Implementation Issues in Spectrum Sensing

for Cognitive Radios

Danijela Cabric, Shridhar Mubaraq Mishra, Robert W. Brodersen
Berkeley Wireless Research Center, University of California, Berkeley

Abstract- There are new system implementation challenges

involved in the design of cognitive radios, which have both the
ability to sense the spectral environment and the flexibility to
adapt transmission parameters to maximize system capacity
while co-existing with legacy wireless networks. The critical
design problem is the need to process multi-gigahertz wide
bandwidth and reliably detect presence of primary users. This
places severe requirements on sensitivity, linearity, and dynamic
range of the circuitry in the RF front-end. To improve radio
sensitivity of the sensing function through processing gain we
investigated three digital signal processing techniques: matched
filtering, energy detection, and cyclostationary feature detection.
Our analysis shows that cyclostationary feature detection has
advantages due to its ability to differentiate modulated signals,
interference and noise in low signal to noise ratios. In addition, to Figure 1. Measurement of 0-6 GHz spectrum utilization at BWRC
further improve the sensing reliability, the advantage of a MAC creating time and location dependent in “virtual unlicensed
protocol that exploits cooperation among many cognitive users is
investigated. bands”, i.e. bands that are shared with primary users. Unique
to cognitive radio operation is the requirement that the radio is
able to sense the environment over huge swaths of spectrum
I. INTRODUCTION
and adapt to it since the radio does not have primary rights to
It is commonly believed that there is a spectrum scarcity at any pre-assigned frequencies. This new radio functionality
will involve the design of various analog, digital, and network
frequencies that can be economically used for wireless
communications. This concern has arisen from the intense processing techniques in order to meet challenging radio
competition for use of spectra at frequencies below 3 GHz. sensitivity requirements and wideband frequency agility.
Spectrum sensing is best addressed as a cross-layer design
The Federal Communications Commission’s (FCC) frequency
allocation chart indicates overlapping allocations over all of problem. Cognitive radio sensitivity can be improved by
the frequency bands, which reinforces the scarcity mindset. enhancing radio RF front-end sensitivity, exploiting digital
signal processing gain for specific primary user signal, and
On the other hand, actual measurements taken in downtown
Berkeley are believed to be typical and indicate low network cooperation where users share their spectrum sensing
utilization, especially in the 3-6 MHz bands. Figure 1 shows measurements.
The paper is organized as follows; Section II defines
the power spectral density (PSD) of the received 6 GHz wide
spectrum sensing function and proposes a cross-layer
signal collected for a span of 50?s sampled at 20 GS/s [12].
approach for its implementation. Section III considers RF
This view is supported by recent studies of the FCC’s
front-end and A/D requirements for spectrum sensing and
Spectrum Policy Task Force who reported [1] vast temporal
analog techniques for feasible implementations. In section IV
and geographic variations in the usage of allocated spectrum
we investigate digital signal processing techniques that can
with utilization ranging from 15% to 85%. In order to utilize
improve radio sensitivity and detect primary users’ presence.
these spectrum ‘white spaces’, the FCC has issued a Notice of
Section V presents the results from a cooperative sensing
Proposed Rule Making (NPRM – FCC 03-322 [2]) advancing
scheme, achievable gains and implementation issues . Finally,
Cognitive Radio (CR) technology as a candidate to implement
conclusions are presented in Section VI.
negotiated or opportunistic spectrum sharing.
Wireless systems today are characterized by wasteful static II. SPECTRUM SENSING
spectrum allocations, fixed radio functions, and limited
network coordination. Some systems in unlicensed frequency A “Cognitive Radio” is a radio that is able to sense the
bands have achieved great spectrum efficiency, but are faced spectral environment over a wide frequency band and exploit
with increasing interference that limits network capacity and this information to opportunistically provide wireless links
scalability. Cognitive radio systems offer the opportunity to that best meet the user communications requirements [2].
use dynamic spectrum management techniques to help prevent While many other characteristics have also been discussed as
interference, adapt to immediate local spectrum availability by possible additional capabilities, we will use this more
 rate, accuracy and power, so that digital signal processing
techniques can be utilized for spectrum sensing, cognition, and
adaptation. This also motivates research of signal processing
techniques that can relax challenging requirements for analog,
specifically wideband amplification, mixing and A/D
conversion of over a GHz or more of bandwidth, and enhance
overall radio sensitivity.

