The efficient and successful recovery of hydrocarbon resources depends heavily on the fast and high fidelity transmission of log (telemetry) signals to the surface. Regardless of the type of telemetry, wireline or wireless, the ultimate quality criterion for communication of log signals is the accuracy to which these signals can be reconstructed at the surface. An optimally designed system should be capable of transmitting as many log signals as possible, with each achieving a minimally acceptable fidelity level. In digital communication theory, the traditional metric for quantifying the transmission fidelity is the bit error rate (BER): the probability of bits being erroneously received at the receiver because of communication channel disturbances. Therefore, the commonly used criterion for designing digital communication systems is based on minimizing the average BER. This approach optimizes the system performance if all bits are equally important and need to be protected equally against transmission errors. Unfortunately, this criterion does not mean that high fidelity signals result.
This paper first presents the mean-squared error (MSE) distortion as a fidelity measure for the end-to-end performance of a digital telemetry system. We then introduce a specific transmission scheme—orthogonal frequency division multiplexing (OFDM)—that enables a simple implementation of a design criterion based on maximizing the end-to-end fidelity. We use the logging-while-drilling (LWD) acoustic telemetry scenario as a case study to demonstrate that, in contrast to optimizing BER, our design not only improves the quality of log signals but also results in significant gains in system throughput. Consequently, many more real-time log signals can be transmitted with acceptable fidelity than other techniques.