A Unified QoS-Aware Multiplexing Framework for Next Generation Immersive Communication with Legacy Wireless Applications

TL;DR

This work tackles the challenge of co-deploying immersive communication with legacy wireless applications by proposing a unified QoS-aware framework that operates over dual-timescale network slicing and resource allocation. It leverages Lyapunov drift to convert long-term throughput optimization into a tractable, backlog-aware short-term problem and introduces adaptive adversarial slicing (Ad2S) along with a Kalman-filter-based non-stationary refinement (Ad2S-NR) to handle time-varying channels. The methods include a frame-scale PBRA for fine-grained resource allocation and a super-frame slicing strategy guided by contextual bandits, with ME-KF tracking providing robust non-stationary adaptation. Key results show throughput gains up to 3.86 Mbps, latency reductions around 63.96%, and sub-linear regret under both stationary and non-stationary conditions, demonstrating practical impact for future immersive-mixed networks.

Abstract

Immersive communication, including emerging augmented reality, virtual reality, and holographic telepresence, has been identified as a key service for enabling next-generation wireless applications. To align with legacy wireless applications, such as enhanced mobile broadband or ultra-reliable low-latency communication, network slicing has been widely adopted. However, attempting to statistically isolate the above types of wireless applications through different network slices may lead to throughput degradation and increased queue backlog. To address these challenges, we establish a unified QoS-aware framework that supports immersive communication and legacy wireless applications simultaneously. Based on the Lyapunov drift theorem, we transform the original long-term throughput maximization problem into an equivalent short-term throughput maximization weighted by virtual queue length. Moreover, to cope with the challenges introduced by the interaction between large-timescale network slicing and short-timescale resource allocation, we propose an adaptive adversarial slicing (Ad2S) scheme for networks with invarying channel statistics. To track the network channel variations, we also propose a measurement extrapolation-Kalman filter (ME-KF)-based method and refine our scheme into Ad2S-non-stationary refinement (Ad2S-NR). Through extended numerical examples, we demonstrate that our proposed schemes achieve 3.86 Mbps throughput improvement and 63.96% latency reduction with 24.36% convergence time reduction. Within our framework, the trade-off between total throughput and user service experience can be achieved by tuning systematic parameters.

Figures (17)

• Figure 1: (a) System model; (b) Dual-scale resource grid for network slicing and resource allocation.
• Figure 2: Slice configuration $\mathcal{F}_l$ and other system dynamics (e.g., $Q_e(k)$, $Q_M(k)$ denoting average backlog for eMBB and MBBLL users, respectively, and their time-average $\mathbb{E}[Q_e(k)]$, $\mathbb{E}[Q_M(k)]$). After entering exploitation stage, as $\textcircled{1}$ the $\mathbb{E}[Q_e(k)]$ approaching $\delta_e$, $\textcircled{2}$ our proposed Ad2S algorithm adaptively slices in favor of honoring legacy QoS, $\textcircled{3}$$Q_e(k)$ decreases in response.
• Figure 3: Estimation error, e.g. $|\hat{\mu_n}(l) - \mu_n(l)|$ of a priori based vinogradova_estimating_2022 and proposed ME-KF scheme for each user.
• Figure 4: Empirical cumulative regret.
• Figure 5: (a) Impact of different chunk sizes on exploration time and average reward for Ad2S and EXP3 under identical parameters (e.g., $\eta = 1, \gamma = 0.5$). (b) Slice configuration behavior of Ad2S under chunk size of 1 sub-channel. (c) Slice configuration behavior of EXP3 under chunk size of 1 sub-channel.

Theorems & Definitions (9)

• Lemma 1: Lyapunov Drift neely_mj_stochastic_2010
• proof
• Lemma 2: Convergence of Alg. \ref{['alg_psum_bcd']}
• proof
• Lemma 3: Convergence of ME-KF
• Theorem 1: Asymptotical Optimality for Ad2S and Ad2S-NR
• proof
• Lemma 4: Computational complexity for Ad2S and Ad2S-NR
• proof