A method and network for optimizing a connection from a end user device

The present invention relates to a method of operating a network in which at least one end user device is connected to an access network.

The present invention also relates to a network in which at least one end-user device is connected to an access network, preferably performing the above method.

Access networks such as today's broadband or x DSL (x Digital Subscriber Line) networks and the devices connected to them go to the network and the Internet from the perspective of applications running on end-user devices such as laptop computers and PCs. It has very special properties that can significantly affect the access performance of the network. End-user devices or systems today have limited means to address this issue. This is because no information is available to help adjust the network stack parameters accordingly. The reason is that they are not directly connected to the access network, but usually to a home gateway or modem. Home gateways and modems connect to the access network on the one hand and the home network on the other (often a wireless LAN (Local Area)). Network), 802.11). Somewhat similar problems can be found in today's access networks. In the access network, the DSL modem negotiates a rate with the MSAN / DSLAM (Multi Service Access Node / Digital Subscriber Line Access Multiplexer). BRAS (Broadband Remote Access Server) requires this lower layer information for QoS (Quality of Service) provisioning and billing. To deal with this problem, a protocol called ANCP (Access Node Control Protocol) has been devised (ANCP, See http://tools.ietf.org/wg/ancp/). Other protocols in the access network space, such as TR-69 (see Non-Patent Document 1), are far from playing this role. This is because they are used to operate and manage home gateways, which are still (at least) one IP hop away from end-user equipment with no access network visibility.

The problem described here is, for example, the problem of resource consumption of background applications such as P2P (Peer-to-Peer) and interactive applications such as VoIP (Voice over Internet Protocol), web browsing, and games. .. Of particular concern is that background traffic should always be compromised to interactive traffic. That is, when an interactive application needs resources, background traffic should stop or slow down resource consumption (eg, traffic transmission). In the IETF (Internet Engineering Task Force), the LEDBAT (Low Extra Delay Background Traffic) working group deals with such problems (see Non-Patent Document 2). However, the IETF's approach is a so-called end-to-end approach and does not rely on the information available to the access network and its internal or attached devices.

The congestion control algorithm proposed in Non-Patent Document 3 attempts to minimize the extra delay that a background application such as P2P adds to an interactive application such as VoIP in LEDBAT. This algorithm does this by estimating the delay to introduce. Today, this delay usually comes from the modem buffer in the home gateway. Furthermore, the basic technology of the access network is important. For example, as is also quite common today, when ATM (Asynchronous Transfer Mode) is used, the algorithm may be optimized to send a segment that just fills the ATM cell. Assuming a narrow upstream bandwidth (less than about 500kbps), this can help reduce the probability of unwanted headblocking. All of this is referred to as lower layer information of the access network itself (eg technology and bandwidth and cell size) and its internal devices or devices attached to it (eg buffer size).

Patent Document 1 (System and method for automatic detection and configuration of network parameters) deals with a problem different from that of the present invention. This Patent Document 1 deals with changing networks and the parameter resetting necessary for them, and discloses a case where one has a static IP and the other has a dynamic IP in a home network and a corporate network. The user must manually switch from one to the other.

In PON (Passive Optical Network), upstream transmitter time slots on shared PON are assigned by a very advanced algorithm that considers, for example, frame level (Layer 2) measurements (Non-Patent Document 4 and its references and non-patent documents). See Patent Document 5 and its references). According to this method, the access network can notify the endpoint that does not know the access network of its specific current parameter settings. This allows the IP endpoint to accurately adapt its transmission behavior to the notified boundary conditions. For example, to prevent the packet burst from dynamically increasing or decreasing, send a packet burst that just fits in the time slot.

<p><patcit num="1"><text>U.S. Pat. No. 7051087</text></patcit></p>

<p><nplcit num="1"><text>http://www.broadband-forum.org/technical/download/TR-069Amendment2.pdfl</text></nplcit><nplcit num="2"><text>IETF LEDBAT working group, http://www.ietf.org/html.charters/ledbat-charter.html</text></nplcit><nplcit num="3"><text>S. Shalunov, "Low Extra Delay Background Transport (LEDBAT)", http://tools.ietf.org/html/draft-shalunov-ledbat-congestion</text></nplcit><nplcit num="4"><text>Cauvin et al., Common Technical Specification of the G-PON System among Major Worldwide Access Carriers, IEEE ComMag October 2006</text></nplcit><nplcit num="5"><text>Assi et al., "Toward Quality of Service Protection in Ethernet Passive Optical Networks: Challenges and Solutions" in IEEE ComMag Sept / Oct 2007</text></nplcit></p>

