Traditional wireless communication design is built on the foundation of Shannon’s capacity theorem, which assumes that data packets are infinitely long. In this asymptotic regime, channel noise and fading average out perfectly, allowing for error-free transmission at any rate below the theoretical limit. However, as Changyang She et al. argue in Cross-layer Transmission Design for Tactile Internet, the rise of the Tactile Internet and Ultra-Reliable Low-Latency Communications (URLLC) shatters this assumption. To achieve sub-millisecond latency, packets must be extremely short, leaving no time for noise to "average out."
The Coding Rate Penalty
When blocklengths are finite, the achievable transmission rate is no longer a fixed "capacity" but a function of the desired reliability. As Mohammad Robat Mili et al. explain in Joint Sum Rate and Error Probability Optimization: Finite Blocklength Analysis, the achievable rate in the short-packet regime is roughly the Shannon capacity minus a penalty term. This penalty is proportional to the square root of the "channel dispersion"—a measure of how much the channel varies—and inversely proportional to the square root of the blocklength ($1/\sqrt{n}$).
This mathematical relationship creates a fundamental trade-off: to maintain ultra-high reliability with a short packet, one must significantly reduce the transmission rate or increase the transmit power. Relying on traditional Shannon models in these scenarios leads to a dangerous underestimation of packet loss and energy requirements, as noted by Shengfeng Xu et al. in Energy-Efficient Packet Scheduling With Finite Blocklength Codes: Convexity Analysis and Efficient Algorithms.
Taming the Reliability Tail
In safety-critical applications like autonomous driving, "average" performance is irrelevant; what matters is the "tail" of the distribution—the rare, extreme delay events that could lead to a collision. Mohamed K. Abdel-Aziz et al. demonstrate in Ultra-Reliable Low-Latency Vehicular Networks: Taming the Age of Information Tail that finite blocklength analysis is essential for managing the Age of Information (AoI).
Because short packets are more sensitive to interference and noise, the authors utilize Extreme Value Theory (EVT) to model these rare failures. By accounting for the finite nature of the transmission, they can optimize the network to ensure that even the worst-case delays stay within safe bounds.
Scaling to Massive MIMO and Beyond
The transition to finite blocklength analysis also changes how we view advanced physical layer technologies. For instance, while Massive MIMO is known for its high throughput, Alejandro Lancho et al. show in A Finite-Blocklength Analysis for URLLC with Massive MIMO that simply adding antennas is not enough for URLLC. In the short-packet regime, pilot contamination and channel estimation errors become dominant bottlenecks.
The authors utilize the Random-Coding Union (RCU) bound and saddlepoint approximations to provide more precise error estimates than the standard normal approximation. These tools allow researchers to calculate the exact point where the overhead of learning the channel (pilots) outweighs the diversity gain of the antennas, a critical balance for low-latency systems described by Johan Östman et al. in Finite-Blocklength Bounds on the Maximum Coding Rate of Rician Fading Channels with Applications to Pilot-Assisted Transmission.
Energy and Resource Efficiency
Finally, finite blocklength constraints redefine energy efficiency. In the infinite-length world, efficiency is theoretically maximized as the transmission rate approaches zero. However, Vineeth S. Varma et al. prove in An Energy-Efficient Framework for the Analysis of MIMO Slow Fading Channels that when blocklengths are finite, the most energy-efficient communication occurs at a specific, non-zero spectral efficiency. This is because the fixed costs of keeping the radio hardware active and the increased risk of outage for short packets create a "sweet spot" for power consumption that traditional theory misses.
By integrating these finite-length bounds into practical scheduling—such as the contract-based mechanisms proposed by Aunas Manzoor et al. in Contract-Based Scheduling of URLLC Packets in Incumbent EMBB Traffic—networks can finally support the simultaneous demands of high-speed broadband and mission-critical control.