arXiv subjects

Soohyun Park

Publications and source records attributed to Soohyun Park.

At least 55 records · Page 3Linked to original sources

arXiv2022

Age-of-Information Aware Contents Caching and Distribution for Connected Vehicles

To support rapid and accurate autonomous driving services, road environment information, which is difficult to obtain through vehicle sensors themselves, is collected and utilized through communication with surrounding infrastructure in connected vehicle networks. For this reason, we consider a scenario that utilizes infrastructure such as road side units (RSUs) and macro base station (MBS) in situations where caching of road environment information is required. Due to the rapidly changed road environment, a concept which represents a freshness of the road content, age of information (AoI), is important. Based on the AoI value, in the connected vehicle system, it is essential to keep appropriate content in the RSUs in advance, update it before the content is expired, and send the content to the vehicles which want to use it. However, too frequent content transmission for the minimum AoI leads to indiscriminate use of network resources. Furthermore, a transmission control, that content AoI and service delay are not properly considered adversely, affects user service. Therefore, it is important to find an appropriate compromise. For these reasons, the objective of this paper is about to reduce the system cost used for content delivery through the proposed system while minimizing the content AoI presented in MBS, RSUs and UVs. The transmission process, which is able to be divided into two states, i.e., content caching and service, is approached using Markov decision process (MDP) and Lyapunov optimization framework, respectively, which guarantee optimal solutions, as verified via data-intensive performance evaluation.

cs.DC↗

arXiv2022

Self-Configurable Stabilized Real-Time Detection Learning for Autonomous Driving Applications

Guaranteeing real-time and accurate object detection simultaneously is paramount in autonomous driving environments. However, the existing object detection neural network systems are characterized by a tradeoff between computation time and accuracy, making it essential to optimize such a tradeoff. Fortunately, in many autonomous driving environments, images come in a continuous form, providing an opportunity to use optical flow. In this paper, we improve the performance of an object detection neural network utilizing optical flow estimation. In addition, we propose a Lyapunov optimization framework for time-average performance maximization subject to stability. It adaptively determines whether to use optical flow to suit the dynamic vehicle environment, thereby ensuring the vehicle's queue stability and the time-average maximum performance simultaneously. To verify the key ideas, we conduct numerical experiments with various object detection neural networks and optical flow estimation networks. In addition, we demonstrate the self-configurable stabilized detection with YOLOv3-tiny and FlowNet2-S, which are the real-time object detection network and an optical flow estimation network, respectively. In the demonstration, our proposed framework improves the accuracy by 3.02%, the number of detected objects by 59.6%, and the queue stability for computing capabilities.

cs.CV↗

arXiv2022

Characterizing cubic hypersurfaces via projective geometry

We use the cut and paste relation $[Y^{[2]}] = [Y][\mathbb{P}^m] + \mathbb{L}^2 [F(Y)]$ in $K_0(\text{Var}_k)$ of Galkin--Shinder for cubic hypersurfaces arising from projective geometry to characterize cubic hypersurfaces of sufficiently high dimension under certain numerical or genericity conditions. Removing the conditions involving the middle Betti number from the numerical conditions used extends the possible $Y$ to cubic hypersurfaces, complete intersections of two quadric hypersurfaces, or complete intersections of two quartic hypersurfaces. The same method also gives a family of other cut and paste relations that can only possibly be satisfied by cubic hypersurfaces.

math.AG↗

arXiv2022

Simplicial chromatic polynomials as Hilbert series of Stanley--Reisner rings

We find families of simplicial complexes where the simplicial chromatic polynomials defined by Cooper--de Silva--Sazdanovic \cite{CdSS} are Hilbert series of Stanley--Reisner rings of auxiliary simplicial complexes. As a result, such generalized chromatic polynomials are determined by $h$-vectors of auxiliary simplicial complexes. In addition to generalizing related results on graphs and matroids, the simplicial complexes used allow us to consider problems that are not necessarily analogues of those considered for graphs. Some examples include supports of cyclotomic polynomials, log concavity properties of a polynomial or some translate of the polynomial, and symmetry relations between a polynomial and its reciprocal polynomial. If the $h$-vectors involed have sufficiently large entries, the Hilbert series are Hilbert polynomials of some $k$-algebra. As a consequence of connections between $h$-vectors and simplicial chromatic polynomials, we also find simplicial complexes whose $h$-vectors are determined by addition-contraction relations of simplicial complexes analogous to deletion-contraction relations of graphs. The constructions used involve generalizations of relations Euler characteristics of configuration spaces and chromatic polynomials of graphs.