<p> Therefore, an object of the present invention is to improve and further develop the operation method of the network and the corresponding network so as to enable high resource consumption in the network.</p>

<p> According to the present invention, the above object is achieved by the method according to claim 1. As described in claim 1, the method comprises at least one end-user device or an application on the end-user device attached to at least one network element of the access network and / or the access network. It is characterized in that the network element is inquired about the lower layer parameter information, and the connection and / or protocol parameters are optimized after receiving the inquired lower layer parameter information.</p><p> In addition, the above object is achieved by the network according to claim 22. As described in claim 22, the network is a lower layer with respect to at least one network element of the access network and / or at least one network element attached to the access network by the at least one end user device. It is characterized by providing means or an application on the end-user apparatus that inquires about parameter information and optimizes connection and / or protocol parameters after receiving the inquired lower layer parameter information.</p><p> As recognized by the present invention, the achievement of the above object is made possible in a very simple manner by a query procedure initiated from at least one end-user device or from an application on the end-user device. Specifically, the at least one end-user device or an application on the end-user device is a lower layer with respect to at least one network element of the access network and / or at least one network element attached to the access network. Inquire about parameter information. The lower layer parameters are preferably lower layer parameters in the OSI (Open Systems Interconnection) model. After receiving the lower layer parameter information inquiry, the end user instrumentation location or application, it is possible to optimize the connections and / or protocol parameters.</p><p> In this way, a simple inquiry procedure from the end user device enables a high level of resource consumption in the network.</p><p> In a specific embodiment, the lower layer parameter information may refer to the network element of the query. This allows the end-user device or application to directly acquire lower layer parameter information regarding the network element to be queried.</p><p> Alternatively, or in addition, the lower layer parameter information may refer to at least one other network element. This allows the inquiring end-user device or application to obtain information about at least one other network element rather than with respect to the inquired network element.</p><p> In a more specific embodiment, the end-user device or application may receive the queried lower layer parameter information via the queried network element and / or another network element. Thereby, in the simplest embodiment, the end-user device or the application on the end-user device can obtain the requested information directly with respect to the network element of the query by the network element of the query.</p><p> In a more preferred embodiment, a given network element or application on that network element is subordinate to at least one other network element in the access network and / or at least one other network element attached to the access network. You may inquire about layer parameter information. In this case, a given network element or an application on that network element may collect lower layer parameter information from the end user device or another network element that provides this information to the application on the end user device. Preferably, after the query by the end user device or the application on the end user device, the predetermined network element or the application on the network element queries the end user device or the application on the end user device for lower layer parameter information. May be sent. This also allows the method proposed by the present invention to chain requirements.</p><p> Preferably, the network element or application on the network element may adapt the received information according to other collected data and / or other available settings. That is, a network element or an application on that network element may calculate a "trade-off" from different information collected from segments within the network along their respective network paths.</p><p> As a very reliable and simple inquiry / response procedure, the end-user device or the network element of the inquiry source and the network element of the inquiry destination may each have a communication manager that implements the inquiry / response mechanism. Such modules or features ensure reliable implementation of the methods proposed by the present invention.</p><p> In a more preferred embodiment, the communication manager in the contacted network element may combine lower layer parameter information with subscriber data and / or parameters. Such subscriber data and / or parameters may preferably be based on a given policy. This can provide a very sophisticated and personalized method.</p><p> In another preferred embodiment, the communication manager in the contacted network element calculates the aggregated trade-offs by combining information collected or received from other network elements in the network path and queries the network element. You may reply to it.</p><p> Preferably, the contacted network element may have a protective function to avoid overloading the network element in the event of an attack, for example. Further, for security reasons, an appropriate means such as a digital signature may be used. The actual means may depend on the operator policy and the network attachment procedure used.</p><p> Further, in order to provide a very reliable network operation method, the communication manager in the end user equipment or the network element of the inquiry source may store the received lower layer parameter information in the local storage, preferably. This allows the stored information to be used by the end-user device as long as the information is stored in the storage.</p><p> In a more preferred embodiment, the communication manager may provide an update function for changing lower layer parameters. This allows the communication manager to exchange new information in the form of updates when the parameters change. Depending on the individual circumstances and network, the update function may be initiated by the push method, ie, the network element of the query, or by the pull method, ie, the end-user device or the network element of the query source. ..</p><p> For very reliable operation of the network, the network element to be queried may include a compliance detector. Such a compliance detector can monitor whether lower layer information is being used correctly by the end user equipment. In a more preferred and very convenient way, the compliance detector triggers the communication manager to notify the end-user device if the lower layer parameter information is not used correctly or in a given way. May be good.