History of Innovation
Sentar has a strong history of innovation. For years, Sentar has been known for developing technologies and pursuing concepts that were “ahead of their time,” including Artificial Intelligence, Machine Learning, and agent-based algorithms.
We especially have a long legacy of domain experience in Artificial Intelligence (AI). We started work in AI in the early 1990s, participating in the AIAA committee for AI standards, and leading the publication of an ANSI Guideline for the development of Knowledge Based Systems (KBS). Sentar principals also led the design of one of the earliest industrial grade, classified, AI systems that was fielded and used by the US Intel Community (IC).
Quick Planner for Verification and Validation of Distributed Hybrid Systems
As an early adopter of Artificial Intelligence, Sentar saw the need for automating and verifying a catalog of Verification and Validation (V&V) tools used in Distributed Hybrid Systems. We developed methodology for AI-enabled Knowledge Based Systems to determine appropriate V&V approaches.
Fault-Tolerant Cooperative Intelligent Agent Systems for Distributed C3I
In 1997, in concert with the rapidly evolving C3I System of Systems in the Air Force, Sentar used AI to design a fault-tolerant cooperative intelligent agent for management of distributed data objects in decentralized and distributed C3I system(s) architecture.
An Agent Architecture for Continuously Evolving Large-Scale Knowledge Bases
Supporting the rapid evolution of object-oriented technologies and their use in C3I architecture, Sentar developed an Agent Architecture for continuously evolving large-scale knowledge bases. This revolutionary Agent Architecture served as a catalyst for the evolution of large-scale knowledge based conventional and hybrid systems.
Agent-based Knowledge-design Assistance (AKA)
As a way to help the Missile Defense Agency build and share knowledge across their growing portfolio of complex integrated weapon systems, Sentar developed Agent-based Knowledge-design Assistance (AKA). Sentar's AKA produced knowledge modules for use in the Ballistic Missile Defense System (BMDS) C3I architecture.
Agent Enabled Advanced Intrusion Detection System
As an early enabler of cybersecurity, Sentar developed an Intrusion Detection System (IDS) that used intelligent agents to recognize and respond to the unique challenges of cyber attacks in weapon systems. Designed for the Ground Based Midcourse Defense (GMD) System architecture, Sentar leveraged DARPA developments to create an agile and high confidence Computer Network Defense (CND) capability.
Agent-based Knowledge-design Assistant (AKA) Technology for Computer Network Defense (CND)
Also in 2003, Sentar leveraged earlier AI work in intelligent agents to fuse new capabilities into a weapon-systems centric Computer Network Defense (CND) construct. Sentar's agile CND solution allowed security operators to continuously adapt and evolve the CND to rapidly evolving conditions.
Work-Centered Interface Technology
Building on the Computer Network Defense (CND) work for Ground Based Midcourse Defense (GMD), Sentar tackled the problem of data overload that could cripple an Intrusion Detection System, along with effective human-in-the-loop response. The resulting Work Centered Interface (WCI) for the Computer Network Defense (CND) enabled automated, policy triggered responses to network anomalies and cyber attacks within a weapons system construct.
Active Resource Manager (ARM)
In 2006, Sentar responded to the cybersecurity needs of mission-defined networks by building the Active Resource Manager (ARM). ARM provided an integrated portable console containing firewalls and Intrusion Detection and Response systems that could be rapidly configured to support "on the fly" network configurations.
Legerdemain: An Illusion-based System to Thwart Covert Access and Exfiltration
Recognizing the value of obfuscation as a means to thwart cyber attackers, our 2009 Ledgerdemain provided protection strategies that used combinations of deceptive techniques that could either be anticipatory or reactive. The Legerdemain manipulated both the items to be protected and the protection mechanisms in order to secure CPI against network-based attacks, host-based attacks, and reverse engineering.
Structured Application Protection Process (SAPP)
In 2009, we shifted our focus to cyber attack modeling against websites with our Structured Application Protection Process (SAPP). SAPP created and executed threat models and attacks against domain classes of software functionalities. By modeling attacks, defenders have greater insight into needed protective mechanisms.
Reflective Annotations for Functional Test (RAFT)
Responding to V&V (Software Verification & Validation) needs for complex system of systems, Sentar developed the Reflective Annotations for Functional Test (RAFT). RAFT used symbolic execution to verify that all possible executable paths of code meet the required functional integrity thresholds.
Eunomia- A Process and Toolset to Implement Big Code Approaches for Vulnerability Detection
In 2016, Sentar refocused on using symbolic execution to explore code vulnerabilities and malware. RAFT used machine learning to mine code and automate the detection of vulnerabilities and malware.
In 2017, Sentar updated its static code scanning tool veriScan with new capabilities for FORTRAN, and map findings to CWEs and STIGs. We also conducted a proof of concept utilizing the RAFT/Eunomia technologies to dynamically assess code and map to CWEs.
Specula (Multiplexing Delta Transfer)
The proposed concept is a software utility tool that resides on each of the client and server. The coordinating apps implement a delta transfer protocol, plus multiplexing data transfers, over multiple network paths. It is a concept and tool for a system made for prepositioning and synchronizing data between file systems deployed across austere environments, along with commercial or private cloud enterprises.