140 of FIG. 1C) for MAPC message transmission. In the example from FIG. 1C, this may include determining the transit times for messages sent over the air from RAP 135 to every MAP 130 in the CG.

In some embodiments, the AP/WLC 210 may then evaluate the current MAPC model along with various factors to select the most suitable transport mode (or the transport mode that offers an improved performance for MAPC). These factors may include transit time, packet loss rate, and other metrics reflecting the conditions of both wired and wireless paths. For example, when the CG is operating under the C-TDMA mode, wired mode may be selected to ensure precise timing and synchronized communication. However, wireless mode may be chosen under the C-TDMA mode in embodiments where wired connections cannot reach every AP, such as when cables only exist between the RAP (e.g., 135 of FIG. 1C) and first-tier MAPs (e.g., MAP 1 (130-1), MAP 2 (130-2), or MAP 3 (130-3) of FIG. 1C). Even when the desired path reaches every AP, such as when cables exist both between the RAP (e.g., 135 of FIG. 1C) and first-tier MAPs (e.g., MAP 1 (130-1), MAP 2 (130-2), or MAP 3 (130-3) of FIG. 1C), and between first-tier MAPs and second-tier MAPs (e.g., MAP 4 (130-4), MAP 5 (130-5), or MAP 6 (130-6) of FIG. 1C), wireless mode may still be selected if the calculation results reveal a significant hop-delay at a relay point (e.g., MAP 1 (130-1) of FIG. 1C). This is because the significant hop-delay may cause the transit time between RAP (e.g., 135 of FIG. 1C) and a second-tier MAP (requiring a relay at a first-tier MAP) (e.g., MAP 4 (130-4) of FIG. 1C) much longer than that between RAP and a first-tier MAP (e.g., MAP 2 (130-2), rendering wired transmission no longer suitable for the C-TDMA mode (which requires high reliability and synchronization across all APs in the same CG).

Alternatively, if C-SR mode is in use, which prioritizes speed and flexibility over reliability, wireless mode may be preferred for its ability to quickly adapt to changing network conditions (even when the estimates show a slight increase in latency compared to wired connections). However, the system may switch to wired mode if it detects that the wireless network is experiencing congestion, high interference levels, or reduced signal strength, making wireless transmission no longer reliable.

Upon determining the transport mode (wired or wireless), as depicted, AP/WLC 210 communicates the determined mode back to AP 205 (step 225). The communication instructs AP 205 to follow the determined transport mechanism for sending and receiving MAPC coordination messages. In some embodiments, the message conveying the selected transport mode may be transmitted using specific management frames, such as a beacon or probe response.

As depicted, AP 205 sends a transmission request to AP/WLC 210 using the determined transport mode (step 230). In some embodiments, the request may include information such as AP's 205 need to transmit data using a shared channel, along with details about the type of data transmission (e.g., uplink or downlink data), the desired timing, priority (e.g., best-effort, voice, and video), and any other relevant parameters. Upon receiving the transmission request, AP/WLC 210 evaluates the request based on the current network conditions and MACP mode. AP/WLC 210 then sends back a response to AP 205 using the determined transport mode (step 235). In some embodiments, the response may include the instructions for proceeding with the data transmission. More specifically, the response may detail the timing, channel access parameters, and any specific protocols to follow. In embodiments where the instruction from AP/WLC 210 is for a downlink transmission, AP 205 may directly follow the instruction and send the downlink data to its associated STA 215 (step 240). The data may be transmitted according to the parameters set by AP/WLC 210 and align with the synchronized communication strategy within the CG. In embodiments where the instruction from AP/WLC 210 is for an uplink transmission, AP 205 may forward the instruction to the STA 215 (step 240). The STA 215 may then follow the instruction and send the uplink data to AP 205 (step 245).

FIG. 3 depicts a wired path configuration 300 where a CG has a leader AP CG, according to some embodiments of the present disclosure. In some embodiments, the leader AP 405-1 may correspond to AP 105-1 as depicted in FIG. 1A, WLC 120 as depicted in FIG. 1B, or RAP 135 as depicted in FIG. 1C.

