Matching Items (7)
Filtering by
- All Subjects: Neuroscience
- All Subjects: neural net
- Creators: Bliss, Daniel
- Creators: Marballi, Ketan
- Member of: Barrett, The Honors College Thesis/Creative Project Collection
- Resource Type: Text
Leveraging Machine Learning and Wireless Sensing for Robot Localization - Location Variance Analysis
Description
Object localization is used to determine the location of a device, an important aspect of applications ranging from autonomous driving to augmented reality. Commonly-used localization techniques include global positioning systems (GPS), simultaneous localization and mapping (SLAM), and positional tracking, but all of these methodologies have drawbacks, especially in high traffic indoor or urban environments. Using recent improvements in the field of machine learning, this project proposes a new method of localization using networks with several wireless transceivers and implemented without heavy computational loads or high costs. This project aims to build a proof-of-concept prototype and demonstrate that the proposed technique is feasible and accurate.
Modern communication networks heavily depend upon an estimate of the communication channel, which represents the distortions that a transmitted signal takes as it moves towards a receiver. A channel can become quite complicated due to signal reflections, delays, and other undesirable effects and, as a result, varies significantly with each different location. This localization system seeks to take advantage of this distinctness by feeding channel information into a machine learning algorithm, which will be trained to associate channels with their respective locations. A device in need of localization would then only need to calculate a channel estimate and pose it to this algorithm to obtain its location.
As an additional step, the effect of location noise is investigated in this report. Once the localization system described above demonstrates promising results, the team demonstrates that the system is robust to noise on its location labels. In doing so, the team demonstrates that this system could be implemented in a continued learning environment, in which some user agents report their estimated (noisy) location over a wireless communication network, such that the model can be implemented in an environment without extensive data collection prior to release.
Modern communication networks heavily depend upon an estimate of the communication channel, which represents the distortions that a transmitted signal takes as it moves towards a receiver. A channel can become quite complicated due to signal reflections, delays, and other undesirable effects and, as a result, varies significantly with each different location. This localization system seeks to take advantage of this distinctness by feeding channel information into a machine learning algorithm, which will be trained to associate channels with their respective locations. A device in need of localization would then only need to calculate a channel estimate and pose it to this algorithm to obtain its location.
As an additional step, the effect of location noise is investigated in this report. Once the localization system described above demonstrates promising results, the team demonstrates that the system is robust to noise on its location labels. In doing so, the team demonstrates that this system could be implemented in a continued learning environment, in which some user agents report their estimated (noisy) location over a wireless communication network, such that the model can be implemented in an environment without extensive data collection prior to release.
ContributorsChang, Roger (Co-author) / Kann, Trevor (Co-author) / Alkhateeb, Ahmed (Thesis director) / Bliss, Daniel (Committee member) / Electrical Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2020-05
Description
Epilepsy affects numerous people around the world and is characterized by recurring seizures, prompting the ability to predict them so precautionary measures may be employed. One promising algorithm extracts spatiotemporal correlation based features from intracranial electroencephalography signals for use with support vector machines. The robustness of this methodology is tested through a sensitivity analysis. Doing so also provides insight about how to construct more effective feature vectors.
