Towards a communication-theoretic understanding of system-level power consumption

Traditional communication theory focuses on minimizing transmit power. However, communication links are increasingly operating at shorter ranges where transmit power can be significantly smaller than the power consumed in decoding. This paper models the required decoding power and investigates the minimization of total system power from two complementary perspectives. First, an isolated point-to-point link is considered. Using new lower bounds on the complexity of message-passing decoding, lower bounds are derived on decoding power. These bounds show that 1) there is a fundamental tradeoff between transmit and decoding power; 2) unlike the implications of the traditional “waterfall” curve which focuses on transmit power, the total power must diverge to infinity as error probability goes to zero; 3) Regular LDPCs, and not their known capacity-achieving irregular counterparts, can be shown to be power order optimal in some cases; and 4) the optimizing transmit power is bounded away from the Shannon limit. Second, we consider a collection of links. When systems both generate and face interference, coding allows a system to support a higher density of transmitter-receiver pairs (assuming interference is treated as noise). However, at low densities, uncoded transmission may be more power-efficient in some cases.

💡 Research Summary

The paper revisits the classic communication‑theoretic focus on minimizing transmit power and argues that, for short‑range links, the power consumed by the decoder can dominate the total energy budget. It develops a system‑level power model that explicitly includes decoding power and investigates how to minimize the sum of transmit and decoding power under two complementary scenarios.

In the first scenario, an isolated point‑to‑point link is examined. By deriving new lower bounds on the computational complexity of message‑passing decoders (e.g., belief‑propagation on LDPC graphs), the authors translate these bounds into fundamental lower limits on decoding power. The analysis yields four key insights. (1) There exists an intrinsic trade‑off between transmit power and decoding power: reducing transmit power forces the decoder to perform more iterations, thereby increasing its energy consumption. (2) Unlike the traditional “water‑fall” curve that only considers transmit power, the total system power diverges to infinity as the target error probability approaches zero; achieving arbitrarily low error rates is energetically prohibitive once decoding power is accounted for. (3) Regular LDPC codes, rather than their capacity‑approaching irregular counterparts, can be order‑optimal with respect to total power in certain regimes because they require fewer decoding operations for a given error target. (4) The transmit power that minimizes total power is bounded away from the Shannon limit; operating too close to capacity forces the decoder into a high‑complexity regime that outweighs any transmit‑power savings. These results reshape the design space: the optimal operating point is a balance between channel coding gain and decoder energy cost.

The second scenario extends the analysis to a collection of links that generate and experience mutual interference. Assuming interference is treated as Gaussian noise, the authors show that coding enables a higher spatial density of transmitter‑receiver pairs because each link can tolerate a lower signal‑to‑interference‑plus‑noise ratio (SINR) while still meeting the error‑rate requirement. However, at low link densities—where interference is negligible—the additional decoding power required by coding may make uncoded transmission more energy‑efficient. The paper quantifies this density‑dependent crossover by deriving expressions for the total power per unit area as a function of link density, coding rate, and decoder complexity.

Overall, the work provides a rigorous communication‑theoretic framework that integrates decoding energy into the classic power‑vs‑performance trade‑off. It demonstrates that minimizing transmit power alone is insufficient for modern low‑power, short‑range wireless systems. By exposing the fundamental limits of decoding power, the authors highlight the need for jointly optimizing coding schemes, decoder architectures, and transmit power. The findings have immediate implications for the design of IoT devices, sensor networks, and emerging ultra‑dense wireless deployments, where energy efficiency must be evaluated at the system level rather than at the link level alone.