2015 IR&D Annual Report

Telemetry System Manager Automatic Program Synthesis, 10-R8532

Principal Investigators
Patrick Noonan
Austin Whittington
Ben Abbott

Inclusive Dates: 01/14/15 – 05/14/15

Background — The current telemetry component provider landscape is diverse in approach and capability, and this landscape continues to evolve as new test articles require new components, which leads to new system management requirements and complexities, especially in the area of configuration. To combat this evolving landscape, the integrated network enhanced telemetry (iNET) program has defined an extensible markup language (XML) grammar that defines how telemetry devices are to configure themselves, the metadata description language (MDL). This vendor-independent grammar is being increasingly adopted across the board by the telemetry industry. As such, a general method for creating MDL while considering the constraints (or capabilities) of individual devices is needed. The only viable approach to this problem is for the vendors to supply the new or updated constraints along with any new or updated devices. However, there are various classes of users, or layers of users, that require the ability to add further constraints on top of the vendor constraints.

Approach — Standardized XML technologies were used to develop an experimental framework that automatically generated XForms, or user interface elements, directly from an XML schema, while also constraining the relevant fields. The generated user interface was then capable of providing immediate feedback, in the form of error messages describing the failed constraint(s), as the user input configuration data. The supplied constraints were written using XPath expressions, providing interoperability and straightforward conversion to other XML-based languages. Once the XForms were created, templates that contained basic data and partial frameworks for devices were used as initial instance data to feed into the interface. This allowed for defined specialization of the interfaces, as the initializing data combined with the constraints meant that portions of the user-editable data could be intelligently completed without user input. After the XForms were filled out, populated XML fragments containing all the information were stored in a database. Then, using marshalling techniques to combine all necessary devices/elements of the desired output, a final XML output file was produced. This approach resulted in an XML file that described a valid device configuration, which was used to successfully configure the target device.

Accomplishments — The experimental framework demonstrated that the hierarchical automatic program synthesis approach would work for the telemetry domain by combining standardized XML technologies in a specialized way to synthesize system management based on a hierarchy of constraints from multiple sources. However, the generated XForms were not user friendly because the user interface had an XML-centric look-and-feel. To address this, a method was developed for styling the XForms with JavaScript, while also retaining their functionality. In doing so, the user was presented with a rich user interface experience containing a custom layout and exposing only the fields relevant for their task.

Benefiting government, industry and the public through innovative science and technology
Southwest Research Institute® (SwRI®), headquartered in San Antonio, Texas, is a multidisciplinary, independent, nonprofit, applied engineering and physical sciences research and development organization with 10 technical divisions.