A Unifying Framework for Local Throughput in Wireless Networks

With the increased competition for the electromagnetic spectrum, it is important to characterize the impact of interference in the performance of a wireless network, which is traditionally measured by its throughput. This paper presents a unifying framework for characterizing the local throughput in wireless networks. We first analyze the throughput of a probe link from a connectivity perspective, in which a packet is successfully received if it does not collide with other packets from nodes within its reach (called the audible interferers). We then characterize the throughput from a signal-to-interference-plus-noise ratio (SINR) perspective, in which a packet is successfully received if the SINR exceeds some threshold, considering the interference from all emitting nodes in the network. Our main contribution is to generalize and unify various results scattered throughout the literature. In particular, the proposed framework encompasses arbitrary wireless propagation effects (e.g, Nakagami-m fading, Rician fading, or log-normal shadowing), as well as arbitrary traffic patterns (e.g., slotted-synchronous, slotted-asynchronous, or exponential-interarrivals traffic), allowing us to draw more general conclusions about network performance than previously available in the literature.

💡 Research Summary

The paper develops a comprehensive analytical framework for evaluating the local throughput of a single “probe” link in a wireless network. Two complementary perspectives are considered. The first, a connectivity‑based view, treats a packet as successfully received if no other transmission from nodes that are within the receiver’s audible range collides with it during the same time slot. The second, an SINR‑based view, declares success when the signal‑to‑interference‑plus‑noise ratio at the receiver exceeds a predefined threshold, taking into account interference from every active transmitter in the network.

The authors model the spatial distribution of nodes as a homogeneous Poisson point process (PPP) on the plane with density λ. Each node attempts to transmit in a given slot with probability p, which can be adapted to represent slotted‑synchronous, slotted‑asynchronous, or even continuous‑time exponential‑arrival traffic. For the connectivity model, the set of “audible interferers” is defined as all transmitters whose received power at the probe receiver would be above a detection level. Assuming independent Bernoulli transmission decisions, the probability of a collision‑free transmission is derived in closed form as an exponential function of λ, p and the effective audible area. In the asynchronous case, partial overlap of slots is accounted for by expanding the collision region, leading to a slightly larger exponent.

The SINR analysis incorporates arbitrary propagation effects. Path loss follows the standard distance‑dependent law r^–α, while small‑scale fading can be Nakagami‑m, Rician (with K‑factor), or Rayleigh. Large‑scale shadowing is modeled as a log‑normal random variable with standard deviation σ_sh. By exploiting the Laplace transform of the aggregate interference generated by a PPP, the authors obtain a generic expression for the success probability P