Energy saving dynamic bandwidth assignment

Abstract

The present disclosure provides apparatus and methods for energy efficient dynamic bandwidth assignment. Example embodiments describe an optical line terminal (110), OLT, configured to communicate with an optical network unit (130, 140, 150), ONU, in a passive optical network (100), PON; wherein the OLT comprises a module configured to dynamically assign to traffic bearing entities (131, 132) within the ONU a bandwidth equal to an estimated bandwidth demand of a respective traffic bearing entity by allocating uplink transmission opportunities (201, 204, 207, 210) to the traffic bearing entities; and wherein the module is further configured to perform assigning a bandwidth margin (203, 206, 209, 212) to a respective traffic bearing entity in addition to estimating a bandwidth demand (202, 205, 208, 211); and wherein the OLT further comprises a receiver configured to receive an uplink optical signal from the traffic bearing entity during the allocated uplink transmission opportunity; and wherein the receiver is further configured to turn off the one or more functions during an unallocated period (213, 214, 215) without uplink optical signals.

Description

Technical Field

[0001] Various example embodiments relate to an apparatus and method for dynamic bandwidth assignment in a passive optical network. Background Art

[0002] Dynamic bandwidth assignment (DBA) is a function in a passive optical network (PON) that dynamically assigns upstream transmission opportunities to service bearers of optical network units (ONUs). An operator typically provides one or more service parameters for traffic flows, which guarantee the quality of service by imposing constraints on the assigned bandwidth and latency. The assigned bandwidth or the assigned data rate of a traffic flow is typically determined based on the current activity of the service bearer (i.e., the bandwidth demand) and bandwidth-related service parameters (such as fixed bandwidth, guaranteed bandwidth, and maximum bandwidth). When the upstream traffic load is low, some DBA algorithms can result in useless power consumption within the PON. Other DBA algorithms can result in a degradation of the quality of service. Summary of the Invention

[0003] The scope of protection of various embodiments of the present invention is set forth in the independent claims. Embodiments and features described in this specification that are not within the scope of the independent claims (if any) should be construed as examples that contribute to the understanding of various embodiments of the present invention.

[0004] In particular, an object of embodiments of the present invention is to reduce power consumption in a passive optical network without negatively affecting the quality of service.

[0005] According to a first example aspect of the present disclosure, this object is achieved by an optical line terminal (OLT) configured to communicate with an optical network unit (ONU) in a passive optical network (PON); wherein the OLT includes means configured to dynamically assign to the service bearer a bandwidth equal to the estimated bandwidth demand of the corresponding service bearer by allocating an upstream transmission opportunity to the service bearer within the ONU; and wherein the means is further configured to assign a bandwidth margin to the corresponding service bearer in addition to the estimated bandwidth demand; wherein the OLT further includes a receiver configured to receive an upstream optical signal from the service bearer during the allocated upstream transmission opportunity; and wherein the receiver is further configured to turn off one or more functions during unallocated periods without an upstream optical signal.

[0006] Therefore, the bandwidth assigned to a service bearer entity is the sum of the estimated bandwidth requirement of the service bearer entity and the bandwidth margin. In other words, the length of the uplink transmission opportunity assigned to the service bearer entity for transmitting an uplink burst signal is defined by the bandwidth margin and the estimated bandwidth requirement. For example, the estimated bandwidth requirement may be determined or estimated based on the buffer occupancy reported within the service bearer entity requested by the OLT and provided by the ONU, such as Dynamic Bandwidth Report Uplink (DBRu) according to ITU-T G.9807.1 standard. Alternatively or additionally, the estimated bandwidth requirement may be determined by monitoring the uplink frames received by the OLT, such as by measuring the amount of payload data transmitted during one or more transmission opportunities and / or by comparing the number of idle XGEM frames with the bandwidth map. The respective service bearer entities may have different bandwidth margins assigned to them, for example depending on their respective estimated bandwidth requirements. Alternatively or additionally, some or all of the service bearer entities may have substantially the same bandwidth margin assigned to them.

[0007] By assigning a bandwidth equal to the sum of the estimated bandwidth requirement and the bandwidth margin, when the uplink traffic load within the PON is below the effective available bandwidth capacity, only a portion of the effective available bandwidth capacity of the PON is assigned to the service bearer entity. Thus, the uplink transmission window includes periods of uplink transmission opportunities that are not allocated, i.e., unallocated periods. During these periods, turning off one or more functions of the OLT receiver reduces the power consumption of the OLT by preventing unnecessary power consumption for receiving idle data or idle XGEM frames within the uplink optical signal. The one or more functions of the receiver may include analog circuits and / or digital signal processing circuits, such as analog-to-digital converters, clock and data recovery devices, equalizers, and / or decoders. Turning off one or more receiver functions means adjusting the operating state of at least a portion of the OLT receiver from a first operating state for receiving and decoding the uplink optical signal to a second operating state that consumes less power than the first operating state.

[0008] Assigning the bandwidth margin above the estimated bandwidth ensures that sufficient bandwidth is assigned to the service bearer entity to avoid or limit negative impacts on the quality of service, as the estimated bandwidth requirement may underestimate the effective bandwidth requirement, i.e., the true or actual bandwidth requirement. Doing so can avoid or limit queue fill build-up, increased latency, jitter, increased packet loss probability, and / or reduced TCP throughput.

[0009] Thus, in addition to the estimated bandwidth requirements, the bandwidth margin is assigned to the respective service bearer entity, and one or more receiver functions are turned off during the unallocated period, allowing the power consumption of the PON to be reduced while avoiding negative impacts on the quality of service. The advantage is that this can be easily incorporated into existing PONs because it only requires limited (e.g., to the DBA algorithm) software modifications.

[0010] According to an example embodiment, the receiver is configured to decode an upstream optical signal according to a low-density parity-check (LDPC) error decoding; wherein, the LDPC error decoding is turned off during the unallocated period.