math.CO↗

arXiv2022

Matroidal Cayley-Bacharach and independence/dependence of geometric properties of matroids

We consider the relationship between a matroidal analogue of the degree $a$ Cayley-Bacharach property (finite sets of points failing to impose independent conditions on degree $a$ hypersurfaces) and geometric properties of matroids. If the matroid polytopes in question are nestohedra, we show that the minimal degree matroidal Cayley-Bacharach property denoted $MCB(a)$ is determined by the structure of the building sets used to construct them. This analysis also applies for other degrees $a$. Also, it does not seem to affect the combinatorial equivalence class of the matroid polytope. However, there are close connections to minimal nontrivial degrees $a$ and the geometry of the matroids in question for paving matroids (which are conjecturally generic among matroids of a given rank) and matroids constructed out of supersolvable hyperplane arrangements. The case of paving matroids is still related to with properties of building sets since it is closely connected to (Hilbert series of) Chow rings of matroids, which are combinatorial models of the cohomology of wonderful compactifications. Finally, our analysis of supersolvable line and hyperplane arrangements give a family of matroids which are natrually related to independence conditions imposed by points one plane curves or can be analyzed recursively.

math.CO↗

arXiv2022

Carleson embedding theorem for an exponential Bergman space on the unit ball

We characterize the Carleson measures for an exponential Bergman space on the unit ball of $\mathbb C^n$ in terms of the ball induced by the complex Hessian of the logarithm of the weight function. The boundedness (or compactness) of integral operators, Cesàro operators and Toeplitz operators, is given using the Carleson measure (or vanishing Carleson measure) characterization.

math.CV↗

arXiv2022

Anti-Ramsey theory problems, lattice point counts on polytopes, and Hodge structures on the cohomology of toric varieties

We find families of graphs $G$ and subgraphs $H$ of $G$ such that the number of edge colorings of $G$ avoiding a monochromatic coloring of $H$ is determined by lattice point counts or a Hodge structure on the cohomology of a certain toric variety. In general, this gives a class of ``anti-Ramsey theory problems'' with a geometric structure. For example, we find one for Ramsey numbers of classes of such graphs. The key observation is that our previous result expressing simplicial chromatic polynomials in terms of $h$-vectors of auxiliary simplicial complexes can be reinterpreted as one on edge colorings of graphs avoiding monochromatic colorings of specified forbidden subgraphs. Specializing to simplicial complexes arising from triangulations of polytopes (e.g. unimodular triangulations), we obtain families of graphs and forbidden subgraphs where edge colorings avoiding monochromatic colorings of the forbidden subgraphs depend on lattice point counts or Hodge structures on the cohomology of toric varieties.

math.CO↗

arXiv2022

Tutorial on Course-of-Action (COA) Attack Search Methods in Computer Networks

In the literature of modern network security research, deriving effective and efficient course-of-action (COA) attach search methods are of interests in industry and academia. As the network size grows, the traditional COA attack search methods can suffer from the limitations to computing and communication resources. Therefore, various methods have been developed to solve these problems, and reinforcement learning (RL)-based intelligent algorithms are one of the most effective solutions. Therefore, we review the RL-based COA attack search methods for network attack scenarios in terms of the trends and their contrib

cs.AI↗

arXiv2022

Quality-Aware Real-Time Augmented Reality Visualization under Delay Constraints

Augmented reality (AR) is one of emerging applications in modern multimedia systems research. Due to intensive time-consuming computations for AR visualization in mobile devices, quality-aware real-time computing under delay constraints is essentially required. Inspired by Lyapunov optimization framework, this paper proposes a time-average quality maximization method for the AR visualization under delay considerations.

