arXiv subjects

Toshiaki Koike-Akino

Publications and source records attributed to Toshiaki Koike-Akino.

86 records · Page 5Linked to original sources

arXiv2019

Adversarial Deep Learning in EEG Biometrics

Deep learning methods for person identification based on electroencephalographic (EEG) brain activity encounters the problem of exploiting the temporally correlated structures or recording session specific variability within EEG. Furthermore, recent methods have mostly trained and evaluated based on single session EEG data. We address this problem from an invariant representation learning perspective. We propose an adversarial inference approach to extend such deep learning models to learn session-invariant person-discriminative representations that can provide robustness in terms of longitudinal usability. Using adversarial learning within a deep convolutional network, we empirically assess and show improvements with our approach based on longitudinally collected EEG data for person identification from half-second EEG epochs.

cs.LG↗

arXiv2019

Model-Driven Deep Learning Method for Jammer Suppression in Massive Connectivity Systems

We present a method for separating collided signals from multiple users in the presence of strong and wideband interference/jamming signal. More specifically, we consider a massive connectivity setup where few, out of a large number of users, equipped with spreading codes, synchronously transmit symbols. The received signal is a noisy mixture of symbols transmitted through users' flat fading channels, impaired by fast frequency hopping jamming signal of relatively large power. In the absence of any conventional technique suitable for the considered setup, we propose a "model-driven" deep learning method, based on convolution neural network, to suppress jamming signal from the received signal, and detect active users together with their transmitted symbols. A numerical study of the proposed method confirms its effectiveness in scenarios where classical techniques fail. As such, in a two user scenario with wideband jamming signal of power $20$ dB above the power any active user, the proposed algorithm achieves error rates $10^{-2}$ for a wide range of AWGN variances.

eess.SP↗

arXiv2019

Polar Coding with Chemical Reaction Networks

In this paper, we propose a new polar coding scheme with molecular programming, which is capable of highly parallel implementation at a nano-scale without a need of electrical power sources. We designed chemical reaction networks (CRN) to employ either successive cancellation (SC) or maximum-likelihood (ML) decoding schemes for short polar codes. From ordinary differential equation (ODE) analysis of the proposed CRNs, we demonstrate that SC and ML decoding achieve accurate computations across fully-parallel chemical reactions. We also make a comparison in terms of the number of required chemical reactions and species, where the superiority of ML decoder over SC decoder is observed for very short block lengths.

cs.IT↗

arXiv2019

Learning to Modulate for Non-coherent MIMO

The deep learning trend has recently impacted a variety of fields, including communication systems, where various approaches have explored the application of neural networks in place of traditional designs. Neural networks flexibly allow for data/simulation-driven optimization, but are often employed as black boxes detached from direct application of domain knowledge. Our work considers learning-based approaches addressing modulation and signal detection design for the non-coherent MIMO channel. We demonstrate that simulation-driven optimization can be performed while entirely avoiding neural networks, yet still perform comparably. Additionally, we show the feasibility of MIMO communications over extremely short coherence windows (i.e., channel coefficient stability period), with as few as two time slots.

eess.SP↗

arXiv2019

Deep Learning-Based Constellation Optimization for Physical Network Coding in Two-Way Relay Networks

This paper studies a new application of deep learning (DL) for optimizing constellations in two-way relaying with physical-layer network coding (PNC), where deep neural network (DNN)-based modulation and demodulation are employed at each terminal and relay node. We train DNNs such that the cross entropy loss is directly minimized, and thus it maximizes the likelihood, rather than considering the Euclidean distance of the constellations. The proposed scheme can be extended to higher level constellations with slight modification of the DNN structure. Simulation results demonstrate a significant performance gain in terms of the achievable sum rate over conventional relaying schemes. Furthermore, since our DNN demodulator directly outputs bit-wise probabilities, it is straightforward to concatenate with soft-decision channel decoding.