Six APs are depicted in FIG. 3, including AP 305-1, AP 305-2, AP 305-3, AP 305-4, AP 305-5, and AP 305-6. AP 305-1 directly connects to AP 305-2, AP 305-3, and AP 305-4 through wired cables (e.g., Ethernet cables). Additionally, AP 305-2 is wired to AP 305-5, and AP 305-4 is wired to AP 305-6. Since the AP 305-1 is acting as the leader AP, a total of (1×N) wired paths may be estimated for exchanging MAPC messages (e.g., discovery, negotiation, or coordination messages), where “1” represents the leader AP, and “N” represents the number of participating APs. As depicted, when AP 305-1, AP 305-3, AP 305-5, and AP 305-6 form (or intend to form) a CG, three wired paths may be considered: path 1 (310) connecting the leader AP 305-1 to AP 305-5 via AP 305-2, path 2 (315) directly connecting the leader AP 305-1 to AP 305-3, and path 3 (320) connecting the AP 305-1 to AP 305-6 via AP 305-4. Each AP may broadcast its capability of being a transport node and its role to the leader AP 305-1. For example, AP 305-5 may indicate that it is an endpoint node in the network topology, while AP 305-2 may dictate its role as a relay node (forwarding messages between the leader AP 305-1 and AP 305-5).

In some embodiments, the transit time for each wired path within the CG may be calculated by considering both termination-delays and hop-delays along the paths. For path 1 (310), the transit time may include calculating the termination-delay at AP 305-5 (or AP 305-1) and the hop-delay at AP 305-2 (where the signal is relayed). For path 2 (315), the transit time may only include the termination delay at AP 305-3 (or AP 305-1), as there are no intermediate relay points (e.g., no hop-delays). For path 3, the transit time may include calculating the termination-delay at AP 305-6 (or AP 305-1) and the hop-delay at AP 305-4.

The transit times for wired paths may then be used to determine the suitable transport mode within the CG, considering the MAPC mode in use. For example, when C-TDMA mode is used in the CG that consists of AP 305-1, AP 305-3, AP 305-5 and AP 305-6, wired transport mode may be selected since the wired paths (including paths 1, 2 and 3) can reach every participating AP and wired communication has high reliability and low latency. However, the selection may shift to wireless mode if the calculation reveals that the hop-delay is significant, making the transit time for paths 1 and 3 much longer than that for path 2. This discrepancy represents that messages may reach AP 305-3 much earlier than AP 305-5 or AP 305-6. In this configuration, wired mode may not be selected since C-TDMA mode requires synchronized communication across the CG.

In embodiments where C-SR mode is used, wireless transport mode may be preferred for their flexibility and speed. However, if the calculation reveals that transit times for wired paths (including paths 1, 2 and 3) are much shorter than for wireless communications, potentially due to wireless congestion, low signal strength, or high interference levels, the selection may shift to wired mode to ensure the communication remains efficient and reliable within the CG.

FIGS. 4A-4C show a wired path configuration 400 for a decentralized CG, according to some embodiments of the present disclosure. The network setup in FIGS. 4A-4C is similar to that in FIG. 3, with AP 405-1, AP 405-2, AP 405-3, AP 405-4, AP 405-5, and AP 405-6 connected through wired connections. The only difference is that AP 405-1 is no longer functioning as the leader AP. Instead, the network operates under a decentralized MAPC. In the decentralized setup, each AP sends MAPC messages to other APs in the same CG. The total number of paths in use may be represented by N×N, where N represents the total number of APs in the CG minus 1. As depicted, when the CG includes AP 405-1, AP 405-3, AP 405-5, and AP 405-6, the total wired paths used for MAPC may include nine paths in total.

