Prior art

The following prior art was cited by the examiner in US patent 7577099.

1. US20020019870A1

• Full Citation: US20020019870A1, "Proactive on-line diagnostics in a manageable network", filed June 29, 2000, published February 14, 2002, assigned to International Business Machines Corporation.
• Publication/Filing Date:
  • Filing Date: June 29, 2000
  • Publication Date: February 14, 2002
• Brief Description: This patent application describes a system and method for proactive on-line diagnostics in a manageable network. It focuses on diagnosing potential problems before they lead to catastrophic failures. The system identifies network components that could potentially cause network problems and determines their relationships. It then performs diagnostic actions on these components to determine their status and predict future failures, generating diagnostic reports with suggested corrective actions. This involves collecting monitoring data from network elements, analyzing dependencies between components, and using historical data to predict faults.
• Potential Anticipation (35 U.S.C. § 102): US20020019870A1 generally discloses methods for network fault diagnosis and monitoring data collection. It covers the broad concept of "generating an observation from monitoring data associated with network elements in the network" and implicitly "querying a database for a plurality of models of risks" by analyzing component dependencies. However, it does not explicitly disclose the specific greedy approximation algorithm used in US7577099, nor the specific mathematical definitions of "hit ratio" and "coverage ratio" or their maximization as defined in claims 1, 2, and 3. Therefore, while it provides a foundational understanding of network diagnostics, it is unlikely to anticipate the specific algorithmic steps for hypothesis calculation defined in US7577099.

2. WO2006005665A2

• Full Citation: WO2006005665A2, "Method for reacting to changes in context by means of a neural network, and neural network used for reacting to changes in context", filed July 9, 2004, published January 19, 2006, assigned to Siemens Aktiengesellschaft.
• Publication/Filing Date:
  • Filing Date: July 9, 2004
  • Publication Date: January 19, 2006
• Brief Description: This international patent application describes a method for reacting to changes in context using a neural network. The neural network detects changes in context, for example, in technical systems, by processing input data and adapting its behavior based on these changes. The context is represented by specific variables or parameters, and the neural network learns to recognize patterns indicative of context changes. This is broadly applicable to various systems where adaptation to changing conditions is required.
• Potential Anticipation (35 U.S.C. § 102): WO2006005665A2 describes the use of neural networks for detecting changes in context, a general pattern recognition problem. While network fault localization could be seen as a form of change detection, this prior art does not specifically address network fault localization, shared risk link groups, or any method of calculating a hypothesis using a greedy approximation with defined hit and coverage ratios as described in US7577099. The core of WO2006005665A2 is about neural networks and context adaptation, which is distinct from the specific problem and algorithmic solution presented in US7577099. Therefore, it is unlikely to anticipate any of the claims of US7577099.