cs.IT↗

arXiv2019

HoloCast: Graph Signal Processing for Graceful Point Cloud Delivery

In conventional point cloud delivery, a sender uses octree-based digital video compression to stream three-dimensional (3D) points and the corresponding color attributes over band-limited links, e.g., wireless channels, for 3D scene reconstructions. However, the digital-based delivery schemes have an issue called cliff effect, where the 3D reconstruction quality is a step function in terms of wireless channel quality. We propose a novel scheme of point cloud delivery, called HoloCast, to gracefully improve the reconstruction quality with the improvement of wireless channel quality. HoloCast regards the 3D points and color components as graph signals and directly transmits linear-transformed signals based on graph Fourier transform (GFT), without digital quantization and entropy coding operations. One of main contributions in HoloCast is that the use of GFT can deal with non-ordered and non-uniformly distributed multi-dimensional signals such as holographic data unlike conventional delivery schemes. Performance results with point cloud data show that HoloCast yields better 3D reconstruction quality compared to digital-based methods in noisy wireless environment.

cs.MM↗

arXiv2019

Channel Decoding with Quantum Approximate Optimization Algorithm

Motivated by the recent advancement of quantum processors, we investigate quantum approximate optimization algorithm (QAOA) to employ quasi-maximum-likelihood (ML) decoding of classical channel codes. QAOA is a hybrid quantum-classical variational algorithm, which is advantageous for the near-term noisy intermediate-scale quantum (NISQ) devices, where the fidelity of quantum gates is limited by noise and decoherence. We first describe how to construct Ising Hamiltonian model to realize quasi-ML decoding with QAOA. For level-1 QAOA, we derive the systematic way to generate theoretical expressions of cost expectation for arbitrary binary linear codes. Focusing on (7, 4) Hamming code as an example, we analyze the impact of the degree distribution in associated generator matrix on the quantum decoding performance. The excellent performance of higher-level QAOA decoding is verified when Pauli rotation angles are optimized through meta-heuristic variational quantum eigensolver (VQE). Furthermore, we demonstrate the QAOA decoding performance in a real quantum device.

cs.IT↗

arXiv2019

Parallel-Amplitude Architecture and Subset Ranking for Fast Distribution Matching

A distribution matcher (DM) maps a binary input sequence into a block of nonuniformly distributed symbols. To facilitate the implementation of shaped signaling, fast DM solutions with high throughput and low serialism are required. We propose a novel DM architecture with parallel amplitudes (PADM) for which m component DMs, each with a different binary output alphabet, are operated in parallel in order to generate a shaped sequence with m amplitudes. With negligible rate loss compared to a single nonbinary DM, PA-DM has a parallelization factor that grows linearly with m, and the component DMs have reduced output lengths. For such binary-output DMs, a novel constant-composition DM (CCDM) algorithm based on subset ranking (SR) is proposed. We present SR-CCDM algorithms that are serial in the minimum number of occurrences of either binary symbol for mapping and fully parallel in demapping. For distributions that are optimized for the additive white Gaussian noise (AWGN) channel, we numerically show that PA-DM combined with SR-CCDM can reduce the number of sequential processing steps by more than an order of magnitude, while having a rate loss that is comparable to conventional nonbinary CCDM with arithmetic coding.

eess.SP↗

arXiv2018

Transfer Learning in Brain-Computer Interfaces with Adversarial Variational Autoencoders

We introduce adversarial neural networks for representation learning as a novel approach to transfer learning in brain-computer interfaces (BCIs). The proposed approach aims to learn subject-invariant representations by simultaneously training a conditional variational autoencoder (cVAE) and an adversarial network. We use shallow convolutional architectures to realize the cVAE, and the learned encoder is transferred to extract subject-invariant features from unseen BCI users' data for decoding. We demonstrate a proof-of-concept of our approach based on analyses of electroencephalographic (EEG) data recorded during a motor imagery BCI experiment.

cs.LG↗

arXiv2018

Deep Neural Network Inverse Design of Integrated Nanophotonic Devices

Predicting physical response of an artificially structured material is of particular interest for scientific and engineering applications. Here we use deep learning to predict optical response of artificially engineered nanophotonic devices. In addition to predicting forward approximation of transmission response for any given topology, this approach allows us to inversely approximate designs for a targeted optical response. Our Deep Neural Network (DNN) could design compact (2.6x2.6 μm2) silicon-on-insulator (SOI)-based 1 X 2 power splitters with various target splitting ratios in a fraction of a second. This model is trained to minimize the reflection (smaller than 20 dB) while achieving maximum transmission efficiency (above 90%) and target splitting specifications. This approach paves the way for rapid design of integrated photonic components relying on complex nanostructures.