In the decentralized setup, each AP participates in the path estimation and transit time calculation process. As depicted, each AP may broadcast its capability of being a transport node and its role or locations in the network. For example, AP 405-5 may indicate that it is an endpoint node in the network, while AP 405-2 may dictate its role as a relay node (forwarding messages between the leader AP 405-1 and AP 405-5). When the CG is formed or potentially formed (not yet fully established) between AP 405-1, AP 405-3, AP 405-5, and AP 405-6, AP 405-1 may estimate the following three paths as depicted in FIG. 4A: path 1 (410) connecting the AP 405-1 to AP 405-5 via AP 405-2, path 2 (415) directly connecting the AP 405-1 to AP 405-3, and path 3 (420) connecting the AP 405-1 to AP 405-6 via AP 405-4. The transit time for path 1 (410) may include a termination delay at AP 405-5 (or AP 405-1) and a hop-delay at AP 405-2. The transit time for path 2 (415) may include a termination delay at AP 405-3 (or AP 405-1), with no hop-delays. The transit time for path 3 (420) may include a termination delay at AP 405-6 (or AP 405-1) and a hop-delay at AP 405-4.

Within the CG, AP 405-5 may estimate the following three paths as depicted in FIG. 4B: path 4 (425) connecting the AP 405-5 to AP 405-1 via AP 405-2, path 5 (430) connecting the AP 405-5 to AP 405-3 via AP 405-2 and AP 405-1, and path 6 (435) connecting the AP 405-5 to AP 405-6 via AP 405-2, AP 405-1, and AP 405-4. The transit time for path 4 (425) may be the same as that for path 1 (410), including a termination delay at AP 405-5 (or AP 405-1) and a hop-delay at AP 405-2. The transit time for path 5 (430) may include a termination delay at AP 405-5 (or AP 405-3), with hop-delays at AP 405-2 and AP 405-1. The transit time for path 6 (435) may include a termination delay at AP 405-5 (or AP 405-6), with hop-delays at AP 405-2, AP 405-2, and AP 405-4.

Within the CG, AP 405-6 may estimate the following three paths as depicted in FIG. 4C: path 7 (440) connecting the AP 405-6 to AP 405-1 via AP 405-4, path 8 (445) connecting the AP 405-6 to AP 405-3 via AP 405-4 and AP 405-1, and path 9 (450) connecting the AP 405-6 to AP 405-5 via AP 405-4, AP 405-1, and AP 405-2. The transit time for path 7 (440) may include a termination delay at AP 405-6 (or AP 405-1) and a hop-delay at AP 405-4. The transit time for path 8 (445) may include a termination delay at AP 405-6 (or AP 405-3), with hop-delays at AP 405-4 and AP 405-1. The transit time for path 9 (450) may be the same as that for path 6 (435), including a termination delay at AP 405-6 (or AP 405-5), with hop-delays at AP 405-4, AP 405-1, and AP 405-2.

In some embodiments, the total number of paths may be represented by N×N×P, where “P” represents additional factors such as redundancy or failover paths that need to be considered in more complex network topologies. As a simplification, the CG operation may be limited to an L2 broadcast domain (whether it is a wired LAN or Wi-Fi), in which embodiment the N×N×P may be simplified to N×1×2. This simplification assumes that a broadcast domain (BCAST) is representative of any link, which may not be strictly accurate for a virtual LAN (VLAN), where broadcast domains may be segmented and not all links are universally accessible.

In the decentralized MAPC setup, each AP in the CG, such as AP 405-1, AP 405-3, AP 405-5, and AP 405-6, may actively participate in path estimation and calculate the corresponding transit time for each path. After these calculations, in some embodiments, each AP may report the transit times to an AP assigned for transport mode selection. The assigned AP may aggregate the calculated times from all APs in the CG, evaluate the relevant transit times for wireless paths, and check the MAPC mode in use. Based on these factors, the assigned AP may then determine the most suitable transport mode (or the transport mode that offers an improved performance) for the CG for multi-AP coordination.

In the network depicted in FIGS. 4A-4C, where wired connections reach each AP in the CG (including AP 405-1, AP 405-3, AP 405-5, and AP 405-6), wired connections may be preferred when the CG operates under C-TDMA mode. However, as discussed above, the system may switch to wireless mode if the hop-delays are significant, making the transit time for certain paths, such as path 2 (415), much shorter than that for other paths, such as paths 1, 3, and 4-9. When C-SR mode is in use, wireless mode may be selected for their speed and flexibility, but if the calculation reveals that wired paths offer significantly faster transmission than wireless paths (e.g., potentially due to factors like congestion or interference in the wireless network), the system may switch to wired mode for maintaining efficient coordination within the CG.