III. COGNITIVE RADIO FRONTEND

There are two frequency bands where the cognitive radios

might operate in a near future: 400-800 MHz (UHF TV bands)
and 3-10 GHz. The FCC has noted that in the lower UHF
bands almost every geographical area has several unused 6
Figure 2. Cross layer functionalities related to spectrum sensing MHz wide TV channels. This frequency band is particularly
restricted definition and consider physical (PHY) and medium appealing due to good propagation properties for long-range
access control (MAC) functions that are linked to spectrum communications. Furthermore, given the static TV channel
sensing as illustrated in Figure 2. allocations, the timing requirements for spectrum sensing are
Since cognitive radios are considered lower priority or very relaxed. The FCC approval of UWB underlay networks
secondary users of spectrum allocated to a primary user, a in 3-10 GHz indicates that this frequency range might be
fundamental requirement is to avoid interference to potential opened for opportunistic use. Furthermore, this band has very
primary users in their vicinity. On the other hand, primary user low spectral utilization, as indicated in Figure 1.
networks have no requirement to change their infrastructure Regardless of operating frequency range, a wideband front-
for spectrum sharing with cognitive networks. Therefore, end for a cognitive radio could have an architecture as
cognitive radios should be able to independently detect depicted in Figure 3. The wideband RF signal presented at the
primary user presence through continuous spectrum sensing. antenna of a cognitive radio includes signals from close and
Different classes of primary users would require different widely separated transmitters and from transmitters operating
sensitivity and rate of sensing for the detection. For example, at widely different power levels and channel bandwidths. As a
TV broadcast signals are much easer to detect than GPS result, detection of weak signals must frequently be performed
signals, since the TV receivers’ sensitivity is tens of dBs in the presence of very strong signals. Thus, there will be
worse than GPS receiver. extremely stringent requirements placed on the linearity of the
In general, cognitive radio sensitivity should outperform RF analog circuits as well as their ability to operate over wide
primary user receiver by a large margin in order to prevent bandwidths. In order to keep the requirements on the final
what is essentially a hidden terminal problem. This is the key analog to digital (A/D) converter at a reasonable level in a
issue that makes spectrum sensing very challenging research mostly digital architecture, front-end design needs a tunable
problem. Meeting the sensitivity requirement of each primary notch analog processing block that would provide a dynamic
receiver with a wideband radio would be difficult enough, but range control.
the problem becomes even more challenging if the sensitivity Reducing the in-band interference to a manageable level is a
requirement is raised by additional 30-40 dB. This margin is critical design problem, since the traditional strategy of narrow
required because cognitive radio does not have a direct band analog frequency selective filtering to avoid the wide
measurement of a channel between primary user receiver and dynamic range of interfering signals is not viable. The
transmitter and must base its decision on its local channel ultimate solution to this problem would involve a combination
measurement to a primary user transmitter. This type of of techniques, including adaptive notch filtering such as
detection is referred to as local spectrum sensing and the worst employed in UWB designs, banks of on chip RF filters
case hidden terminal problem would occur when the cognitive possibly using MEMS technology such as FBAR’s, and spatial
radio is shadowed, in severe multipath fading, or inside filtering using RF beam-forming through adaptive antenna
buildings with high penetration loss while in a close
neighborhood there is a primary user whose is at the marginal
reception, due to its more favorable channel conditions. Even
though the probability of this scenario is low, cognitive radio
should not cause interference to such primary user.
The implementation of the spectrum sensing function also
requires a high degree of flexibility since the radio
environment is highly variable, both because of different types
of primary user systems, propagation losses, and interference.
The main design challenge is to define RF and analog
architecture with right trade-offs between linearity, sampling
Figure 3 . Wideband RF/analog front-end architecture for cognitive radio
arrays. Other more sophisticated approaches could involve
active cancellation, because in the situation in which the
interfering signal is extremely strong, it is then possible to
decode the signal and provide an active canceling signal
before the A/D conversion process. While the active
cancellation approach will consume significantly more
hardware, it has the important advantage of ultimately being Figure 4. Implementation of an energy detector using Welch periodogram
more flexible. averaging
The spatial dimension provides several new opportunities.
The sensitivity of the sensing receiver can be increased by the a spectrum analyzer by averaging frequency bins of a Fast
exploitation of multiple antennas through diversity increase Fourier Transform (FFT), as outlined in Figure 4 [3].
