Noise cancellation for continuous wave mud pulse telemetry based on empirical mode decomposition and particle swarm optimization

Mud pulse telemetry (MPT) is the most widely implemented communication method in measurement while drilling (MWD) systems. However, the signal-to-noise ratio of received signals on surface in MPT systems is relatively low because of the various background noises the signals have been exposed to during the drilling process. Due to the reciprocating action of the mud pump, pump noise could be the most detrimental factor to the mud pulse signal. Hence, pump noise cancellation plays an important role in the mud pulse telemetry. This paper presents a novel approach for pump noise cancellation by making use of the optimal signal reconstruction based on the combination of the empirical mode decomposition (EMD) and the particle swarm optimization (PSO). The proposed method comprises the pump noise recovery and the pump noise subtraction. The pump noise recovery is implemented by EMD decomposing the pump noise into a series of subcomponents and by PSO searching for the optimal weights to reconstruct the pump noise. Then the pump noise is subtracted from the received signal. To validate the effectiveness of the proposed method, both simulation tests and field experiment were carried out. The noise cancellation results of the simulation tests showed a good agreement with the modulated signal. In the experiment, the downhole information data was successfully demodulated after the implementation of the EMD-PSO based noise cancellation method. Meanwhile, the proposed method could achieve a lower bit error rate than the pump noise filter method. The findings of this study indicate that the optimal signal reconstruction has a promising application in MPT systems in the future.

Automated summary

This paper presents a novel approach for pump noise cancellation in mud pulse telemetry systems, utilizing the combination of empirical mode decomposition (EMD) and particle swarm optimization (PSO) for optimal signal reconstruction. The proposed method successfully reduced pump noise in simulation tests and field experiments, demonstrating a promising application for future MPT systems.