A flexible filter for synchronous angular resampling with a wireless sensor network

Many tasks in the context of data acquisition with a wireless sensor network, e.g. angular resampling or jitter compensation, can be solved by reconstructing a uniformly sampled signal from a nonuniformly sampled one. A filter structure that is capable of performing this task has been adopted from the literature. It has been implemented in fixed-point arithmetic on a wireless sensor node. Compared to a floating-point implementation, this fixed point implementation has been shown to provide the same signal quality at only about one-tenth of computation time. Furthermore, it has been proven to be suitable for real-time operation on a resource constraint wireless sensor node. The implementation has been tested for the application example of synchronous angular resampling, but it can also easily be used in other resampling applications. Experiments with simulated data, a waveform generator and at an induction motor test bench show that the implemented filter is effective in resampling signals over a wide range of speeds and speed changes. In all cases its signal quality in terms of amplitude and frequency accuracy is at least as good that obtained using sample-and-hold resampling and surpasses it clearly in terms of signal-to-noise and distortion ratio (SINAD).