[0011] LDPC error decoding is typically present in recent PON technologies, such as 25GS-PON, 50G-PON according to the ITU-T G.9804 standard, or PON for optical transmission above 50 Gb / s according to the ITU-T G.suppl.V HSP work plan. LDPC error decoding can achieve a lower bit error rate at a given energy noise power ratio per useful bit, for example, compared to Reed-Solomon error decoding. Turning off or not performing the LDPC error decoding during periods without upstream transmission opportunities can avoid the relatively large power consumption associated with LDPC error decoding of idle data or frames. Thus, the energy efficiency of the PON can be improved.

[0012] According to an example embodiment, one or more bandwidth margins can be a predetermined value regardless of the estimated bandwidth requirements.

[0013] In other words, the bandwidth margin of one or more service bearer entities can be a fixed value assigned in addition to the estimated bandwidth requirements of these service bearer entities.

[0014] According to an example embodiment, one or more bandwidth margins can be a dynamic value; and wherein, the module can be further configured to determine the one or more bandwidth margins of the one or more service bearer entities based on the estimated bandwidth requirements of the one or more service bearer entities.

[0015] Therefore, the size or value of the bandwidth margin for the respective service bearer entity can vary over time according to the estimated bandwidth requirements of the service bearer entity.

[0016] According to an example embodiment, determining one or more bandwidth margins can include determining the one or more bandwidth margins based on a proportional relationship with the estimated bandwidth requirements.

[0017] When doing so, the sum of the estimated bandwidth and the bandwidth margin (i.e., the assigned bandwidth) is proportional to the estimated bandwidth demand in the PON. For example, the assigned bandwidth margin can increase linearly with the estimated bandwidth demand. Thus, a small bandwidth demand results in a relatively small assigned bandwidth, i.e., with respect to XGEM frames, the uplink bandwidth amount is limited. A large bandwidth demand results in a relatively large assigned bandwidth, i.e., with respect to XGEM frames, the uplink bandwidth amount is large.

[0018] According to an example embodiment, determining one or more bandwidth margins can include determining the one or more bandwidth margins such that the bandwidth assigned to the corresponding one or more service bearer entities corresponds to a stepwise function of the estimated bandwidth demand.

[0019] In other words, the bandwidth margin of a service bearer entity is determined such that the assigned bandwidth (i.e., the sum of the bandwidth margin and the estimated bandwidth demand) follows a stepwise function of the estimated bandwidth demand. Thus, the assigned bandwidth can be determined as a piecewise constant function or a series of steps between a minimum and a maximum estimated bandwidth demand (e.g., 0% and 100% of the maximum bandwidth of a service bearer entity). To this end, intervals of the estimated bandwidth demand can correspond to different values of the assigned bandwidth.

[0020] According to an example embodiment, the stepwise function can have hysteresis.

[0021] Thus, if the estimated bandwidth demand exceeds an upper threshold, the assigned bandwidth (i.e., the sum of the estimated bandwidth demand and the bandwidth margin) can be determined to have a high value, and if the estimated bandwidth demand is below a lower threshold, the assigned bandwidth (i.e., the sum of the estimated bandwidth demand and the bandwidth margin) can be determined to have a low value. When the assigned bandwidth has a high value and the estimated bandwidth demand exceeds the lower threshold, or when the assigned bandwidth has a low value and the estimated bandwidth demand is below the upper threshold, no switching occurs, thus providing hysteresis when determining the bandwidth margin. This avoids unstable switching between assigned bandwidths. Clearly, the upper threshold and the lower threshold define a single-step function. The stepwise function can also be a multi-step function defined by multiple upper and lower thresholds.

[0022] According to an example embodiment, determining one or more bandwidth margins for one or more service bearer entities can also be based on corresponding service parameters of the one or more service bearer entities, which indicate a bias towards quality of service or energy saving.

[0023] Therefore, dedicated service parameters can be provided for the service-carrying entity, and the dedicated service parameters indicate preferences for quality of service or energy saving. Preferences or biases for quality of service may result in the assignment of a larger bandwidth margin, while preferences or biases for energy saving may result in the assignment of a smaller bandwidth margin. In addition to typical service parameters (such as fixed bandwidth, guaranteed bandwidth, and maximum bandwidth according to the ITU-T G.9807.1 standard), this dedicated service parameter can also be provided. The dedicated service parameter can indicate qualitative preferences or biases, such as "quality of service" or "energy saving". Alternatively or additionally, the dedicated service parameter can specify quantitative preferences or biases, such as maximum energy saving, minimum energy saving, threshold bandwidth estimation, minimum assigned bandwidth, maximum assigned bandwidth, or guaranteed latency. This dedicated service parameter allows control of the trade-off between quality of service impact and energy saving. Alternatively or additionally, the bias for quality of service or energy saving can be derived from a specific configuration of typical service parameters.

[0024] According to an example embodiment, the module can also be configured to optimize the bandwidth margin based on the buffer occupancy of the corresponding service-carrying entity.

[0025] The buffer occupancy or queue filling can be inferred through status reports and / or local monitoring. This feedback allows the establishment of a control loop that optimizes the bandwidth safety margin over time. For example, a generous bandwidth margin can be initially assigned to avoid negative impacts on the quality of service. The generous bandwidth margin can be gradually reduced to increase energy saving until the buffer occupancy feedback indicates a non-negligible increase in queue filling, i.e., a decrease in the quality of service.

[0026] According to an example embodiment, the module can also be configured to execute an instruction for one or more ONUs to turn off one or more transmitter functions for encoding an upstream optical signal and transmitting the upstream optical signal to the OLT.

[0027] The one or more transmitter functions can be, for example, a laser or a serializer / deserializer (SERDES) link of one or more ONUs. These one or more ONUs can be instructed to turn off these functions through a control message (such as a physical layer operation, administration, and maintenance (PLOAM) message). In addition to turning off one or more functions of the OLT receiver, turning off these transmitter functions allows further improvement of the energy efficiency of the PON by reducing the unnecessary power consumption of the ONUs.

[0028] According to a second exemplary aspect, an optical network unit (ONU) is disclosed that is configured to communicate with an optical line terminal (OLT) in a passive optical network (PON) according to the first aspect; and wherein the ONU includes a transmitter configured to send an upstream optical signal to the OLT during an upstream transmission opportunity allocated to the ONU; and wherein the transmitter is further configured to turn off one or more functions during the unallocated period.

