On the Performance of Single LDGM Codes for Iterative Data Fusion over the Multiple Access Channel

One of the applications of wireless sensor networks currently undergoing active research focuses on the scenario where the information generated by a data source S is simultaneously sensed by N nodes and therefrom transmitted to a common receiver. Based on the received information from such N nodes, such receiver infers the original information from S potentially more accurately than in the case of a single sensor. Often referred to as the CEO (Central Estimating Officer) problem [1], in this scenario we propose the use of single Low Density Generator Matrix (LDGM) codes for the transmission of the information registered by the nodes over the Multiple Access Channel (MAC). The corresponding receiver iterates between a soft demodulator, the set of N LDGM decoders and a soft-information fusion stage. Simulation results for the AWGN MAC channel show that 1) the proposed coding scheme outperforms the suboptimum limit assuming separated Slepian-Wolf distributed coding and capacity-approachi...

• Type: Journal
• Year: 2010
• Where: MACOM
• Authors: Javier Del Ser, Javier Garcia-Frias, Pedro M. Crespo, Diana Manjarres, Ignacio (Iñaki) Olabarrieta
• Keywords: Communications, Mac Channel, MACOM 2010, Multiple Access Channel, Wireless Sensor Networks