Department Publications

Capsule Network is a promising concept in deep learning, yet its true potential is not fully realized thus far, providing sub-par performance on several key benchmark datasets with complex data. Drawing intuition from the success achieved by Convolutional Neural Networks (CNNs) by going deeper, we introduce DeepCaps, a deep capsule network architecture which uses a novel 3D convolution based dynamic routing algorithm. With DeepCaps, we surpass the state-of-the-art capsule domain networks results on CIFAR10, SVHN and Fashion MNIST, while achieving a 68% reduction in the number of parameters. Further, we propose a class independent decoder network, which strengthens the use of reconstruction loss as a regularization term. This leads to an interesting property of the decoder, which allows us to identify and control the physical attributes of the images represented by the instantiation parameters.

Accurately estimating the remaining useful life (RUL) of industrial machinery is beneficial in many real-world applications. Estimation techniques have mainly utilized linear models or neural network based approaches with a focus on short term time dependencies. This paper, introduces a system model that incorporates temporal convolutions with both long term and short term time dependencies. The proposed network learns salient features and complex temporal variations in sensor values, and predicts the RUL. A data augmentation method is used for increased accuracy. The proposed method is compared with several state-of-the-art algorithms on publicly available datasets. It demonstrates promising results, with superior results for datasets obtained from complex environments.

Many localized languages struggle to reap the benefits of recent advancements in character recognition systems due to the lack of substantial amount of labeled training data. This is due to the difficulty in generating large amounts of labeled data for such languages and inability of deep learning techniques to properly learn from small number of training samples. We solve this problem by introducing a technique of generating new training samples from the existing samples, with realistic augmentations which reflect actual variations that are present in human hand writing, by adding random controlled noise to their corresponding instantiation parameters. Our results with a mere 200 training samples per class surpass existing character recognition results in the EMNIST-letter dataset while achieving the existing results in the three datasets: EMNIST-balanced, EMNIST-digits, and MNIST. We also develop a strategy to …

A variety of applications and forwarding protocols have been proposed for opportunistic networks (OppNets) in the literature. However, the methodology of evaluation, testing and comparing these forwarding protocols are not standardized yet, which leads to large levels of ambiguity in performance evaluation studies. Performance results depend largely on the evaluation environment, and on the used parameters and models. More comparability in evaluation scenarios and methodologies would largely improve also the availability of protocols and the repeatability of studies, and thus would accelerate the development of this research topic. In this survey paper, we focus our attention on how various OppNets data forwarding protocols are evaluated rather than what they actually achieve. We explore the models, parameters and the evaluation environments and make observations about their scalability, realism and comparability. Finally, we deduce some best practices on how to achieve the largest impact of future evaluation studies of OppNets data dissemination/forwarding protocols.

This paper presents the design, implementation, and the experimental verifications of a 2.4-GHz multibeam array receiver based on two-dimensional (2-D) spatially bandpass (SBP) digital plane wave (PW) filters. The digital array receiver consists of 2.4-GHz 16-element uniform linear array, and quadrature sampling is employed to process the broadband PWs at the baseband. Both 2-D finite impulse response (FIR) and infinite impulse response (IIR) digital filters having multiple trapezoidal-shaped passbands are employed as the 2-D SBP digital PW filters. Both 2-D FIR and IIR trapezoidal filters are implemented on a Xilinx Virtex 6 sx475t field programmable gate array chip that comes with the reconfigurable open architecture computing hardware version-2 (ROACH-2) platform. The 16 in-phase and quadrature-phase downconverted signals are sampled using the ROACH-2 platform and are processed using the …

Activity recognition in videos in a deep-learning setting-or otherwise-uses both static and pre-computed motion components. The method of combining the two components, while keeping the burden on the deep network less, still remains uninvestigated. Moreover, it is not clear what the level of contribution of individual components is, and how to control the contribution. In this paper, we use a combination of convolutional-neural-network-generated static features and motion features in the form of motion tubes. We propose three schemas for combining static and motion components: based on a variance ratio, principal components, and Cholesky decomposition. The Cholesky-decomposition-based method allows the control of contributions. The ratio given by variance analysis of static and motion features matches well with the experimental optimal ratio used in the Cholesky decomposition-based method. The …