[0029] According to an example embodiment, the transmitter may be further configured to turn off one or more functions during an upstream transmission opportunity allocated to other ONUs.

[0030] According to a third exemplary aspect, a computer-implemented method is disclosed for dynamically assigning bandwidth to traffic-bearing entities within an optical network unit (ONU) by an optical line terminal (OLT), the optical line terminal (OLT) being configured to communicate with the ONU in a passive optical network (PON); wherein the OLT includes a receiver configured to receive an upstream optical signal from the ONU; the computer-implemented method includes:

[0031] . Dynamically assigning to a corresponding traffic-bearing entity a bandwidth equal to an estimated bandwidth demand and a bandwidth margin of the corresponding traffic-bearing entity by allocating an upstream transmission opportunity to the traffic-bearing entity; and

[0032] . Turning off one or more functions of the receiver during an unallocated period in which there is no upstream optical signal.

[0033] According to a third exemplary aspect, a data processing system is disclosed that is configured to execute the computer-implemented method according to the second aspect.

[0034] According to a fourth exemplary aspect, a computer program product is disclosed that includes computer-executable instructions that, when executed by a computer, cause the computer to execute the computer-implemented method according to the second aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A schematic block diagram of an example passive optical network is shown;

[0036] Figure 2 An embodiment of allocating a transmission opportunity to a traffic-bearing entity by dynamic bandwidth assignment is shown, which improves the energy-saving operation of the PON without negatively affecting the quality of service within the PON;

[0037] Figure 3AIllustrates an example of the difference in power consumption of an OLT according to an example embodiment by dynamically assigning bandwidth as the sum of the estimated bandwidth demand of a service - bearing entity and the determined bandwidth margin;

[0038] Figure 3B Illustrates an example embodiment in which the assigned bandwidth corresponds to a step - like function with hysteresis;

[0039] Figure 3C Illustrates the variation over time of the total dynamically assigned bandwidth equal to the sum of the estimated bandwidth demand and the bandwidth margin according to an example embodiment;

[0040] Figure 4 Illustrates a table according to an example embodiment including the average queue filling of a service - bearing entity and the power consumption of a part of the OLT receiver when applying a typical DBA algorithm that allocates the entire upstream transmission window, a DBA algorithm that allocates a transmission opportunity equal to the estimated bandwidth demand, and a DBA algorithm; and

[0041] Figure 5 Illustrates an example of a suitable computing system according to an embodiment capable of implementing a method for dynamically assigning bandwidth to a service - bearing entity within an ONU. Detailed Description

[0042] Figure 1 Illustrates a schematic block diagram of an example passive optical network (PON) 100, which includes an optical line terminal (OLT) 110 connected to a plurality of optical network units (ONUs) 130, 140, 150 via an optical distribution network (ODN) 120. The ODN 120 may have a tree - like structure, which includes an optical feeder fiber 121, one or more passive optical splitters / multiplexers 123, and a plurality of optical distribution fibers or drop fibers that connect the splitter / multiplexer 123 to the respective ONUs 130, 140, 150. In the downstream direction, the passive optical splitter / multiplexer 123 splits the optical signal from the OLT 110 into lower - power optical signals for the connected ONUs 130, 140, 150, while in the upstream direction, the passive optical splitter / multiplexer 123 multiplexes the optical signals from the connected ONUs 130, 140, 150 into a burst signal for the OLT 110. In this example, the OLT 110 is connected to three ONUs 130, 140, 150, however, the OLT 110 may be connected to fewer or more nodes.

[0043] The passive optical network 100 can be a Gigabit Passive Optical Network (GPON) according to the ITU-T G.984 standard, a 10 Gigabit Passive Optical Network (10G-PON) according to the ITU-T G.987 standard, a 10G symmetric XGS-PON according to the ITU-T G.9807 standard, a four-channel 10G symmetric NG-PON2 according to the ITU-T G.989 standard, a 25G S-PON, a 50G-PON according to the ITU-T G.9804 standard, or a Next Generation Passive Optical Network (NG-PON). The passive optical network 100 can implement Time Division Multiplexing (TDM) or Time Division Wavelength Division Multiplexing (TWDM).

[0044] In Time Division Multiplexing (TDM), the telecommunication medium 121 is shared in time among the ONUs 130, 140, 150 in the upstream direction. For this purpose, recurrent transmission opportunities 133, 142, 143, 144, 154, 155 are assigned to the respective ONUs 130, 140, 150, during which the respective ONUs 130, 140, 150 are allowed to send data to the OLT 110. The transmission opportunity can also be referred to as a time slot or a burst. For example, the ONU 140 is allowed to send upstream data during the recurrent transmission opportunities 142, 143, 144.

[0045] The respective ONUs 130, 140, 150 include one or more service bearer entities 131, 132, 141, 151, 152, 153, where data packets 160 from the connected services 171 - 176 or applications wait, for example in a data queue or buffer, for their turn to be sent to the OLT 110. One or more service bearer entities 131, 132, 141, 151, 152, 153 can be transmission containers, also referred to as T-CONTs. A transmission container is an ONU object that represents a set of logical connections within the ONUs 130, 140, 150, which appear as a single entity in the passive optical network for the purpose of upstream bandwidth assignment.

[0046] Therefore, the data sent during the recurrent transmission opportunities 133, 142, 143, 144, 154, 155 may originate from the service bearer entities 131, 132, 141, 151, 152, 153 within the associated ONUs 130, 140, 150. The respective service bearer entities 131, 141, 151 are allowed to send data to the OLT 110 during dedicated transmission opportunities 133, 142, 143, 144, 154, 155 that occur repeatedly in time (i.e., during recurrent time slots). One or more non-overlapping recurrent transmission opportunities (such as 143 and 144) associated with different transmission queues can be assigned between consecutive transmission opportunities (such as 154 and 155) associated with a certain service bearer entity.

[0047] Transmission opportunities associated with different service bearer entities within the same ONU (such as transmission queues 131 and 132) can occur subsequently ( Figure 1 not shown in []. That is, they can follow each other in time. This can allow for sending only one preamble for those subsequent service bearer entities. For example, an upstream burst can include a single preamble for one or more cyclic transmission opportunities associated with different service bearer entities within the same optical network unit (ONU). Obviously, Figure 1 the transmission opportunities 133, 142, 143, 144, 154, 155 shown in [] are associated with a single service bearer entity in the corresponding ONUs 130, 140, 150.

