KEYWORDS: Radar, Education and training, Deep learning, Data modeling, Sensors, Cognitive modeling, Velocity measurements, Systems modeling, Signal to noise ratio, Remote sensing
Cognitive radar systems are radar systems that can self-adjust themselves to respond to changes in the environment. Developing cognitive radar systems relies on their ability to detect these changes in operational conditions and use this knowledge to change the operating characteristics of the system, to optimally solve a selected task. Engineers must have an expert level knowledge of radar systems in order to solve these problems as they arise. The goals of the system can be easily stated to engineers in the form of natural language, but are very difficult for computers to analyze. Previous work has shown that Natural Language Processing (NLP) models can be developed to extract radar parameters, values, and units from text. Language Based Cost Functions (LBCFs) can then utilize this extracted information to develop constraints on specific radar parameters. In this work, we propose to combine these language models with LBCFs to define a objective function for optimization tasks using natural language.
When performing automatic target recognition it is common to train models using synthetically generated data. This is because synthetically generated data is plentiful, and cheap to produce. Once trained on synthetic data machine learning models are often testing on measured or real-world SAR images. These models do not perform as well when analyzing measured SAR images. This problem is known as the synthetic-measured gap. In this work we explore training generative and contrastive models to close this gap. We train our models on synthetically generated data with the goal of being able to classify measured SAR images. We utilize segmentation masks as well fully-formed SAR images. In the generative approach we explore using an auto-encoder to generate segmentation masks of input SAR images. The auto-encoders architecture includes a classifier which is trained using shared features between the raw image and the segmentation mask. This model is capable of generating a segmentation mask from a SAR image. The contrastive approach uses the Sim-Siam architecture, which utilizes segmentation masks and SAR images. The contrastive model makes a classification decision, by learning features that are shared between the two input types. The goal of this work is to improve classification performance when training on synthetic data, and evaluating on measured data.
Determining what radar parameters to use for a given scenario is a non-trivial task. When working in a new radar domain, it is quite common to turn to published literature to understand how to approach a new problem. When reviewing research, there can be such a wide range of values used in a radar design that it can become difficult to determine what values to use when designing a new system. An ideal scenario would be to turn to a single source that provides base listings for different radar parameters, but at the time of writing no source is known. In this work, we aim to statistically analyze published radar literature to determine a base set of radar parameter values for a given domain. These parameters include things such as the carrier frequency, bandwidth, pulse repetition frequency, and target range, among many others. To do this, a base set of parameters that are included in nearly all radar systems design will need to be established. Then, by selecting published research in specific domains (ground penetrating radar, atmospheric sensing, imaging, etc . . . ), we can determine the most common values for these parameters. In this paper, we examine the most common values for a given domain, as well as analyze the relationships between these parameters. This information could then be used to develop simulations, optimization problems, or provide insight when developing a new radar system.
This paper introduces the radar text data set (RadarTD) for technical language modeling. This data set is comprised of sentences containing radar parameters, values, and units determined from real-world values. This data set is created based on values determined from published academic research. Additionally, each statement is assigned a sentiment label and goal priority label. Preliminary investigations into the applicability of this data set are explored using the BERT model and several bi-directional LSTM models. These models are evaluated on text classification and named entity recognition tasks. This study evaluates the applicability of technical language modeling using neural networks to analyze input statements for cognitive radar applications. These findings suggest that this data set can be used to achieve reasonable performance for both text classification and named entity recognition for autonomous radar applications.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.