IEEE Consumer Electronics Magazine - July/August 2021 - 97

Table 1. Characterization of the System.
Server
IoMT device
PUF
Avg. time to
generate the key
Avg. time to
connect to network
Avg. time to send
packets
Average power
consumed
PUF Uniqueness
PUF Reliability
PUF Randomness
Arduino MKR
VIDOR 4000
FPGA
1 s
3.5 s
5ms
0.58 W
49.49%
48.2%
62.02%
SBC
Table 2. Comparison With Existing Protocols.
Protocol
Advantages
Raspberry
Pi + FPGA
Uses an
FPGA
3.45 s
3.5 s
5ms
3.1 W
49.49%
50.02%
50.29%
storing the data on the server. As shown in
Figure 8(a), only the public key and the signature
are stored on the device but no identification or
tracking information is stored that can be traced
back to the user. Figure 8(b) shows the output of
the server, the SBC in the case of the prototype,
the number of people that are present in the
area at the given point of time. As shown in the
Figure 8(b), once a person with PIM enters the
area, the count increases and once the person
exits, the count decreases keeping track of the
hosts in the area. Table 1 shows the characterization
of the prototype. As presented in the
table, the time taken by the system to connect to
a Wi-Fi network once it has entered the area. The
average time taken to generate the key is 1 s in
the first testbench with PUF integrated and 3.2 s
to generate the key in the second testbench with
a modular design. This time depends on the
design of PUF. Once the keys are generated, the
same key is used for the session in progress. All
further time delay depends only on the latency
added by the network. The power consumption
of both the testbenches was measured. For the
first testbench, the FPGA is present on the Arduino
board. The average power consumption of
the board is 58 mW. The second testbench
has the SBC connected to the FPGA board. The
average power consumption of the SBC with
FPGA is 3.1 W.
July/August 2021
Deepsort
techniques7
Social distancing
between patients
is monitored
through cameras.
PIM (This
article)
PIM uses a
standalone
device to track
the number of
patients in a
location.
Access to the
cameras in
location has to be
granted for the
protocol.
Mobile
phone
location
Tracking10
Number of
patients in a
given location is
tracked using
mobile data.
Patient mobile
phone
information is
stored at a
central location.
Mobile
phone
application6
application
installed on a
mobile phone for
contact tracing.
Reasons for less
adaptability
Patient personal
information is
stored on the
mobile phone
and location data
are transmitted
to the cloud
-
Comparison of PIM With Existing Protocols
Many protocols to track a COVID-19 patient
were already proposed as presented in Table 2. A
contact tracing application has been proposed to
track the patient movements throughout the day
and inform the number of contacts during the
period.6 A protocol that uses the mobile phone
location to track the number of patients in a given
location has been proposed in the work by Rahman
et al.10 A deep-learning-based tracking protocol
was also presented in the work by Punn et al.7
Compared to PIM, these protocols use the data
from the patient and the user is hesitant in providing
the respective information for the application.
PIM uses a standalone device for patient
tracking and does not disclose any patient information.
The device itself cannot contain any
information pertaining to patient. This helps in
maintaining the privacy of the patient compared
to the other proposed protocols.
CONCLUSION AND FUTURE
RESEARCH
This article presents a novel PIM protocol to
track the host carrying the virus in a crowded
area. A novel PUF architecture is proposed for
97
IEEE Consumer Electronics Magazine - July/August 2021

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - July/August 2021
