Bayesian Attack Model (BAM) User Story

Maccarone, Lee T., Valme, Romuald, Anaya, Ted R. · 2025

Read Bayesian Attack Model (BAM) User Story by Maccarone, Lee T., Valme, Romuald, Anaya, Ted R. (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/doe-osti-2589620-bayesian-attack-model-bam-user-story.txt.

Introduction

The user story of the Bayesian Attack Model, a tool for defenders of industrial control systems, is told in Sandia report SAND2025-11801R. Authors Lee Maccarone, Romuald Valme, and Ted Anaya describe software funded through the Department's cybersecurity office by way of Idaho National Laboratory. Control systems in power and water plants face attacks that unfold across many small clues.

The tool gathers attack observables, the traces an intruder leaves, from many sources into one analyst view. Bayesian reasoning, which updates odds as evidence arrives, helps rank what the traces mean. Analysts can then characterize an attack while it is still under way and interrupt it before damage lands. The document is written as a user story, showing the defender's workflow rather than the code.

Catalog record

CREATORMaccarone, Lee T., Valme, Romuald, Anaya, Ted R.
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE34 KB
LICENSEPublic Domain
RIGHTSPUBLIC DOMAIN
SOURCEDOE OSTI

Source: https://www.osti.gov/biblio/2589620

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

SAND2025-11801R Bayesian Attack Model (BAM) User Story Lee T. Maccarone, Romuald Valme, Ted R. Anaya Revision 0 2025 Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525. ABSTRACT This document presents a user story for the Bayesian Attack Model (BAM) tool designed to aggregate and analyze cyber-attack observables for operational technology (OT) systems. BAM aims to empower cybersecurity analysts by providing a streamlined interface for collecting observable data from various sources, enabling real-time analysis of potential adversary activity. By enhancing the response capabilities of security teams, BAM facilitates risk-informed decision-making and improves organizational security posture. This user story outlines the key functionalities, user interactions, and requirements necessary to successfully integrate BAM with other security information and event management (SIEM) technology and cybersecurity operations centers (CSOCs). VALUE PROPOSITION BAM aggregates observer experiences of anomalous events across OT environments to provide actionable decision intelligence. BAM can comprehensively analyze anomalous host-based, network-based, and physical environment observables. The primary benefit of aggregating a wide range of observables is to enhance the cybersecurity incident response process via earlier detection of adversarial activity and more effective response. Successful interruption of an OT cyber-attack may result in averting significant consequences including loss of revenue, damage to property, and loss of safety to facility staff and the public. BACKGROUND BAM is a Bayesian network developed to characterize the stage of cyber-attacks on OT systems given observable evidence. A Bayesian network is a graphical method for representing the probabilistic relationship between variables and enables inference about the state of some variables given the known state of other variables. As shown in Figure 1, observable evidence propagates through BAM based on the MITRE ATT&CK for ICS taxonomy [1]. Observables may come from a variety of sources, including hosts, networks, and the physical environment. The overall cyber- attack is characterized by the probabilities of four adversary behavior phases: Early, Middle, Late, and Impact. Inference can also be performed on the intermediate MITRE ATT&CK for ICS techniques and tactics. For more information on BAM, readers are encouraged to refer to [2, 3]. Figure 1. BAM's Structure [3] 2 USERS For simplicity, the BAM user groups were defined based on the SANS ICS job role to competency level recommendations [4]. These recommendations divide job roles into seven groups: Engineering, OT Staff, OT Cybersecurity, Management, Support Staff, IT Cybersecurity, and IT Staff (we have divided the original OT group into OT Staff and OT Cybersecurity to mirror the IT groups defined by SANS). Individuals with the titles defined by these groups may perform one or more of the work roles defined by the U.S. National Institute of Standards and Technology (NIST) National Initiative for Cybersecurity Education (NICE) Workforce Framework [5]. For brevity, we will discuss the NICE Framework alignment only within the context of the BAM user groups. BAM users are categorized into two groups: Observers and Analysts. These groups are not mutually exclusive – one individual may perform functions corresponding to both groups. Observers often perform the Implementation and Operation (IO) and/or Protection and Defense (PD) work role categories in the NICE Framework. Analysts often perform the PD and/or Investigation (IN) work role categories in the NICE Framework. These work role categories are defined below: • Implementation and Operation (IO): “Provides implementation, administration, configuration, operation, and maintenance to ensure effective and efficient technology system performance and security.” [6] • Protection and Defense (PD): “Protects against, identifies, and analyzes risks to technology systems or networks. Includes investigation of cybersecurity events or crimes related to technology systems and networks.” [6] • Investigation (IN): “Conducts national cybersecurity and cybercrime investigations, including the collection, management, and analysis of digital evidence.” [6] Observers are responsible for providing observable evidence to BAM. Observables may come from a variety of sources, so observers may support a variety of roles within the organization. Examples of observers are provided in Table I. This table is intended to communicate BAM’s primary users and is not an exhaustive list. Table I. BAM Users – Observers IT Host- IT Network- OT Host- OT Network- Physical Based Based Based Based Environment Job Role Observables Observables Observables Observables Observables IT Staff X X IT Cybersecurity X X OT Staff X X X OT Cybersecurity X X Engineering X x X Support Staff x x x x x Management x X = greater frequency of observations, x = lesser frequency of observations 3 The core user story for the Observer is: As an Observer, I want to document the anomalous events that I see in my workplace so that cybersecurity experts in my organization can investigate them and respond appropriately. Analysts are responsible for performing inference in BAM, and for interpreting and communicating BAM’s output to decision-makers. BAM may be used to support decision-making at the operational, tactical, and strategic levels. The specific Analyst users are OT Cybersecurity Analysts and IT Cybersecurity Analysts, and they are responsible for communicating their findings and recommending courses of action to Management. The core user story for the Analyst is: As an Analyst, I want to identify and understand adversarial activity in my environments earlier so that I can begin response activities earlier, more effectively, and with data-supported justification. As noted previously, the Analyst and Observer user groups are not mutually exclusive. It is anticipated that when a user belongs to both groups, the user’s primary role is that of the Analyst. PRECONDITIONS This section describes the preconditions that should be achieved before a user can successfully begin the main BAM workflow. These preconditions are primarily focused on operational aspects of BAM’s implementation. Installation requirements will be provided in the BAM User’s Guide. 1. Observer is equipped with the processes and technology to enable the perception of observables pertinent to the Observer’s job role 2. Observer is equipped with the processes and technology to enable the logging of observables in a SIEM tool applicable to their job role and responsibilities 3. Data pipelines exist for the passing of observable data in Structured Threat Intelligence eXpression (STIX) format from their SIEM tool collection source to BAM 4. Data pipelines exist for the passing of BAM output data in STIX format to other SIEM tools as appropriate 5. Processes exist for the Analyst to communicate cybersecurity decision intelligence to Management POSTCONDITIONS This section describes the postconditions that should occur after the main BAM workflow has been executed. 1. Analyst has successfully generated a JSON file containing the probability of all levels of adversary activity (MITRE ATT&CK for ICS Techniques and Tactics, and adversary behavior phase) as a function of observable evidence accumulated over time 2. Analyst has successfully generated graphical representations of the MITRE ATT&CK for ICS Techniques and adversary behavior phase data contained in the JSON file 3. Analyst has identified if any risk thresholds have been exceeded at the MITRE ATT&CK for ICS Technique or adversary behavior levels 4 4. Analyst has identified the most critical observables corresponding to each MITRE ATT&CK for ICS Technique 5. Analyst is equipped with knowledge to inform risk-informed decision-making at the tactical and operational levels.

Read the complete text