Improved pipelined wavelet implementation for filtering ECG signals

Abstract

We present a novel algorithm for digital filtering of an electrocardiogram (ECG) signal received by both stationary and non-stationary sensors. The basic idea of digital ECG signal processing is to extract heartbeat frequencies, which are found to be normal in the range between 50 and 200 beats per min. The extracted heartbeat frequency is found to be irregular if the rate increases or decreases and serves as evidence for a diagnosis of a complex physiological condition. A lot of noise can be generated from the environment, including the electrical energy supply (50 or 60 Hz), breathing, physical movement, muscles etc. We experimented with several digital band pass filters including the finite response filter, Butterworth filter and a filter implementing the digital wavelet transformation. Classical programming of a digital wavelet transformation includes a lot of operations, which increase the algorithm complexity and, therefore, cannot be recommended for smartphones or other mobile devices, due to their limited resources, such as battery life, storage capacity, and processing power. In order to realize a solution that will run on a smartphone and wearable ECG sensor, we faced a challenge to enable a sufficient quality of service and develop an efficient algorithm that will preserve the mobile phone resources. We implemented a new improved wavelet filter, which uses a circular buffer and, when compared with the classical solution, it reduces the number of memory accesses and instructions that transfer values between data arrays. This algorithm is a highly efficient solution for a smartphone (mobile device) since it decreases the processing time 15–20 times and also saves battery life.