Multi-Mode, Multi-Band Spectrum Sensor for Cognitive Radios Embedded to a Mobile Phone Sami Kallioinen**, Mikko Vääräkangas*, Ping Hui, Jani Ollikainen, Ilari Teikari, Aarno Pärssinen** Nokia Research Center, Otaniemi, Espoo, Finland *Nokia Research Center, Helsinki, Finland **Currently with Renesas Mobile, Helsinki, Finland Vesa Turunen, Marko Kosunen, Jussi Ryynänen Aalto University School of Electrical Engineering Department of Micro- and Nanosciences SMARAD CoE Espoo, Finland Abstract— Localized spectrum sensing is an alternate for database centric approaches to solve secondary use of spectrum in cognitive radios. This can be carried out by using collaborative spectrum sensing where larger amount of small devices is utilized for local spectrum sensing. This paper describes an mobile device scale implementation of multi-mode, multi-band spectrum sensor for cognitive radio. Cyclostationary feature detector algorithm is utilized to detect digital television (DVB-T/H) on UHF band and IEEE802.11a/g on 2.4/5 GHz (ISM/WLAN) bands. A miniaturized spectrum sensing device encounters physical challenges; like limited size and battery capacity, but provides opportunities to establish a dense network and fast response to dynamic chances in signal conditions. Keywords – cognitive radio; spectrum sensing; mobile phone; white space devices; digital television; DVB; UHF; WLAN; cyclostationary; feature detector; intermodulation; radio frequency; broadband I. INTRODUCTION Spectrum sensing is broadly studied area in cognitive radios and many different algorithms have been developed to find energy [1] or specific protocols, especially TV transmissions and wireless microphones [2], in order to improve spectrum utilization. However, issues related to portability and small mobile phone sized devices are not addressed thoroughly in previous spectrum sensing literature [3]-[7]. Spectrum sensing capability especially in portable, battery limited devices is one of the reasons why database centric approaches are currently considered as core solutions for secondary use of spectrum [8],[9]. Attractiveness of more localized spectrum detection has not vanished [10] and various aspects need to be addressed in the field trials using realistic devices to complement trials using rack-sized equipment [3].The requirements proposed and later revised by Federal Communications Commission (FCC) in the US have been the first concrete step towards opening licensed frequency band for secondary use [11]. Sensitivity requirement in the FCC ruling is -114 dBm at 6 MHz channel. Required sensitivity level is achievable using known spectrum sensing algorithms as shown later. However it has been questioned that is current sensitivity requirement enough to protect primary users especially in the UHF band [12]. On the other hand, opportunities to determine spectrum collaboratively using larger amount of small devices have not been included in the current rulings. Another problem and key point in flexible spectrum use is to really find unused spectrum. This relates at least to exact positions of communicating devices and linearity requirements of spectrum sensors. We show by field measurements the importance of the problem when there are strong primary signals in the vicinity. By making sensitivity requirement stricter we decrease possibility of interference caused by white space devices (WSD) to primary signals but this increases unintentional false alarms caused by intermodulation results. Miniaturization of spectrum sensing devices is not straightforward for small, battery-operated devices. Because signal detection needs to have sufficient margin to protect primaries in the band of interest sensors need to observe signals from significantly lower levels than the actual communication typically requires. Although many algorithms can perform the task at very low SNR values noise of the receiver and especially losses in the antenna and front-end filter pose major challenges for practical implementations. As antenna form factors are larger at lower frequencies and relative bandwidth is typically proportional to receiver front-end losses sensing at UHF band will be performance limited mostly due to antenna losses and intermodulation performance of the receiver ASIC. These aspects will be discussed further in the paper with practical examples from field tests. In addition to UHF band the demonstrated device is also capable of detecting signals at 2.4 GHz ISM and 5 GHz WLAN bands. Although not discussed in detail power consumption will also be an issue if multitude of channels needs to be monitored frequently. The mechanism does not differ significantly from the principles of cellular paging although individual detection times may be relatively long and in many cases the detection over a larger bandwidth needs to be done sequentially. II. SPECTRUM SENSOR EMBEDDED TO A MOBILE DEVICE In order to conduct field tests using a device with realistic form factor a spectrum sensor was embedded into a Nokia N900 mobile computer with all functionalities. The choice caused some extra challenges since the N900 has not been designed for a mobile TV receiver. Spectrum sensor hardware has been designed on a separate printed circuit board (PCB) and it has been equipped with hardware which enables to receive desired frequency bands and realize all spectrum sensor functionality (Fig. 1). Fig. 2 shows the two complete signal paths that have This research is supported by Tekes (the Finnish Funding Agency for Technology and Innovation). CROWNCOM 2011, June 01-03, Osaka, Japan Copyright © 2012 ICST DOI 10.4108/icst.crowncom.2011.245877