The present disclosure relates to multidrop bus networks, and in particular concerns an apparatus and associated method of collision avoidance for a node of a multidrop bus network.
Collisions on the bus of a multidrop bus network can cause a decrease in bus efficiency and throughput. The apparatus and associated method described herein may address this issue.
According to a first aspect of the present disclosure, there is provided a reconciliation module for a node of a multidrop bus network, the node comprising a MAC module and a PHY module, the reconciliation module comprising circuitry configured to:
In one or more embodiments, the reconciliation module may be configured to delay de-assertion of the sense signal until a predefined wait time of the MAC module in advance of the next transmit opportunity of the node in the absence of intervening transmissions from the other nodes of the multidrop bus network.
In one or more embodiments, in the event of an intervening transmission from another node of the multidrop bus network after transmission of the data, the reconciliation module may be configured to delay de-assertion of the sense signal after the intervening transmission until the predefined wait time in advance of the next transmit opportunity. This assumes that the time between the end of the intervening transmission and the start of the next transmit opportunity is longer than the wait time, otherwise it would not be possible to utilise the next transmit opportunity.
In one or more embodiments, in the event of an intervening transmission from another node of the multidrop bus network after transmission of the data, the reconciliation module may be configured to:
In one or more embodiments, in the event of an intervening transmission from another node of the multidrop bus network after transmission of the data, the reconciliation module may be configured to:
In one or more embodiments, the reconciliation module may be configured to:
In one or more embodiments, the MAC module may be configured to delay resending of the further data by a wait time after de-assertion of the sense signal, and the reconciliation module may be configured to transmit a filler sequence during the wait time of the MAC module to prevent the next transmit opportunity of the node from timing out.
In one or more embodiments, the time between the transmit opportunity and next transmit opportunity of the node in the absence of intervening transmissions from all other nodes may be equal to or greater than the wait time of the MAC module.
In one or more embodiments, the wait time of the MAC module may be equal to an inter-packet-gap length.
In one or more embodiments, each transmit opportunity may have a duration of no longer than one inter-packet gap length.
In one or more embodiments, the MAC module may require a recovery time following assertion of the collision signal before it can resend the further data as part of the re-transmission attempt, and the reconciliation module may be configured to wait at least the recovery time before de-asserting the sense signal.
In one or more embodiments, the reconciliation module may be configured to determine the recovery time of the MAC module.
In one or more embodiments, the reconciliation module may be configured to determine the recovery time of the MAC module by monitoring timing properties of the MAC module or by obtaining them from a register.
In one or more embodiments, the reconciliation module may be configured to receive a detection signal from the PHY module indicative of activity from other nodes on the multidrop bus network, and de-assert the sense signal when the detection signal indicates that an immediately preceding node in a bus schedule of the multidrop bus network is no longer active or has timed out to enable transmission of the further data as part of a (re-) transmission attempt.
In one or more embodiments, the multidrop bus network may comprise at least three nodes.
In one or more embodiments, the data received from the MAC module may comprise a plurality of data frames within the same sequence for transmission at the transmit opportunity, and the reconciliation module may be configured to insert an active-idle sequence between adjacent data frames to prevent the multidrop bus network from becoming idle before the plurality of data frames have been transmitted.
In one or more embodiments, the reconciliation module may be bufferless.
According to a second aspect of the present disclosure, there is provided a node of a multidrop bus network, the node comprising the reconciliation module of the first aspect, the MAC module and the PHY module.
In one or more embodiments, the multidrop bus network may be an Ethernet network, the MAC module may be an Ethernet CSMA/CD MAC layer and the PHY module may be a 10BASE-T1S PHY layer.
According to a third aspect of the present disclosure, there is provided a multidrop bus network comprising a plurality of the nodes of the second aspect.
According to a fourth aspect of the present disclosure, there is provided a method of collision avoidance for a node of a multidrop bus network, the node comprising a MAC module and a PHY module, the method comprising:
According to a fifth aspect of the present disclosure, there is provided a computer program comprising computer code configured to control the reconciliation module of the first aspect, control the node of the second aspect, control the multidrop bus network of the third aspect, or perform the method of the fourth aspect.