[0048] Based on the bandwidth assigned to service bearer entities 131, 132, 141, 151, 152, 153 by dynamic bandwidth assignment (DBA) (sometimes also referred to as dynamic bandwidth allocation), transmission opportunities 133, 142, 143, 144, 154, 155 are allocated within the upstream transmission window 170. For this purpose, the OLT 110 can include a DBA engine 111 or a DBA functional module that executes the DBA algorithm. The DBA engine 111 can be configured to determine or estimate the bandwidth requirements of service bearer entities 131, 132, 141, 151, 152, 153 by collecting in-band status reports, by monitoring upstream frames, or a combination thereof. The DBA algorithm can implement a DBA model that defines how to determine the assigned bandwidth for service bearer entities 131, 132, 141, 151, 152, 153, such as the reference DBA model according to the ITU-T G.9807.1 standard. Based on the determined bandwidth, the DBA engine 111 also determines the size and timing of the upstream transmission opportunities 133, 142, 143, 144, 154, 155 such that the determined bandwidth allocation is for service bearer entities 131, 132, 141, 151, 152, 153. Then, the transmission opportunities are transmitted to the ONUs in-band within the bandwidth map.

[0049] The OLT 110 also includes a receiver 112, sometimes also referred to as a burst-mode receiver, which is configured to receive and decode the upstream optical burst signals transmitted by the connected ONUs 130, 140, 150. For this purpose, the receiver 112 typically includes burst-mode receiver components. The burst-mode receiver components can be analog circuits and / or digital circuits. The burst-mode receiver components are typically interconnected to form a sequence or pipeline such that each burst-mode receiver component contributes to decoding the upstream optical burst signal. Thus, the receiver 112 generates a decoded output signal, i.e., a digital signal.

[0050] Typical DBA algorithms assign the entire available upstream PON capacity by allocating the entire upstream transmission window 170 to the corresponding service bearer entities 131, 132, 141, 151, 152, 153, even if the upstream services in PON 100 do not require the entire upstream PON capacity. In other words, even if the sum of the bandwidth requirements of the service bearer entities 131, 132, 141, 151, 152, 153 within PON 100 is less than the effectively available upstream PON capacity, the entire upstream transmission window 170 is divided into upstream transmission opportunities 133, 142, 143, 144, 154, 155.

[0051] The problem with this is that when the upstream services in PON 100 are low, for example, when a limited number of data packets 160 come from the connected services 171 - 176, a large part of the allocated transmission opportunities 133, 142, 143, 144, 154, 155 includes idle data or idle XGEM frames. This results in excessive power consumption within PON 100 because the OLT receiver 112 consumes useless power to decode the idle data or idle XGEM frames. For next-generation PONs, this useless power consumption can be particularly high because next-generation PONs include receiver functions with relatively high power consumption due to relatively high computational complexity, such as low-density parity-check (LDPC) error decoding. Such a next-generation PON can be, for example, 25GS-PON, 50G-PON according to the ITU-T G.9804 standard, or a PON with optical transmission above 50 Gb / s according to the ITU-T G.suppl.V HSP work plan.

[0052] Other typical DBA algorithms assign bandwidth equal to their estimated bandwidth requirements to the service bearer entities 131, 132, 141, 151, 152, 153. For example, this estimated bandwidth requirement can be determined based on in-band status reports or by monitoring the upstream frames. When doing so, when the upstream service demand is low, a portion of the upstream transmission window 170 may have time periods without upstream transmission opportunities ( Figure 1 not shown in the figure). This can allow some OLT functions to be turned off during these time periods, thereby reducing power consumption. However, the estimated bandwidth requirements are usually inaccurate because the bandwidth estimation algorithm may not be optimal, the in-band status reports (such as DBRu) are inaccurate, the monitoring of the upstream frames or data is inaccurate, some components in the PON limit the achievable accuracy, or the PON standards hinder the implementation of more accurate bandwidth estimation. There is a problem that the bandwidth requirement may be underestimated, resulting in insufficiently assigned bandwidth. There is also a problem that the quality of service may degrade due to queue filling accumulation, increased latency, jitter, increased packet loss probability, and / or reduced TCP throughput.

[0053] Therefore, it is preferable to dynamically assign bandwidth to the service bearer entities 131, 132, 141, 151, 152, 153 within the PON 100, so that the PON can operate more energy-efficiently without negatively affecting the quality of service within the PON, such as the quality of service provided to the connection services 171 - 176.

[0054] Figure 2 Exemplary embodiments of transmission opportunities 201, 207, 204, 210 are shown, which are assigned to the service bearer entities 131, 132 through dynamic bandwidth assignment, thereby improving the energy-efficient operation of the PON without negatively affecting the quality of service within the PON. To this end, the OLT 110 in the PON 100 includes a module configured to dynamically assign bandwidth equal to the estimated bandwidth requirements 202, 208, 205, 211 of the corresponding service bearer entities 131, 132 to the service bearer entities 131, 132. The module is also configured to assign bandwidth margins 203, 209, 206, 212 to the corresponding service bearer entities 131, 132 in addition to the estimated bandwidth requirements 202, 208, 205, 211. Therefore, the bandwidth assigned to a service bearer entity (such as 131) is the sum of the estimated bandwidth requirements 202, 205 of the service bearer entity and the bandwidth margins 203, 206. In other words, the lengths of the upstream transmission opportunities 201, 204, 207, 210 assigned to the service bearer entities 131, 132 are defined not only by the estimated bandwidth requirements 202, 205, 208, 211 but also by the bandwidth margins 203, 206, 209, 212. Obviously, the corresponding service bearer entities in the PON may have different bandwidth margins assigned to them according to their estimated bandwidth requirements. Alternatively or additionally, some or all of the service bearer entities may have substantially the same bandwidth margin assigned to them.