Processing gain is proportional to FFT size N and
and range extension, which in effect could make it much more
sensitive than the primary users which it is trying to detect. observation/averaging time T. Increasing N improves
frequency resolution which helps narrowband signal detection.
Also, longer averaging time reduces the noise power thus
IV. SIGNAL PROCESSING TECHNIQUES FOR SPECTRUM SENSING
improves SNR. However, due to non-coherent processing
A key advantage of CMOS integration is that digital signal O(1/SNR2) samples are required to meet a probability of
processing can be used to assist the analog circuits. In case of detection constraint [5].
spectrum sensing the need for signal processing is two-fold: There are several drawbacks of energy detectors that might
improvement of radio front-end sensitivity by processing gain diminish their simplicity in implementation. First, a threshold
and primary user identification based on knowledge of the used for primary user detection is highly susceptible to
signal characteristics. In this section we discuss advantages unknown or changing noise levels. Even if the threshold
and disadvantages of three techniques that are used in would be set adaptively, presence of any in-band interference
traditional systems: matched filter, energy detector and would confuse the energy detector. Furthermore, in frequency
cyclostationary feature detector. selective fading it is not clear how to set the threshold with
respect to channel notches. Second, energy detector does not
A. Matched Filter differentiate between modulated signals, noise and
interference. Since, it cannot recognize the interference, it
The optimal way for any signal detection is a matched filter cannot benefit from adaptive signal processing for canceling
[4], since it maximizes received signal-to-noise ratio. the interferer. Furthermore, spectrum policy for using the band
However, a matched filter effectively requires demodulation is constrained only to primary users, so a cognitive user should
of a primary user signal. This means that cognitive radio has a treat noise and other secondary users differently. Lastly, an
priori knowledge of primary user signal at both PHY and energy detector does not work for spread spectrum signals:
MAC layers, e.g. modulation type and order, pulse shaping, direct sequence and frequency hopping signals, for which
packet format. Such information might be pre-stored in CR more sophisticated signal processing algorithms need to be
memory, but the cumbersome part is that for demodulation it devised. In general, we could increase detector robustness by
has to achieve coherency with primary user signal by looking into a primary signal footprint such as modulation
performing timing and carrier synchronization, even channel type, data rate, or other signal feature.
equalization. This is still possible since most primary users
have pilots, preambles, synchronization words or spreading C. Cyclostationary Feature Detection
codes that can be used for coherent detection. For examples:
TV signal has narrowband pilot for audio and video carriers; Modulated signals are in general coupled with sine wave
CDMA systems have dedicated spreading codes for pilot and carriers, pulse trains, repeating spreading, hoping sequences,
synchronization channels; OFDM packets have preambles for or cyclic prefixes which result in built-in periodicity. Even
packet acquisition. The main advantage of matched filter is though the data is a stationary random process, these
that due to coherency it requires less time to achieve high modulated signals are characterized as cyclostationary, since
processing gain since only O(1/SNR) samples are needed to their statistics, mean and autocorrelation, exhibit periodicity.
meet a given probability of detection constraint [5]. However, This periodicity is typically introduced intentionally in the
a significant drawback of a matched filter is that a cognitive signal format so that a receiver can exploit it for: parameter
radio would need a dedicated receiver for every primary user estimation such as carrier phase, pulse timing, or direction of
class. arrival. This can then be used for detection of a random signal
with a particular modulation type in a background of noise and
B. Energy Detector other modulated signals.
Common analysis of stationary random signals is based on
One approach to simplify matched filtering approach is to autocorrelation function and power spectral density. On the
perform non-coherent detection through energy detection. This other hand, cyclostationary signals exhibit correlation between
sub-optimal technique has been extensively used in widely separated spectral components due to spectral
radiometry. An energy detector can be implemented similar to redundancy caused by periodicity [6]. By analogy with the
 Signal processing techniques studied in this paper motivate
the need to study other feature detection techniques that can
improve sensing detection and recognize modulation, number
Figure 5. Implementation of a cyclostationary feature detector and type of signals in a low SNR regimes.
definition of conventional autocorrelation, one can define
spectral correlation function (SCF): V. COOPERATIVE SPECTRUM SENSING
1 ?? t/ 2 1
? X ?t, f ? ? / 2?X ?t, f ? ? / 2?dt
(1)
Sx ( f ) ? lim
?
lim *
In previous sections we have reviewed RF and Digital Signal
? t ??t / 2 T
T ? ? ?t? ? T T

