| 1. |
- Chen, Bo, et al.
(författare)
-
Deep Contextualized Compressive Offloading for Images
- 2021
-
Ingår i: SenSys 2021 - Proceedings of the 2021 19th ACM Conference on Embedded Networked Sensor Systems. - New York, NY, USA : ACM. ; , s. 467-473
-
Konferensbidrag (refereegranskat)abstract
- Recent years have witnessed sensors becoming an indispensable part of our life with the camera being one of the most popular and widely deployed sensors. The camera gives rise to numerous vision-based IoT applications that generate high-level understandings of a live video stream by performing analysis on end devices like mobile or embedded devices. Typically, these applications are built with deep learning (DL) models to conduct complex vision tasks, e.g., image classification and object detection. Due to the prohibitive cost of running DL models on end devices close to the camera and with limited computation capabilities, it is widely adopted to offload the computation to a nearby powerful edge server. However, there is a gap between the restricted offloading bandwidth of the end device and the large volume of image data incurred by the live video stream. In this paper, we present Deep Contextualized Compressive Offloading for Images (DCCOI), a lightweight, context-aware, and bandwidth-efficient offloading framework for images. DCCOI consists of the spatial-adaptive encoder, a lightweight neural network, to spatial-adaptively compress the image, and the generative decoder for reconstructing the image from the compressed data. In contrast to existing DL-based encoders, the spatial-adaptive encoder allows an image region to be encoded into different numbers of feature values based on the information in it. This offers a variable-length coding method for image compression, which is a more optimal way for compression than the fix-length coding method took by existing DL-based compression approaches and demonstrates superior accuracy-compression rate trade-offs. We evaluate DCCOI against several baseline compression techniques while serving an object detection-based application. The results show that DCCOI roughly reduces the offloading size of JPEG by a factor of 9 and DeepCOD, the state-of-the-art offloading approach, by 20% with similar accuracy and a compression overhead less than 50ms.
|
|
| 2. |
- Fan, Jin, et al.
(författare)
-
Design of Ultra-Wideband Phased Array Feed for Radio Telescope
- 2020
-
Ingår i: 2020 33rd General Assembly and Scientific Symposium of the International Union of Radio Science, URSI GASS 2020.
-
Konferensbidrag (refereegranskat)abstract
- This paper presents an ultra-wideband Phased Array Feed (PAF) based on novel flat bend crossed dipole fed by 50Ohm coaxial line. The PAF is dual-polarized and made from all-metal to minimize Ohmic losses and simplify cryogenic integration. It is optimized for 1-2 GHz band and can be good PAF candidate for the Five hundred meter Aperture Spherical Telescope (FAST) and Qi Tai Telescope (QTT) as well as other radio telescopes.
|
|
| 3. |
- Pang, Xiaodan, Dr., et al.
(författare)
-
Free-Space Transmissions in the Upper- and Lower-THz Bands Assisted with Photonics
- 2021
-
Ingår i: 2021 European Conference on Optical Communication (ECOC). - : Institute of Electrical and Electronics Engineers (IEEE). ; , s. 1-4
-
Konferensbidrag (refereegranskat)abstract
- We report our recent studies in photonics-assisted free-space transmissions in both the lower (0.3-0.5 THz) and upper- (~64.5 THz) terahertz bands. We adopt the hybrid electro-optical approach for the lower-THz signal transmission, whereas a directly modulated quantum cascade laser is used for the upper-THz band.
|
|
| 4. |
- Tadesse, Melkie Getnet, 1984-, et al.
(författare)
-
Electromechanical properties of polyamide/lycra fabric treated with PEDOT:PSS
- 2017
-
Ingår i: Electromechanical properties of polyamide/lycra fabric treated with PEDOT:PSS. - : Institute of Physics (IOP). ; 254:7
-
Konferensbidrag (refereegranskat)abstract
- One of the challenges in smart textiles is to develop suitable multifunctional materials that can address simultaneously several characteristics such as durability, stretchability, lightweight, and conductivity. Conductive polymers which showed success in different technological fields like polymer solar cells and light emitting diodes are promising in many smart textile applications. In this work, we treated a common polyamide/lycra knitted fabric with PEDOT:PSS for stretchable e-textiles. PEDOT:PSS, with DMSO as a conductivity enhancer and different ratios of water-based polyurethane dispersions as a binder, was applied to the fabric with simple immersion and coating applications. The effect of different application methods and binder ratio on the surface resistance of the fabric was monitored with four point probe electrical surface resistance measurement systems. Samples prepared by immersion technique are more uniform and have higher conductivity than those prepared by a coating technique. SEM images showed that PEDOT:PSS is incorporated into the structure in the immersion method while in the coating it is majorly present on the surface of the fabric. The tensile measurement showed that the acidic PEDOT:PSS and polyurethane dispersion coating has no adverse effect on the tensile strength of the fabric. The coated samples can be stretched up to 700% while still reasonably conductive. The resistance increases only by a small amount when samples were stretched cyclically by stretching 100%. Generally, samples prepared by the immersion method maintained better conductivity while stretching than those by a coating method. The washing fastness of the samples was also assessed.
|
|
| 5. |
- Vassilev, Vessen, 1969, et al.
(författare)
-
Spectrum Efficient D-band Communication Link for Real-time Multi-gigabit Wireless Transmission
- 2018
-
Ingår i: IEEE MTT-S International Microwave Symposium Digest. - 0149-645X. - 9781538650677 ; , s. 1523-1526
-
Konferensbidrag (refereegranskat)abstract
- This manuscript presents results of wireless real-time data transmission at 143 GHz. The transmitter/receiver (Tx/Rx) front-end circuitry is integrated on a single monolithic microwave integrated circuits (MMICs), realized in a 250-nm indium phosphide (InP) double heterojunction bipolar transistor (DHBT) technology. The Tx module shows gain of 12 dB at 143 GHz with output power of -2.3 dBm at 1 dB gain compression. The Rx module has a gain of 15 dB with noise figure (NF) of 13 dB at 143 GHz. The minimum NF of 10 dB is measured at 132 GHz. The Tx/Rx front-end modules were integrated in two radio units to demonstrate a real time wireless data transmission. At a distance of 10 m and using 40 dBi gain antennas, the highest data rate achieved was 5.3 Gbit/s using 64 QAM modulation over a 1 GHz channel with spectrum efficiency of 5bit/s/Hz.
|
|
