Instrumentation & Measurement Magazine 23-9 - 59

DSA policy allows users of the spectrum resource to dynamically adjust their spectrum usage to achieve more efficient use
of the spectrum, potentially with such dynamic access being
under the oversight of a management system, such as a spectrum database.
In order to avoid harmful interference to the primary users, it is vital to be aware of the spectrum context in which the
user is located. To do so, spectrum sensing can be a preferable approach for evaluating available spectrum opportunities.
With more common application in current regulatory realms,
spectrum sensing might help greatly to populate and enhance
information used in spectrum databases, though empirical propagation modelling. With this need in mind, the IEEE
Standards Association formed a task group with the aim of
developing a standardized system for " Spectrum Characterization and Occupancy Sensing " (SCOS) with a proposal
named 802.15.22.3 [7], [8].
The standardization process has focused on the description of the normative features for characterizing the devices
involved in the spectrum sensing task, by defining the interfaces for entities in the system to interact, the format of the data
to be exchanged and the description of system configurations.
To provide reliable results that can be interpreted consistently
across a distributed network of sensors of varying types or capabilities, the standard describes that sensing data must be
accompanied by relevant metadata gathered at the time of the
sensing activity regarding physical environment conditions,
RF antenna specifications, calibration data, sampling system,
and details of the signal processing technique used.
This manuscript reports on the recent outcomes of this task
group and it outlines the current standard being proposal. In
the second part, a preliminary implementation is reported, in
which commercial Software Defined Radios (SDRs) are employed as sensors and the electromagnetic environment is
emulated using laboratory vector signal generators.

Problem Statement and Proposed Solution
The application of data generated by a managed network of
SCOS nodes is myriad - the most pressing are:
Radio planning: actionable data on actual radio spectrum
utilization by network operators deploying networks, geolocation database operators (GLDB) looking to optimize
available channels for secondary use, and national spectrum
regulators managing licencing.
Spectrum forensics: determination by regulators, law enforcement authorities and network operators of interference
sources or unauthorized use of radio equipment.
Regulatory policy: radio users, academic researchers and
governments needing to understand changes in radio environment to assist in formulating policy, with respect to issues
such as noise floor, band utilization and interference sources.
Each of these use cases would likely involve a different
type of user. To ensure this standard has flexible application,
December 2020	

the user and user requirements are not defined. Users of any
type interact with the system through a standardized interface
to identify spectrum sensing resources and schedule tasks on
these resources, based on credential-based authorization they
receive from the " SCOS Owner, " the operator of the sensing
system.
Currently, the cost of typical commercially available sensing systems is generally prohibitive for the pervasive national
spectrum sensing capability that would be required in an efficient Dynamic Spectrum Access model, or to provide near
real-time, comprehensive SCOS capability for truly effective
spectrum management.
However, the state of the art in sensor hardware and
methods for spectrum characterization has accelerated dramatically in the past few years, with extremely low cost, high
performance hardware becoming available and still evolving
rapidly, both in the radio space and the compute space. This,
coupled with the trend of moving advanced computation, analytics and visualization functions off local devices and into
remotely hosted compute systems ( " the Cloud " ) creates an
opportunity to develop a standardized architecture and data
transfer specification that can bring about a new generation of
pervasive, distributed spectrum sensing networks that can be
deployed at low cost.
To this end, the IEEE 802.22.15.3 Task Group has developed
a draft standard for SCOS sensor networks with the following
key design goals:
Sensor device agnostic: System allows for any device with
sensing capability to be recruited into the SCOS system-
whether for-purpose sensing devices, or devices with in-built
sensing capabilities such as smartphones, Wi-Fi access points
and TVWS base station devices.
Sensing technique agnostic: The standard allows flexible
adoption of new Software Defined Radio (SDR) technologies
and software-based analytic techniques as they develop. Sensing capabilities and functionality can still be accurately and
consistently described through a standardized model, allowing data processing systems to account for variances in data
quality from disparate sensors, and to allow data from distributed and heterogeneous networks of sensors to be aggregated.
Deployment model agnostic: The sensor network can be deployed in multiple topologies ranging from operation as
completely autonomous nodes to a schedule-based managed
service that leverages a multitude of devices formed into a sensor network. This is because the user requesting the sensing
activity, and the data generated from it, is independent of the
operation of the sensor network (i.e., the data acquisition and
analytics methods are abstracted from the underlying operational system).
These system requirements are enabled through the following functional requirements:
◗◗ The protocol for creating and managing the sensor
network, and the protocol for establishing sensing tasks

IEEE Instrumentation & Measurement Magazine	59
Instrumentation & Measurement Magazine 23-9

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 23-9
