XFdtd's system sensor collates discrete frequency data, and is generated by XF without the need for a specific sensor. The following results are calculated when there is a single active port in the simulation space.

When S-parameters are requested while creating a simulation, each active port has both a corresponding run and associated output. The tabular results available for the system sensor are consistent with a single active component. In some cases, the results are equivalent to those computed for a circuit component.

Net available power, $P_{av}$, is equivalent to the power available from either the active circuit component or waveguide.

$$P_{av} = P_{av\text{,active source}}$$

Net input power, $P_{in}$, is equivalent to the input power from either the active circuit component or waveguide.

$$P_{in} = P_{in\text{,active source}}$$

Net component loss, $P_{cl}$, is the sum of the component losses for each inactive circuit component in the simulation. It is written as

$$P_{cl} = \sum_i^MP_{cl,i}$$

where $M$ is the number of inactive components in the simulation.

Net waveguide loss, $P_{wl}$, is the sum of the waveguide losses for each inactive waveguide in the simulation. It is written as

$$P_{wl} = \sum_i^NP_{wl,i}$$

where $N$ is the number of inactive waveguides in the simulation.

Net additional component loss, $P_{acl}$, is only computed for distributed components. It is the sum of the additional component losses for each distributed circuit component in the simulation. It is written as

$$P_{acl} = \sum_i^OP_{acl,i}$$

where $O$ is the number of active and inactive distributed components in the simulation.

Net matching circuit loss, $P_{mcl}$, is only computed when a feed includes a matching circuit. It is the sum of the matching circuit losses for each feed—either active or passive— in the simulation. It is written as

$$P_{mcl} = \sum_i^QP_{mcl,i}$$

where $Q$ is the number of feeds containing a netlist matching circuit definition in the simulation.

Dissipated power, $P_d$, is a measure of the amount of energy lost in materials, both electric and magnetic. When the near-field method is specified, XF computes $P_d$ per cell edge based on the associated electric field value, $E$, material conductivity, $\sigma$, and volume, $\Delta v$.

$$P_{d,\text{per cell edge}} = \frac{1}{2}\sigma|E|^2\Delta v$$

The system sensor reports total dissipated power in all materials, dissipated power per material, and dissipated power in tissues versus non-tissues. When the far-field method is specified, XF computes only the total dissipated power based on

$$P_d = P_{in} - P_{cl} - P_{wl} - P_{acl} - P_{mcl} - P_r$$

Radiated power, $P_r$, is the amount of power exiting the simulation space. When the near-field method is specified, $P_r$ is computed as

$$P_r = P_{in} - P_{cl} - P_{wl} - P_{acl} - P_{mcl} - P_d$$

When the far-field method is specified, XF calculates radiated power based on the power exiting the six surfaces of the simulation space.

$$P_r = \frac{1}{2}{\oint}_S\text{Re}\{E\times H^*\}\cdot ds$$

System efficiency, $e_s$, is the ratio of radiated power to net available power.

$$e_s = \frac{P_r}{P_{av}}$$

Radiation efficiency, $e_r$, is the ratio of radiated power to net input power.

$$e_r = \frac{P_r}{P_{in}}$$

Standalone radiation efficiency, $e_a$, is a variation of radiation efficiency, but does not include the losses from inactive feeds and waveguides. The primary use case is to directly feed the system's antenna ports and export the efficiencies and S-parameters to an external tool. Each port has a matching circuit inserted, and the losses are not included in those ports because they change depending on the matching circuit at that port. If there are other simulation components that are not changed out, they effectively become part of the system. It is written as

$$e_a = \frac{P_r}{P_{in}-P_{cl}-P_{acl}-P_{mcl}-P_{wl}}$$