Capsule networks excel in understanding spatial relationships in 2D data for vision related tasks. Even though they are not designed to capture 1D temporal relationships, with TimeCaps we demonstrate that given the ability, capsule networks excel in understanding temporal relationships. To this end, we generate capsules along the temporal and channel dimensions creating two temporal feature detectors which learn contrasting relationships. TimeCaps surpasses the state-of-the-art results by achieving 96.21% accuracy on identifying 13 Electrocardiogram (ECG) signal beat categories, while achieving on-par results on identifying 30 classes of short audio commands. Further, the instantiation parameters inherently learnt by the capsule networks allow us to completely parameterize 1D signals which opens various possibilities in signal processing.

In this paper, we are concerned with indoor localization based on multiple-antenna channel measurements. Indoor localization is an active area of research due to its great importance in many applications. We propose a hybrid algorithm which combines the benefits of two techniques, namely signal processing and machine learning. We validate our algorithm based on real measurements acquired from two practical setups. Our approach shows a very promising performance in the IEEE CTW 2019 - Positioning Algorithm Competition where the algorithm achieves an accuracy within RMSE values below 10 cm. We further build a setup in another indoor environment, where the algorithm still proves a very good performance compared to state-of-the art techniques used in indoor localization tasks.

The paper considers physical layer security (PLS) of an untrusted unmanned aerial vehicle (UAV) network, where a multitude of UAVs communicate in full-duplex (FD) mode. A source-based jamming (SBJ) scheme is exploited for secure communication without utilizing any external jammers. Firstly, the optimal power allocation between the confidential signal and the jamming signal is derived to maximize the secrecy rate of each link between the source and the destination. Then, the best UAV selection scheme is proposed to maximize the overall secrecy rate of the network. The corresponding secrecy outage probability (SOP) and the average secrecy rate (ASR) of the network are analyzed based on the proposed UAV selection and the optimal power allocation schemes. Asymptotic results are also obtained to derive the achievable diversity order. Results are validated through numerical evaluations while …

Humanoid robots that act as human-robot interfaces equipped with social skills can assist people in many of their daily activities. Receptionist robots are one such application where social skills and appearance are of utmost importance. Many existing robot receptionist systems suffer from high cost and they do not disclose internal architectures for further development for robot researchers. Moreover, there does not exist customizable open-source robot receptionist frameworks to be deployed for any given application. In this paper we present an open-source robot receptionist intelligence core-“DEVI”, that provides researchers with ease of creating customized robot receptionists according to the requirements (cost, external appearance, and required processing power). Moreover, this paper also presents details on a prototype implementation of a physical robot using the DEVI system. The robot can give directional …

Extremum seeking control (ESC) is a data-driven optimization technique that can steer a dynamic plant toward an extremum of an unknown but measurable input to steady-state output. In the context of networked control systems, a new implementation method for ESC inspired by the well-known Luus-Jaakola algorithm is proposed. The main motivation is to minimize the communication burden associated with the search phase of ESC. In the proposed method, the controller only requires a notification of a change registered at the sensor, rather than the full information available at the sensor. This event-based approach leads to sporadic packet transmission. In addition, the proposed method is able to directly account for constraints while seeking the desired extremum. The constraints may be of the inequality or equality type. The algorithm's behavior is illustrated on a networked water pump control system.

Occlusion removal is an interesting application of image enhancement, for which, existing work suggests manually-annotated or domain-specific occlusion removal. No work tries to address automatic occlusion detection and removal as a context-aware generic problem. In this paper, we present a novel methodology to identify objects that do not relate to the image context as occlusions and remove them, reconstructing the space occupied coherently. The proposed system detects occlusions by considering the relation between foreground and background object classes represented as vector embeddings, and removes them through inpainting. We test our system on COCO-Stuff dataset and conduct a user study to establish a baseline in context-aware automatic occlusion removal.