In some embodiments, a software-defined-networking (SDN)-based controller (e.g., a Catalyst Center) may be established. The SDN-based controller may operate at the IP layers, which allows it to centralize transport mode selection across the entire network, including multiple CGs. In some embodiments, the SDN-based controller may collect transport capability data from all APs across the network. For example, in the depicted network setup, assuming that CG 1 is formed to include AP 405-1, AP 405-3, AP 405-5, and AP 405-6, and CG 2 is formed to include AP 405-2 and AP 405-4, the controller may perform calculations for all paths within each CG (whether 1×N or N×N) and determine the most suitable transport mode (or the transport mode that offers an improved performance) for each CG. In some embodiments, each AP may conduct its own path estimation and transit time calculation, then report these results to the SDN-based controller. The controller may aggregate the data and determine the transport mode for each CG.

The network setup as depicted in FIGS. 3 and 4A-4C, which includes six APs wired to each other with a CG formed between AP 405-1, AP 405-3, AP 405-5, and AP 405-6, is provided for conceptual clarity. In some embodiments, any number of APs may be presented within the network, and the APs may be connected either through wired connections, wireless connections, or a combination of both.

FIG. 5 depicts an example method 500 for a leader AP deciding transport mode for a CG, according to some embodiments of the present disclosure. In some embodiments, the leader AP may correspond to AP 105-1 of FIG. 1C, RAP 135 of FIG. 1C, AP/WLC 210 of FIG. 2, or AP 305-1 of FIG. 3. In some embodiments, the example method 500 may be performed by a WLC, such as WLC 120 as depicted in FIG. 1B.

At block 505, a leader AP (or WLC) initializes its settings within a CG after being selected as the leader. In some embodiments, the initialization may include setting up the required configurations and defining its responsibility for coordinating communication between the APs in the CG.

At block 510, the leader AP (or WLC) receives capability information from the other APs in the CG. In some embodiments, the capability information may include details such as each AP's capability to act as a transport port in wired connections and its role or location within the network (e.g., relay or endpoint). In some embodiments, the capability information may be sent through beacon or probe response frames. By processing the received information, the leader AP (or WLC) may develop a comprehensive understanding of the network's topology, each AP's capability, and its potential roles in the MAPC communication. This information may later be used to estimate wired paths, calculate transit times (or delays), and select the appropriate transport mode for the CG.

At block 515, the leader AP (or WLC) verifies whether every AP in the CG has provided the required data, such as its transport capability, role, and other performance metrics. If capability information has been received from all other APs in the CG, the method 500 proceeds to block 520. If some APs have not yet sent their capability information, the method 500 returns to block 510, where the leader AP (or WLC) may wait for the remaining data to arrive. In some embodiments, the leader AP (or WLC) may send additional requests to the APs to collect their capability data.

At block 520, the leader AP (or WLC) discovers wired paths and calculates or estimates the transit time (or delay) for each path. In some embodiments, the leader AP (or WLC) may use protocols such as Internet Control Message Protocol (e.g., traceroute) or Address Resolution Protocol to identify the available wired routes. Since the CG has a leader AP (or WLC) configured to coordinate data transmission, only the paths from the leader AP (or WLC) to each of the participating APs (also be referred to in some embodiments as peer APs) need to be estimated, resulting in a total of 1×N wired paths (where “1” represents the leader AP, and “N” represents the number of participating APs in the CG). Once the wired paths are identified, the leader AP (or WLC) calculates the transit time (or delay) for each path, considering factors like hop-delays and termination delays. As used herein, the transit time (or delay) for a wired path may include the time for a signal to travel over cables between APs and the time that the signal is processed at the endpoint for multi-AP coordination. These transit times may then be used to assess the suitability of the wired paths for MAPC communication and coordination within the CG.

