skip to main content

Search for: All records

Creators/Authors contains: "Dandekar, Kapil R."

Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

  1. Passive ultra high frequency (UHF) radio frequency identification (RFID) tags have the potential to find ubiquitous use in indoor object tracking, localization, and contact tracing. We propose a machine learning-based method for RFID indoor localization using a pattern reconfigurable UHF RFID reader antenna array. The received signal strength indicator (RSSI) values (from 10,000 tags) recorded at the reader antenna units are used as features to evaluate the machine learning models with a train-test split of 75%-25%. The training and testing data is generated by a wireless ray tracing simulator. Five machine learning models: random forest regressor, decision tree regressor, Nu support vector regressor, k nearest regressor, and kernel ridge regressor are compared. Random forest regressor has the lowest localization error both in terms of average Euclidean distance (AED) and root-mean-square error (RMSE). For random forest regressor, localization error results show that 90% of the tags are within 1 meter of their true position, and 67% are within 50 cm of their true position based on Euclidean distance.
    Free, publicly-accessible full text available April 27, 2023
  2. Modern-day radar is used extensively in applications such as autonomous driving, robotics, air traffic control, and maritime operations. The commonality between the aforementioned examples is the underlying tracking filter used to process ambiguous detections and track multiple targets. In this paper, we present a Software-Defined Radio-based radar testbed that leverages controllable and repeatable large-scale wireless channel emulation to evaluate diverse radar applications experimentally without the complexity and expense of field testing. Through over-the-air (OTA) and emulated evaluation, we demonstrate the capa-bilities of this testbed to perform multiple-target tracking (MTT) via Joint Probabilistic Data Association (JPDA) filtering. This testbed features the use of flexible sub-6 GHz or mmWave operation, electromagnetic ray tracing for site-specific emulation, and software reconfigurable radar waveforms and processing. Although the testbed is designed generalizable, for this paper we demonstrate its capabilities using an advanced driver-assistance system radar application.
    Free, publicly-accessible full text available March 21, 2023
  3. Precise monitoring of respiratory rate in premature newborn infants is essential to initiating medical interventions as required. Wired technologies can be invasive and obtrusive to the patients. We propose a deep-learning-enabled wearable monitoring system for premature newborn infants, where respiratory cessation is predicted using signals that are collected wirelessly from a non-invasive wearable Bellypatch put on the infant’s body. We propose a five-stage design pipeline involving data collection and labeling, feature scaling, deep learning model selection with hyperparameter tuning, model training and validation, and model testing and deployment. The model used is a 1-D convolutional neural network (1DCNN) architecture with one convolution layer, one pooling layer, and three fully-connected layers, achieving 97.15% classification accuracy. To address the energy limitations of wearable processing, several quantization techniques are explored, and their performance and energy consumption are analyzed for the respiratory classification task. Results demonstrate a reduction of energy footprints and model storage overhead with a considerable degradation of the classification accuracy, meaning that quantization and other model compression techniques are not the best solution for respiratory classification problem on wearable devices. To improve accuracy while reducing the energy consumption, we propose a novel spiking neural network (SNN)-based respiratory classification solution, which canmore »be implemented on event-driven neuromorphic hardware platforms. To this end, we propose an approach to convert the analog operations of our baseline trained 1DCNN to their spiking equivalent. We perform a design-space exploration using the parameters of the converted SNN to generate inference solutions having different accuracy and energy footprints. We select a solution that achieves an accuracy of 93.33% with 18x lower energy compared to the baseline 1DCNN model. Additionally, the proposed SNN solution achieves similar accuracy as the quantized model with a 4× lower energy.« less
    Free, publicly-accessible full text available March 1, 2023
  4. We present a pattern reconfigurable conformal mmWave antenna at 28 GHz for 5G applications. Using a T-junction power divider and PIN diodes, eight 2×2 sub-arrays of microstrip patches are configured to radiate independently creating separate states capable of 360° of discrete coverage.
  5. We present a step-by-step approach for designing a recofigurable Alford loop antenna (RALA). The design of an 3.5 GHz RALA is shown. The antenna is fabricated using a 1.6 mm thick double-sided FR4 substrate. We sweep antenna geometrical parameters and show the effect on antenna input impedance, reflection coefficient (S 11 ), and radiation patterns. The final antenna structure resonates at 3.5 GHz with eight directional and one omnidirectional radiation patterns. We also present a simplistic control circuit responsible for activating the antenna elements. Tri-state impedance matching- a major challenge in the design of RALA is also discussed and analyzed along with a proposed method for mitigation. 3D radiation patterns of the RALA was measured using an EMScan and a maximum gain of 4.5 dBi is found.
  6. Currently, wired respiratory rate sensors tether patients to a location and can potentially obscure their body from medical staff. In addition, current wired respiratory rate sensors are either inaccurate or invasive. Spurred by these deficiencies, we have developed the Bellyband, a less invasive smart garment sensor, which uses wireless, passive Radio Frequency Identification (RFID) to detect bio-signals. Though the Bellyband solves many physical problems, it creates a signal processing challenge, due to its noisy, quantized signal. Here, we present an algorithm by which to estimate respiratory rate from the Bellyband. The algorithm uses an adaptively parameterized Savitzky-Golay (SG) filter to smooth the signal. The adaptive parameterization enables the algorithm to be effective on a wide range of respiratory frequencies, even when the frequencies change sharply. Further, the algorithm is three times faster and three times more accurate than the current Bellyband respiratory rate detection algorithm and is able to run in real time. Using an off-the-shelf respiratory monitor and metronome-synchronized breathing, we gathered 25 sets of data and tested the algorithm against these trials. The algorithm’s respiratory rate estimates diverged from ground truth by an average Root Mean Square Error (RMSE) of 4.1 breaths per minute (BPM) over all 25more »trials. Further, preliminary results suggest that the algorithm could be made as or more accurate than widely used algorithms that detect the respiratory rate of non-ventilated patients using data from an Electrocardiogram (ECG) or Impedance Plethysmography (IP).« less