intrauterine growth restriction (IUGR) without any change in placental weight and lipid transport function. The diabetic placentas however exhibited abnormal morphology. Gene expression profiling identified some genes that might contribute to diabetic pathology. In another study, it was found that the heterochromatin protein CBX1 is required for normal placentation, as deletion of the gene caused consistent spongiotrophoblast and labyrinthine phenotypes. Gene expression profiling and spatio-temporal expression analysis showed that several genes with known function in placental development were deregulated in the Cbx1 null placenta.
|
|
| 5. |
|
|
| 6. |
- Yu, Yang, 1987-
(författare)
-
Understanding Composition–Structure–Bioactivity Correlations in Bioactive Glasses
- 2018
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Bioactive glasses integrate with bone/tooth tissues by forming a layer of hydroxy-carbonate apatite (HCA), which mimics the composition of bone mineral. In the current thesis, we investigated composition–structure–bioactivity correlations of phosphosilicate and borophosphosilicate (BPS) glasses. Bioactive phosphosilicate glasses extend the compositional space of the ”45S5 Bioglass®”, which has been in clinical use for decades. Recently developed bioactive BPS glasses with SiO2→B2O3 substitutions transform more completely into HCA and their glass dissolution behaviors can be tuned by varying the relative contents of B and Si. It is known that the average silicate network connectivity NSi and the phosphate content (x(P2O5)) affect the apatite formation (in vitro bioactivity) of phosphosilicate glasses, but the details remain poorly explored. Three series of phosphosilicate glasses were designed by independently varying NSi and x(P2O5). After immersion of the glasses in a simulated body fluid (SBF) for 24 hours, different degrees of their apatite formation were quantified by Fourier-transform infrared (FTIR) spectroscopy. The results revealed that a high P content widened the NSi range that generated optimum amounts of apatite and also mitigated the detrimental effects associated with using glass particles with < 50 μm. The amounts of apatite derived from FTIR agreed with those from 31P magic angle spinning (MAS) nuclear magnetic resonance (NMR) spectroscopy. The growth of apatite at bioactive glass surfaces was found to follow a sigmoidal growth model, in which the precursor phase, amorphous calcium phosphate (ACP), formed in the induction period and then crystallized into HCA in the following proliferation period, with an improvement in the structural ordering of HCA in the maturation period. This formation process closely resembles the apatite precipitated spontaneously from supersaturated Ca/P-containing solutions. The simultaneous growth of ACP and HCA is discussed in conjunction with a previously proposed mechanism for explaining in vitro bioactivity and apatite growth from bioactive glasses. The short- and medium- range structures of bioactive borophosphosilicate (BPS) glasses were investigated by solid-state MAS NMR. Two series of BPS glasses were designed by gradually replacing SiO2 with B2O3 in the 45S5 glass, as well as another base glass featuring a more condensed glass network. As the B2O3 content is increased, the glass networks become more polymerized, together with decreased fractions of the dominating BO3 and orthophosphate units. Borate groups are homogeneously mixed with the isolated orthophosphate groups, while the remaining phosphate groups exhibit a slight preference for bonding to BO4 over SiO4 units. Linkages among borate groups are dominated by B[3]–O–B[4] linkages at the expenses of B[3]–O–B[3] and B[4]–O–B[4] linkages, with the latter B[4]–O–B[4] motifs disfavored yet abundant. A similar fashion of borate mixing was observed in P-free Na/Ca-based borosilicate glasses that span a large compositional space. The content of B[4]–O–B[4] linkages was found to be controlled by the relative fractions of BO4 groups and non-bridging oxygen ions.
|
|
