Discussion on Development of Wi-Fi Multi-AP Home Network Technology

Broadband networks are strategic public infrastructure for social-economic development. Globally, they have driven a new wave of informatization development, with countries prioritizing broadband development as a key strategic action area. Fixed broadband is evolving towards F5G such as PON, Wi-Fi, and 400G, accelerating the interconnection of everything with fiber, from fiber-to-the-building (FTTB) to fiber-to-the-home (FTTH), fiber-to-the-room (FTTR), and even to the future fiber-to-the-machine (FTTM). Home broadband services have shifted from early one fiber scenario to an integrated one home solution that combines fiber and Wi-Fi networking.

However, the significant increase in bandwidth has not resulted in a corresponding growth in average revenue per user (ARPU). From 2017 to 2022, China’s broadband speeds surged by 3000%, yet ARPU decreased by 25%. User experience has not improved with the increase in bandwidth speeds. Expectations for broadband services have shifted from merely increasing internet speed to enhancing the quality of experience (QoE). Operators need to improve the Wi-Fi experience to reverse the downward trend in bandwidth pricing. Factors affecting QoE include not only bandwidth but also latency, reliability, coverage, and fault troubleshooting.

To enhance Wi-Fi QoE, a deeper overall management of Wi-Fi is necessary to ensure that ultra-gigabit capabilities cover every room in the home. For this purpose, organizations in the standards and applications domain such as IEEE, ITU, CCSA, and the China Academy of Information and Communications Technology (CAICT) have designed and discussed various solutions based on the FTTR+Wi-Fi Multi-AP home network architecture. The home Wi-Fi network is gradually evolving into a more manageable system.

FTTR+Wi-Fi All-Optical Multi-AP Networking

The all-optical multi-AP home networking is inherently suited for centralized multi-AP control and distributed AP collaborative management, aimed at optimizing Wi-Fi networking.

The “FTTR White Paper 2022” by CAICT states that the all-optical centralized/cloud wireless-optical access network (C-WAN) architecture primarily addresses the ineffective coordination among home networking technologies, disordered competition in the air interface, and challenges in meeting user demands for emerging business experiences. The overall design approach involves the main device collecting information and making decisions, centrally controlling both optical and Wi-Fi transmissions as a unified network, and achieving unified and coordinated configuration of optical and air interface link resources. The main device provides real-time control of subordinate devices to ensure orderly collaboration of air interface Wi-Fi within the FTTR network (managing the air interface Wi-Fi from CSMA to C-WAN), dynamic AI-based link management, and roaming control through seamless handoff technology. This maximizes air interface performance and enhances user QoE.

The C-WAN architecture defined by ITU FTTR (G.fin), as shown in Fig. 1, aims to optimize latency-sensitive services, enhance real-time communication tuning, and enable seamless roaming. It ensures low-latency transmission of channel information by establishing prioritized information channels. In scenarios with stringent air interface collaboration requirements, the round-trip signaling delay should be less than 120 µs.

In the CCSA FTTR DLL draft, the Wi-Fi management and control channel (WMCC) and the Wi-Fi management and control interface (WMCI) are defined for multi-AP Wi-Fi management. WMCC carries messages of the WMCI protocol, providing an interface for Wi-Fi management functions, primarily handling multi-AP air interface scheduling (in time, frequency, and space), roaming control, and energy-saving measures.

The main purpose of C-WAN is to centrally manage, configure, and control Wi-Fi network resources distributed across various devices (air interface time, channel/width, and spatial SR BSS color/transmit power). This control includes QoS mapping, air interface scheduling, STA roaming, and energy saving, all aiming at enhancing user QoE.

IEEE Wi-Fi 8 Multi-AP Solution

The IEEE agenda has included extensive discussions on optimizing multi-AP scenarios for home environments. The exploration of multi-AP collaborative solutions began in the originally planned Release 2 of the Wi-Fi 7 802.11be standard. However, due to the complexity of the solutions, the progress of the standard, and the development and commercialization of chips, the multi-AP functionalities have been postponed to the Wi-Fi 8 802.11bn UHR specification. Potential solutions for multi-AP may include C-OFDMA, C-beamforming, C-spatial reuse (CSR), joint transmission, and seamless roaming (Fig. 2).

• C-OFDMA: When multi-APs synchronize their respective resource status, different APs can simultaneously transmit data to their associated STAs using the same or different resource units (RUs). Simulations indicate that this feature significantly enhances overall system capacity in medium to large networks.
• C-beamforming: When an AP uses beamforming for directional transmission to its associated STA, it also minimizes interference for specific adjacent STAs, avoiding mutual interference between neighboring APs. However, implementing this solution requires the APs to obtain channel state information (CSI) for STAs associated with neighboring APs, which poses a challenge.
• C-spatial reuse: CSR is the simplest coordination mechanism for multiple APs and can be used when interference among BSSs is weak, but the channel state is perceived as busy. To ensure all STAs have sufficient signal-to-noise ratios, APs collaboratively control transmission power to reduce interference. CSR can be achieved with minimal inter-AP collaboration information.
• Joint transmission: By creating a dynamic distributed MU-MIMO system, joint transmission allows multiple APs to serve the same STA. The system operates jointly across multiple APs. Experiments have shown that the benefits of joint transmission and reception are greatest in the downlink.
• Seamless roaming: The aim is to exchange frames between a STA and a group of cooperative APs without adding negotiation overhead. To achieve this, the authentication and association process should be completed with all centralized APs.

The required level of synchronization among the multi-AP methods varies: CSR can operate with coarse frame-level synchronization, CBF and Co-OFDMA require symbol-level synchronization, while JTR needs tight time and phase synchronization. This implies a need for highly reliable, low-latency backhaul links, such as fiber-based FTTR, which presents the greatest implementation challenges.

For the seamless roaming technology in multi-AP architecture, two main architectural schemes are currently under discussion in the Wi-Fi 8 specification process: the non-co-located UHR AP MLD architecture and the UHR AP MLD architecture with Virtual-MLD.

Compared to the 802.11be MLD architecture, the non-co-located UHR AP MLD architecture adds a UHR UMAC entity to provide services related to UHR MAC-level functionality. While this architecture addresses some roaming and technical issues, it also introduces complexity. In certain scenarios, the non-co-located UHR AP MLD architecture may result in additional latency and increased complexity in infrastructure and chipsets.

The UHR AP MLD architecture with virtual-MLD makes minimal changes to the existing EHT AP MLD architecture. However, due to the lack of a unified management entity similar to the UHR AP MLD UMAC, it may require significant inter-AP communication to transmit the context needed for roaming and multi-AP cooperation.

In the era of ultra-gigabit whole-home networks, home networking is centered on enhancing user experience. The deep integration of FTTR all-optical foundation with Wi-Fi coverage has made speed, latency, concurrency, and reliability key areas of development for Wi-Fi technology. New technologies such as FTTR+Wi-Fi, Wi-Fi 8, and AI will further evolve home networks into manageable, controllable, and operable intelligent systems for operators.