Heart rate monitoring is crucial for detecting possible changes in human health. Therefore, in recent years, the development of applications and devices that monitor vital signs has evolved significantly. This work proposes a non-contact approach to heart rate monitoring through the processing of channel state information from a commodity Wi-Fi network. Various settings for signal processing parameters were evaluated. The experimental analysis counted on over 50 participants, each performing 17 different positions and activities. Results confirm that it is possible to achieve accurate heart rate detection for each position or activity, provided proper parameter configuration.