The following papers have been published for the NSF CNS Smart Farm project
Analysis of Deep Learning Models Towards High
Performance Digital Predistortion for RF Power
Amplifiers
Rajesh Kudupudi, Fariborz Lohrabi Pour, Dong Sam Ha, Sook Shin Ha, and Keyvan Ramezanpour
Multifunctional Integrated Circuits and System (MICS) Group
Bradley Department of Electrical and Computer Engineering
Virginia Tech, Blacksburg, Virginia, 24061, USA
E-mail: {rajeshk19, fariborzlp, ha, sook, rkeyvan8} @vt.edu
Abstract: This paper investigates direct and indirect learning methods to develop deep learning digital predistortion (DL-DPD) models and apply the models to improve the linearity of a power amplifier (PA). The two methods are applied to class-AB and class-F−1 PAs designed with gallium nitride (GaN) on silicon carbide (SiC) high electron mobility transistors (HEMTs). The simulation results show that both direct and indirect DL-DPD methods improve the linearity of the class-AB PA by about 12 dB and the class-F^−1 PA by 11 dB, while the indirect method offers marginally better performance. The paper shows the direct learning method leads to significant improvement of the DL-DPD method based over the memory polynomial. It also presents the advantages of a BiLSTM based on the neural network architecture to design direct/indirect DPDs. Finally, it demonstrates that the DL-DPD can improve the linearity of class-AB and class-F^-1 PAs without architectural changes.
R. Kudupudi, F. Lohrabi Pour, D. S. Ha, S. S. Ha, and K. Ramezanpour, “Analysis of Deep Learning Models Towards High Performance Digital Predistortion for RF Power Amplifiers,” International Symposium on Circuits and Systems (ISCAS), 5 pages, May 2022.
Power Efficient Wireless Sensor Node through Edge
Intelligence
Abhishek Priyadarshan Damle*, Sook Shin Ha*, Zhuqing Zhao *, Barbara Roqueto dos Reis**, Robin White**, and Dong Sam Ha*
*Multifunctional Integrated Circuits and System (MICS) Group
Bradley Department of Electrical and Computer Engineering
**Dept of Animal and Poultry Sciences
Virginia Tech, Blacksburg, Virginia, 24061, USA
E-mail: {adamle, sook, zhuqing8, barbarar, rrwhite, ha} @vt.edu
Abstract: Edge intelligence can reduce power dissipation to enable power-hungry long-range wireless applications. This work applies edge intelligence to quantify the reduction in power dissipation. We designed a wireless sensor node with a LoRa radio and implemented a decision tree classifier, in situ, to classify behaviors of cattle. We estimate that employing edge intelligence on our wireless sensor node reduces its average power dissipation by up to a factor of 50, from 20.10 mW to 0.41 mW. We also observe that edge intelligence increases the link budget without significantly affecting average power dissipation.
A. P. Damle, S. S. Ha, Z. Zhao, B. R. dos Reis, R. White, and D. S. Ha, “Power Efficient Wireless Sensor Node through Edge Intelligence,” International Symposium on Circuits and Systems (ISCAS), 5 pages, May 2022.
An Attack-Resilient and Energy-Adaptive
Monitoring System for Smart Farms
Qisheng Zhang†, Yash Mahajan†, Ing-Ray Chen†, Dong Sam Ha‡, and Jin-Hee Cho†
†Department of Computer Science, Virginia Tech, Falls Church, VA, USA
‡Bradley Department of Electrical and Computer Engineering, Virginia Tech, Blacksburg, VA, USA
{qishengz19, yashmahajan, irchen, dha, jicho}@vt.edu
Abstract: In this work, we propose an energy-adaptive monitoring system for a solar sensor-based smart animal farm (e.g., cattle). The proposed smart farm system aims to maintain high-quality monitoring services by solar sensors with limited and fluctuating energy against a full set of cyberattack behaviors including false data injection, message dropping, or protocol non-compliance. We leverage Subjective Logic (SL) as the belief model to consider different types of uncertainties in opinions about sensed data. We develop two Deep Reinforcement Learning (DRL) schemes leveraging the design concept of uncertainty maximization in SL for DRL agents running on gateways to collect high-quality sensed data with low uncertainty and high freshness. We assess the performance of the proposed energy-adaptive smart farm system in terms of accumulated reward, monitoring error, system overload, and battery maintenance level. We compare the performance of the two DRL schemes developed (i.e., multi-agent deep Q-learning, MADQN, and multi-agent proximal policy optimization, MAPPO) with greedy and random baseline schemes in choosing the set of sensed data to be updated to collect high-quality sensed data to achieve resilience against attacks. Our experiments demonstrate that MAPPO with the uncertainty maximization technique outperforms its counterparts.
