Impulse Noise Modeling in an Indoor Narrowband Power Line Communication Channel using M-QAM and a Software-Defined Radio Approach

Noise is generated on the indoor Narrowband Power Line Communication channel by the uncoordinated activities of several electrical devices connected to the power line networks, and by radiated environmental disturbances. There is a need to overcome these noise and channel impairments for reliable communication to be achieved. Statistical channel modeling of noise will facilitate the development and optimization of dependable PLC systems. Thus, in this work, we have developed a Software-Defined-PLC transceiver and test-bed that adopts 4, 8 and 16 Quadrature amplitude modulation schemes and uses the Universal Software Radio Peripheral and MATLAB/Simulink platforms to facilitate impulsive noise error measurement and modeling using the three-state Fritchman Markov Model (FMM) and the Baum-Welch Algorithm. The statistical channel models obtained are accurate derived channel models based on experimental measurement. The close match between the experimental and model error-free run distribution and error probabilities justify the modeling of the PLC memory channel using the three-state FMM. The model results obtained will assist in the implementation of error correction systems and novel optimization techniques in an impulsive noise PLC channel environment.