While the disclosure is amenable to various modifications and alternative forms, specifics thereof have been shown by way of example in the drawings and will be described in detail. It should be understood, however, that other embodiments, beyond the particular embodiments described, are possible as well. All modifications, equivalents, and alternative embodiments falling within the spirit and scope of the appended claims are covered as well.
The above discussion is not intended to represent every example embodiment or every implementation within the scope of the current or future Claim sets. The figures and Detailed Description that follow also exemplify various example embodiments. Various example embodiments may be more completely understood in consideration of the following Detailed Description in connection with the accompanying Drawings.
One or more embodiments will now be described by way of example only with reference to the accompanying drawings in which:
While much of the focus in recent Ethernet development has centred on high data rates, not every application requires speeds of up to 400 Gbps. For some applications, including Internet of Things (IoT), industrial and automotive, 10 Mbps is sufficient for certain functions. Factors like cost, weight, distance and the space required for cables are more important for these use cases.
Recognizing these evolving requirements, IEEE began work in early 2017 to define IEEE 802.3cg, a standard for single-pair Ethernet that supports 10 Mbps. The goals of IEEE 802.3cg were to define a point-to-point and a multidrop short-distance standard with a maximum length of 25 meters, and a long-distance point-to-point standard that supports distances up to 1,000 meters. The resulting IEEE 802.3cg specification includes two link-layer standards: 10BASE-T1S and 10BASE-T1L.
The 10BASE-T1S short-range standard is primarily targeted at automotive and industrial applications. Multiple nodes on the network can share a cable in half-duplex shared-medium mode (multidrop mode) using the standard Ethernet Carrier-Sense Multiple Access with Collision Detection (CSMA/CD) access method or operate using PHY-Level Collision Avoidance (PLCA). The cable might be an unshielded twisted pair (UTP) that may have multiple nodes with their medium dependent interfaces (MDIs) attached thereto. As such, 10BASE-T1S may also be referred to as Multidrop Single Pair Ethernet.
The 10BASE-T1L long-range option is designed for IoT and industrial control applications. The 1,000-meter range is sufficient for use in large factories or warehouses, and 10 Mbps is sufficient for gathering data from sensors and to monitor and control many types of industrial machinery. It shares the advantages of the short network variant: MAC-level compatibility with multi-pair Ethernet and lower cost, weight and required space.
The following description relates to the 10BASE-T1S standard but may be also applicable to other (including future) networking standards and is therefore not necessarily limited to Ethernet or 10BASE-T1S.
A problem with CSMA/CD is that multiple nodes may attempt to transmit data on the bus simultaneously when the bus is silent. This leads to collisions, corrupted data, time-outs and loss of bus capacity, which in turn can result in severe degradation of effective bus throughput or even a complete collapse as the total bus load gets closer to (e.g. around 70% of) the maximum bus capacity.
PLCA uses a round-robin scheme in which the nodes are numbered and given the opportunity to transmit one at a time starting at a dedicated time slot (or transmit opportunity TO) in a cyclic manner based on their numbering. One node is assigned as the head node (or PLCA coordinator) and starts each cycle with a beacon. If no data is transmitted by a node at its transmit opportunity, the transmit opportunity is passed to the next node. The reconciliation module knows when the transmit opportunity of its node occurs based on the node number and manages the MAC module using collision (COL-M) and sense (CRS-M) signal manipulations to prevent the MAC module from sending data that cannot be transmitted by the PHY module. In this way, the reconciliation module only passes data to the PHY module when there is a transmit opportunity. The combination of the round-robin scheme with the reconciliation module therefore avoids collisions on the bus. Since there should be no collision events for the PHY module to report, the COL-P pin becomes functionally redundant (and is hence greyed out in
The reconciliation module receives a detect signal (CRS-P) from the PHY module indicative of transmission activity detected on the bus. The detect signal substantially follows the bus schedule with a slight delay depending on the responsiveness of the PHY module to detect and report the bus activity. The reconciliation module is configured to de-assert the sense signal (CRS-M) each time the detect signal (CRS-P) is de-asserted, unless there has been an earlier transmission attempt by the MAC module for the same frame as described below.
