Presence detection based on link parameter data and network parameter data

A method for determining human presence comprising:

• receiving, from a first device connected to a home network at a location, a first received signal strength indicator (RSSI) value and a first channel state information (CSI) value, the first RSSI value and the first CSI value corresponding to an acknowledgment packet sent from a second device to the first device in response to a CSI packet sent from the first device to the second device;
• generating, using a first prediction model and the first RSSI value, a first confidence value associated with first presence detection at the location;
• determining, from a first plurality of predefined weights associated with the first prediction model, a first weight, the first weight determined by at least comparing the first confidence value to a first threshold value;
• generating a first weighted confidence value associated with the first presence detection by multiplying the first confidence value with the first weight;
• generating, using a second prediction model and the first CSI value, a second confidence value associated with second presence detection at the location;
• determining, from a second plurality of predefined weights associated with the second prediction model, a second weight, the second weight determined by at least comparing the second confidence value to a second threshold value;
• generating a second weighted confidence value associated with the second presence detection by multiplying the second confidence value with the second weight;
• generating a weighted confidence value associated with presence detection at the location by adding the first weighted confidence value and the second weighted confidence value;
• determining that the weighted confidence value is greater than a third threshold value;
• and
• determining, presence of a human at the location.