ContributorsMa, Owen (Author) / Bliss, Daniel (Thesis director) / Berisha, Visar (Committee member) / Barrett, The Honors College (Contributor) / Electrical Engineering Program (Contributor)
Created2015-05
Description
Egr3 is an immediate early gene transcription factor that shows genetic association with schizophrenia, and is found in decreased levels in the brains of schizophrenia patients. Schizophrenia patients also exhibit cognitive and memory deficits, both of which Egr3 has been shown to play a crucial role in. Additionally, high levels of DNA damage are found in the brains of schizophrenia patients. A recent study has shown that DNA damage occurs as a result of normal physiological activity in neurons and is required for induction of gene expression of a subset of early response genes. Also, failure to repair this damage can lead to gene expression in a constitutive switched on state. Egr3 knockout (Egr3-/-) mice show deficits in hippocampal synaptic plasticity and memory. We were interested in characterizing downstream targets of EGR3 in the hippocampus. To determine these targets, electroconvulsive seizure (ECS) was carried out in Egr3 -/- versus wild type (WT) mice, and a microarray study was first done in our lab. ECS maximally stimulates Egr3 expression and we hypothesized that there would be gene targets that are differentially expressed between Egr3 -/- and WT mice that had been subjected to ECS. Two separate analyses of the microarray yielded 65 common genes that were determined as being differentially expressed between WT and Egr3 -/- mice after ECS. Further Ingenuity Pathway Analysis of these 65 genes indicated the Gadd45 signaling pathway to be the top canonical pathway, with the top four pathways all being associated with DNA damage or DNA repair. A literature survey was conducted for these 65 genes and their associated pathways, and 12 of the 65 genes were found to be involved in DNA damage response and/or DNA repair. Validation of differential expression was then conducted for each of the 12 genes, in both the original male cohort used for microarray studies and an additional female cohort of mice. 7 of these genes validated through quantitative real time PCR (qRT-PCR) in the original male cohort used for the microarray study, and 4 validated in both the original male cohort and an independent female cohort. Bioinformatics analysis yielded predicted EGR3 binding sites in promoters of these 12 genes, validating their role as potential transcription targets of EGR3. These data reveal EGR3 to be a novel regulator of DNA repair. Further studies will be needed to characterize the role of Egr3 in repairing DNA damage.
ContributorsBarkatullah, Arhem Fatima (Author) / Newbern, Jason (Thesis director) / Gallitano, Amelia (Committee member) / Marballi, Ketan (Committee member) / School of Life Sciences (Contributor) / Barrett, The Honors College (Contributor)
Created2018-05
Description
Multiple-channel detection is considered in the context of a sensor network where data can be exchanged directly between sensor nodes that share a common edge in the network graph. Optimal statistical tests used for signal source detection with multiple noisy sensors, such as the Generalized Coherence (GC) estimate, use pairwise measurements from every pair of sensors in the network and are thus only applicable when the network graph is completely connected, or when data are accumulated at a common fusion center. This thesis presents and exploits a new method that uses maximum-entropy techniques to estimate measurements between pairs of sensors that are not in direct communication, thereby enabling the use of the GC estimate in incompletely connected sensor networks. The research in this thesis culminates in a main conjecture supported by statistical tests regarding the topology of the incomplete network graphs.
ContributorsCrider, Lauren Nicole (Author) / Cochran, Douglas (Thesis director) / Renaut, Rosemary (Committee member) / Kosut, Oliver (Committee member) / Barrett, The Honors College (Contributor) / School of Mathematical and Statistical Sciences (Contributor)
Created2014-05
Description
Lossy compression is a form of compression that slightly degrades a signal in ways that are ideally not detectable to the human ear. This is opposite to lossless compression, in which the sample is not degraded at all. While lossless compression may seem like the best option, lossy compression, which is used in most audio and video, reduces transmission time and results in much smaller file sizes. However, this compression can affect quality if it goes too far. The more compression there is on a waveform, the more degradation there is, and once a file is lossy compressed, this process is not reversible. This project will observe the degradation of an audio signal after the application of Singular Value Decomposition compression, a lossy compression that eliminates singular values from a signal’s matrix.