</p><p> More preferably, the end-user equipment may include adaptive modules that optimize connection and / or protocol parameters and / or provide lower layer parameter information to at least one given application or protocol. This allows the application or upper layer protocol to adapt its behavior according to this information. That is, such an adaptive module may configure the network stack with the received information, or may be queried by the application and protocol, for example for the application and protocol to use that information.</p><p> In a more preferred embodiment, the end user device may specify or select the type of lower layer parameter information to be provided to the end user device. This allows only the specified or selected type of lower layer parameter information to be transmitted to the end user device.</p><p> In a specific embodiment, the lower layer parameter information indicates the available bandwidth and / or buffer size and / or preferably the maximum frame size in the upstream and / or downstream directions, and / or whether the frame size is static or not. May include. Other lower layer parameter information may be provided upon request by the end user device.</p><p> Optimization of connection and / or protocol parameters may include selecting or parameterizing the protocol or IP stack of the calling end-user device. However, other connection and / or protocol parameters can be optimized.</p><p> In a specific embodiment, the network element may be a home gateway, an MSAN (Multi Service Access Node) or a DSLAM (Digital Subscriber Line Access Multiplexer). The access network is xDSL (x Digital Subscriber Line) network, xPON (x Passive Optical Network), WiFi network, WiMAX (Worldwide Interoperability for Microwave Access) network or LTE (Long Term Evolution) network or such network or wireless. It may be a further extension of the network. However, other network types can also be access networks in the sense of the method proposed by the present invention.</p><p> Conventionally, the end user device is TCP (Transmission Control). Like the Protocol congestion control mechanism, control decisions are based on an end-to-end perspective on the network, but many future applications and transport services will have knowledge of lower layer parameters of the access network (eg, provisioning). It will benefit greatly from some information (eg, queue length) about the bandwidth being used) and the parameters of the devices present in the access network or attached to the access network. So far, these parameters are usually not available to the end device. This is because the end device is no longer directly connected to the access network (for example, broadband, DSL, as well as wireless access networks like WiMAX), and today usually home gateways are directly connected. This is because they are connected and the end-user equipment is connected to, for example, a local area network based on 802.11. However, both device and network properties can significantly affect the performance characteristics of end-user device access to the network or the Internet. If this information is available, the end system can positively influence these characteristics, for example by parameterizing the local IP stack or protocol accordingly.</p><p> According to the methods and networks according to the invention, the end-user device or end host or the application on or on the end-user device knows the above parameters. According to the present invention, end-user equipment uses this information to adjust network stack parameters to improve network efficiency in light of changing network and application requirements.</p><p> The present invention provides a request / response mechanism for inquiring lower layer information of an access network and a request / response mechanism for inquiring network element information of a network element in the access network or a network element attached to the access network. It also provides a way to change the request / response and a mechanism to handle changing access network and network element information. Also provided are modules that configure the IP protocol stack to accept lower layer information and / or ensure compliance with end-host or end-user device parameter settings. Finally, the mechanism for discovering the response module is disclosed.</p><p> Knowing these parameters can help optimize higher layer stack parameters (eg in congestion control) and can even determine protocol choices. For example, if you find that the modem buffer is very small, you can use regular TCP instead of the new background transport protocol.</p><p> The present invention is not limited to the xDSL environment. This approach is useful for other access / transport technologies (eg x-PON, WiMAX, WiFi) and especially for combinations thereof (eg GPON-fed WiMAX).</p><p> That is, the present invention provides the following embodiments. 1) End-user equipment that does not directly attach information about the characteristics of the access network and its internal equipment to the access network with respect to lower layers such as Layer 1 and Layer 2 (which may be located in an untrusted domain). How to optimize IP stack parameter settings for these characteristics by notifying (high). 2) A device that adapts the network protocol stack in such an end-user device to parameters learned from the access network. 3) A method of dynamically updating changed lower layer parameters and updating at related events in the access network. 4) A means of controlling transmitted information and extending it with parameters provided by the network operator, based on the specific policies imposed. 5) A method of chaining requests using the above method. 6) A means of detecting the compliant behavior of an IP endpoint or end-user device (eg, by a module within an access node or contacted network element).</p><p> There are a number of possibilities for implementing the present invention in a preferred manner. To this end, refer to claims that are subordinate to claim 1 on the one hand and the following description of the preferred embodiments of the invention illustrated in the drawings on the other hand. When a preferred embodiment of the present invention is described with reference to the drawings, a general preferred embodiment according to the teaching of the present invention and a modification thereof will be described.</p>

<figref num="1">It is a schematic diagram which shows the example of a typical access network in the form of a DSL configuration.</figref><figref num="2">It is a figure which shows the exemplary structure of this invention.</figref><figref num="3">It is a figure which shows the network element by the preferable embodiment of this invention.</figref>

In the following, general problems are illustrated using examples of DSL access networks (see Fig. 1), home gateways, and P2P applications, which are popular applications. The access network consists of BRAS, MSAN / access node and home gateway attached to it. For the purposes of this specification, only the case where a single end system is attached to the gateway without loss of generality is illustrated.

