Dynamic multi-access wireless network virtualization

DESCRIPTION

Virtualization of wireless network with multiple dynamic access

Countryside

The present application refers to wireless broadband communication systems, cellular systems, small cells and network virtualization.

Background

The concept of virtualization can be defined in general terms as a methodology by which an underlying resource is shared among multiple consumers, while providing each of the consumers with the illusion of owning the entire resource independent of the other consumers. Virtualization can be applied through wired and wireless networks. In the wireless field, virtualization has been carried out in IP networks and in wireless communication networks. Even within these two categories, there are several types of virtualization. For example, for IP networks, U.S. Patent Publication No. 2013/0094486 entitled "Wireless Network Virtualization for Wireless Local Area Networks" describes an access point for airtime guarantees to a group of customers who share a point of access.

In another IP network patent application, WO 2011/144538 A1 entitled "Method and System for Network Virtualization", the inventors analyze sharing a common network infrastructure by dividing it into several logical network instances, called "segments", composed of "slices" node of virtual nodes and virtual links. The request describes how to isolate the wireless resources that coexist at the same time to ensure minimal interference between the resources, as well as control the use of wireless resources to ensure that one segment does not infringe the resources of other segments.

As for wireless communication networks, US patent application number 2009/0170472 entitled "Shared Network Infrastructure" describes a wireless communication network in which the "[v] ualualization can provide an abstraction layer that allows multiple virtual machines run in isolation from each other, in parallel on the same physical machine. "

Although these virtualization versions differ in the medium in which they are implemented, they share the common point that they are a "one to several" type of virtualization. By "one to several", it means that a common set of hardware is shared by more than one customer or subscriber. In a way, this type of arrangement is like multiple tenancy in a building: many people live in individual apartments within a single building infrastructure.

These previous applications also represent multiple tenure in the sense that people living within a given building are always assigned the same amount of living space. For example, a tenant in a multi-tenancy building can rent a one-bedroom apartment. If at any time you have six relatives who visit you from afar, you are still assigned only one one-bedroom apartment. Your greatest demand for living space is not currently in this model. As a result, their relatives will have to sleep in armchairs, the floor, provided they can find space because the nature of a multiple tenure model is static. It does not respond to changing environmental conditions, and there is no real-time intelligence associated with the allocation of resources.

To the extent that virtualization techniques of the prior art encompass intelligent decision making in real time, they typically report alarms experienced within the system. These alarms alert customers within the network that network performance may be compromised, but they do not make dynamic decisions about how to respond to the alarm. In terms of pooled resources, some virtualization techniques of the prior art include the idea of ​​clustered processing capabilities. But they do not include the idea of ​​pooled resources. In the case of a heterogeneous multi-RAT mesh network, there may be hundreds of pooled resources available for use depending on the network conditions.

Going back to the definition of virtualization, each of these "many" users believes that they have exclusive use and control of a set of hardware. Figure 1 is an example of how a virtualized base station 100 of the prior art can operate in accordance with this definition of virtualization. The base station 100 is composed of a single set of hardware components. Additionally, the prior art performs the virtualization of physical form, in the hardware. In that unique set of hardware components, virtualization techniques are used so that AT&T and Verizon could share without knowing the hardware components of this base station, each of which believes it was the only network operator using the base station 100.

While this type of virtualization has the advantage of being able to use common hardware efficiently, it is a one-way street in terms of the perspective from which virtualization is seen. Specifically, each service provider considers that the base station 100 is dedicated exclusively to it. In this regard, the base station 100 is virtualized when it goes from the central network to the base station. In the opposite direction, however, from the base station 100 to the central network, there is no virtualization. And that is, in some aspects, a function of the static nature of virtualization of the prior art. The resources in the base station 100 are statically divided into a form of multiple tenure, which means that any change in the network, for example, the capacity, the operating conditions, the latency experienced by a network operator, cannot be addressed dynamically Therefore, it is necessary to create a dynamic wireless communication network.

This need is felt in many different types of wireless communication networks. Historically, wireless communication has been done on 3G or Wi-Fi networks using macrocells and access points for local Wi-Fi data delivery. Looking ahead, the selection of networks available for wireless communication is increasing to include LTE, White Space TV, small cell solutions integrated in macro-networks, etc. The base stations that support this heterogeneous network, which is a network that integrates multiple radio technologies, will require more sophisticated management techniques to handle the increasing demands of the networks.

By focusing, for example, on LTE networks, LTE is designed to support only switched packet services, in contrast to the circuit switching model of previous cellular systems. One of the objectives of LTE is to provide an Internet protocol (IP) connectivity without interruptions between the user equipment (UE) and the data packet network (PDN), without interrupting the applications of the end users during mobility. See, generally, "The LTE Network Architecture: A comprehensive tutorial", Strategic White Paper, Alcatel-Lucent.

While the term "LTE" encompasses the evolution of radio access of the universal mobile telecommunications system (UMTS) through the evolved UTRAN (E-UTRAN), it is accompanied by an evolution of non-radio related aspects under the term "system architecture evolution" (SAE), which includes the evolved packet core network (EPC). Together, LTE and SAE comprise the evolved package system (EPS).