At block 525, the leader AP (or WLC) evaluates wireless paths between itself and other APs in the CG. In some embodiments, the speed and reliability of wireless transmission may be affected by metrics like signal strength, traffic load, or the level of interference from other devices. By considering these factors, the leader AP (or WLC) calculates the overall transit time (or delay) for each wireless path. As used herein, the transit time (or delay) for a wireless path may refer to the time for a signal to travel over the air between APs and the time that the signal is processed at the endpoint for multi-AP coordination.

At block 530, the leader AP (or WLC) assesses the MAPC mode (e.g., C-TDMA, C-SR, or other modes) being used by the CG and determines the respective requirements for speed, synchronization, and/or reliability.

At block 535, the leader AP (or WLC) selects the suitable transport mode (wired or wireless) for multi-AP coordination. In some embodiments, the leader AP (or WLC) may compare the transit times for both wired and wireless paths. If the wired paths offer lower transit times, lower packet loss rates, and higher reliability, wired transport mode may be selected. Additionally, the specific MAPC mode in use may also influence the transport mode selection. For example, when C-TDMA mode is in use, which requires low latency, low packet loss, and high reliability, wired paths may be preferred to maintain precise timing and synchronization. If CG is operating under C-SR mode, which prioritizes speed and flexibility over reliability, wireless paths may be selected, even if the transit times are slightly higher than that for wired paths. Furthermore, the leader AP (or WLC) may adjust the transport mode based on real-time changes in network conditions. For example, if wireless transport mode is initially selected for a CG operating under C-SR mode, but the leader AP (or WLC) detects a significant increase in network congestion and interference (e.g., increased packet loss rate) or reduced signal strength, the leader AP (or WLC) may switch to wired connections to maintain stable and efficient MAPC communication.

At block 540, the leader AP (or WLC) communicates the selected transport mode to all participating APs in the CG. In some embodiments, the selected transport mode may be sent through beacon or probe response frames.

At block 545, the leader AP (or WLC) performs centralized multi-AP coordination using the selected transport mode. If wired mode is selected, the leader AP (or WLC) may send and receive MAPC messages through the wired cables. The messages may include data transmission requests (e.g., 230 of FIG. 2) received from participating APs and instructions (e.g., 235 of FIG. 2) to the APs for timing synchronization and data transmission. If wireless mode is selected, the leader AP (or WLC) may use wireless connections (e.g., Wi-Fi links) to coordinate MAPC communication. The leader AP (or WLC) may broadcast and receive MAPC messages over the air.

FIG. 6 depicts an example method 600 for a participating AP estimating path and transit time for transport mode selection, according to some embodiments of the present disclosure. In some embodiments, the participating AP may correspond to APs 105-2 or 105-3 of FIG. 1A, APs 105 of FIG. 1B, MAP 130 of FIG. 1C, AP 205 of FIG. 2, APs 305-2, 305-3, 305-4, 305-5, or 305-6 of FIG. 3, or AP 405 of FIG. 4.

At block 605, a participating AP (also referred to in some embodiments as a peer AP) initialize its settings and prepares to communicate with other APs in the CG.

As used herein, the participating AP in a centralized MAPC configuration refers to an AP that relies on a leader AP or WLC (e.g., 105-1 of FIG. 1A or 120 of FIG. 1B) to coordinate data transmission using a shared channel. The participating AP may send requests to the leader AP (or WLC), indicating its need for coordination, and the leader AP (or WLC) may respond with instructions for timing synchronization and data transmission. As used herein, the participating AP in a decentralized MAPC setup refers to any AP in the CG. The participating AP communicates directly with other APs in the CG for coordination and data transmission, without depending on a centralized leader AP (or WLC).

At block 610, the participating AP broadcasts its transport capability information to other APs in the CG, including its ability to act as a transport node in wired connections, and its role in the network (e.g., relay or endpoint).

At block 615, the participating AP listens for and receives transport capability information from other APs in the same CG.

At block 620, the participating AP checks if it has received capability information from every other AP in the CG. If any data is missing, the method 600 returns to block 615, where the participating AP may either wait for the remaining data or send additional requests to ensure complete information is gathered.