physics.app-ph↗

arXiv2018

Multiset-Partition Distribution Matching

Distribution matching is a fixed-length invertible mapping from a uniformly distributed bit sequence to shaped amplitudes and plays an important role in the probabilistic amplitude shaping framework. With conventional constantcomposition distribution matching (CCDM), all output sequences have identical composition. In this paper, we propose multisetpartition distribution matching (MPDM) where the composition is constant over all output sequences. When considering the desired distribution as a multiset, MPDM corresponds to partitioning this multiset into equal-size subsets. We show that MPDM allows to address more output sequences and thus has lower rate loss than CCDM in all nontrivial cases. By imposing some constraints on the partitioning, a constructive MPDM algorithm is proposed which comprises two parts. A variable-length prefix of the binary data word determines the composition to be used, and the remainder of the input word is mapped with a conventional CCDM algorithm, such as arithmetic coding, according to the chosen composition. Simulations of 64-ary quadrature amplitude modulation over the additive white Gaussian noise channel demonstrate that the block-length saving of MPDM over CCDM for a fixed gap to capacity is approximately a factor of 2.5 to 5 at medium to high signal-to-noise ratios (SNRs).

eess.SP↗

arXiv2018

Invariant Representations from Adversarially Censored Autoencoders

We combine conditional variational autoencoders (VAE) with adversarial censoring in order to learn invariant representations that are disentangled from nuisance/sensitive variations. In this method, an adversarial network attempts to recover the nuisance variable from the representation, which the VAE is trained to prevent. Conditioning the decoder on the nuisance variable enables clean separation of the representation, since they are recombined for model learning and data reconstruction. We show this natural approach is theoretically well-founded with information-theoretic arguments. Experiments demonstrate that this method achieves invariance while preserving model learning performance, and results in visually improved performance for style transfer and generative sampling tasks.

cs.LG↗

arXiv2017

Co-design of Safe and Efficient Networked Control Systems in Factory Automation with State-dependent Wireless Fading Channels

In factory automation, heterogeneous manufacturing processes need to be coordinated over wireless networks to achieve safety and efficiency. These wireless networks, however, are inherently unreliable due to shadow fading induced by the physical motion of the machinery. To assure both safety and efficiency, this paper proposes a state-dependent channel model that captures the interaction between the physical and communication systems. By adopting this channel model, sufficient conditions on the maximum allowable transmission interval are then derived to ensure stochastic safety for a nonlinear physical system controlled over a state-dependent wireless fading channel. Under these sufficient conditions, the safety and efficiency co-design problem is formulated as a constrained cooperative game, whose equilibria represent optimal control and transmission power policies that minimize a discounted joint-cost in an infinite horizon. This paper shows that the equilibria of the constrained game are solutions to a non-convex generalized geometric program, which are approximated by solving two convex programs. The optimality gap is quantified as a function of the size of the approximation region in convex programs, and asymptotically converges to zero by adopting a branch-bound algorithm. Simulation results of a networked robotic arm and a forklift truck are presented to verify the proposed co-design method.

math.OC↗

arXiv2014

Threshold Analysis of Non-Binary Spatially-Coupled LDPC Codes with Windowed Decoding

In this paper we study the iterative decoding threshold performance of non-binary spatially-coupled low-density parity-check (NB-SC-LDPC) code ensembles for both the binary erasure channel (BEC) and the binary-input additive white Gaussian noise channel (BIAWGNC), with particular emphasis on windowed decoding (WD). We consider both (2,4)-regular and (3,6)-regular NB-SC-LDPC code ensembles constructed using protographs and compute their thresholds using protograph versions of NB density evolution and NB extrinsic information transfer analysis. For these code ensembles, we show that WD of NB-SC-LDPC codes, which provides a significant decrease in latency and complexity compared to decoding across the entire parity-check matrix, results in a negligible decrease in the near-capacity performance for a sufficiently large window size W on both the BEC and the BIAWGNC. Also, we show that NB-SC-LDPC code ensembles exhibit gains in the WD threshold compared to the corresponding block code ensembles decoded across the entire parity-check matrix, and that the gains increase as the finite field size q increases. Moreover, from the viewpoint of decoding complexity, we see that (3,6)-regular NB-SC-LDPC codes are particularly attractive due to the fact that they achieve near-capacity thresholds even for small q and W.

cs.IT↗