How does multidrop SPE work?

Multidrop capability is included in several IEEE single-pair Ethernet (SPE) standards, including the 802.3cg substandards 10BASE-T1S and 10BASE-T1M, and the newer 802.3da, which provides enhancements to 10BASE-T1M. This article begins with a brief review of SPE multidrop standards and then explores the bus timing requirements and the use of Physical Layer Collision Avoidance (PLCA) and dynamic PLCA.

SPE standards define two basic architectures: point-to-point (P2P) and multidrop. For example, 802.3bp, 1000BASE-T1 supports data rates up to 1 Gbps and a P2P architecture. In the 802.3cg series, 10BASE-T1S supports data rates up to 10 Mbps and 15 m P2P or 25 m multidrop applications, and 10BASE-T1M supports data rates up to 10 Mbps and multidrop applications with a reach of at least 25 m (Figure 1). 802.3da provides enhancements to 10BASE-T1M, including doubling the reach to 50 m.

Managing multidrop communications

Nodes, also called stations or devices, on an SPE multidrop network can share a single pair cable using the standard Ethernet carrier-sense multiple access with collision detection (CSMA/CD) technology or operate using PLCA. SPE with PLCA does not require an Ethernet switch since all nodes are connected to a single bus. One of the communication nodes is also defined as the server or PLCA coordinator and uses PLCA to prevent data collisions.

Each node is configured with a node ID number, and the node assigned ID0 becomes the coordinator. A communications cycle is initiated when the coordinator sends a beacon message that other nodes use to coordinate their clocks.

The process starts when the server opens a 25 μs time widow for transmitting the first node (ID1). The node is permitted to transmit all its data. The nodes listen to determine when the node with an ID number one lower than its ID has stopped transmitting or has reached the end of its assigned 25 μs time window, at which time it can transmit data.

The process of listening is repeated until all the nodes have had an opportunity to send data. The cycle is completed at that time, and the PLCA coordinator will initiate the next cycle by sending a beacon signal. Communication latency in PLCA cycles depends on the number of nodes transmitting data. There are three cases (Figure 2):

• Minimum latency occurs when no nodes transmit any data during a cycle.
• Intermediate latency occurs when some fraction of the nodes transmits some amount of data.
• Maximum latency occurs when all nodes have a maximum packet size and transmit data.

PLCA and CSMA/CD

PLCA defines a way to operate a multidrop network with a high utilization level and is generally used in combination with CSMA/CD. It also avoids collisions that can occur and reduces throughput when a CSMA/CD-only link operates near maximum capacity (Figure 3).

Multidrop and energy-efficient Ethernet

In a multidrop network, a PHY that’s not transmitting puts the corresponding physical medium-dependent (PMD) sublayer into a high-impedance mode, which saves energy. As a result, there is no need for the dedicated Energy Efficient Ethernet (EEE) protocol with SPE multidrop networks.

Summary

Several standards define SPE multidrop implementations, including 802.3cg 10BASE-T1S, 10BASE-T1M, and 802.3da. The enhancements included in 802.3da include doubling the maximum transmission distance of 10BASE-T1M to 50 m. SPE multidrop communications are implemented using a combination of PLCA and CSMA/CD. The addition of PLCA ensures the highest level of network throughput. SPE multidrop networks inherently need the requirements of EEE without a separate protocol.

References

A technical paper of the Single Pair Ethernet System Alliance
Physical-Layer Collision Avoidance in 10Base-T1S Automotive Ethernet, Teledyne Lecroy
Single pair Ethernet: Applications and use cases studies, Single Pair Ethernet
Single pair Ethernet Data and Power for the Wired World, IEEE
Single Pair Ethernet Technology Brief, AimValley