[0055] The module of the OLT can be, for example, a DBA engine or a DBA functional module that executes a DBA algorithm, such as Figure 1 the DBA engine 111 in. The estimated bandwidth requirements can be determined, for example, based on reports of buffer occupancy requested by the OLT and provided by the ONU, such as the dynamic bandwidth report upstream (DBRu) according to the ITU-T G.9807.1 standard. Alternatively or additionally, the estimated bandwidth requirements can be determined by monitoring the upstream frames received by the OLT, such as by measuring the amount of payload data sent during one or more transmission opportunities and / or by comparing the number of idle XGEM frames with the bandwidth map.

[0056] The OLT also includes a receiver configured to receive and decode the upstream optical signal from the ONU within the PON, such as Figure 1The receiver 112 therein. The receiver is also configured to turn off one or more functions during periods when no uplink transmission opportunity is allocated (i.e., the unallocated periods 213, 214, 215 between the subsequently scheduled transmission opportunities 201, 207, 204, 210). During these unallocated periods 213, 214, 215, no uplink frames are sent from the service bearer entities 131, 132 to the OLT. Turning off one or more receiver functions during these unallocated periods 213, 214, 215 allows reducing the energy consumption of the OLT by preventing unnecessary power consumption for decoding idle data or idle XGEM frames.

[0057] One or more functions of the receiver that are turned off may include analog circuits and / or digital signal processing circuits, such as an analog-to-digital converter, a clock and data recovery device, an equalizer, and / or a decoder. Turning off one or more receiver functions means adjusting the operating state of at least a part of the OLT receiver from a first operating state for receiving and decoding an uplink optical signal to a second operating state that consumes less power than the first operating state. This second operating state may be, for example, in particular a standby mode, a sleep mode, a low-power mode, a disabled mode, a switched-off mode, or a mode in which one or more receiver functions are not performed. Turning off one or more functions of the receiver may be achieved by instructing the receiver, for example, by providing a control signal to the receiver at the start of the unallocated periods 213, 214, 215. Alternatively or additionally, turning off one or more functions of the receiver may be achieved without actively instructing the receiver. For example, when an input indicating the absence of an uplink optical signal is received during the unallocated periods 213, 214, 215, one or more functions of the receiver may be turned off.

[0058] Assigning bandwidth margins 203, 209, 206, 212 above the estimated bandwidths 202, 208, 205, 211 can ensure that sufficient bandwidth is assigned to the service bearer entities 131, 132 to avoid or limit the negative impact on the quality of service, because the estimated bandwidth requirements 202, 208, 205, 211 may underestimate the effective bandwidth requirements. Doing so can avoid or limit queue fill accumulation, increased latency, jitter, increased packet loss probability, and / or reduced TCP throughput. In other words, the bandwidth margins 203, 209, 206, 212 can provide a safety margin or buffer to cope with the estimation errors in the estimated bandwidth requirements 202, 208, 205, 211.

[0059] This is further illustrated by the header 201a, payload 201b, and idle data 201c sent during the allocated transmission opportunity 201. As Figure 2As shown, the estimated bandwidth requirement 202 of the service bearer entity 131 does not accurately correspond to the header 201a and the payload 201b transmitted during the transmission opportunity 201 (i.e., the effective or actual bandwidth requirement of the T-CONT 131). Therefore, simply assigning the estimated bandwidth requirement 202 will result in a degradation of the quality of service because a portion of the payload 201b will not be transmitted, leading to queue accumulation in the associated service bearer entity.

[0060] Therefore, in addition to the estimated bandwidth requirements 202, 208, 205, 211, bandwidth margins 203, 209, 206, 212 are also assigned to the corresponding service bearer entities 131, 132, and turning off one or more OLT receiver functions during the unallocated periods 213, 214, 215 can reduce the power consumption of the PON while avoiding a negative impact on the quality of service. The advantage is that this can be easily incorporated into existing PONs because it only requires limited software modifications and no changes to the existing hardware. Another advantage is that network operators can safely provide the energy-saving function of the present disclosure without risking a degradation of the quality of service.

[0061] It is obvious that Figure 2 An example embodiment is shown where the estimated bandwidths 202, 208, 205, 211 are less than the sum of the headers 201a, 204a, 207a, 210a and the payloads 201b, 204b, 207b, 210b; however, the estimated bandwidth of one or more service bearer entities can also be greater than the header and the payload.

[0062] One or more functions of the receiver that are turned off during the unallocated periods 213, 214, 215 can include low-density parity-check (LDPC) error decoding. Turning off LDPC error decoding during the unallocated periods 213, 214, 215 allows for a significant reduction in the power consumption of the OLT because LDPC error decoding typically consumes a large amount of power due to its relatively high computational complexity, for example, compared to Reed-Solomon error decoding. Doing so can further improve the energy efficiency of the PON. Turning off LDPC error decoding can be achieved by disabling or shutting down LDPC error decoding or not performing it. Alternatively, turning off LDPC error decoding can be achieved by not performing at least a portion of the LDPC error decoding instructions. Turning off LDPC error decoding can be achieved by instructing the LDPC decoder within the receiver, for example, by providing a control signal to the LDPC decoder at the start of the unallocated periods 213, 214, 215. Alternatively or additionally, turning off LDPC error decoding can be achieved without actively instructing the LDPC decoder, for example, by turning off upon receiving an input indicating the absence of an upstream optical signal during the unallocated periods 213, 214, 215.

[0063] An ONU typically includes a transmitter that is configured to encode an upstream optical signal and send it to an OLT. The OLT's module can also be configured to instruct one or more ONUs to turn off one or more transmitter functions during unassigned time periods 213, 214, 215. This can be achieved by sending a control signal or message from the OLT to the ONU, such as a Physical Layer Operations, Administration, and Maintenance (PLOAM) message or an ONU Management and Control Interface (oMCI) message. The one or more transmitter functions instructed to be turned off can be, for example, a laser or a serializer / deserializer (SERDES) link of one or more ONUs. Turning off these transmitter functions, in addition to turning off one or more receiver functions of the OLT, also allows for further improvement in the energy efficiency of the PON by reducing unnecessary power consumption of the ONUs.