Making a single network effectively address diverse contexts---learning the variations within a dataset or multiple datasets---is an intriguing step towards achieving generalized intelligence. Existing approaches of deepening, widening, and assembling networks are not cost effective in general. In view of this, networks which can allocate resources according to the context of the input and regulate flow of information across the network are effective. In this paper, we present Context-Aware Multipath Network (CAMNet), a multi-path neural network with data-dependant routing between parallel tensors. We show that our model performs as a generalized model capturing variations in individual datasets and multiple different datasets, both simultaneously and sequentially. CAMNet surpasses the performance of classification and pixel-labeling tasks in comparison with the equivalent single-path, multi-path, and deeper single-path networks, considering datasets individually, sequentially, and in combination. The data-dependent routing between tensors in CAMNet enables the model to control the flow of information end-to-end, deciding which resources to be common or domain-specific.

In Electroencephalography (EEG) classifiers, processing of EEG signals consists of extracting features from EEG signal and then processing those features to classify the signals. As of today, in most EEG processing systems, the classification part is done on software platform even though the feature extraction is done on hardware. Hence this require an operating system which limits the development of portable medical devices. This paper presents the design and implementation of feature classification in EEG signal processing system on Field Programmable Gate Array (FPGA) hardware platform. The classification is done with two-class Linear Discriminant Analysis (LDA), based on the features extracted using Discrete Wavelet Transform (DWT), for EEG signals obtained through PhysioNet website. The final design has minimum resource utilization, hence is able to implement on Basys 3 Artix-7 FPGA Board …

The response time of emergency vehicles (EVs) determines the outcome of many emergencies, thus improving the traversal time of EVs is of paramount importance. Vehicular communication is a key enabler of such an improvement. This paper studies two EV traversal algorithms, focusing mainly on their communication aspects. Fast moving dense traffic is modeled, and the traversal algorithms are implemented on top of the dedicated short-range communication (DSRC) / wireless access in vehicular environments (WAVE) protocol stack, while accounting for channel impairments such as path loss and fading. Algorithms that highlight the required packet transfers for the traversal and for safe lane changes are presented, and simulated in the VEINS framework for different traffic conditions. Simulation results show that the suitability of the EV traversal algorithms defer depending on the speed distribution of the …

Due to the increasing heterogeneity in network user requirements, dynamically varying day to day network traffic patterns and delay in-network service deployment, there is a huge demand for scalability and flexibility in modern networking infrastructure, which in return has paved way for the introduction of Software Defined Networking (SDN) in core networks. In this paper, we present an FPGA-based switch that is fully compliant with OpenFlow; the pioneering protocol for southbound interface of SDN. The switch architecture is completely implemented on hardware. The design consists of an OpenFlow Southbound agent which can process OpenFlow packets at a rate of 10Gbps. The proposed architecture speed scales up to 400Gbps while it consumes only 60% of resources on a Xilinx Virtex-7 featuring XC7VX485T FPGA.

This paper considers a low-resolution wireless communication system in which transmitted signals are corrupted by fading and additive noise. First, a universal lower bound on the average symbol error probability (SEP), correct for all M-ary modulation schemes, is obtained when the number of quantization bits is not enough to resolve M signal points. Second, in the special case of M-ary phase shift keying (M-PSK), the optimum maximum likelihood detector for equi-probable signal points is derived. Third, utilizing the structure of the derived optimum receiver, a general average SEP expression for the M-PSK modulation with n-bit quantization is obtained when the wireless channel is subject to fading with a circularly-symmetric distribution. Finally, an extensive simulation study of the derived analytical results is presented for general Nakagami-m fading channels. It is observed that a transceiver architecture with n-bit …

Positron emission tomography is used to diagnose gastric cancer and identify lymph node metastasis preoperatively. However, locating these lymph node metastases during surgery is difficult and usually leads to unnecessary node removal and results in a deterioration of a patient’s quality of life. To improve the quality of life, intraoperative identification of lymph node metastasis is necessary. A positron emission tomography system consisting of an external detector array connected to the patient’s surgical bed and a movable detector probe which is inserted into the patient’s stomach along with a laparoscopic instrument has been proposed to identify lymph node metastasis intraoperatively. This study presents the image reconstruction of a system involving data acquisition using a movable detector. A simulation-based investigation on the refinement of the detector design is also presented. The image is …