Processing techniques to increase the probability of primary

where finite time Fourier transform is given by: user detection. The performance of these techniques is limited
t? T / 2
(2)
X T (t , v ) ? ? x(u )e ? j 2 ?vu du by received signal strength which may be severely degraded
t? T / 2 due to multipath fading and shadowing. Digital TV
Spectral correlation function is also termed as cyclic spectrum. measurements report standard deviations of 2.0 to 4.0 for log-
Unlike PSD which is real-valued one dimensional transform, normal shadowing effects [8]. In such a scenario cooperative
the SCF is two dimensional transform, in general complex- sensing may alleviate the problem of detecting the primary
valued and the parameter ? is called cycle frequency. Power user by reducing the probability of interference to a primary
spectral density is a special case of a spectral correlation user. In cooperative sensing we rely on the variability of signal
function for ? =0. strength at various locations. We expect that a large network
The distinctive character of spectral redundancy makes of cognitive radios with sensing information exchanged
signal selectivity possible. Signal analysis in cyclic spectrum between neighbors would have a better chance of detecting the
domain preserves phase and frequency information related to primary user compared to individual sensing.
timing parameters in modulated signals [6]. As a result,
overlapping features in the power spectrum density are non- There are three main questions regarding cooperative sensing:
overlapping feature in the cyclic spectrum. Different types of (a) How much can be gained from cooperation?
modulated signals (such as BPSK, QPSK, SQPSK) that have (b) How can cognitive radios cooperate?
identical power spectral density functions can have highly (c) What is the overhead associated with cooperation?
distinct spectral correlation functions. Furthermore, stationary
noise and interference exhibit no spectral correlation. To answer the first of these questions we designed a
Implementation of a spectrum correlation function for simulation environment where a group of cognitive radios
cyclostationary feature detection is depicted in Figure 5. It can attempt to detect a TV transmitter in the 700MHz band. Each
be designed as augmentation of the energy detector from radio may transmit if it decides (either individually or in
Figure 4 with a single correlator block. Detected features are cooperation with other users) that a primary user is not
number of signals, their modulation types, symbol rates and present. We do not assume any particular medium access
presence of interferers. Figure 6 illustrates the advantages of scheme used by this radio group and are interested in the
cyclostationary detection versus energy detection for maximum interference caused by any potential cognitive radio
continuous phase 4-FSK modulated signals. Distinct pattern of transmitter.
4-FSK modulation in a spectral correlation function is
preserved even in low SNR=-20dB while energy detector is Digital TV receivers are required to receive signals as low as -
limited by the large noise. 83 dBm without significant errors with a typical CNR of 15
dB [9, 10]. We assumed that any interference from the
cognitive radio network would appear as white noise to the
TV receiver and interference levels in the order of -98 dBm
(minimum received signal – typical CNR = -83-15 = -98dBm)
would significantly degrade receiver performance. We
assumed a cognitive radio network spread out in a circle of
radius 100m, located in a building 200m in height. Each radio
a) PSD of 4-FSK SNR=10dB b) SCF of 4-FSK SNR=10dB
can transmit with a power of 20dBm. The TV receiver was
located at a height of 3m, 10 km from the radio network. This
allowed us to use the standard Hata-Okumura model for
suburban environments [11]. Each cognitive radio performs
local sensing and decides on the presence of primary user
using sensing results from a certain fraction of cognitive
radios in the network. Figure 7 shows the probability of
interference to the TV receiver from the cognitive radio
network. The fraction of the network consulted by each
c) PSD of 4-FSK SNR=-20dB d) SCF of 4-FSK SNR=-20dB cognitive radio is varied between 0 (no cooperation), 10% and
Figure 6. Detection of a continious-phase 4-FSK using energy detection 20% . From the figure we see a drastic reduction in probability
and cyclostationary feature detection.
of interference as the fraction of radios consulted is increased. VI. CONCLUSION
A particularly noteworthy aspect is the reduction in
probability of interference as the number of cognitive radios in In the paper, we explore the new field of cognitive radios with
the network is increased. a special emphasis on one unique aspect of these radios -
spectrum sensing. We motivate the strong need for
sophisticated sensing techniques and established sensing to be
a cross-layer function. Firstly, we identify two key issues
related to the cognitive radio frontend - dynamic range
reduction and wideband frequency agility. Primary user
detection can be further improved by advanced feature
detection schemes like cyclostationary detectors which utilize
the inherent periodicity of modulated signals. Further,
individual sensing is not adequate for reliable detection of
primary users due to shadowing and multipath effects. In such
a case cooperative decision making is the key to reducing the
probability of interference to primary users.