[0064] The transmitter of the ONU can also be configured to turn off one or more functions during the upstream transmission opportunities assigned to other ONUs. In other words, one or more transmitter functions of the ONU are only activated during the upstream transmission opportunities assigned to that ONU to encode and send the upstream optical signal.

[0065] Figure 3A Example 300 shows the power consumption difference ΔP of the OLT resulting from dynamically allocating bandwidth as the sum of the estimated bandwidth demands of the service-bearing entities and the determined bandwidth margins 304, 305, 306 according to an example embodiment. The power consumption difference ΔP indicates the power consumption difference between allocating the entire available upstream PON capacity (shown by curve 303) by allocating the entire upstream transmission window and allocating the minimum upstream PON capacity (shown by the minimum value on the vertical axis at the origin) by allocating the minimum upstream transmission window. The minimum value can correspond to zero or another small value. As described above, when the total estimated bandwidth demand 302 is low relative to the maximum allocated bandwidth, this results in excessive power consumption savings 301. The power consumption difference may further illustrate that the power consumed by the OLT is independent of the allocated bandwidth or frame, i.e., the baseline power consumption. Therefore, the vertical axis 301 indicates the power consumption difference of the OLT receiver relative to the baseline power consumption, which is related to the allocated bandwidth. The horizontal axis 302 indicates the total estimated bandwidth demand at one PON port of the OLT in the PON.

[0066] One or more bandwidth margins can be predetermined values, independent of the estimated bandwidth demand. In other words, the bandwidth margins of one or more service-bearing entities can be fixed values over time, which are allocated in addition to the estimated bandwidth demands of these service-bearing entities. Figure 3A Curve 304 in [the figure] shows an example of the resulting power consumption difference.

[0067] Alternatively or in addition, one or more bandwidth margins may be dynamic values that vary over time according to the estimated bandwidth requirements. Thus, the modules of the OLT may be configured to determine the bandwidth margin based on the estimated bandwidth requirements of one or more service-bearing entities.

[0068] The bandwidth margin may be determined based on a proportional relationship with the estimated bandwidth requirements. For example, the bandwidth margin BW M (t) may be determined proportionally to the estimated bandwidth requirements, i.e., BW M (t) = A x Bw d (t), where A is a constant and BW d (t) is the estimated bandwidth requirement at time step t. In another example, the bandwidth margin BW M (t) may be determined as BW M (t) = B + A x Bw d (t), where B is a fixed value. By determining the bandwidth margin based on a proportional relationship with the estimated bandwidth requirements, the sum of the estimated bandwidth and the bandwidth margin (i.e., the assigned bandwidth) may be proportional to the total estimated bandwidth requirements in the PON. Figure 3A The curve 305 in shows an example of the resulting OLT power consumption difference. In this example, the bandwidth margin may increase linearly with the total estimated bandwidth requirements in the PON. Thus, a smaller bandwidth requirement results in a relatively smaller assigned bandwidth, i.e., with respect to the XGEM frame, the uplink bandwidth amount is limited. A larger bandwidth requirement results in a relatively larger assigned bandwidth, i.e., with respect to the XGEM frame, the uplink bandwidth amount is larger.

[0069] Alternatively, the bandwidth margin may be determined such that the bandwidth assigned to the corresponding service-bearing entity corresponds to a step function of the estimated bandwidth requirements. Figure 3A The curve 306 in shows an example of the resulting OLT power consumption difference. Thus, the assigned bandwidth may be determined as a piecewise constant function or a series of steps. To this end, the bandwidth requirement intervals, i.e., the steps of the curve 306, may correspond to different values of the assigned bandwidth.

[0070] Alternatively, the assigned bandwidth may be determined as a step function of the estimated bandwidth requirements with hysteresis. This is shown in Figure 3B In this example embodiment 310, the bandwidth margin may be determined such that the assigned bandwidth has a high value 313 or a low value 312, respectively. If the total estimated bandwidth requirement 316 exceeds the upper threshold Th up 314, e.g., 1.25 Gbps, then a high assigned bandwidth is assigned. The value of the assigned bandwidth remains high until the total estimated bandwidth requirement 316 drops below the lower threshold Th low 315, which has a lower value than the upper threshold Th upSmaller values, such as 0.625 Gbps. This hysteresis in the switch between the high assigned bandwidth 313 and the low assigned bandwidth 312 avoids unstable switching between the assigned bandwidths 312 and 313, thereby improving the stability of the assigned bandwidth. The switch between the high bandwidth margin value and the low bandwidth margin value, in addition to the thresholds Th low 314 and Th up 315, can also be conditional on a time threshold. For example, if the upper threshold Th up 314 is exceeded for a predetermined period of time, the assigned bandwidth can only switch to the high value, or if the total estimated bandwidth demand is below the lower threshold Th low 315 for a predetermined period of time, the assigned bandwidth can only switch to the low value. The high value 313 can enable the assignment of the entire uplink bandwidth capacity. Doing so can avoid having a negative impact on the quality of service.

[0071] Obviously, although Figure 3B shows an example of a single-step function with hysteresis, the bandwidth margin can also be determined such that the assigned bandwidth corresponds to a multi-step function with hysteresis, for example by adding hysteresis to 306. It is further obvious that the bandwidth margins of some traffic-bearing entities within the PON can be predetermined values, while the bandwidth margins of other traffic-bearing entities can be dynamic values determined based on their estimated bandwidth demands, i.e., based on a proportional relationship with the estimated bandwidth demand, as a stepwise function of the estimated bandwidth demand, as a single-step function of the estimated bandwidth demand with hysteresis, and / or according to another relationship of the estimated bandwidth demand.