The MAC module is configured to send data to the reconciliation module whenever it has a frame pending for transmission and the sense signal (CRS-M) has been de-asserted by the reconciliation module. To ensure a minimum gap between frames for most Ethernet PHY types, however, the MAC module will delay sending the data to the reconciliation module by a wait time after de-assertion of the sense signal (CRS-M). In this example shown, the wait time of the MAC module is equal to an inter-packet-gap (IPG) length (e.g. 12 bytes). The reconciliation module then re-asserts the sense signal (CRS-M) again on receipt of the data from the MAC module or on detecting activity on the bus via the detect signal (CRS-P). In the examples shown herein, de-assertion of the sense signal (CRS-M) is directly correlated with de-assertion of the detect signal (CRS-P), but re-assertion of the sense signal (CRS-M) is slightly delayed relative to re-assertion of the detect signal (CRS-P). This is intended to help ensure that the MAC module sends the pending data after the wait time even if the transmit opportunities are shorter than 1 IPG, but may not be necessary. As such, re-assertion of the sense signal (CRS-M) could also be directly correlated with re-assertion of the detect signal (CRS-P).
While the sense signal (CRS-M) is de-asserted, the data from the MAC module is buffered in the reconciliation module provided the buffer does not overflow. If a transmission attempt by the MAC module cannot be sufficiently aligned with a transmit opportunity for that node using the buffer, the reconciliation module is configured to assert a collision signal (COL-M) to stop the MAC module from continuing to send the data. This happens by default when another node starts transmitting a data frame on the bus and the detect signal (CRS-P) is asserted. The reconciliation module then de-asserts the collision signal (COL-M). The collision signal (COL-M) need only be asserted long enough to be recognised by the MAC module, and should ideally be as short as possible (e.g. one clock cycle) to avoid any loss of time until recovery.
Following assertion of the collision signal (COL-M) by the reconciliation module, the sense signal (CRS-M) is not immediately de-asserted. This is for two reasons. Firstly, the MAC module requires a recovery time (CR) following assertion of the collision signal before it can resend the data as part of the upcoming transmission attempt. Secondly, it is more efficient for the bus to try and ensure that the re-transmission attempt is synchronised with the upcoming transmit opportunity to avoid any further delays in transmitting the data frame. To achieve this, the sense signal (CRS-M) is kept asserted until the data can be directly transmitted without buffering at the upcoming transmit opportunity and then de-asserted to enable said transmission after the recovery time of the MAC module.
The data frame is then passed by the reconciliation module to the PHY module for transmission (PHY-TXDATA) starting at the upcoming transmit opportunity. It should be mentioned that transmit opportunities can be relatively short (e.g. shorter than the IPG length) and typically do not provide enough time for a node to transmit complete frames. Rather, they provide a time slot or window for the node to claim the bus before the opportunity times out and the next node has the opportunity. A node can claim the bus during a transmit opportunity by starting to transmit a data frame or a filler sequence (SYNC symbols). For example, the reconciliation module may be configured to transmit SYNC symbols on the bus via the PHY module during the wait time of the MAC module. Once the bus has been claimed by the node, it can then proceed to transmit any frame size or even multiple data frames if configured to allow this.
In the example of
As shown in the figure, the MAC module of node 2 attempts to transmit frame 1 an IPG length after the sense signal (CRS-M) is de-asserted following transmission of the first frame by node 1. Since this is too late for transmission at the upcoming transmit opportunity of node 2, the reconciliation module temporarily stores frame 1 in the buffer until the next transmit opportunity arrives, at which point it is transmitted on the bus by the PHY module. This is only possible because there are no intervening transmissions by other nodes, otherwise the buffer would have been flushed and a collision signal asserted (as per the example in
Although the PLCA mechanism successfully avoids collisions on the bus by blocking transmission outside of the scheduled transmit opportunities for the node, the use of a data buffer in the reconciliation module complicates the system and loses its benefit when the MAC module tries to send more data—before it can be transmitted on the bus—than the buffer can temporarily store or when any other node transmits a frame. Furthermore, the time-period during which it is too early to transmit the data from the MAC module and yet too late for re-transmission after a virtual collision is relatively short in relation to the total bus cycle time. In view of the above, it can be difficult to justify the additional cost and complexity associated with the data buffer.
