IEEE Technology and Society Magazine - Winter 2013 - 60

countries that have not signed on to
certain international agreements.

PTM in the Future: Sensors
Everywhere, and an
Interconnected World
The amount and public accessibility of data already extends beyond
that of STM which operate under
official guidelines and treaties
between governments that define
formal limits on data acquisition
and protocols for sharing with participating nations. PTM exists in
numerous academic and national
seismic research institutions that
supplement the IMS for detecting
and locating low-yield underground
nuclear tests that violate the Comprehensive Test Ban Treaty. Another
example is the Institute for Science
and International Security based in
washington, DC, which analyzes
unclassified data from commercial
photo reconnaissance satellites to
monitor activities such as nuclear
fuel enrichment or missile test preparations that portend developments
of concern to arms control and nonproliferation efforts.
with the rapid advances in information technology that can be, and
are being, introduced into ubiquitous
mobile sensors, many more pathways
for data are rapidly opening up for a
less structured form of PTM - i.e.,
societal monitoring. In particular,
the rapid increase in the market for
smartphones can be very useful for
PTM. There is no need to require that
a single smartphone provide the sensitivity and discrimination needed for
treaty verification. It is the PTM sensor network's capability and capacity
that are relevant. redundancy and a
broad spatial distribution of sensors
provide both robustness and discrimination advantages. Moreover,
the sensor technology insertion time
frame is surprisingly short. The typical turnover/replacement time for
smartphone users is 18-24 months,
and so new sensors could be implemented rapidly.
The overall net effect of the rapidly evolving potential of public
60

and societal monitoring is to make
it much more difficult for governments to implement policies that
violate established agreements and
get away with such deception for
very long. Many more people will
know what is going on, around
them and around the globe.

Some Challenges of
Implementing PTM Treaty
Verification
Here we identify and discuss some
of the challenges associated with
fully implementing the PTM verification concept. Addressing any
of these issues in full is beyond the
scope of this paper, and we hasten to point out that some of these
challenges pose significant hurdles to PTM implementation. we
highlight these points as potential
research and development areas for
the technical community.

The Data Integrity Problem:
Detecting Spoofing and
Deception
An issue that arises immediately for any PTM-generated
sensor data is the data integrity
problem. How can the PTM system detect instances of spoofing
and deception, where a nation (or
an individual) modifies the data or
the accompanying header information or metadata (such as sensor
type, calibration information, etc.)?
This could arise in various ways.
A nation could attempt to conceal
or obscure activity through PTM
information manipulation. Or a
nation could attempt to modify the
PTM data stream emanating from a
rival country, to implant deceptive
information that implies a treaty
violation. Or a malicious hacker
might find the PTM sensor data
stream to be an irresistible target.
A substantial body of cybersecurity analysis has been done [8], [9]
on the problem of treaty sensor data
authentication, from the perspective
of public-private key exchange, as
well as steganography (embedding
a digital watermark in the noisy

portions of the data itself). PTM
sensors differ from more traditional
verification measurement systems
in that neither the physical data collection hardware nor the acquisition
software is under close supervision.
The severity of the data integrity problem depends to a large
extent on the fraction of the PTM
sensors that are deceitful or compromised. It would certainly be
difficult to contend with the situation where a nation floods a country (either theirs or someone else's)
with pseudo-PTM sensors that
were fiendishly clever in generating
fake data. This would essentially
amount to PTM denial-of-service
attack, however, and would be readily detectable. On the other hand, if
the legitimate PTM sensors comprise the dominant fraction of the
sensing network, we can envision
identifying the deceitful ones since
they would be anomalous outliers.
we must also guard against
malicious data modification by
individuals or groups, as opposed
to nations. There are numerous
examples in the press of instances
where commercial or government
data systems are infiltrated by
hackers. In this regard we don't see
the PTM data as being particularly
different from other information,
and employing best practices for
cybersecurity is as important for
PTM as it is in other domains.
There is one straightforward validation method that is intrinsic to the
PTM sensor approach. Any member of the public with local access
to their own sensor data can always
check to make sure the data stored on
a central repository are in fact identical to the local copy, which is under
their control. This cross-validation
could be carried out in a secure, password-protected automated certificate
exchange process.

Signal Discrimination
Problem
Sifting through PTM data harvested from a variable and dynamic
collection of sensors to extract

IEEE TECHNOLOGY AND SOCIETY MAGAZINE

wINTEr 2013

Table of Contents for the Digital Edition of IEEE Technology and Society Magazine - Winter 2013