At block 625, the participating AP estimates the wired paths between itself and every other AP in the CG. In some embodiments, the participating AP may use protocols such as ICMP (e.g., traceroute) or ARP to identify the available wired routes. In a decentralized configuration, each participating AP may directly communicate with every other AP in the CG for data transmission. Therefore, for each participating AP, the paths from that AP to every other AP may be estimated, resulting in a total of 1×N wired paths (where “1” represents the participating AP, and “N” represents the number of APs in the CG minus 1). Each participating AP may then report the estimated paths to a coordinator, which leads to a total of N×N wired paths being estimated for the entire CG.

Once the paths are determined, the participating AP calculates or estimates the transit time (or delay) for each wired path based on factors like hop-delays and termination-delays.

At block 635, the participating AP evaluates wireless paths by measuring transit times, considering factors such as signal strength, traffic load, and interference levels.

At block 640, the participating AP reports the estimated transit times for both wired and wireless paths to a coordinator. The coordinator may be a central AP or WLC (e.g., 105-1 of FIG. 1A or 120 of FIG. 1B) in a centralized MAPC configuration. In a decentralized MAPC configuration, the coordinator may be any AP (e.g., 405-5 of FIG. 4A) or network device assigned for transport mode selection in the CG. In some embodiments, the coordinator may evaluate the MAPC mode (e.g., C-TDMA or C-SR) in use within the CG and analyze the reported transit times. Based on the evaluation, the coordinator may decide the suitable transport mode (wired or wireless) for the CG.

At block 645, the participating AP receives the selected transport mode (wired or wireless) from the coordinator.

At block 650, the participating AP sends and/or receive MAPC messages using the selected transport mode.

In some embodiments, operations at blocks 615-640 may be performed entirely by the coordinator. In such configurations, the participating AP may simply broadcast its capability information to the coordinator, which then performs the N×N wired path estimation, transit time calculation, and/or evaluation of wireless link speed and quality.

In some embodiments, the coordinator may be an SDN-based controller that collects metrics across the entire network that includes multiple CGs. The SND-based controller may operate at the IP layers and be configured to determine the transport mode for each CG within the network.

FIG. 7 is a flow diagram depicting an example method 700 for transport mode selection based on MAPC deployment modes, according to some embodiments of the present disclosure.

At block 705, a network device (e.g., AP/WLC 210 of FIG. 2) receives information from one or more other network devices within a coordination group (CG) (as depicted by step 220 of FIG. 2), where the CG is configured to operate using a Multi-AP Coordination Protocol (MAPC) mode, and the information indicates, for each of the one or more other network devices, a support of acting as a transport node over wired connections and a MAPC transport role within the CG.

At block 710, the network device estimates a wired path (e.g., paths 1-3 of FIG. 3) between the network device and at least one of the one or more other network devices in the CG based on the received information.

At block 715, the network device determines a transit time required for communication over at least one of the wired paths.

At block 720, the network device determines a transit time required for communication over at least one of wireless paths between the network device and at least one of the one or more other network devices in the CG.

At block 725, the network device selects a transport mode for at least one of the one or more other network devices in the CG based on at least one of the transit times for the wired paths, the transit times for the wireless paths, and the MAPC mode.

At block 730, the network device communities the transport mode to at least one of the one or more other network devices in the CG (as depicted by step 225 of FIG. 2).

At block 735, the network device performs a network coordination operation with the one or more other network devices in the CG using the transport mode (as depicted by steps 230-235 of FIG. 2).

In some embodiments, the network device may broadcast, to at least one of the one or more other network devices in the CG, information indicating a support of acting as a transport node over wired connections and a MAPC transport role of the network device within the CG.

In some embodiments, the network device may comprise a leader access point (AP) (e.g., AP/WLC 210 of FIG. 2, AP 305-1 of FIG. 3) in the CG. In some embodiments, the network device may comprise a wireless local area network (LAN) controller (WLC) (e.g., WLC 120 of FIG. 1B, AP/WLC 210 of FIG. 2).