REFERENCES

[1] FCC, Spectrum Policy Task Force Report, ET Docket No. 02-155, Nov
Figure7 Probability of Interference to TV receiver by a cognitive radio 02, 2002.
network with individual sensing and cooperative decision making [2] FCC. ET Docket No. 03-322. Notice of Proposed Rule Making and
Order, December 2003.
[3] A. V. Oppenheim, R. W. Schafer and J. R. Buck, Discrete-Tme Signal
While we can minimize interference to the primary receiver by Processing, Prentice Hall, 1999.
never transmitting, more sophisticated cooperation schemes [4] J. Proakis, Digital Communications, 3rd edition, Mc Graw Hill
[5] A. Sahai, N. Hoven, R. Tandra, “Some Fundamental Limits on
have to be designed to achieve optimal tradeoff between Cognitive Radio”, Proc. of Allerton Conference, Monticello, Oct 2004.
network capacity and probability of interference. In [7] a [6] W.A.Gardner, “Signal Interception: A Unifying Theoretical Framework
centralized network is proposed where the access point for Feature Detection”, IEEE Trans. on Communications, vol. 36, no. 8.
collects sensing results from all users. The access point sounds August 1988
[7] R.W. Brodersen, A.Wolisz, D.Cabric, S.M.Mishra, D. Willkomm, 2004
the channel and then performs channel allocation so as to meet White Paper: “CORVUS-A Cognitive Radio Approach for Usage of
the requested data rates of each user. The overhead associated Virtual Unlicensed Spectrum”, available online
with this scheme is in providing sensing results to the access http://www.bwrc.eecs.berkeley.edu/MCMA
point every time the channel conditions change. If channel [8] http://www.broadcastpapers.com/tvtran/BSDValidateDVBT02.htm
[9] ATSC, “ATSC Recommended Practice: Receiver Performance
coherency time is small, increment updates need to be Guidelines”, June 2004.
performed so as to reduce bandwidth requirements on the [10] FCC, “FCC OET Bulletin No. 69: Longley-Rice methodology for
control channel. A distributed cooperation scheme (as used in Evaluating TV Coverage and Interference”, Feb., 2004
the simulation environment presented in Figure 7) where [11] T. S. Rappaport, “Wireless Communications: Principles and Practice”,
2 nd ed., Pearson Education International, 2002
neighbors are chosen randomly may be easier to implement [12] Jing Yang, “Spatial Channel Characterization for Co gnitive Radios”, MS
but may not achieve the capacity of the centralized scheme. Thesis, UC Berkeley, 2004

One of the problems in cooperation is in combining the results

of various users which may have different sensitivities and
sensing times. Some form of weighted combining needs to be
performed in order to take this into account.

Cooperation also introduces the need for a control channel. A

control channel can either be implemented as a dedicated
frequency channel or as an underlay UWB channel. Wideband
RF frontend tuners/filters can be shared between the UWB
control channel and normal cognitive radio
reception/transmission. Furthermore, with multiple cognitive
radio groups active simultaneously, the control channel
bandwidth needs to be shared. With a dedicated frequency
band, a CSMA scheme may be desirable. For a spread
spectrum UWB control channel, different spreading
sequencing could be allocated to different groups of users.