cs.DC↗

arXiv2022

AoI-Aware Markov Decision Policies for Caching

We consider a scenario that utilizes road side units (RSUs) as distributed caches in connected vehicular networks. The goal of the use of caches in our scenario is for rapidly providing contents to connected vehicles under various traffic conditions. During this operation, due to the rapidly changed road environment and user mobility, the concept of age-of-information (AoI) is considered for (1) updating the cached information as well as (2) maintaining the freshness of cached information. The frequent updates of cached information maintain the freshness of the information at the expense of network resources. Here, the frequent updates increase the number of data transmissions between RSUs and MBS; and thus, it increases system costs, consequently. Therefore, the tradeoff exists between the AoI of cached information and the system costs. Based on this observation, the proposed algorithm in this paper aims at the system cost reduction which is fundamentally required for content delivery while minimizing the content AoI, based on Markov Decision Process (MDP) and Lyapunov optimization.

cs.DC↗

arXiv2022

Motivic limits for Fano varieties of $k$-planes

We study the probability that an $(n - m)$-dimensional linear subspace in $\mathbb{P}^n$ or a collection of points spanning such a linear subspace is contained in an $m$-dimensional variety $Y \subset \mathbb{P}^n$. This involves a strategy used by Galkin--Shinder to connect properties of a cubic hypersurface to its Fano variety of lines via cut and paste relations in the Grothendieck ring of varieties. Generalizing this idea to varieties of higher codimension and degree, we can measure growth rates of weighted probabilities of $k$-planes contained in a sequence of varieties with varying initial parameters over a finite field. In the course of doing this, we move an identity motivated by rationality problems involving cubic hypersurfaces to a motivic statistics setting associated with cohomological stability.

math.AG↗

arXiv2022

Graph coloring-related properties of (generating functions of) Hodge-Deligne polynomials

Motivated by a connection between the topology of (generalized) configuration spaces and chromatic polynomials, we show that generating functions of Hodge-Deligne polynomials of quasiprojective varieties and colorings of acyclic directed graphs with the complete graph $K_n$ as the underlying undirected graph. In order to do this, we combine an interpretation of Crew-Spirkl for plethysms involving chromatic symmetric polynomials using colorings of directed acyclic graphs to plethystic exponentials often appearing Hodge-Deligne polynomials of varieties. Applying this to a recent result of Florentino-Nozad-Zamora, we find a connection between $GL_n$-character varieties of finitely presented groups and colorings of these "complete" directed graphs by (signed) variables used in the Hodge-Deligne polynomials of the irreducible representations. As a consequence of the constructions used, generators of Hodge-Deligne polynomials carry over to generators of the plethystic exponentials as a power series in a way that is compatible with this graph coloring interpretation and gives interesting combinatorial information that simplifies when we consider Hodge-Deligne polynomials of birationally equivalent varieties to be equivalent. Finally, we give an application of methods used to symmetries of Hodge-Deligne polynomials of varieties and their configuration spaces and their relation to chromatic symmetric polynomials.

math.CO↗

arXiv2022

Cooperative Multi-Agent Deep Reinforcement Learning for Reliable Surveillance via Autonomous Multi-UAV Control

CCTV-based surveillance using unmanned aerial vehicles (UAVs) is considered a key technology for security in smart city environments. This paper creates a case where the UAVs with CCTV-cameras fly over the city area for flexible and reliable surveillance services. UAVs should be deployed to cover a large area while minimize overlapping and shadow areas for a reliable surveillance system. However, the operation of UAVs is subject to high uncertainty, necessitating autonomous recovery systems. This work develops a multi-agent deep reinforcement learning-based management scheme for reliable industry surveillance in smart city applications. The core idea this paper employs is autonomously replenishing the UAV's deficient network requirements with communications. Via intensive simulations, our proposed algorithm outperforms the state-of-the-art algorithms in terms of surveillance coverage, user support capability, and computational costs.