[0072] Determining the bandwidth margin of a traffic-bearing entity can also be based on the service parameters of the traffic-bearing entity, which indicate a preference for quality of service or energy saving. Therefore, dedicated service parameters can be provided for the traffic-bearing entity, which indicate a preference for quality of service or energy saving. A preference or bias for quality of service may result in the assignment of a larger bandwidth margin, while a preference or bias for energy saving may result in the assignment of a smaller bandwidth margin. In addition to typical service parameters (e.g., fixed bandwidth, guaranteed bandwidth, and maximum bandwidth according to the ITU.TG.9807.1 standard), this dedicated service parameter can also be provided. The dedicated service parameter can indicate a qualitative preference or bias, such as "quality of service" or "energy saving". Alternatively or additionally, the dedicated service parameter can specify a quantitative preference or bias, such as maximum energy saving, minimum energy saving, threshold bandwidth estimation, minimum assigned bandwidth, maximum assigned bandwidth, or guaranteed latency. This dedicated service parameter allows for controlling the trade-off between quality of service impact and energy saving, for example, depending on the service or application connected to the traffic-bearing entity. Alternatively or additionally, the preference for quality of service or energy saving can be derived from a specific configuration of typical service parameters.

[0073] The OLT module can also be configured to optimize the determined bandwidth margin based on feedback from the service-bearing entity. The feedback can include in-band status reports indicating buffer occupancy or queue filling included in the upstream frame, such as DBRu according to the ITU.TG.9807.1 standard. Alternatively or additionally, feedback can be obtained by monitoring the upstream frames received by the OLT, whereby the buffer occupancy of the service-bearing entity can be inferred, for example, by measuring the amount of payload data transmitted during one or more transmission opportunities and / or by comparing the number of idle XGEM frames with the bandwidth map. For example, if the feedback indicates that the buffer occupancy or queue filling of the service-bearing entity is increasing or exceeding a threshold, the assigned bandwidth margin can be increased to avoid negatively impacting the quality of service at the cost of reduced energy savings. Conversely, if the feedback indicates that the buffer occupancy of the service-bearing entity is decreasing or below a threshold, the assigned bandwidth margin can be reduced to increase energy savings. As another example, a sufficient bandwidth margin can be initially assigned to avoid negatively impacting the quality of service. This sufficient bandwidth margin can be gradually reduced over time to increase energy savings until the buffer occupancy feedback indicates a non-negligible increase in queue filling, i.e., a degradation of the quality of service. Thus, this feedback allows for the establishment of a control loop that optimizes the bandwidth safety margin over time.

[0074] Figure 3C Further shown is the variation of the total dynamically assigned bandwidth 322 over time according to an example embodiment, where the total dynamically assigned bandwidth 322 is equal to the sum of the estimated bandwidth demand 323 and the bandwidth margin 324. Clearly, regardless of how the bandwidth margin is determined, or even when the bandwidth margin is fixed, the assigned bandwidth 322 varies over time with the upstream traffic load. In contrast, the assigned bandwidth 321 of a typical DBA algorithm that assigns the entire available upstream PON capacity by allocating the entire upstream transmission window is static over time.

[0075] Figure 4 Table 400 is shown, which includes the average queue filling 440 of the service-bearing entity and the power consumption 450 of a part of the OLT receiver when applying a typical DBA algorithm 410 that allocates the entire upstream transmission window, a DBA algorithm 420 that allocates transmission opportunities equal to the estimated bandwidth demand, and a DBA algorithm 430 (also referred to as green DBA) according to an example embodiment. These queue fillings 440 and power consumptions 450 are obtained by simulating a PON with ten T-CONTs; the traffic source corresponds to a typical Poisson distribution packet arrival interval with an exponentially distributed packet size; the average traffic load increases from 50 Mbps to 100 Mbps between T-CONT 1 and T-CONT 10; the effective available bandwidth capacity of the PON corresponds to 1 Gbps.

[0076] Table 400 shows that full bandwidth assignment 410 results in the lowest average queue fill, but consumes a large amount of power 450, i.e., 2W. In contrast, Table 400 shows that allocating transmission opportunities equal to the estimated bandwidth requirement 420 results in low power consumption 450, i.e., 1.47W, but significantly increases the average queue fill 440. Table 400 further shows that GreenDBA, which assigns a bandwidth margin in addition to the estimated bandwidth requirement, results in a significant power savings compared to full bandwidth allocation 410, i.e., a power consumption of 1.47W, without a significant increase in the average queue fill 440.

[0077] Figure 5FIG. 500 shows a suitable computing system capable of implementing an embodiment of a method for an OLT to dynamically assign bandwidth to service-bearing entities within an ONU. The computing system 500 can generally be formed as a suitable general-purpose computer and includes a bus 510, a processor 502, a local memory 504, one or more optional input interfaces 514, one or more optional output interfaces 516, a communication interface 512, a storage unit interface 506, and one or more storage units 508. The bus 510 can include one or more conductors that allow communication between components of the computing system 500. The processor 502 can include any type of conventional processor or microprocessor that interprets and executes programming instructions. The local memory 504 can include random access memory (RAM) or another type of dynamic storage device that stores information and instructions for execution by the processor 502, and / or read-only memory (ROM) or another type of static storage device that stores static information and instructions for use by the processor 502. The input interface 514 can include one or more conventional mechanisms that allow an operator or user to input information into the computing device 500, such as a keyboard 520, a mouse 530, a pen, a voice recognition and / or biometric mechanism, a camera, etc. The output interface 516 can include one or more conventional mechanisms that output information to an operator or user, such as a display 540, etc. The communication interface 512 can include any transceiver-like mechanism, such as one or more Ethernet interfaces, that enables the computing system 500 to communicate with other devices and / or systems, such as with an OLT 110 and one or more ONUs 130, 140, 150. The communication interface 512 of the computing system 500 can be connected to such source nodes or destination nodes via a local area network (LAN) or a wide area network (WAN) (such as the Internet). The storage unit interface 506 can include a storage interface (such as a Serial Advanced Technology Attachment (SATA) interface or a Small Computer System Interface (SCSI)) for connecting the bus 510 to one or more storage units 508 (such as one or more local disks, such as SATA disk drives) and controlling the reading and writing of data from and / or to these storage units 508. Although the above-described storage units 508 are described as local disks, any other suitable computer-readable medium can generally be used, such as a removable disk, an optical storage medium (such as a CD or DVD), a ROM, a disk, a solid-state drive, a flash card, etc. Thus, the computing system 500 can correspond to Figure 1 module 111 of the OLT 110 shown.