The EPS uses the concept of EPS bearers to route IP traffic from a gateway in the PDN to the UE. A bearer is an IP packet stream with a defined quality of service (QoS) between the gateway and the UE. Together, E-UTRAN and EPC configure and release carriers as required by applications. An EPS carrier is typically associated with a QoS. Multiple bearers can be established for a user in order to provide different QoS flows or connectivity to different PDNs. For example, a user can participate in a voice call (VoIP), while at the same time performing a web browsing or downloading fTp. A VoIP carrier would provide the necessary QoS for the voice call, while a better effort carrier would be suitable for web browsing or an FTP session.

Figure 2 shows the general network architecture, including network elements and standardized interfaces. The core network (called EPC in SAE) is responsible for the overall control of the UE and for the establishment of carriers. The main logical nodes of the EPC are: (1) PDN Gateway (P-GW); (2) Service gateway (S-GW); and (3) Mobility Management Entity (MME). Currently, wireless base stations, such as eNodoB LTE and its return network infrastructure are managed on a component basis.

As an example, if an eNodeB used three different microwave return links, each would be dedicated to, and managed by, a different provider. As a result, there would be three different sets of operator policies and network policies that should be translated into three unique configurations unique to the individual pieces of return equipment. Similarly, on the access side, there are four main network operators in the US: AT&T, Verizon, Sprint, and T-Mobile. Each of these network operators implements its own patented architecture similar to that shown in Figure 2, but importantly, it lacks interoperability between carriers. The term "eNodeB" is used in the art to represent a standard, as opposed to a non-customized LTE base station. This term is used to have that meaning, which is different from our custom multiRAT nodes. Our multi-RAT nodes can function as standardized LTE base stations, but they also have much broader functionality.

The packet core signaling volumes in the first large-scale LTE network deployments are significantly higher than in existing 2G / 3G core networks. This is partly due to the flattest and fully IP architecture of LTE, where the macro and the metrocell are connected directly to the MME, the dedicated control plane element in the EPC. Analysis of the field data of several large LTE network implementations found that an MME can experience a sustained signaling load of more than 500 800 messages per user equipment (UE) during normal peak peak hours and up to 1500 messages per user per hour in adverse conditions. See, generally, Managing LTE Core Network Signaling Traffic ", July 30, 2013, Alcatel Lucent, www2.alcatel-lucent.com/techzine/managing -lte-core-network-signaling -traffic.

The increase in core signaling can also be attributed to a general increase in the use of the network by subscribers to LTE. In some of the major US metropolitan markets Where LTE is available, maximum network usage is as high as 45 service requests per UE per hour at peak peak hours. As LTE grows in popularity, signaling in the EPC will continue to increase, which increases the congestion potential of the control plane and signaling storms if not managed properly. In addition, when small cells, which are becoming more ubiquitous, are added to the network, the EPC is requested to administer 100 to 1000 times more cells. Some of the functions that the EPC has to administer for each small cell include: (1) providing return links; (2) dynamic configuration; (3) power level management; (4) assignment of physical cell ID; (5) signal management; and (6) greater transfers within the network due to the lower transmission range of small cells. Therefore, MNOs must implement a next-generation MME platform / Next-generation GPRS Service Support Nodes (SGSN) that not only has the capacity, scalability and CPU processing performance, but also the ability to manage in a way Smart this traffic to reduce global core signaling.

Two examples of where signaling efficiencies can be obtained through the use of virtualized networks are: (1) paging / tracking management procedures; and (2) transfer management. When you first return to paging / tracking, a UE enters the INACTIVE mode when its radio connection is released. When the UE is in INACTIVE mode and must be reached by the network, for example, if it has an incoming call, the LTE standards, as well as the inherited standards, define a PAGINATION process to reach the UE. Under these paging / tracking scenarios, a PAGE is sent on a control channel to all base stations in the last known tracking area for the UE.

When the core network was mainly composed of macro cells, there could be 10 macros within a tracking area. In current networks, the number of base stations within a tracking area can typically be 100 or more. This is because service providers are increasing their networks by incorporating small cells that ultimately connect to the central network to keep pace with the growing number of wireless communications users and the demands they put on the network. Given the large number of macro cells and small cells existing in current networks, the cost of transmission to the EPC and the network can generally be very high if the EPC does not know exactly where the UE is at any given time. As small cells become more integrated into existing networks, this transmission cost will only increase. At any time within a particular tracking area, there are many UEs that are in INACTIVE mode. A subset of these will require PAGE at any given time. Therefore, it is advantageous to design a method and an apparatus for efficiently managing the network resources with respect to the PAGE.

Each eNodeB within the control of the EPC must coordinate with its EPC when making a transfer of one of the UEs within its sector. When small cells are part of the network, the number of transfers between these small cells increases compared to macro cells, because the coverage area for small cells is much smaller than for macros. In addition to the system assessment, each network operator has its own EPC. Therefore, it is desirable, from the point of view of network resources, to incorporate an aggregating node that can virtualize a part, or the entire network, so that, from the perspective of the central network, it seems that it is only giving service to an eNodeB when in reality, behind that eNodeB there is a large network of multi-RAT nodes. The embodiments described herein include an aggregator node, referred to as a "cloud computing."

WO 2012/044148 describes a network architecture for an intelligent cloud base station.

Aspects of the invention are described in the independent claims, and preferred features are mentioned in the dependent claims.

