Geo-Intelligence based fiber fault localization system for rural India

Abstract

This study paper proposes Geo-Intelligence based algorithmic development of fiber fault localization system (FFLS) for rural India. This research is to support fiber fault localization for 2, 50,000 Gram Panchayats (GP) under Govt. Of India NOFN (National Optical Fiber Network) project. In absence of basic infrastructure facilities in rural areas e.g. approaching road etc., the biggest challenge for this technical work is to digitize existing fiber based infrastructure i.e. fiber with attributes, join boxes, telephone exchanges along with incremental fiber approaching towards GPs. Geo-Intelligence based FFLS require accurate Geographical Information System (GIS) mapping of fiber with junction boxes and other important places i.e. exchanges, BTSs, FPOI(Fiber Point Of Interconnect) etc. For PON (Passive Optical Network) (as GPON network has been selected as choice for network), the fiber link is point to multipoint topology. Traditional OTDR cannot be used for fault localization in GPON (tree topology) because OTDR is designed for point to point topology. Since GPON is point to multipoint (tree topology) technology, hence present algorithmic development for fault detection is useful for traditional OTDR. The development of FFLS algorithms includes a). Extraction of technology based topological GIS mapped fiber sections (from central office to CPE) b). Extracting equipment components (active and passive) inventory c). Predicting extra fiber length between two nodes i.e. fiber spool. d). Co-relation of OTDR traces with alarm based information and Utilizing images for referencing above information. In present study ITU-T G.984.x Gigabit Passive Optical Network (GPON) based practical FFLS system has been demonstrated in lab environment.