IEEE - Aerospace and Electronic Systems - May 2021 - 18

A Systematic Literature Review: Is Military Cognitive Radio System on the Brink of the " Valley of Death " ?

Figure 1.
CCMM viewpoints within respective communities of interest.
Adopted from [19].

proof-of-concept [16], [17]. Therefore, articles describing
the use of simulation have been coded to TRL 3. In
this category, a significantly high number (63) of
papers have been included. If the paper shows specific
traits being a very narrow or early research paper or if
the paper is part of a broader research effort, the coding to TRL 3 may have been adjusted by a level þ-1.
As the fourth coding stage, the CCMM, depicted in
Figure 1, was applied. The model has been developed
from the Zachman framework for enterprise architecture
to support the military capability-based planning domain.
Koivisto and Tuukkanen [18] is an initial attempt to apply
CCMM to military wireless communications R&D. For
this survey, we shall apply capability viewpoints with
emphasis on left-hand side viewpoints in Figure 1. The
idea is to map our source data to the structure of the design
artifacts of military capability viewpoints.

RESULTS
Within military CR research, the keynote by Fette [9] should
already be considered as seminal. In this article, Dr. B. Fette,
at that time with DARPA, outlines the first 14 years of the
cognitive radio research and development since the inception of the notion of CR. The paper argues that, for the military, the essential feature of CR is not necessarily the
enhanced user experience, nor spectrum access alone, but
the full range adaptivity and flexibility in support of warfighter communications. Another important, yet easily overlooked, observation is that, although cognition has been
widely researched from the spectrum access perspective, the
military communication system needs to implement cognition at several layers of the communication protocol stack to
provide the expected performance improvement.
18

Out of the total of 193 articles1 collected for this
review, seven are doctoral dissertations, 14 surveys, 33
journal articles, and 139 conference papers. However, the
distribution of articles by publication type turned out not to
be a useful metric. The minimal number of doctoral dissertations is in dire contrast to generic civilian CR research,
whereby a large number of dissertations were excluded
because they did not have any recognizable bearing or relation to military CRSs. The few military CRS oriented dissertations included, focus on security-related topics mostly
involving electronic warfare measures. Of the 14 surveys
included in our data, we point to Tang and Watson [10].
This report summarizes military cognitive radio research
by the end of 2014. The report covers different national and
organizational research roadmaps on military CRs and considers the significance of cognitive radio networks for
future tactical wireless communications. According to the
report, CRNs are expected to provide benefits such as
dynamic spectrum access (DSA), increased communications resilience in dynamic, heterogeneous environments,
and may provide the basis for tactical electronic warfare.
Many multinational research task groups of North
Atlantic Treaty Organization's (NATO) Science and
Technology Organization (STO) present publicly available conference papers in addition to formal research
reports. Some of the reports are also available in the public
domain, e.g., [11]. NATO STO reports used for this article
have been tagged as surveys due to their contents and the
way they address the topic. All of the included NATO
papers address dynamic spectrum access one way or
another and point toward the need to adjust current rigid
administrative and bureaucratic mechanisms.

RESULTS BY TOPICS
As can be seen in Table 1, topics of waveforms, security,
modeling spectrum use, and spectrum sharing have
received the most attention (71%) in our dataset. In contrast, the topic of economics has a single reference.
Within the topic of machine learning, we observe that
most years yield a steady two to three papers on the topic,
whereas the sudden late increase in 2019 to 8 papers is noteworthy. As an example, the research team of the Royal Military Academy of Belgium suggests the use of machine
learning techniques at the cognitive radio to avoid intentional, hostile jamming or to support channel selection.
The category of theoretical analysis of dynamic spectrum access contains two papers. Potentially the most exciting category of papers is related to security and electronic
warfare. This is the second most populous category within
the dataset with altogether 41 papers. Within this category,
1

Full listing of the reviewed articles is available from the corresponding author: topi.tuukkanen@mil.fi

IEEE A&E SYSTEMS MAGAZINE

MAY 2021

IEEE - Aerospace and Electronic Systems - May 2021

Table of Contents for the Digital Edition of IEEE - Aerospace and Electronic Systems - May 2021