Summary of the Invention

The invention is set forth in the set of appended claims. The embodiments and / or examples of the following description that are not covered by the appended claims are not considered as part of the present invention. The embodiments of this invention include methods, systems, devices and architecture designed to provide wireless services by virtualizing various radio access technologies in a unique architecture controlled by an intelligent radio controller. One embodiment describes a paging strategy for a virtualized wireless communication network connected to a packet core network. A further embodiment describes a delivery method using network connection information.

In one embodiment, there is a virtualized wireless communication network comprising a heterogeneous multi-RAT mesh network coupled in communication to a computer cloud component, the computer cloud component further comprising a general purpose processor and an accelerator. In a further embodiment, there is a virtualized wireless communication network comprising a heterogeneous multi-RAT mesh network coupled in communication to a computer cloud component, the computer cloud component further comprising a general purpose processor and an accelerator, in which the cloud computing component acts as an S1-GW proxy and a WAG / MAG and an Iu proxy to facilitate the transfer of LTE to Wi-Fi. In an alternative embodiment, there is a virtualized wireless communication network comprising a heterogeneous multi-RAT mesh network coupled in communication to a computer cloud component, the computer cloud component further comprising a general purpose processor and an accelerator, in which the cloud computing component further comprises a memory for storing GTP-U data that can be used in an X2 handover. In an alternative embodiment, there is a virtualized wireless communication network comprising a heterogeneous multi-RAT mesh network coupled in communication to a computer cloud component, the computer cloud component further comprising a general purpose processor and an accelerator that It also includes an SDN controller.

In an alternative embodiment, there is a virtualized wireless communication network comprising a heterogeneous multi-RAT mesh network coupled in communication to a computer cloud component, the computer cloud component further comprising a general purpose processor and an accelerator, in which the cloud computing component further comprises a memory for storing executable code by computer which, when executed, causes the processor to update a network topology and resource adjustment based on a capacity requirement of the heterogeneous multi-RAT mesh network.

In an alternative embodiment, there is a virtualized wireless communication network comprising a heterogeneous multi-RAT mesh network coupled in communication to a computer cloud component, the computer cloud component further comprising a general purpose processor and an accelerator, in which the additional cloud computing component comprises a memory for storing a GPS location of a heterogeneous multi-RAT node, which is a part of the heterogeneous multi-RAT mesh network, the GPS location being analyzed by the computer cloud component to determine if the heterogeneous multi-RAT node must be connected to a second computer cloud component to receive a better quality of service or a lower data path latency. In still an alternative embodiment, there is a virtualized wireless communication network comprising a heterogeneous multi-RAT mesh network coupled in communication to a computer cloud component, the computer cloud component further comprising a general purpose processor and an accelerator further comprising a database coupled in communication to the computer cloud component in which the database stores a plurality of environmental conditions of the multi-mesh network. Heterogeneous RAT.

In an alternative embodiment, there is a virtualized wireless communication network comprising a heterogeneous multi-RAT mesh network coupled in communication to a computer cloud component, the computer cloud component further comprising a general purpose processor and an accelerator, in which the cloud computing component is coupled in communication to an upstream network device and the cloud computing component further comprises a memory for storing computer executable code which, when executed, causes the cloud computing component to perform a calculation associated with the allocation of resources within the heterogeneous multi-RAT mesh network to the upstream network device, which could be, to by way of example and without limitation, an MME, an SGW or a PGW.

In an alternative embodiment, there is a virtualized wireless communication network comprising a heterogeneous multi-RAT mesh network coupled in communication to a computer cloud component, the computer cloud component further comprising a general purpose processor and an accelerator, in which the cloud computing component is coupled in communication to an upstream network device and the cloud computing component is virtualized so that it appears on the upstream network device as a base station.

In still a further alternative embodiment, there is a virtualized wireless communication network comprising a heterogeneous multi-RAT mesh network coupled in communication to a computer cloud component, the computer cloud component further comprising a general purpose processor and a accelerator, in which the upstream network device is an MME, an SGW, or a macrocell. In a variation of this embodiment, and without limiting the breadth of this embodiment, the base station could be an eNodeB.

In an alternative embodiment, there is a virtualized wireless communication network comprising a heterogeneous multi-RAT mesh network coupled in communication to a computer cloud component, the computer cloud component further comprising a general purpose processor and an accelerator, in which the heterogeneous multi-RAT mesh network further comprises a non-meshed return connection to the computer cloud component. In a variation of this embodiment, there could be a wired return connection that is part of the network. In an alternative embodiment, the return connection that is part of the network could be a wireless return connection.

In an alternate embodiment, there is a computerized virtualization method comprising the steps of: (a) Grouping a first radio resource of a first multi-RAT node with a second radio resource of a second multi-RAT node wherein the first and second multi-RAT node are coupled in communication to form a mesh network; (b) Maintain a connection to an upstream network device; and (c) Manage the pooled resources so that the upstream network device interconnects with the pooled resources as a single base station.

In an alternate embodiment, there is a computerized virtualization method comprising the steps of: (a) Grouping a first radio resource of a first multi-RAT node with a second radio resource of a second multi-RAT node wherein the first and second multi-RAT node are coupled in communication to form a mesh network; (b) Maintain a connection to an upstream network device; and (c) Manage the pooled resources so that the upstream network device interconnects with the clustered resources as a single base station that further comprises storing anchor data in a memory device.

