Detecting faults in IP-PBX with deterministic dendritic cell algorithm

作者：Lu Lin^*; Liang Yiwen; Yang Chao; Song Shiwei

来源：Advances in Information Sciences and Service Sciences, 2011, 3(11).

DOI：10.4156/AISS.vol3.issue11.57

• Anomaly detection
• Artificial Immune System
• Call detail records
• Dendritic cells
• IP-PBX

The data of Voice over Internet Protocol (VoIP) is transmitted over the packet switching network, so the latency, packet loss and jitter caused by network congestion or equipments failure also infect the quality of VoIP service, especially in the IP-PBX, and finally obstruct the development of telecommunication technique and markets. The real-time analysis of calling log is one of the most useful methods to detect problems. We propose a method to use the Deterministic Dendritic Cell Algorithm (dDCA) to analysis a typical call log Call Detail Record (CDR). In this paper, we discuss the characters of DCA for the real-time analysis of large time-ordered data, as it has very low CPU processing requirements and does not need extensive training period, and it has good results for this kind of data. Moreover, CDR's features and processing requirement make it a very suitable processing object for DCA. At last, we choose the dDCA to process the detection and analysis, as it is an extension of DCA by removed some useless randomize parameters and procedures for engineering implementation and applications. As a novel algorithm inspired by both theoretical immunology and observed immune functions, principles and models, dDCA would be a beneficial method to help improve the quality of service in VoIP development.

• 出版日期2011