With the advent of neural networks, users at the tactical edge have started experimenting with AI enabled intelligent mission applications. Autonomy stacks have been proposed for the tactical environments for sensing, reasoning and computing the situational awareness to provide the human in the loop actionable intelligence in mission time. Tactical edge computing platforms must employ small-form-factor modules for compute, storage, and networking functions that conform to strict size, weight, and power constraints (SWaP). Many of the neural network models proposed for the tactical AI stack are computationally complex and may not be deployable without modifications. In this paper we discuss deep neural network optimization approaches for resource constrained tactical unmanned ground vehicles.
Normalizations used in model reduction can be chosen to emphasize anything from computation reduction to parameter reduction. Choosing a normalization that emphasizes a model with a small number of parameters is useful when deploying a model onto machines with a limited communication rate, while choosing a normalization that emphasizes a model with a small computational cost is useful when deploying a model onto a machine for real-time sensor analysis. As such, we explore the effect of various normalizations used to prune kernel parameters on models trained on the ImageNet database.
The goal of this effort is to train Deep Learning (DL) models using synthetic Orthogonal Frequency-Division Multiplexing (OFDM) datasets to predict the modulation schemes of real OFDM signals without transfer learning. To facilitate our study, we generated a synthetic dataset, OFDM-O, that consists of 480k instances across four different modulations which include BP SK, QP SK, QAM16, and QAM64. Each instance with 16 OFDM symbols consists of 1280 IQ symbols. Since real OFDM instances have lengths of [2, 5, 44] OFDM symbols, the DL models are trained using short instances in order to overcome the instance length mismatch. Two datasets generated dynamically during training, OFDM-ro and OFDM-riq, are derived from dataset OFDM-O, by randomly choosing 5 consecutive OFDM symbols or 400 consecutive IQ symbols from each instance in OFDM-O at each epoch. 1-D Residual Neural Network (ResNet) models trained using three datasets achieve overall accuracies of 97.8%, 84.5% and 77.6% for OFDM-O, OFDM-ro and OFDM-riq, respectively. Cross validation of the three datasets shows that the ResNet model trained using OFDM-riq predicts the validation datasets of OFDM-O and OFDM-ro with high accuracy. Furthermore, a two-step validation is proposed during training of DL models where DL models are first validated with a synthetic validation dataset and then validated with real OFDM instances. Including a validation set with real signal allows us to terminate training before the DL model is over fit to the synthetic signals. The ResNet model trained using OFDM-riq correctly predicts 5 out of 7 short instances and all 5 long instances in the testing dataset of real signals. Both mis-classifications come from short instances of 2 OFDM symbols. Overall, the ResNet model trained using OFDM-riq can successfully predict the modulation schemes of real OFDM signals with high accuracy.
Machine Learning (ML) technologies have been widely used for blind signal classification problem (Modulation Classification) to automatically identify the modulation schemes of Radio Frequency (RF) signal from complex- valued IQ (In-phase Quadrature) samples. Traditional ML methods usually have two stages where the first stage is to manually extract the features of the IQ symbols by subject matter experts and the second stage is to feed the features to an ML algorithm (e.g., a support vector machine) to develop the classifier. The state- of-art technology is to apply Deep Learning (DL) technologies such as Convolutional Neural Network (CNN) or Recurrent Neural Network (RNN) directly to the complex-value IQ symbols to train a multi-class classifier. In this effort we are focused on the modulation and coding rate classification problems for multi-carrier Orthogonal Frequency-Division Multiplexing (OFDM) signals. The generated OFDM data set consists of over 480k instances across four different modulations and three different coding rates which include BPSK + 1=2, BPSK+3=4, QPSK+1=2, QPSK+3=4, QAM16+1=2, QAM16+3=4, QAM64+2=3 and QAM64+3=4. Deep Learning (DL) models can successfully catch the features of OFDM modulations and identify the modulation scheme from the baseband IQ symbols. Four DL models including Residual Neural Network 34(Resnet34), Resnet18, Squeezenet and Long Short-Term Memory (LSTM) trained over the data set can achieve over 94% overall accuracy for signals across 10dB to 10dB with step size of 4dB. Among them, Squeezenet and LSTM models have much smaller model sizes which can be easily loaded in resource-limited edge computing platforms. To the best of our knowledge, coding rate classification of OFDM signals has not been studied in previous works. Therefore, three different DL models including a single-stage coding rate classifier, a combination of modulation and coding rate classifier, and a two-stage coding rate classifier are developed to identify the coding rate. However, coding rates can only be identified for low-order modulation schemes BPSK and QPSK but not for high-order modulation schemes QAM16 and QAM64. Further investigation is needed to understand the coding rate classification for high-order QAM signals.
The Army Rapid Capabilities Office (RCO) sponsored a Blind Signal Classification Competition seeking algorithms to automatically identify the modulation schemes of RF signal from complex-valued IQ (in-phase quadrature) samples. Traditional spectrum sensing technology uses energy detection to detect the existence of RF signals but the RCO competition further aimed to detect the modulation scheme of signals without prior information. Machine Learning (ML) technologies have been widely used for blind signal classification problem. Traditional ML methods usually have two stages where the first stage is to manually extract the features of the IQ symbols by subject matter experts and the second stage is to feed the features to an ML algorithm (e.g., a support vector machine) to develop the classifier. The state-of-art technology is to apply deep learning technologies such as Convolutional Neural Network (CNN) or Recurrent Neural Network (RNN) directly to the complex-value IQ symbols to train a multi-class classifier. Our team, dubbed Deep Dreamers, participated in the RCO competition and placed 3rd out of 42 active teams across industry, academia, and government. In this work we share our experience and lessons learned from the competition. Deep learning methods such as CNN, Residual Neural Network (ResNet), and Long Short-Term Memory (LSTM) are the fundamental neural network layers we used to develop a multi-class classifier. None of our individual models were able to achieve a competitively high ranking in the competition. The key to our success was to use ensemble learning to average the outputs of multiple diverse classifiers. In order for ensemble methods to be more accurate than any of its base models; the base learners have to be as accurate as possible. We found that while ResNet was more accurate than the LSTM; the LSTM was less sensitive to deviations in the test set.
The benefits of blockchain technology are evident in the banking industry as the underlying technology for cryptocurrencies. Recently, research has been focused on novel, non-cryptocurrency uses of blockchain for other industries such as national defense. Application of these technologies in military use cases requires special consideration of the limitations inherent to tactical military operations, namely the network communication technologies. In this work we explore the performance of blockchain technologies on network environments representative of those available in Department of Defense (DoD) tactical operations. Our experimentation with the Ethereum blockchain on a mobile network emulation reveals a series of verbose blockchain network communication protocols as well as a heavy reliance on Transmission Control Protocol (TCP) for block transfer and synchronization that may limit the effectiveness of blockchain on current DoD mobile ad hoc networks.