Abstract-The distribution of the aggregate interference power in large wireless networks has gained increasing attention with the emergence of different types of wireless networks such as ad hoc networks, sensor networks, and cognitive radio networks. The interference in such networks is often characterized using the Poisson point process (PPP). As the number of interfering nodes increases, there might be a tendency to approximate the distribution of the aggregate interference power by a Gaussian random variable, given that the individual interference signals are independent. However, some observations in the literature suggest that this Gaussian approximation is not valid, except under some specific scenarios. In this paper, we cast these observations in a single mathematical framework and express the conditions for which the Gaussian approximation will be valid for the aggregate interference power generated by a Poisson field of interferers. Furthermore, we discuss the effect of different system and channel parameters on the convergence of the distribution of the aggregate interference to a Gaussian distribution.Index Terms-Berry-Esseen bound, cochannel interference, cumulants, fading, Poisson point process (PPP).

Abstract-The importance of characterizing the aggregate interference power generated by a wireless network has increased with the emergence of different types of wireless networks such as ad-hoc networks, sensor networks, and cognitive radios. A cumulant-based characterization of this aggregate interference is an attractive approach. A number or recent papers in literature have dealt with cumulants of the aggregate interference but under specific scenarios. In this paper, we introduce a simple yet comprehensive method to determine the cumulants of the aggregate interference power originating from a wireless network. This method is quite general and applicable for finite and infinite network sizes, and it is flexible to encompass different system and propagation parameters such as large-scale fading, small-scale fading or even composite fading. We also investigate the behavior of these cumulants with respect to changes in the network size and fading distributions.

Abstract-Previous works studied the effect of many system parameters on spectrum sharing opportunities where secondary users access the spectrum of primary users. However, a parameter that has received little attention is the spatial size of the field of secondary users. Usually, the field size is assumed to be infinite. Using results developed for infinite fields might be too pessimistic leading to missing spectrum sharing opportunities. This paper studies the effect of the field size on spectrum sharing opportunities. We verify that asymptotic results obtained for infinite fields are applicable for finite but relatively large fields as well, i.e., when the radial depth of the field is much greater than the minimum distance to the primary user. We demonstrate that in some cases, however, asymptotic results are too pessimistic hiding some spectrum sharing opportunities. Moreover, the paper shows that in certain situations a small reduction in the field size may create spectrum sharing opportunities while in certain other situations a huge increase in the field size may not eliminate spectrum sharing opportunities. Our results also suggest the possibility of a secondary network to concurrently share the spectrum with a primary user without the need for spectrum sensing techniques or other cognitive radio functionalities.

Abstract-Massive deployments of Wireless Sensor Networks (WSNs) are expected in near future. In one of the most likely scenarios, these WSNs would share a licensed frequency band with a primary user. So, it is essential to understand the behavior of the interference generated by a WSN towards the primary user. This paper provides an asymptotic analysis of the average interference power generated by a WSN. The analysis is extended to a special but important shape of a sensor field. This shape can be used to provide an upper bound of the average interference power generated by any sensor field with an arbitrary shape. The paper shows that the expansion of the sensor field does not necessarily cause an increase in the average interference power. For most practical values of path loss exponent, the average interference power asymptotically approaches constant levels with the increase in the field size provided that the minimum distance from the field to the primary user is fixed. The paper provides expressions for these constants. Moreover, results indicate that a key parameter in determining the average interference power is the ratio of the radial depth of the field to the minimum distance from the field to the primary user. Also, this paper illustrates how a WSN can be equivalently represented by a single virtual node producing the same level of average interference power.