Eelbrain - ADS Now on home page ads icon ads Enable full ADS view ADS Eelbrain Brodbeck, Christian ; Das, Proloy ; Brooks, Teon L ; Reddigari, Samir ; jpkulasingham Abstract No description provided. Publication: Zenodo Pub Date: DOI: 10.5281/zenodo.7951251 Bibcode: 2023zndo...7951251B full text sources Zenodo | Related Materials (2) Version 0.36 Version v0.32 © The SAO Astrophysics Data System adshelp[at]cfa.harvard.edu The ADS is operated by the Smithsonian Astrophysical Observatory under NASA Cooperative Agreement 80NSSC21M0056 Smithsonian logo Harvard Center for Astrophysics logo NASA logo *The material contained in this document is based upon work supported by a National Aeronautics and Space Administration (NASA) grant or cooperative agreement. Any opinions, findings, conclusions or recommendations expressed in this material are those of the author and do not …

SMTPE ST 2110 is a professional video over IP standard which enables uncompressed video to be transported as separate video, audio and ancillary data streams, allowing true flexibility in media workflows. Uncompressed video transmission allows professional video environments to achieve high quality video distribution while meeting strict latency requirements. Although this standard can be implemented on software, hardware solutions can achieve superior performance, lower latency and lower power consumption while efficiently handling jitter and packet bursts. This paper presents an FPGA based hardware implementation of a SMPTE ST 2110 compliant end to end solution supporting uncompressed video resolutions of up to 4K at 30fps. Due to its scalability, the architecture can be easily extended to support higher video resolutions and/or multiple video streams. Additionally, the interface modularity of the …

Application of depth-selective filtering to modulated-sparse light fields towards reducing the DSP and memory complexities in real-time light field processing is investigated. A modulated-sparse light field is obtained by spatially windowing an original 4-D light-field signal, which is subsequently processed by 4-D depth-selective filters to achieve planar focus and occlusion suppression. Multidimensional spectral properties of modulated-sparse light fields are explored and demonstrative examples with real light fields are provided to show that almost similar depth filtering and occlusion suppression performance can be achieved by processing such modulated-sparse light fields, leading to significant reduction in DSP hardware and memory complexities.

This paper presents a method for localizing an Unmanned Aerial Vehicle (UAV) using geo-referenced aerial images, in case of GPS failure. When flying in urban environments in low altitudes the inaccurate GPS localization has become a major problem. The proposed method fuses measurements from inertial sensors and visual odometer and perform image registration process using the images captured by the onboard down looking camera. A pre-built local orthomosaic map with geotagged aerial images is used for the registration. The problem of unreliable matches caused by the dynamic objects is addressed in our approach. The experiment results show that it is possible to localize the UAV in urban environments with a low margin of error by using our method.

This paper describes a hardware and software codesign approach for object detection and tracking algorithm which is based on modified ViBe background subtraction method and scale adaptable particle filtering algorithm. The modified ViBe performs with a close accuracy to the original work of ViBe but with less complexity and it outperforms the Mixture of Gaussian(MoG) method. Since the modified algorithm has a low off chip memory access requirement, the real time object detection in embedded level with the help of hardware acceleration techniques could be achieved. The implemented particle filtering algorithm is capable of adapting to the scale variation of the tracking objects. After careful investigation of the time consumption profile of each function in our algorithm, the complete algorithm was implemented in IoT edge devices targeted Avnet Ultra 96 development board which is based on Xilinx Zynq …

Indoor positioning by wireless means is of significant interest in a wide range of applications; Wi-Fi and Bluetooth Low Energy (BLE) are popular candidate wireless technologies for this. We present a novel BLE beacon placement algorithm that complements existing Wi-Fi infrastructure in a hybrid Wi-Fi/BLE positioning system. Composite signal strength measurements thus obtained, are used to derive machine learning models for location estimation. Our results show that positioning estimates at table-level granularity (0.9 m × 1.8 m) in a computer laboratory can be achieved with 97% accuracy. This is an improvement of 20% and 30% compared to Wi-Fi and BLE only techniques respectively.