The paper presents an overview of the need address the spectrum crunch to deliver higher data rates for future wireless networks. Two techniques which are currently being investigated are the use of dynamic spectrum access and new frequency bands in the mm wave. Both techniques require thorough propagation studies. In the case of dynamic spectrum access it is necessary to ensure that the technique used to identify the primary user is sufficiently robust to reduce the possibility of interference. Two possible approaches are sensing the spectrum to identify unused frequency bands and the second approach involves the setting up of a large data base identifying the availability of frequency bands in space i.e. geographic location and time. To evaluate the feasibility of sensing, it is necessary to identify a minimum detection threshold level to detect the primary user. To evaluate the sensing approach a number of sensing devices have been evaluated under the EU funded project, CREW at Durham University with respect to a number of transmissions ranging from simple CW transmissions to more demanding WiFi transmissions, frequency hopping and frequency sweeps. Low cost and high cost sensing engines were evaluated including commercial off the shelf equipment and custom designed devices such as the receiver of the frequency swept channel sounder at Durham University. The test which was performed in the large anechoic chamber at Durham University with devices tested simultaneously to ensure uniformity identified high end devices as the only ones that are capable of detecting the primary user for the different types of modulation. This is an important result since for the sensing approach to be successful it would need the capability to detect the primary user with a high level of confidence at a low cost. The second approach which involves setting up a data base requires extensive radio propagation measurements and accurate modelling and is the subject of current research. At Durham University a high end channel sounder based on the digital chirp technique has been designed and implemented to study both approaches. The sounder is capable of both radio propagation measurements in various frequency bands and of sensing since it uses a high rate frequency sweep which covers 750 MHz bandwidth in 204.8 microseconds. This enables the detection of the signal at the packet level for accurate modelling of occupancy and the feasibility of using the idle time. To investigate the use of the mm wave band for future radio networks, it is necessary to characterise the radio channel in the expected bands. Candidate bands include the 60 GHz and the 28 GHz band. To enable such studies, a wideband channel sounder which is capable of measuring the 60 GHz band with 6 GHz bandwidth has been developed at Durham with measurements performed in different scenarios including on body networks, indoor and outdoor environments to estimate both the path loss and delay spread. Currently the sounder architecture is being upgraded to the 28 GHz band. The upgraded sounder will then be used for comparative studies between different bands as it enables multi-band channel characterisation which is of importance for multi-tier multi-frequency cellular networks where the mm wave band provides high data rates and the lower frequency bands provide coverage.
2