In some embodiments, the network device may comprise a participating access point (AP) (e.g., AP 205 of FIG. 2) in the CG, and the selection of the transport mode further considers at least one of transit times for wired paths between each of the one or more other network devices in the CG (e.g., paths 1-9 of FIGS. 4A-4C), or transit times for wireless paths between each of the one or more other network devices in the CG.

In some embodiments, the network device may transmit the transit times for the wired paths to a central controller, where the central controller collects transit times from one or more other CGs in a network, and receive a second transport mode for the CG determined by the central controller.

In some embodiments, the MAPC transport role may comprise one of a relay or an endpoint within the CG.

In some embodiments, broadcasting the information may be performed using over-the-air (OTA) transmission, and the information may be transmitted in frames comprising at least one of a beacon or a probe response.

In some embodiments, the transit time for communication over at least one of the wired paths is determined based on a termination-delay associated with a termination node and a hop-delay associated with one or more relay nodes.

In some embodiments, the selection of the transport mode may further consider at least one of packet loss for the wired paths and packet loss of the wireless paths.

FIG. 8 depicts an example network device 800 configured to perform various aspects of the present disclosure, according to some aspects of the present disclosure. In some embodiments, the example network device 900 may correspond to a leader AP, such as AP 105-1 as depicted in FIG. 1A, RAP 135 as depicted in FIG. 1C, or AP/WLC 210 as depicted in FIG. 2, or AP 305-1 as depicted in FIG. 3. In some embodiments, the example network device 900 may correspond to a WLC, such as WLC 120 as despite in FIG. 1B, or AP/WLC 210 as depicted in FIG. 2. In some embodiments, the example network device 900 may correspond to a participating AP, such as AP 105-2 or 105-3 as depicted in FIG. 1A, AP 105 as depicted in FIG. 1B, MAP 130 as depicted in FIG. 1C, AP 205 as depicted in FIG. 2, AP 305-2, 305-3, 305-4, 305-5, or 305-6 as depicted in FIG. 3, or AP 405 as depicted in FIGS. 4A-4C.

As illustrated, the example network device 800 includes a processor 805, memory 810, storage 815, one or more transceivers 820, one or more I/O interfaces 880, and one or more network interfaces 825. In some embodiments, I/O devices 840 are connected via the I/O interface(s) 880. Further, via the network interface 825, the network device 800 can be communicatively coupled with one or more other devices and components (e.g., via a network, which may include the Internet, local network(s), and the like). Each of the components is communicatively coupled by one or more buses 830. In some embodiments, one or more antennas 835 may be coupled to the transceivers 820 for transmitting and receiving wireless signals.

The processor 805 is generally representative of a single central processing unit (CPU) and/or graphic processing unit (GPU), multiple CPUs and/or GPUs, a microcontroller, an application-specific integrated circuit (ASIC), or a programmable logic device (PLD), among others. The processor 805 processes information received through the transceiver 820, I/O interfaces 880, and the network interfaces 825. The processor 805 retrieves and executes programming instructions stored in memory 810, as well as stores and retrieves application data residing in storage 815.

The storage 815 may be any combination of disk drives, flash-based storage devices, and the like, and may include fixed and/or removable storage devices, such as fixed disk drives, removable memory cards, caches, optical storage, network attached storage (NAS), or storage area networks (SAN). The storage 815 may store a variety of data for the efficient functioning of the system.

The memory 810 may include random access memory (RAM) and read-only memory (ROM). The memory 810 may store processor-executable software code containing instructions that, when executed by the processor 805, enable the network device 800 to perform various functions described herein for wireless communication. In the illustrated example, the memory 810 includes four software components: the transport information management component 845, the path estimation component 850, the transport mode selection component 855, and the coordination control component 860.

In one embodiment, the transport information management component 845 may broadcast the device's 800 transport capability (e.g., wired support) and transport role within the network (e.g., relay or endpoint) to other APs or WLCs, and receive transport information from other APs. In some embodiments, the transport information management component 845 may be part of a messaging and communication component that handles network discovery and coordination communication using specific frames like beacons, probe responses, or customized MAPC frames.