Communications

Our goal is to find whether a convolutional neural network (CNN) performs better than the existing blind algorithms for image denoising, and, if yes, whether the noise statistics has an effect on the performance gap. We performed automatic identification of noise distribution, over a set of nine possible distributions, namely, Gaussian, log-normal, uniform, exponential, Poisson, salt and pepper, Rayleigh, speckle and Erlang. Next, for each of these noisy image sets, we compared the performance of FFDNet, a CNN based denoising method, with noise clinic, a blind denoising algorithm.

Our Signing in the Wild dataset consists of various videos harvested from YouTube containing people signing in various sign languages and doing so in diverse settings, environments, under complex signer and camera motion, and even group signing. This dataset is intended to be used for sign language detection.

Device identification using network traffic analysis is being researched for IoT and non-IoT devices against cyber-attacks. The idea is to define a device specific unique fingerprint by analyzing the solely inter-arrival time (IAT) of packets as feature to identify a device. Deep learning is used on IAT signature for device fingerprinting of 58 non-IoT devices. We observed maximum recall and accuracy of 97.9% and 97.7% to identify device. A comparitive research GTID found using defined IAT signature that models of device identification are better than device type identification.

Design of novel RF front-end hardware architectures and their associated measurement algorithms.
Research objectives, includes:
RO1: Novel architecture based upon Adaptive Wavelet Band-pass Sampling (AWBS) of RF Analog-to-Information Conversion (AIC).
RO2: Integration of AWBS for increasing the wideband sensing capabilities of real-time spectrum analyzers by using AICs.
RO3: Propose online calibration methods and algorithms for front-end hardware non-idealities compensation.