In an alternate embodiment, there is a computerized virtualization method comprising the steps of: (a) Grouping a first radio resource of a first multi-RAT node with a second radio resource of a second multi-RAT node , in which the first and second multi-RAT nodes are coupled in communication to form a mesh network; (b) Maintain a connection to an upstream network device; and (c) Manage the pooled resources so that the upstream network device interconnects with the pooled resources as a single base station, which further comprises choosing the first radio resource or the second radio resource of the pooled resources to lead to carry out a communication task based on an environmental condition. In a variation of this embodiment, there could be an alternative embodiment in which the transfer of a data transmission from a first radio resource to a second radio resource is based on an environmental condition.

In an alternate embodiment, there is a computerized virtualization method comprising the steps of: (a) Grouping a first radio resource of a first multi-RAT node with a second radio resource of a second multi-RAT node , in which the first and second multi-RAT nodes are coupled in communication to form a mesh network; (b) Maintain a connection to an upstream network device; and (c) Managing the pooled resources so that the upstream network device interconnects with the pooled resources as a single base station, which further comprises the transfer of a data transmission from the first radio resource to the second radio resource.

In an alternative embodiment, there is a virtualization method implemented by computer that It comprises the steps of: (a) Grouping a first radio resource of a first multi-RAT node with a second radio resource of a second multi-RAT node, in which the first and second multi-RAT nodes are coupled in communication to form a mesh network; (b) Maintain a connection to an upstream network device; and (c) Manage the pooled resources so that the upstream network device interconnects with the pooled resources as a single base station, in which the management of the pooled resources further comprises assigning a radio bearer by analyzing a quality of the service value, a radio resource utilization value, or a data priority metric.

In an alternate embodiment, there is a computerized virtualization method comprising the steps of: (a) Grouping a first radio resource of a first multi-RAT node with a second radio resource of a second multi-RAT node wherein the first and second multi-RAT node are coupled in communication to form a mesh network; (b) Maintain a connection to an upstream network device; and (c) Manage the pooled resources so that the upstream network device interconnects with the pooled resources as a single base station, which further comprises: (a) dynamically inspecting a data packet; (b) determine a data type of the data packet; and (c) choose the first or second radio resource according to the type of data.

In an alternate embodiment, there is a computerized virtualization method comprising the steps of: (a) Grouping a first radio resource of a first multi-RAT node with a second radio resource of a second multi-RAT node , in which the first and second multi-RAT nodes are coupled in communication to form a mesh network; (b) Maintain a connection to an upstream network device; and (c) Manage the pooled resources so that the upstream network device interconnects with the clustered resources as a single base station, which further comprises caching the content on a local content server based on a data traffic feature within the network

In an alternative embodiment, there is a method implemented by computer for the paging of a "UE" user equipment within a heterogeneous multi-RAT mesh network comprising the steps of: (a) receiving a location request for a UE from a first network; (b) use a network connection information for a second network to locate the UE; and (c) send a paging message to the UE based on the network connection information for the second network.

In an alternative embodiment, there is a method implemented by computer for the paging of a "UE" user equipment within a heterogeneous multi-RAT mesh network comprising the steps of: (a) receiving a location request for a UE from a first network; (b) use a network connection information for a second network to locate the UE; and (c) send a paging message to the UE based on the network connection information for the second network, in which the second network is a Wi-Fi network and the network connection information is network connection information Wifi.

Brief description of the drawings

Figure 1 is an illustration of a virtualized base station of the prior art.

Figure 2 is an illustration of the prior art LTE network architecture.

Figure 3 is an illustration of a heterogeneous mesh network on which embodiments of the present invention can be executed.

Figure 4 illustrates the internal architecture of a multi-RAT node.

Figure 5 is an illustration of an architecture of a computer cloud in embodiments of the present invention having virtualized 3G, LTE and Wi-Fi resources.

Figure 6 is an illustration of the bidirectional nature of the virtualization systems and methods described in this document.

Figure 7 illustrates the bi-directional nature of virtualization of the embodiments of this invention.

Figure 8 illustrates a virtualization method of a wireless communication network.

Definitions

The following definitions are included to complement additional descriptions within the specification. These terms should be interpreted as an expert in the art would understand them.

Accelerator is a hardware device designed to accelerate the process, for example, a web accelerator is a Proxy server that accelerates access to the website.

eNodeB is an abbreviation for a Node B E-UTRAN. An eNodeB is an element in the E-UTRA, which is the aerial interface of the Long Term Evolution "LTE" of 3GPP. An eNodeB interacts with the UE and hosts the physical (PHY), media access control (MAC), radio link control (RLC) and packet data control protocol (PDCP) layers. It also hosts the radio resource control (RRC) functionality corresponding to the control plane. It performs many functions, including radio resource management, admission control, programming, compliance with the negotiated UL QoS, cell information dissemination, encryption / decryption of user and control plane data, and compression / decompression of the packet headers of the DL / UL user plane.

GTP-U is used to transport user data within the central network of the general radio packet service "GPRS" and between the radio access network and the central network. The transported user data can be packets in any of the IPv4, IPv6 or PPP formats. The GTP-U protocol is used on the S1-U, X2, S4, s 5 and S8 interfaces of the evolved packet system (EPS). GTP-U tunnels are used to transport encapsulated T-PDUs and signaling messages between a given pair of GTP-U tunnel endpoints. The tunnel endpoint ID (TEID) that is present in the GTP header indicates to which tunnel a particular T-PDU belongs. The transport bearer is identified by the GTP-U TEID and the IP address (source TEID, destination TEID, source IP address, destination IP address).