Q. Zhang, Y. Mahajan, I.R. Chen, D. Ha, and J.H. Cho, “An Attack-Resilient and Energy-Adaptive Monitoring System for Smart Farms,” IEEE GLOBECOM, Dec. 2022
MiLTOn: Sensing Product Integrity without Opening the Box using Non-Invasive Acoustic Vibrometry
Akshay Gadre
Carnegie Mellon University
agadre@andrew.cmu.edu
Deepak Vasisht
Microsoft and UIUC
deepakv@illinois.edu
Nikunj Raghuvanshi, Bodhi Priyantha, Manikanta Kotaru
{nikunjr,bodhip,mkotaru}@microsoft.com
Swarun Kumar
Carnegie Mellon University
swarun@cmu.edu
Ranveer Chandra
Microsoft
ranveer@microsoft.com
Abstract: This paper asks: “Can we detect whether a fragile product, made of porcelain or glass is damaged as it travels along the supply chain, without opening its packaging?” We ask this question in the context of the multi-billion dollar global supply chain industry of fragile products that experience large overheads due to product returns. This paper presents MiLTOn, a novel acoustic and mm-wave based solution for through-box non-invasive product integrity sensing that is sensitive to even minute sub-mm cracks in the object. MiLTOn is inspired by acoustic vibrometry used for instance to monitor cracks in railroads. Unlike traditional vibrometry, MiLTOn is unique in its ability to sense products non-invasively using an external transducer and microphone, neither of which are in direct physical contact of the object within the box. MiLTOn processes measurements from the microphone to design a robust and environment-independent product signature that can be used to sense presence of product defects. Our extensive evaluation on a large number of fragile products of diverse materials demonstrates 97% accuracy in identifying product damage.
A. Gadre et al., “MiLTOn: Sensing Product Integrity without Opening the Box using Non-Invasive Acoustic Vibrometry,” 2022 21st ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN), 2022, pp. 390-402, doi: 10.1109/IPSN54338.2022.00038.
NFCapsule: An Ingestible Sensor Pill for Eosinophilic Esophagitis
Detection Based on Near-field Coupling
Junbo Zhang, Gaurav Balakrishnan, Sruti Srinidhi, Arnav Bhat, Swarun Kumar, Christopher Bettinger
Carnegie Mellon University Pittsburgh, PA, USA
{junboz2, gbalakr1, ssrinidh, abhat, swarun, cbetting}@andrew.cmu.edu
Abstract: This paper presents NFCapsule, a light-weight, battery-free, and ingestible biomedical sensor that can potentially enable non-invasive detection of active eosinophilic esophagitis (EoE). EoE is an allergen-induced inflammatory condition of the esophagus; its diagnosis generally involves invasive, wired, and time-consuming endoscopy. In contrast, NFCapsule aims to wirelessly detect active EoE by tracking tissue impedance through an ingestible pill that the patient swallows. Specifically, recent biomedical research has shown that active EoE induces observable changes in the electrochemical impedance of the esophagus tissue due to an increase in its intercellular spacing. We design the NFCapsule pill based on RLC resonant circuits and model the target tissue as an impedance component that changes the resonant properties of the pill circuit. Further, the NFCapsule reader identifies the resonant properties of the pill by
consistently monitoring the amount of energy transferred to the pill as it goes through the esophagus, and converts this information to estimates of bio-impedance. We implement NFCapsule pill prototypes with flexible polyimide PCBs and gelatin capsules (27 mm in height and 10 mm in diameter) and evaluated NFCapsule with both ionic agarose hydrogel models and ex vivo porcine esophageal tissues (no human patients involved). We show that NFCapsule maintains high classification accuracy under various practical scenarios (e.g., blockage, bending, movement, etc.) and achieves 85% average accuracy between healthy and unhealthy tissue samples.
Junbo Zhang, Gaurav Balakrishnan, Sruti Srinidhi, Arnav Bhat, Swarun Kumar, and Christopher Bettinger. 2023. NFCapsule: An Ingestible Sensor Pill for Eosinophilic Esophagitis Detection Based on near-Field Coupling. In Proceedings of the 20th ACM Conference on Embedded Networked Sensor Systems (SenSys ’22). Association for Computing Machinery, New York, NY, USA, 75–90. https://doi.org/10.1145/3560905.3568523