Abstract:More refined resource management and Quality of Service (QoS) provisioning is a critical goal of wireless communication technologies. In this paper, we propose a novel Business-Centric Network (BCN) aimed at enabling scalable QoS provisioning, based on a cross-layer framework that captures the relationship between application, transport parameters, and channels. We investigate both continuous flow and event-driven flow models, presenting key QoS metrics such as throughput, delay, and reliability. By jointly considering power and bandwidth allocation, transmission parameters, and AP network topology across layers, we optimize weighted resource efficiency with statistical QoS provisioning. To address the coupling among parameters, we propose a novel deep reinforcement learning (DRL) framework, which is Collaborative Optimization among Heterogeneous Actors with Experience Sharing (COHA-ES). Power and sub-channel (SC) Actors representing multiple APs are jointly optimized under the unified guidance of a common critic. Additionally, we introduce a novel multithreaded experience-sharing mechanism to accelerate training and enhance rewards. Extensive comparative experiments validate the effectiveness of our DRL framework in terms of convergence and efficiency. Moreover, comparative analyses demonstrate the comprehensive advantages of the BCN structure in enhancing both spectral and energy efficiency.
What problem does this paper attempt to address?
### Problems the Paper Attempts to Solve
This paper aims to address the critical issues of resource management and Quality of Service (QoS) assurance in wireless communication networks. Specifically, the paper proposes a Business-Centric Network (BCN) architecture that achieves scalable QoS assurance through a cross-layer framework. The main objectives of the paper include:
1. **Providing High-Quality Service (QoS) on Demand**:
- Different services have varying QoS requirements for latency, reliability, and transmission rate. Ensuring that these diverse QoS needs can be met on demand is a key challenge, especially as network architecture evolves towards user-centric design.
2. **Resource Efficiency**:
- Efficient utilization of spectrum and power resources is another major focus, particularly in the era of green communications. As networks become increasingly complex, balancing resource efficiency while ensuring QoS becomes more challenging.
### Main Contributions
1. **Proposing a Business-Centric Network Architecture**:
- From the perspective of cross-layer optimization, a scalable QoS provisioning framework is proposed. This framework achieves optimal resource efficiency and provides statistical QoS assurance by jointly considering power and bandwidth resource allocation at the physical layer, transmission parameters at the link layer, and AP network modes at the network layer. The framework also considers factors such as queuing delay, delay violations, transmission errors, and retransmissions to characterize QoS metrics (e.g., throughput, delay, and reliability).
2. **Building a Collaborative Deep Reinforcement Learning Framework**:
- To manage the complex dependencies between resource allocation and QoS assurance, a collaborative Deep Reinforcement Learning (DRL) framework is constructed. In this framework, heterogeneous power agents and sub-channel (SC) agents represent multiple access points (APs) and perform collaborative optimization under the guidance of a common critic. Additionally, a multi-threaded experience sharing mechanism is proposed to accelerate offline training and system rewards.
3. **Extensive Experimental Validation**:
- The proposed DRL framework's effectiveness in terms of convergence and efficiency is validated through numerous comparative experiments. Further, extensive ablation experiments demonstrate the comprehensive advantages of the BCN architecture in terms of spectrum efficiency and energy efficiency.
### Solutions
- **Cross-Layer Optimization Scheme**:
- In the downlink scenario, packet transmission delay consists of queuing delay and service delay, the latter including transmission delay and retransmission delay. By optimizing resource allocation and transmission parameters, the maximum delay is ensured to meet QoS requirements.
- **Effective Energy Efficiency (EEE) and Effective Spectrum Efficiency (ESE)**:
- EEE and ESE are defined as the ratios of effective capacity to power and bandwidth, respectively, to more comprehensively describe the relationship between service capability and resource consumption. By studying the trade-offs between EEE and ESE, the performance space of the system is explored.
- **Transmission Parameter Decision**:
- Given resource arrangements, the study focuses on optimal QoS provisioning rather than the ultimate resource efficiency goal. For periodic and non-periodic services, transmission parameter decision (TPD) sub-problems are proposed and the MINLP problem is solved through variable decomposition methods.
### Conclusion
By proposing the BCN architecture and DRL framework, this paper successfully addresses the complex issues of resource management and QoS assurance in wireless communication networks, providing new solutions for future 6G networks.