The heterogeneous mesh network refers to at least two dynamic mesh nodes capable of: using different protocols, different duplexing schemes, or operating in disparate frequency bands, or using different transmission / reception means, such as wired versus wireless . Different protocols may include Wi-Fi, 2G, 3G, 4G, WCDmA, LTE, Advanced LTE, ZigBee, or Bluetooth. The different duplexing schemes may include the time division, code division and frequency division schemes. The disparate frequency bands may include so-called "blanks", VHF and UHF channels, cell phone bands, public safety bands, and the like.

Proxy lu - a proxy is a server, which is a computer system or an application that acts as an intermediary for requests from clients seeking resources from other servers. A client connects to the proxy server, requesting some service, such as a file, connection, web page or other available resource from a different server and the proxy server evaluates the request as a way to simplify and control its complexity. The Iu interface is an external interface that connects the RNC to the central network. An Iu proxy is, therefore, a proxy for the Iu interface.

MME is the acronym for mobility management entity. The MME is the key control node for the LTE access network. It is responsible for the monitoring and paging procedure, including retransmissions, and also for the inactive mode of the user equipment (UE). The MME is also involved in the activation of the bearer and its deactivation procedures, the task of choosing the SGW for an UE in the process of initial connection and when the infra-transfer that involves the relocation of the node is also carried out of the central network (CN). The MME is responsible for authenticating the user to the hSs, if the user is roaming, the MME terminates the interface to the user's local HSS. All non-access stratum (NAS) signaling ends at the MME point, which is also responsible for the generation and assignment of temporary UE identities (GUTI). Its functions also include the EU authorization to the public land mobile network (PLMN) and compliance with the EU roaming restrictions, if any. The MME is also the end point of encryption and integrity protection for NAS signaling. The legal interception (LI) of the signaling could also be supported by the MME entity. It also provides the control plane function for mobility between LTE and 2G / 3G networks through the S3 interface (from SGSN to MME), the functions of the MME include: NAS signaling; NAS signaling security; node signaling between CN for mobility between 3GPP access networks (termination S3); EU reach capability in the ECM-INACTIVE state (including control and execution of paging retransmission); management of the list of monitoring areas; mapping from the UE, location (for example, TAI) with respect to the time zone, and signaling a change in the time zone of the UE associated with mobility; GW PDN and GW service selection; MME selection for transfers with change of MME; SGSN selection for transfers to 2G or 3G 3GPP access networks; roaming (S6a towards the domestic HSS); authentication; authorization; including bearer management functions dedicated carrier establishment; legal interception of signaling traffic; warning message transfer function (including selection of the appropriate eNodeB); EU reach capacity procedures.

The PGW is a "PDN" gateway of the packet data network, and can also be called GW PDN. PGW It provides connectivity to the UE with the external data networks by packet data as it is the point of entry and entry of traffic for the UE. A UE can have simultaneous connectivity with more than one PGW to access multiple PDNs. The PGW carries out the policy application, the filtering of packages for each user, the collection support, the legal interception and the screening of packages.

Proxy is defined in the traditional sense as a computer system or an application that acts as an intermediary for requests from clients seeking resources from other servers. A client connects to the proxy server, requesting some service, such as a file, connection, web page or other available resource from a different server and the proxy server evaluates the request as a way to simplify and control its complexity. The term "proxy" is also used to refer to an aggregator or a backup agent.

The SDN controller is an application in a software-defined network ("SDN") that manages the flow control to enable intelligent networks. SDN controllers are based on protocols, such as OpenFlow, that allow servers to tell switches where to send packets. The controller is the core of an SDN network. It is located between the network devices at one end and the applications at the other end. Any communication between applications and devices has to go through the controller. The controller also uses protocols such as OpenFlow, to configure network devices and choose the optimal network path for application traffic.

The SGW is the acronym for the service gateway. The SGW routes and forwards user data packets, while acting as a mobility anchor for the user plane during eNodeB transfers and as an anchor for mobility between LTE and other 3GPP technologies (terminates the S4 interface and transmits traffic between 2G / 3G and GW PDN systems).

S1-GW is an acronym for an S1 gateway. S1-GW is a patented gateway that aggregates eNodeB communications with the core network. In a typical architecture, each eNodeB could speak directly with the EPC. In the embodiments described herein, these communications are added to virtualize all eNodeB or all base stations on a single eNodeB or base station.

The upstream device is an MME, SGW or any hardware device that is part of the core network.

WAG / MAG is a wireless access gateway / mobility access gateway. The wireless access gateway ("WAG") provides interconnection between the access networks and 3GPP. The WAG manages IP tunnels, QoS measurements and is responsible for roaming. The term TWAG, means trusted wireless access gateway. The nodes of the mobility access gateway ("MAG") provide the role of mobility agent for mobility management, mobility management of the mobile terminal with respect to the local mobility anchor ("LMA") and the establishment and release of a user data transport tunnel for the IP address assigned by the packet data network ("PDN").

The X2 handover is a handover that occurs through an X2 interface. An X2 interface is a new type of interface introduced by the LTE radio access network. Connect the neighboring eNodeB peer-to-peer to facilitate handover and provide a means for rapid coordination of radio resources.

