Source author record

Jong-Ho Lee

Jong-Ho Lee appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Information Theory math.IT eess.SP Neural and Evolutionary Computing

Catalog footprint

What is connected

4works

4topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2022arXiv

Joint Vehicle Tracking and RSU Selection for V2I Communications with Extended Kalman Filter

We develop joint vehicle tracking and road side unit (RSU) selection algorithms suitable for vehicle-to-infrastructure (V2I) communications. We first design an analytical framework for evaluating vehicle tracking systems based on the extended Kalman filter. A simple, yet effective, metric that quantifies the vehicle tracking performance is derived in terms of the angular derivative of a dominant spatial frequency. Second, an RSU selection algorithm is proposed to select a proper RSU that enhances the vehicle tracking performance. A joint vehicle tracking algorithm is also developed to maximize the tracking performance by considering sounding samples at multiple RSUs while minimizing the amount of sample exchange. The numerical results verify that the proposed vehicle tracking algorithms give better performance than conventional signal-to-noise ratio-based tracking systems.

preprint2021arXiv

Hardware Implementation of Spiking Neural Networks Using Time-To-First-Spike Encoding

Hardware-based spiking neural networks (SNNs) are regarded as promising candidates for the cognitive computing system due to low power consumption and highly parallel operation. In this work, we train the SNN in which the firing time carries information using temporal backpropagation. The temporally encoded SNN with 512 hidden neurons showed an accuracy of 96.90% for the MNIST test set. Furthermore, the effect of the device variation on the accuracy in temporally encoded SNN is investigated and compared with that of the rate-encoded network. In a hardware configuration of our SNN, NOR-type analog memory having an asymmetric floating gate is used as a synaptic device. In addition, we propose a neuron circuit including a refractory period generator for temporally encoded SNN. The performance of the 2-layer neural network consisting of synapses and proposed neurons is evaluated through circuit simulation using SPICE. The network with 128 hidden neurons showed an accuracy of 94.9%, a 0.1% reduction compared to that of the system simulation of the MNIST dataset. Finally, the latency and power consumption of each block constituting the temporal network is analyzed and compared with those of the rate-encoded network depending on the total time step. Assuming that the total time step number of the network is 256, the temporal network consumes 15.12 times lower power than the rate-encoded network and can make decisions 5.68 times faster.

preprint2020arXiv

Geometric Sequence Decomposition with $k$-simplexes Transform

This paper presents a computationally efficient technique for decomposing non-orthogonally superposed $k$ geometric sequences. The method, which is named as geometric sequence decomposition with $k$-simplexes transform (GSD-ST), is based on the concept of transforming an observed sequence to multiple $k$-simplexes in a virtual $k$-dimensional space and correlating the volumes of the transformed simplexes. Hence, GSD-ST turns the problem of decomposing $k$ geometric sequences into one of solving a $k$-th order polynomial equation. Our technique has significance for wireless communications because sampled points of a radio wave comprise a geometric sequence. This implies that GSD-ST is capable of demodulating randomly combined radio waves, thereby eliminating the effect of interference. To exemplify the potential of GSD-ST, we propose a new radio access scheme, namely non-orthogonal interference-free radio access (No-INFRA). Herein, GSD-ST enables the collision-free reception of uncoordinated access requests. Numerical results show that No-INFRA effectively resolves the colliding access requests when the interference is dominant.

preprint2010arXiv

Full-Duplex Relay based on Zero-Forcing Beamforming

In this paper, we propose a full-duplex relay (FDR) based on a zero-forcing beamforming (ZFBF) for a multiuser MIMO relay system. The ZFBF is employed at the base station to suppress both the self-interference of the relay and the multiuser interference at the same time. Numerical results show that the proposed FDR can enhance the sum rate performance as compared to the half-duplex relay (HDR), if sufficient isolation between the transmit and receive antennas is ensured at the relay.