eess.SY↗

arXiv2021

Trends in LEO Satellite Handover Algorithms

In this paper, we review well-known handovers algorithms in satellite environment. The modern research trends and contributions are proposed and summarized in order to overcome their considering problems in satellite-air-ground integrated network environment caused by the fast movement of Low Earth Orbit (LEO) satellite and related frequent handover occurrences.

cs.NI↗

arXiv2021

Secure Aerial Surveillance using Split Learning

Personal monitoring devices such as cyclist helmet cameras to record accidents or dash cams to catch collisions have proliferated, with more companies producing smaller and compact recording gadgets. As these devices are becoming a part of citizens' everyday arsenal, concerns over the residents' privacy are progressing. Therefore, this paper presents SASSL, a secure aerial surveillance drone using split learning to classify whether there is a presence of a fire on the streets. This innovative split learning method transfers CCTV footage captured with a drone to a nearby server to run a deep neural network to detect a fire's presence in real-time without exposing the original data. We devise a scenario where surveillance UAVs roam around the suburb, recording any unnatural behavior. The UAV can process the recordings through its on-mobile deep neural network system or transfer the information to a server. Due to the resource limitations of mobile UAVs, the UAV does not have the capacity to run an entire deep neural network on its own. This is where the split learning method comes in handy. The UAV runs the deep neural network only up to the first hidden layer and sends only the feature map to the cloud server, where the rest of the deep neural network is processed. By ensuring that the learning process is divided between the UAV and the server, the privacy of raw data is secured while the UAV does not overexert its minimal resources.

cs.DC↗

arXiv2021

Joint Mobile Charging and Coverage-Time Extension for Unmanned Aerial Vehicles

In modern networks, the use of drones as mobile base stations (MBSs) has been discussed for coverage flexibility. However, the realization of drone-based networks raises several issues. One of the critical issues is drones are extremely power-hungry. To overcome this, we need to characterize a new type of drones, so-called charging drones, which can deliver energy to MBS drones. Motivated by the fact that the charging drones also need to be charged, we deploy ground-mounted charging towers for delivering energy to the charging drones. We introduce a new energy-efficiency maximization problem, which is partitioned into two independently separable tasks. More specifically, as our first optimization task, two-stage charging matching is proposed due to the inherent nature of our network model, where the first matching aims to schedule between charging towers and charging drones while the second matching solves the scheduling between charging drones and MBS drones. We analyze how to convert the formulation containing non-convex terms to another one only with convex terms. As our second optimization task, each MBS drone conducts energy-aware time-average transmit power allocation minimization subject to stability via Lyapunov optimization. Our solutions enable the MBS drones to extend their lifetimes; in turn, network coverage-time can be extended.

eess.SY↗

arXiv2021

Deep learning enables reference-free isotropic super-resolution for volumetric fluorescence microscopy

Volumetric imaging by fluorescence microscopy is often limited by anisotropic spatial resolution from inferior axial resolution compared to the lateral resolution. To address this problem, here we present a deep-learning-enabled unsupervised super-resolution technique that enhances anisotropic images in volumetric fluorescence microscopy. In contrast to the existing deep learning approaches that require matched high-resolution target volume images, our method greatly reduces the effort to put into practice as the training of a network requires as little as a single 3D image stack, without a priori knowledge of the image formation process, registration of training data, or separate acquisition of target data. This is achieved based on the optimal transport driven cycle-consistent generative adversarial network that learns from an unpaired matching between high-resolution 2D images in lateral image plane and low-resolution 2D images in the other planes. Using fluorescence confocal microscopy and light-sheet microscopy, we demonstrate that the trained network not only enhances axial resolution, but also restores suppressed visual details between the imaging planes and removes imaging artifacts.

cs.CV↗

arXiv2020

Reinforced Edge Selection using Deep Learning for Robust Surveillance in Unmanned Aerial Vehicles

In this paper, we propose a novel deep Q-network (DQN)-based edge selection algorithm designed specifically for real-time surveillance in unmanned aerial vehicle (UAV) networks. The proposed algorithm is designed under the consideration of delay, energy, and overflow as optimizations to ensure real-time properties while striking a balance for other environment-related parameters. The merit of the proposed algorithm is verified via simulation-based performance evaluation.

cs.DC↗