Award Data

Despite deep neural networks have demonstrated tremendous success in various commercial and DoD applications, they are susceptible to adversarial attacks with detrimental outcomes to the underlying applications. The generative adversarial network (GAN) provides a good way of defending against adversarial learning attacks, but it is faced with a practical challenge, as classical computers are not a ...

TSA will assess the feasibility of developing or modifying an existing application-based process for creating: a) A functional architecture of a system to identify system missions, mission threads, and Critical Functions (mission model), and b) A physical architecture of a system (system model) in a manner consistent with both TSN Criticality Analysis and the Cybersecurity Risk Assessment Implemen ...

To meet the demands of the SCO182-004 SBIR solicitation (“High Acceleration and Hypervelocity Radiofrequency Communications and Sensing”), MaXentric proposes a dual band software define radio architecture. The design will leverage MaXentric’s expertise in transceiver and radar system design. Beam forming will be used to improve performance and reduce the probability of signal in ...

One key concern of the Strategic Capabilities Office (SCO) is the security component of the “Asia-Pacific Rebalance” where the Chinese anti-access/area denial (A2/AD) policy threatens regional security. In response to A2/AD our forces in the Pacific require more affordable and rapidly deployable alternatives to costly missile defense such as large caliber projectiles for Navy electroma ...

Parry Labs proposes the development and fabrication of a scalable, low-cost, transmit only wideband active electronically scanned array (AESA) operating in Ku band. Operating at Ku band maximizes aperture gain while also minimizing physical size and transmission loss due to rain and other atmospheric effects. The proposed system will be based on a scalable tile building block that can be used to c ...

Development of a prototype application that using automation, transforms an initial Government Client statement of need (SoN) through the use of expert systems and a machine learning engine. The final product will transform broader SoN or Statement of Work (SOW) requirements into work products that are able to leverage the RAD Freelancer concept.

Parry Labs proposes the continued development of a distributed, scalable, heterogeneous system to perform Automated Processing, Exploitation, and Dissemination (APED) of multi-INT sensor feeds. The system will run as a distributed cluster that employs Serverless function execution and containers to enable the reliable, scalable running of applications. The cluster will employ a peer-to-peer overla ...

The vision of truly autonomous, smart, and powerful edge devices requires the ability to perform intensive machine learning tasks with limited size, weight, and power budgets. We propose a highly efficient ASIC design that enables edge devices to achieve this vision.

Rapid strikes upon targets hidden in cluttered background may require a self-guided gun launched high velocity projectile having a unique seeker. A gun hard seeker that employs long wave polarimetry to quickly discriminate man-made objects in natural scenes is proposed. Simple polarimetric schemes that minimizes part count and utilizes no moving parts makes gun launch survival designs accessible. ...

A 1280x1024 on 8 µm pitch seeker with 0.4 to 2.6 µm detection capability will be incorporated with electronics and a Bayer mosaic of polarizing filters to allow real time target discrimination at rapid closure rates. The imager will have peak sensitivity greater than 60% and dark current of less than 6 µA/cm2 at 273 K. The camera will have greater than 99% pixel operability with a 12 bit digita ...