Detailed description

The virtualization embodiments described herein comprise methods and a computer-readable medium configured to execute these methods in a heterogeneous mesh wireless communication network that includes a computer cloud as part of the network. Full details of the operation of the heterogeneous mesh network can be found in United States Patent Application No. 13/889631, filed on May 8, 2013, entitled "Heterogeneous Mesh Network and a Multi-RAT Node Used Therein ". It should be noted that, when analyzing a heterogeneous mesh network, it should be understood that there could be a return link without mesh within the network. The non-mesh return link could be composed of one or more wired connections to the cloud computing or other network device. Similarly, there could be non-combined wireless return links that are part of the heterogeneous mesh network. These variations are changes in the network topology that one skilled in the art would understand how to do and that would not change the inventiveness of the embodiments described herein.

Figure 3 depicts an exemplary heterogeneous mesh network 300 on which the forms of virtualization implementation described in this document. The heterogeneous mesh network 300 is composed of three nodes with multiple radio access technology ("multi-RAT") 310, 320, 330 and a computer cloud component 340. By way of example, and without limitation, each multi node -RAT 310, 320, 330 has an access radio and a return radio. As can be seen in Figure 3, the heterogeneous mesh network 300 includes multi-RAT nodes 310, 320, 330 with varied architecture. This varied architecture allows them to transmit and receive in countless frequencies, protocols, duplexing schemes and media access technologies. While each multi-RAT node 310, 320, 330 is represented with a single access radius and a single return radius, in additional embodiments, these multi-RAT nodes 310, 320, 330 will have more than one access radius and / or return. The return connections 315, 317, 325 between multi-RAT nodes 310, 320, 330 may be wireless in the heterogeneous mesh network 300. In one embodiment, the multi-RAT node 330 may have a wired connection 335 to the cloud computing 340.

Figure 4 illustrates the internal architecture of a multi-RAT 400 node used in the present virtualization embodiments. The multi-RAT node 400 is composed of at least one processor 410, access hardware 420, network hardware 430, an RF front end 440 and a timing source 450. As an example, at least one processor 410 could contain Firmware written in Linux. Additionally, the RF 440 front end can be configured to provide RF capabilities for multiple radio access technologies. In one embodiment, the timing source could be GPS. Alternatively, the timing source could be derived from Ethernet, or an IEEE 1588 source, such as SyncE, PTP / 1588v2, and the like.

Referring again to Figure 3, the computer cloud component 340, or "computer cloud" for short, is composed of software that can be run on any general purpose server. Computer cloud component 340, in some embodiments, can be run on a secure and hosted cloud service, such as those offered by Citrix or Amazon. The computer cloud component 340 includes a memory for storing the computer-readable methods of the described embodiments. The memory can also store additional network information, such as routing tables, environmental conditions, operating parameters and the like. The memory medium can also be a distributed memory medium, for example, for security reasons, or in the situation where more than once the cloud computing 340 is part of the communication network, the data can be distributed among the memory media in each computer cloud component 340 or even in the memory of the multi-RAT nodes 310, 320, 330 that are part of the network. In addition, the memory medium may be one of the networks to which the current network is coupled, for example, a storage area network, a software-defined network controller and the like. The computer cloud is also composed of a processor and a network interface that facilitates communication with at least one multi-RAT node 310, 320, 330.

Figure 5 illustrates a virtualized wireless communication network 500. This virtualized wireless communication network 500 includes multi-RAT nodes 512, 514, 516 coupled in communication to a computer cloud 530. In the embodiment shown in Figure 5, one of the multi-RAT 516 nodes has a 528 wired connection to the computer cloud, while another multi-RAT 512 node has a 524 wireless connection to the computer cloud. The third multi-RAT node 514 is coupled in communication to the computing cloud 530 through its wireless connections 522, 526 to its neighbors, which are in turn directly connected to the computing cloud 530. Although Figure 5 shows three multi nodes -RAT 512, 514, 516, those skilled in the art understand that a network with only two multi-RAT nodes could be created. Similarly, connections 522, 524, 526 between multi-RAT nodes 512, 514, 516 could be wired or wireless. In these embodiments, only one multi-RAT node 512, 514, 516 can be connected in a network to the computing cloud 530; and that connection could be wired or wireless, although a wired connection would be more reliable.

The computing cloud 530 is further composed of a general purpose processor 532 and an accelerator 534. The processor 532 could be a central processing unit. Additionally, although only one 532 processor and one 534 accelerator are shown, those skilled in the art will understand that there could be more than one, the other, or both in the 530 cloud. The computing cloud 530 is connected in communication to a central network 540, which it believes is only connected to a single device, such as a single base station, instead of being connected to more than one multi-RAT node 512, 514, 516 on the other side of the 530 cloud.

It should be understood that the general architecture of the wireless communication network 500 is scalable and can be distributed due to the nature of the computing cloud 530. In terms of scalability, the wireless communication network 500 is scalable in at least two ways. First, although three multi-RAT nodes 512, 514, 516 coupled to the 530 computing cloud are shown, this architecture could support hundreds, thousands, several thousand and millions of multi-RAT nodes without overloading the core network.

