Improving Reliability of Hybrid Bit-Semantic Communications for Cellular Networks

Semantic communications (SemComs) have been considered as a promising solution to reduce the amount of transmitted information, thus paving the way for more energy-and spectrum-efficient wireless networks. Nevertheless, SemComs rely heavily on the utilization of deep neural networks (DNNs) at the transceivers, which limit the accuracy between the original and reconstructed data and are challenging to implement in practice due to increased architecture complexity. Thus, hybrid cellular networks that utilize both conventional bit communications (BitComs) and SemComs have been introduced to bridge the gap between required and existing infrastructure. To facilitate such networks, in this work, we investigate reliability by deriving closed-form expressions for the outage probability of the network. Additionally, we propose a generalized outage probability through which the cell radius that achieves a desired outage threshold for a specific range of users is calculated in closed form. Additionally, to consider the practical limitations caused by the specialized dedicated hardware and the increased memory and computational resources that are required to support SemCom, a semantic utilization metric is proposed. Based on this metric, we express the probability that a specific number of users select SemCom transmission and calculate the optimal cell radius for that number in closed form. Simulation results validate the derived analytical expressions and the characterized design properties of the cell radius found through the proposed metrics, providing useful insights.

💡 Research Summary

This paper addresses the reliability of hybrid cellular networks that combine conventional bit‑based communications (BitCom) with emerging semantic communications (SemCom). While SemCom can dramatically reduce the amount of transmitted data by extracting and transmitting only the meaning of the source content, its reliance on deep neural networks (DNNs) introduces practical challenges: high computational complexity, dedicated hardware, and increased memory requirements. Consequently, a hybrid architecture that leverages the existing BitCom infrastructure while opportunistically employing SemCom where it is most beneficial is proposed.

The authors consider a single‑antenna base station (BS) serving L single‑antenna users in a circular cell. The large‑scale path loss follows a free‑space model pₗ(rₗ)= (λ/4π)²·rₗ^{‑a}, and small‑scale fading is Rayleigh (hₗ∼CN(0,1)). FDMA splits the total bandwidth W equally among the users, and each user receives an SNR gₗ = c_L·|hₗ|²·rₗ^{‑a}, where c_L = (L·P)/(N₀·W)·(λ/4π)².

For SemCom, the DeepSC framework is adopted. Each sentence S_j is first encoded by a semantic encoder (based on BERT) into a vector S′_j, then passed through a channel encoder and transmitted as continuous‑amplitude analog symbols (DTAT). At the receiver, a channel decoder and a semantic decoder reconstruct the original sentence. Reconstruction quality is measured by cosine similarity M_j between the BERT embeddings of the original and reconstructed sentences. Because the exact relationship between M_j and SNR is only available through extensive simulations, the authors employ a highly accurate logistic‑function approximation:

M_{k,l}(gₗ) ≈ A_{k,1}+A_{k,2}−A_{k,1}/