Outage Probability for Multi-Cell Processing under Rayleigh Fading

Multi-cell processing, also called Coordinated Multiple Point (CoMP), is a very promising distributed multi-antennas technique that uses neighbour cell’s antennas. This is expected to be part of next generation cellular networks standards such as LTE-A. Small cell networks in dense urban environment are mainly limited by interferences and CoMP can strongly take advantage of this fact to improve cell-edge users’ throughput. This paper provides an analytical derivation of the capacity outage probability for CoMP experiencing fast Rayleigh fading. Only the average received power (slow varying fading) has to be known, and perfect Channel State Information (CSI) is not required. An optimisation of the successfully received data-rate is then derived with respect to the number of cooperating stations and the outage probability, illustrated by numerical examples.

💡 Research Summary

The paper addresses a fundamental performance metric—capacity outage probability—in coordinated multi‑point (CoMP) or multi‑cell processing systems operating under fast Rayleigh fading. While prior works often assume perfect channel state information (CSI) and rely on extensive Monte‑Carlo simulations, this study derives a closed‑form expression for the outage probability that requires only the average received power (large‑scale fading) from each cooperating base station. This approach eliminates the need for instantaneous CSI, dramatically reducing signaling overhead and making the analysis suitable for real‑time network management.

The system model considers K cooperating base stations serving a single user. Each link experiences independent Rayleigh fading, and the received signal is a coherent sum of the K transmitted streams. The signal‑to‑interference‑plus‑noise ratio (SINR) is expressed as a ratio of a weighted sum of exponential random variables (representing the squared Rayleigh amplitudes) to the noise plus inter‑cell interference. By applying Laplace transforms and partial‑fraction expansion, the authors obtain the exact probability density function of the SINR sum, which leads directly to the outage probability P_out(R)=Pr{C<R}, where C is the instantaneous Shannon capacity and R is the target data rate. The final expression depends only on the target rate R and the average received powers β_i (i=1…K), which encapsulate path loss, shadowing, and transmit power.

Having an analytical outage formula enables the formulation of an optimization problem that simultaneously selects the optimal number of cooperating stations K and the optimal target rate R to maximize the expected successful throughput η = (1−P_out)·R. The problem is mixed‑integer: K is integer‑valued, while R is continuous. The authors employ a Lagrangian approach for the continuous part and a greedy search for the integer part, while incorporating practical constraints such as total transmit power budget and backhaul capacity limits. The resulting solution reveals a clear trade‑off: increasing K reduces outage dramatically for cell‑edge users, but each additional cooperating station consumes backhaul bandwidth and adds processing complexity. In backhaul‑constrained scenarios, the optimal K typically lies between 2 and 3, whereas in an ideal backhaul the benefit continues up to K=4 in the considered dense‑urban layout.

Numerical examples illustrate the theory. Using a dense urban small‑cell layout (cell radius 200 m, transmit power 46 dBm, noise floor –174 dBm/Hz, bandwidth 10 MHz), the authors compare K=1 (no cooperation) to K=4. The outage probability drops from roughly 12 % to below 2 % as K increases, and the optimal data rate rises from about 3.2 Mbps to 4.5 Mbps. When the backhaul capacity is limited to 100 Mbps, the marginal gain of adding a fourth station becomes negligible, confirming the practical relevance of the trade‑off analysis.

In summary, the paper makes three key contributions: (1) an exact, CSI‑free outage probability expression for Rayleigh‑faded CoMP systems; (2) a joint optimization framework for selecting the number of cooperating base stations and the target transmission rate under realistic constraints; and (3) a set of quantitative insights that guide the design of dense small‑cell networks, showing how much cooperation is beneficial before backhaul and processing costs outweigh the outage reduction. The analytical tractability and practical relevance of the results position this work as a valuable reference for both researchers and engineers developing next‑generation cellular standards such as LTE‑Advanced and 5G NR.