To solve the above problems, the end-user equipment is in the lower layer for network elements in the access network, for example in the form of nodes, or network elements attached to the access network, in order to select or parameterize the protocol. A method is provided that allows you to inquire about the parameters. In this method, the end-user device queries a network element that is part of the access network or a network element attached to the access network, such as a home gateway or MSAN. Upon receiving the query, the network element receives the negotiated available bandwidth (both upstream and downstream), buffer size or (maximum) frame size, and whether the frame size is static (eg ATM) or not (eg ATM). Respond to lower layer device settings such as Ethernet) instructions. The end system can also specify whether it is only interested in some information.

The example shown here relates to a relatively slow xDSL using ATM, but the scope of the present invention (providing parameters related to it for an IP endpoint indirectly attached to an access network) is far. It is widespread and can be applied to access networks such as WiMAX and PON (Passive Optical Network).

FIG. 2 shows the network of the present invention. In the present invention, a network attachment procedure that can be used to notify the IP address of a connectable node is assumed. This may be one of the established discovery mechanisms. For home gateways, this may simply be the default gateway address, but DHCP (Dynamic Host Configuration Protocol), PPP (Point-to-Point) Other means such as Protocol), established anycast address, etc. can also be used. The end-user device queries one or all of these devices. In the figure, this may be the case (1) (communicate with the home gateway) or the case (3) (the layer 3 MSAN receives the request or a combination thereof). The method proposed by the present invention can also be a chain of requirements. For example, the home gateway may establish a PPP session to the operator, and the operator may have the IP address of the node to which the lower layer parameter is requested. The home gateway requests such parameters and stores the parameters locally according to (2) in the figure. When the end host queries the parameters, the home gateway can respond for all the parameters it knows, including the one collected in step (2).

FIG. 3 illustrates the network elements and required modules of the method proposed by the present invention. Both the end device and the contact access node (network element) include a communication manager that implements the request / response mechanism. On the access node side, the communication manager collects lower layer information from the system and / or local storage. The communications manager may also combine lower layer information data with subscriber data provided by the policy manager, for example if rates are not allowed by contract or if user-specific settings are applied. It is important that the access node has a protective function to prevent the node from being overloaded in the event of an attack, for example. For security reasons, an appropriate means such as a digital signature may be used. The actual means depend on the operator policy and the network attachment procedure used. On the end device side, the communication manager saves the data from the response in local storage. Moreover, the communication manager may carry out the update function. That is, when the parameters change, the communication manager exchanges new information in the form of updates. This mechanism may be a push method initiated by an access node or a pull method initiated by an end device.

The adaptive module on the end device uses this information to configure the network stack. Alternatively, the adaptive module can be queried by the application and protocol, for example for the application and protocol to use that information. Finally, the access node may include a compliance detector. That is, the access node may perform a function of checking whether or not the lower layer information is used correctly, if possible. If the compliance detector observes that this is not the case, it may trigger the communication manager to notify the end device accordingly.

The present invention provides a deterministic way to parameterize protocols and inform advanced applications and Layer 7 protocols without relying on inaccurate measurements or guesswork. It also provides an easy technical solution to improve the efficiency of the access network from the customer's perspective. This solution can include information that is not only based on network parameters, but also implicitly based on operator policy (eg, not allowing too many time slots). Embodiments can be based on current network attachment procedures for implementing discovery and security.

Embodiments can replace inaccurate and indirect measurements of parameters known to the network. Better utilization of available resources and better response times can improve the user experience of the operator's customers. It can also provide very good QoS compared to known methods. The user can also perform the QoS processing of the application by himself / herself. Indirect endpoint traffic source control is possible according to the operator policy.

Based on the above description and description of the accompanying drawings, one of ordinary skill in the art will be able to conceive of many variations and other embodiments of the present invention. Therefore, the present invention is not limited to the disclosed specific embodiments, and variations and other embodiments should be understood to be included in the appended claims. Although specific terms are used herein, they are used only in a generic and descriptive sense and are not intended to be limiting.