Also, in the event that the data is received from the MAC module while not being able to transmit it via the PHY module, the reconciliation module is configured to assert a collision signal (COL-M), keep the sense signal (CRS-M) asserted until the data can be transmitted at the upcoming transmit opportunity, and then de-assert the sense signal (CRS-M) at the upcoming transmit opportunity to enable said transmission.
A key difference from the earlier collision avoidance mechanism, however, occurs when intervening transmissions from the other nodes of the multidrop bus network preceding the next transmit opportunity of the node are skipped (i.e. timed-out). Note that the reconciliation module is not required to de-assert the sense signal (CRS-M) immediately when the detect signal (CRS-P) from the PHY module is de-asserted for first transmission attempts. With this collision avoidance mechanism, the reconciliation module is configured to delay de-assertion of the sense signal (CRS-M) after transmission of the data to facilitate the transmission of further data at the next transmit opportunity of the node.
As shown in
When compared to the example of
De-assertion of the sense signal (CRS-M) is then delayed after the transmission of frame 1 until one IPG length in advance of the next transmit opportunity of the node. As such, the reconciliation module receives frame 2 immediately after the wait time of the MAC module, which is then passed to the PHY module for transmission on the bus at the next transmit opportunity. This is repeated for frame 3 of node 2. It is also worth noting here that, since frames 2 and 3 are transmitted on the first transmit attempt, the reconciliation module does not claim the bus by sending additional SYNC symbols to the PHY module as it did for the re-transmission of frame 1 (every transmission starts with at least 2 SYNC symbols as part of the standardized format). The use of additional SYNC symbols in this way only occurs for re-transmissions because the reconciliation module knows for sure that a frame is coming from the MAC module and the bus needs to be claimed to utilise the transmit opportunity.
Although the sense signal (CRS-M) is also delayed following the transmission of frame 3 by node 2, the intervening transmission of frame 2 by node 3 causes the sense signal (CRS-M) to follow the detect signal (CRS-P) received from the PHY module. As a result, the sense signal (CRS-M) is de-asserted immediately after the intervening transmission. Since frame 4 of node 2 is available in the MAC module at this point, it is then sent to the reconciliation module after the (one IPG) wait time of the MAC module which triggers a virtual collision (COL-M) and re-transmission of the frame at the upcoming transmit opportunity.
As shown with frames 2 and 3 of node 2 in
Since no collision signal (COL-M) is asserted by the reconciliation module in this scenario, there is no need to fit the recovery time (CR) of the MAC module between adjacent transmit opportunities of the node. As such, the minimum required time between adjacent transmit opportunities of one node is the wait time of the MAC module following de-assertion of the sense signal (CRS-M), which in the illustrated example is only one IPG length.
The minimum time requirement also dictates the minimum number of nodes needed for this collision avoidance mechanism to provide benefit. The duration of a transmit opportunity is a variable to be set by the system integrator and depends on several parameters. In practice, however, it is likely to be around 0.5 IPG lengths. In the example of
The collision avoidance mechanism of
Rather than delaying de-assertion of the sense signal (CRS-M) in the event of an intervening transmission (i.e. no boomerang transmit opportunity), the reconciliation module may instead be configured to de-assert the sense signal (CRS-M) on detection of the intervening transmission and keep it de-asserted until the receipt of further data from the MAC module.