Second, a single computer cloud 530 is shown. The computer cloud component 530 is also scalable, which means that there could be numerous computer clouds 530. Each of these additional computer clouds 530 could contain similar logic or the logic could distribute If the logic were distributed, the two or more 530 computing clouds could communicate with each other using an XML-based protocol. Some of the types of information that could be shared through this protocol are, for example, the environmental conditions in the respective network of each computer cloud 530, the availability of additional resources that will be shared if necessary, a query from a computer cloud 530 to coordinate a transfer, and the like.

Figure 6 shows a virtualized communication network 600 with exemplary radio technologies of: 3G, Wi-Fi and LTE. The choice of these radio technologies is illustrative only and is not intended to limit the scope of the description herein. The general functionality of the virtualized wireless communication network represented in Figure 6 is not intended to be different in a general sense from the functionality represented in Figure 5. Instead, Figure 6 shows a more specific implementation of a virtualized wireless communication network that has three specific wireless technologies within the network: 3G, LTE, and Wi-Fi. In a different virtualized wireless communication network that has different radio technologies, the specific internal contents of the 630 cloud would be altered to adapt to the wireless communication technologies operating within the network.

As can be seen in Figure 6, the three multi-RAT nodes 612, 614, 616 are coupled in communication to the computing cloud 630 through a wired connection from the multi-RAT node 616 to the computing cloud 630. Within In the cloud, there are specific processors and accelerators for LTE, shown as S1633 proxy, Wi-Fi, shown as WAG / MAG 631, and 3G, shown as Iu 632 proxy. Proxy S1 633 in this embodiment is a software entity that virtualizes the existence of all LTE radios within multi-RAT nodes 612, 614, 616 and facilitates transfers to and from LTE. In an alternative embodiment, the S1633 proxy could be further customized by adding proprietary extensions between multi-RAT nodes 612, 614, 616 and the S1 633 proxy. This embodiment of the S1 633 proxy is called an S1 / S1 'proxy because on one side of the proxy, it would be related to the core network 640 in a standard form of S1, but on the other side of the proxy it would be interacting with the LTE portion of multi-RAT nodes 612, 614, 616 in a non-standard way.

The Iu 632 proxy is a software entity that virtualizes 3G radios on multi-RAT nodes 612, 614, 616 and facilitates transfers to and from 3G. The wireless access gateway / mobile access gateway "WAG / MAG" 631 controls multiple access points, provides wireless connectivity, and facilitates the transfer between Wi-Fi networks and LTE networks for multi-RAT nodes 612, 614, 616. In additional embodiments of WAG / MAG 631, you may be able to access the data stored on the SIM card of a UE receiving service from a multi-RAT node 612, 614, 616 to authenticate the UE and allow connectivity without problems with Wi-F access points.

ePDG 634 is a standard-defined interworking function between PDNs and unreliable 3GPP wireless interfaces, such as Wi-Fi. Figure 6 also includes two gateways, an LTE 635 gateway and a 3G / Wi-Fi 636 gateway. These gateways 635, 636 are optional. In this embodiment, the gateways provide a connection to a local network 650 and a local content server 655. In the prior art, content and access to servers were typically performed in a regional or national data center, which was typically located a great distance from most wireless networks. In this embodiment, access to local network 650 and a local content server 655 is provided to reduce traffic costs associated with access to servers or content. In addition, local content could be shared with many users within a network. For example, the schedules of local movies, news, restaurants, social networks, and the like, could be available on the local content server 655.

In alternative embodiments using a similar architecture, the 630 cloud could contain a software module that allows it to carry out a deep inspection of packages. In these embodiments, deep packet inspection could be used to distinguish various types of data, for example, voice versus data, or video vs. voice. This distinction, which can also be a QoS distinction, could be used to make radio bearer determinations, to choose an available network, to change the priority of the data, and the like.

In the prior art, there was no need for gateways that used different protocols to communicate with each other. For example, an MME would not have to communicate with a PDN. These devices were physically separated. In the present virtualized embodiments, it is connected to an MME and a PDN and other gateways in a virtualized manner. In the exemplary architecture of Figure 6, a traditional LTE eNodeB with proxy S1 633 is imitated and a TWAG 631 is housed within the computing cloud 630. In the prior art, an ANDSF would make an initial bearer assignment for a data packet. or a voice pack. This assignment was not dynamic, but was done as an initial matter. In the architecture of the computing cloud 630, a TWAG 631 and an S1 gateway have been placed, which is part of the S1 633 proxy. This placement allows dynamic assignments to be carried out and, most importantly, bearer reallocations in a variety of networks.

In addition, deep packet inspection could be used to aggregate data in a kind of "big data." By this, it refers to the aggregation of data that can be used independently of a wireless communication network. An example of how large data can be used is to set social trends for regional marketing and the like.

In addition to deep packet inspection, the 630 cloud can also perform content caching. In this embodiment, the 630 cloud could determine the ten most popular YouTube clips in real time. The 630 cloud could use this determination and its content caching feature to save local copies of these ten most popular YouTube videos on the local content server 655. Those skilled in the art will recognize that the 630 cloud could make this determination by pointing the web to EU traffic within its coverage area or within a larger area to the extent that the network has been scaled to include multiple 630 clouds. Similarly, those skilled in the art will recognize that this type of content caching could be applied to many different types of content, including television shows, news articles, books or any online content.