[0078] Although the present invention has been illustrated by specific embodiments, those skilled in the art will understand that the present invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention can be embodied in various variations and modifications without departing from its scope. Therefore, this embodiment should be considered illustrative in all respects rather than restrictive, and the scope of the present invention is indicated by the appended claims rather than the foregoing description. Accordingly, all variations falling within the scope of the claims are intended to be included therein.

[0079] Readers of this patent application should also understand that words such as "comprising" or "including" do not exclude other units or steps, words such as "a" or "an" do not exclude a plurality, and a single unit (such as a computer system, a processor, or another integrated unit) can implement the functions of multiple modules described in the claims. Any reference signs in the claims should not be construed as limiting the corresponding claims. Terms such as "first", "second", "third", "a", "b", "c", etc. used in the specification or claims are used to distinguish similar units or steps and do not necessarily describe an order or a time sequence. Similarly, terms such as "top", "bottom", "above", "below", etc. are used for descriptive purposes and do not necessarily indicate a relative position. It should be understood that, where appropriate, the terms so used may be interchanged and that the embodiments of the present invention can operate in other sequences or in directions different from those described or illustrated above according to the present invention.

[0080] As used in this application, the term "circuit" may refer to one or more or all of the following: (a) a pure hardware circuit implementation (e.g., an implementation only in analog and / or digital circuits) and (b) a combination of hardware circuits and software, such as (where applicable): (i) a combination of analog and / or digital hardware circuits and software / firmware, and (ii) any portion of a hardware processor with software (including a digital signal processor), the software, and the memory, which work together to cause a device such as a mobile phone or a server to perform various functions, and (c) a hardware circuit and / or a processor (e.g., a microprocessor or a portion of a microprocessor) that requires software (e.g., firmware) for operation, but the software may not be present when it is not needed for operation. This definition of circuit applies to all uses of the term in this application (including in any claims). As another example, as used in this application, the term circuit also covers an implementation of only a hardware circuit or one or more processors, or a portion of a hardware circuit or processing and its (or their) accompanying software and / or firmware.

Claims

1. An optical line terminal (110) OLT, configured to communicate with an optical network unit (130, 140, 150) ONU in a passive optical network (100) PON; wherein: The OLT comprises a module configured to dynamically assign bandwidth equal to the estimated bandwidth requirement of the corresponding service bearer entity to the service bearer entity (131, 132) within the ONU by allocating upstream transmission opportunities (201, 204, 207, 210); and wherein the module is further configured to perform, in addition to the estimated bandwidth requirement (202, 205, 208, 211), assignment of bandwidth margin (203, 206, 209, 212) to the corresponding service bearer entity; and wherein the OLT further comprises a receiver configured to receive an uplink optical signal from the service bearer entity during the allocated uplink transmission opportunity; and wherein the receiver is further configured to shut down one or more functions during an unallocated period (213, 214, 215) in which there is no uplink optical signal.

2. The optical line terminal OLT according to claim 1, wherein: The receiver is configured to decode an uplink optical signal according to low-density parity check (LDPC) error decoding; and wherein the LDPC error decoding is turned off during the unallocated period (213, 214, 215).

3. An optical line terminal OLT according to any one of the preceding claims, wherein: One or more bandwidth margins (203, 206, 209, 212) are predetermined values ​​(223) regardless of the estimated bandwidth requirement (226).

4. An optical line terminal OLT according to any one of the preceding claims, wherein: One or more bandwidth margins (203, 206, 209, 212) are dynamic values; and wherein the module is further configured to determine one or more bandwidth margins for one or more service bearer entities based on the estimated bandwidth requirements (226) of the one or more service bearer entities.

5. The optical line terminal OLT according to claim 4, wherein: Determining one or more bandwidth margins (203, 206, 209, 212) includes determining the one or more bandwidth margins based on a proportional relationship (222) to the estimated bandwidth requirement (226).

6. The optical line terminal OLT according to claim 4, wherein: Determining one or more bandwidth margins (203, 206, 209, 212) includes determining one or more bandwidth margins such that the bandwidth assigned to the corresponding one or more service bearer entities corresponds to a step-wise function (224) of the estimated bandwidth requirement (226).

7. The optical line terminal OLT according to claim 6, wherein: The step-wise function is characterized by a hysteresis.

8. The optical line terminal OLT according to any one of claims 4 to 7, wherein: Determining one or more bandwidth margins for one or more traffic bearer entities is further based on respective service parameters of the one or more traffic bearer entities, the service parameters indicating a preference for quality of service or energy conservation.

9. An optical line terminal OLT according to any one of the preceding claims, wherein: The module is further configured to optimize the bandwidth margin based on buffer occupancy of the corresponding service bearer entity.

10. The optical line terminal OLT according to any one of the preceding claims, wherein: The module is also configured to execute the function of instructing one or more ONUs (130, 140, 150) to shut down one or more transmitter functions for encoding upstream optical signals and / or transmitting upstream optical signals to the OLT (110).

11. An optical network unit ONU, configured to communicate with an optical line terminal OLT according to any one of claims 1.10 in a passive optical network PON; and wherein, The ONU comprises a transmitter configured to transmit an upstream optical signal to the OLT during an upstream transmission opportunity allocated to the ONU; and wherein the transmitter is further configured to shut down one or more functions during the unallocated period.

12. The optical network unit (ONU) according to claim 11, wherein: The transmitter is further configured to shut down one or more functions during upstream transmission opportunities allocated to other ONUs.

13. A computer-implemented method for dynamically assigning bandwidth to a service-bearing entity within an optical network unit (ONU) via an optical line terminal (OLT), wherein the optical line terminal (OLT) is configured to communicate with the ONU in a passive optical network (PON); wherein: The OLT comprises a receiver configured to receive an upstream optical signal from the ONU; The computer-implemented method comprises: dynamically assigning to the corresponding service bearer entity a bandwidth equal to the estimated bandwidth requirement and bandwidth margin of the corresponding service bearer entity by allocating uplink transmission opportunities to the service bearer entity; as well as One or more functions of the receiver are turned off during an unallocated time period when there is no upstream optical signal.

14. A data processing system configured to perform the computer-implemented method according to claim 13.

15. A computer program product comprising computer executable instructions which, when a computer executes said program, cause said computer to perform the computer-implemented method according to claim 13.

Images