Although the illustrated examples show each transmit opportunity limited to the transmission of a single data frame, it is also possible for a node to transmit a plurality of data frames within a single sequence starting at one transmit opportunity. To allow for this, the reconciliation module may be configured to insert an active-idle sequence between adjacent data frames of the sequence to prevent the bus from becoming idle before the plurality of data frames have been transmitted. Note that in this scenario, the sense signal to the MAC module is de-asserted after every frame while the signal driven on the bus is active-idle. This is done to ensure that the MAC module keeps sending frames up to a predefined maximum number of frames in a burst. Once the maximum number is reached, any further frame will be blocked by a indicating a collision to the MAC module with re-transmission at the next transmit opportunity. If the MAC module has less than the maximum allowed number of frames available to send as a burst, the transmission will time out and stop driving active-idle.
Each node of the multidrop bus network described herein may comprise a MAC module, a PHY module and a reconciliation module. The multidrop bus network may be an Ethernet network, the MAC module may be an Ethernet CSMA/CD MAC layer and the PHY module may be a 10BASE-T1S PHY layer. Furthermore, given that the reconciliation module described in relation to
It should be noted here that the head node of a 10BASE-T1S bus with PLCA transmits beacons to indicate the start of a new bus cycle. This also allows other nodes to synchronize and detect how many nodes are present on the bus (node count before cycle restart). Beacons have not been included in the diagrams for simplicity reasons, but their presence does not preclude the applicability of the ideas/concepts described herein. Nevertheless, the beacon should not be considered as an intervening data transmission, and the de-assertion moment of the boomerang transmit opportunity should take the time duration of the beacon into account.
Although the reconciliation module (or layer) has been described herein with reference to the MAC and PHY modules (or layers), it could be formed independently of the other modules and incorporated into the node during a subsequent modular assembly process. In some cases, the reconciliation module may even be retrofit to an existing node. Alternatively, the reconciliation module may be integrated with the other modules of the node at the time of manufacture (e.g. as different parts/portions of the same chip). One or more (or each) module may comprise circuitry including at least one processor and at least one memory including computer program code. The at least one memory and computer program code may be configured to, with the at least one processor, cause the module to perform the associated functionality described herein.
The instructions and/or flowchart steps in the above figures can be executed in any order, unless a specific order is explicitly stated. Also, those skilled in the art will recognize that while one example set of instructions/method has been discussed, the material in this specification can be combined in a variety of ways to yield other examples as well, and are to be understood within a context provided by this detailed description.
In some example embodiments the set of instructions/method steps described above are implemented as functional and software instructions embodied as a set of executable instructions which are effected on a computer or machine which is programmed with and controlled by said executable instructions. Such instructions are loaded for execution on a processor (such as one or more CPUs). The term processor includes microprocessors, microcontrollers, processor modules or subsystems (including one or more microprocessors or microcontrollers), or other control or computing devices. A processor can refer to a single component or to plural components.
In other examples, the set of instructions/methods illustrated herein and data and instructions associated therewith are stored in respective storage devices, which are implemented as one or more non-transient machine or computer-readable or computer-usable storage media or mediums. Such computer-readable or computer usable storage medium or media is (are) considered to be part of an article (or article of manufacture). An article or article of manufacture can refer to any manufactured single component or multiple components. The non-transient machine or computer usable media or mediums as defined herein excludes signals, but such media or mediums may be capable of receiving and processing information from signals and/or other transient mediums.
Example embodiments of the material discussed in this specification can be implemented in whole or in part through network, computer, or data based devices and/or services. These may include cloud, internet, intranet, mobile, desktop, processor, look-up table, microcontroller, consumer equipment, infrastructure, or other enabling devices and services.
In one example, one or more instructions or steps discussed herein are automated. The terms automated or automatically (and like variations thereof) mean controlled operation of an apparatus, system, and/or process using computers and/or mechanical/electrical devices without the necessity of human intervention, observation, effort and/or decision.
It will be appreciated that any components said to be coupled may be coupled or connected either directly or indirectly. In the case of indirect coupling, additional components may be located between the two components that are said to be coupled.
In this specification, example embodiments have been presented in terms of a selected set of details. However, a person of ordinary skill in the art would understand that many other example embodiments may be practiced which include a different selected set of these details. It is intended that the following claims cover all possible example embodiments.
Number | Date | Country | Kind |
---|---|---|---|
23204207.7 | Oct 2023 | EP | regional |