ASHRAE Journal - May 2021 - 31

TECHNICAL FEATURE

Occupant Sensing Technologies

FIGURE 1 Information grades of occupant data.

Occupant Data

We then present the most promising sensing technologies and data collection mechanisms currently available
to acquire occupant data at these six IGs; introduce key
OCC metrics; present a case study in which we explore
the use of these metrics in indoor climate control; discuss implementation challenges related to OCCs in real
buildings; and provide recommendations.

Preference

5

6

Occupant Count

3

4

Melﬁ, et al., categorize occupant sensing approaches
Presence
1
2
into two groups: implicit and explicit.6 Implicit sensing infers occupancy levels without directly monitoring
occupant mobility or behavior and generally relies on
Building
Zone
CO2, humidity, lighting or plug load measurements.
Spatial Resolution
While implicit sensing provides valuable insights
into occupancy levels, it requires
TABLE 1 Occupant information grades (IG) extractable at different deployment densities of various occupant sensing technologies, with privacy risks and infrastructure availability in existing buildings.
building-speciﬁc calibration with
SENSING
SENSING
OCCUPANT IG PRIVACY INFRASTRUCTURE
ground-truth occupancy data (manDEPLOYMENT DENSITY
TECHNOLOGY
APPROACH
(SEE FIGURE 1)
RISK
AVAILABILITY
ually counting people entering/
At Building Entrances
1
Low
High
leaving a building for a few days)
Motion Detectors Explicit Counting
Inside Each Zone
2 Medium
Medium
or inverse modeling to remove the
Return Air Inlet of an AHU
3
Low
Medium
effects of inanimate processes (e.g.,
CO2 Sensors Implicit Counting
Inside Each Zone
4
Low
Low
the effect of inﬁltration and ventilaWi-Fi
Coverage
Without
tion on CO2 measurements) to esti3
Low
High
Redundancy
mate occupant counts. In contrast,
Wi-Fi Explicit Counting
Low (Require
Wi-Fi Coverage With
explicit sensing approaches detect
4 Medium
Specialized
Redundancy
Software)
occupants directly and provide
At
Building
Entrances
3
Medium
Low
People Counting Explicit Counting
reasonable estimates for occupancy
Cameras
At Zone Entrances
4
High
Low
with minimal preprocessing and
Centralized
Mechanism
without calibration.
to Collect Occupant
5
Low
High
Control and
Input
The list of commercially available
Feedback
Feedback Interfaces FromDirect
Occupants
With IEQ Monitoring
occupant sensing technologies is
Occupant Controllable
6 Medium
High
Thermostats
in Each Zone
7
constantly growing. Table 1 presents a few of the most promising
and widely used, listing three explicit occupant countMotion detectors (passive infrared [PIR] and ultraing technologies (motion detectors, Wi-Fi and peoplesound) are the de facto standard in occupancy sensing,
counting cameras), one implicit occupant counting
to the point that building codes and standards prescribe
technology (CO2 sensors) and one method of collecting
their use in large spaces for lighting controls. Motion
occupant preferences.
detectors output a binary present/absent signal over a
Table 1 categorizes both the IGs and the privacy risks
timeout period of typically 5 to 30 minutes, but someprovided by these technologies at different deployment
times as short as 1 minute or as long as 2 hours. They
densities. Table 1 also categorizes occupant data infraare available in most commercial buildings, although
structure based on availability in existing buildings as
their coverage is often limited to a fraction of the spaces.
low (rarely used, thus will likely require capital investMotion detectors at main entry points can provide data
ment), medium (available in only a fraction of eligible
at IG 1 in Figure 1 (presence/absence at the building level).
buildings) or high (available in most buildings with a
To acquire data at IG 2 (presence/absence at the zone
modern automation system).
level), motion detector coverage must be extended to
M AY 2021

ashrae.org

ASHRAE JOURNAL

31
http://www.ashrae.org
ASHRAE Journal - May 2021

Table of Contents for the Digital Edition of ASHRAE Journal - May 2021