ContributorsHirte, Amanda (Author) / Kosut, Oliver (Thesis director) / Bliss, Daniel (Committee member) / Electrical Engineering Program (Contributor, Contributor) / Barrett, The Honors College (Contributor)
Created2021-05
Description
Immediate early genes (IEGs) are rapidly activated in response to an environmental stimulus, and most code for transcription factors that mediate processes of synaptic plasticity, learning, and memory. EGR3, an immediate early gene transcription factor, is a mediator of biological processes that are disrupted in patients with schizophrenia (SCZ). A microarray experiment conducted by our lab revealed that Egr3 also regulates genes involved in DNA damage response. A recent study revealed that physiological neuronal activity results in the formation of DNA double-stranded breaks (DSBs) in the promoters of IEGs. Additionally, they showed that these DSBs are essential for inducing the expression of IEGs, and failure to repair these DSBs results in the persistent expression of IEGs. We hypothesize that Egr3 plays a role in repairing activity- induced DNA DSBs, and mice lacking Egr3 should have an abnormal accumulation of these DSBs. Before proceeding with that experiment, we conducted a preliminary investigation to determine if electroconvulsive stimulation (ECS) is a reliable method of inducing activity- dependent DNA damage, and to measure this DNA damage in three subregions of the hippocampus: CA1, CA3, and dentate gyrus (DG). We asked the question, are levels of DNA DSBs different between these hippocampal subregions in animals at baseline and following electroconvulsive stimulation (ECS)? To answer this question, we quantified γ-H2AX, a biomarker of DNA DSBs, in the hippocampal subregions of wildtype mice. Due to technical errors and small sample size, we were unable to substantiate our preliminary findings. Despite these shortcomings, our experimental design can be modified in future studies that investigate the role of Egr3 in activity-induced DNA damage repair.
ContributorsKhoshaba, Rami Samuel (Author) / Newbern, Jason (Thesis director) / Gallitano, Amelia (Committee member) / Marballi, Ketan (Committee member) / School of Molecular Sciences (Contributor) / School of Life Sciences (Contributor) / Barrett, The Honors College (Contributor)
Created2020-05
Description
Damage to DNA can affect the genes it encodes; if this damage is not repaired, abnormal proteins may be produced and cellular functions may be disturbed. DNA damage has been implicated in the initiation and progression of a variety of diseases. Conversely, DNA damage has also been discovered to contribute to beneficial biological processes. Madabhushi and colleagues (2015) determined that activity-dependent DNA double strand breaks (DSBs) in the promoter region of immediate early genes (IEGs) induced their expression. EGR3 is an IEG transcription factor which regulates the expression of growth factors and synaptic plasticity-associated genes. In a previously conducted microarray experiment, it was revealed that EGR3 regulates the expression of genes associated with DNA repair such as Cenpa and Nr4a2. These findings inspired us to investigate if EGR3 affects DNA repair in vivo. Before conducting this experiment, we sought to standardize and optimize a method of inducing DNA damage in the hippocampus. Electroconvulsive stimulation (ECS) is utilized to induce neuronal activity. Since neuronal activity leads to the formation of DNA DSBs, we theorized that ECS could be used to induce DNA DSBs in the hippocampus. We predicted that mice that receive ECS would have more DNA DSBs than those that receive the sham treatment. Gamma H2AX, a biomarker for DNA damage, was utilized to quantify DNA DSBs. Gamma H2AX expression in the dentate gyrus, CA1 and CA3 regions of the hippocampus was compared between mice that received the sham treatment and mice that received ECS. Mice that received ECS were sacrificed either 1 or 2 hours post-administration, constituting treatment conditions of 1 hr post-ECS and 2 hrs post-ECS. Our results suggest that ECS has a statistically significant effect exclusively in the CA1 region of the hippocampus. However, our analyses may have been limited due to sample size. A power analysis was conducted, and the results suggest that a sample size of n=4 mice will be sufficient to detect significant differences across treatments in all three regions of the hippocampus. Ultimately, future studies with an increased sample size will need to be conducted to conclusively assess the use of ECS to induce DNA damage within the hippocampus.
ContributorsAden, Aisha Abubakar (Author) / Newbern, Jason (Thesis director) / Gallitano, Amelia (Thesis director) / Marballi, Ketan (Committee member) / School of Life Sciences (Contributor) / Barrett, The Honors College (Contributor)
Created2020-05