The 630 cloud also acts as a security gateway. It ensures network traffic to the MME gateway and the S1 gateway by providing standard security protocols, such as a VPN hub. In addition, the 630 cloud has the intelligence to provide a location-based service, such as weather, traffic, news and the like.

The cloud 630 fulfills the function of grouping the resources of the virtualized wireless communication network 600. Each of the multi-RAT nodes 612, 614, 616 includes hardware for multiple radio technologies. The resources of the multi-RAT nodes are grouped so that the computing cloud 630 can allocate those resources efficiently in real time and dynamically. All this decision making and the combination of resources takes place behind the scenes, so to speak; and the central network 640 is unaware of how resources are allocated or that there is more than one base station behind the 630 cloud. In some embodiments, the determination of how to allocate radio resources more efficiently can be achieved by self-organization, self-optimization, self-recovery methods known as "SON", and are described in more detail in US Patent Application N . ° 14/024717, entitled "Heterogeneous Self-Organizing Network for Access and Backhaul", filed on September 12, 2013, whose content in its entirety is hereby incorporated by reference.

The computer cloud 630 continuously monitors environmental conditions such as signal-to-noise ratio, latency, etc., within the virtualized wireless communication network 600. It has a stored list of operational parameters, such as the various power levels of each multi-RAT node 612, 614, 616, using routing tables for data, using operational protocols and frequencies for each multi-RAT node 612, 614 , 616 that can be dynamically modified to explain less than optimal environmental conditions, such as high signal-to-noise ratio, high latency, or quality of service measures that are below a certain threshold. These environmental conditions are merely illustrative and are not intended to be limiting.

It is well known in the art that certain systemic conditions affect network performance. These conditions are referred to as "environmental conditions". While many of the environmental conditions that affect network performance have been listed, this list is not intended to be exhaustive, but merely illustrative of what those skilled in the art understand about the effect of environmental conditions on network performance. .

At various points in time, mesh networks may include multi-RAT nodes, user equipment coupled in communication to multi-RAT nodes and / or a cloud computing. From the network perspective, therefore, environmental conditions and operational parameters exist for all these participants within the network of mesh.

Cloud component 630 can obtain multiple measurements of environmental conditions. Some of the environmental conditions related to a heterogeneous mesh network include: an interference measurement, a capacity measurement, a spectrum efficiency measurement, a routing route, a network congestion measurement, a performance measurement, a latency measurement, a coverage gap, a signal-to-noise ratio , a quality of service measurement, a radio bearer utilization value, an available portion of the spectrum, a load balance measurement, the status of an operating heterogeneous mesh network, the status of a multi-RAT node within the heterogeneous mesh network, identification information regarding a multi-RAT node, the status of a wired connection within the heterogeneous mesh network, a frequency restriction, a measurement of Signal strength of the neighboring multi-RAT node, a request to join the heterogeneous mesh network, or the existence of a hidden node, and the like. Therefore, unlike prior art virtualized node hardware, measurements of the environmental condition of multiple nodes within the virtualized communication network can be gathered in the cloud component, analyzed and then used to notify, control, adjust or reconfigure one, more than one, or all consumers of the virtualized communication network.

In an alternative embodiment, the SON modules described herein could be harmonized with an external third-party network that has its own set of environmental conditions and operating parameters. These third-party environmental conditions or third-party operating parameters could be any of the environmental conditions or operational parameters described in this document with respect to the SON network. In these embodiments, an XML-based interface could facilitate communication between a cloud computing server or a multi-RAT node containing SON modules described herein and the third-party network. When the SON module receives an environmental condition from a third party or an operational parameter from a third party, such as an operating frequency, it can adjust an operating parameter within its own internal network, for example, by altering the operating frequency of a multi-RAT node You may be experiencing interference related to your proximity to the third party network. The harmonization between SON networks described in this document and third-party networks could serve to better utilize resources between both SON networks and third-party networks, for example, mitigating interference, coordinating transfers, sharing the unused spectrum, and the like

Specific environmental conditions of multi-RAT nodes include: a multi-RAT node identification number, an identification number for software stored in a multi-RAT node, a security parameter, a location of a multi-RAT node , a certificate of configuration of a multi-RAT node, an authentication request, an operational frequency or a transfer request, and the like.

The environmental conditions related to specific user equipment that are served by multi-RAT nodes can also be measured and transmitted to multi-RAT nodes within the mesh network. Some of these environmental conditions, which could also be processed by the described SON embodiments, include: an interval from a user equipment to a multi-RAT node, the travel direction of a user equipment, a travel speed of a user equipment, a signal strength of a user equipment, a location of a user equipment, or a mapping application stored in a user equipment, an operating channel, and the like.

The above-mentioned environmental conditions could be measured by a multi-RAT node 512, 514, 516. They could be calculated in a processor in a multi-RAT node 512, 514, 516 or in a 532 processor in a 530 computing cloud. similarly, they could be stored in a cache within a multi-RAT node 512, 514, 516 or in a 530 computing cloud or distributed throughout the network.

Operating parameters that can be adjusted include: power level, channel, subchannel, frequency band, subcarrier frequency, spectrum allocation, access configuration, return configuration, client, server, routing route, IP address, self-configuration instance, fully qualified fully qualified domain name ( "FQDN"), dynamic host configuration protocol ("DHCP") address, return configuration, network provider (for example, AT&T, Verizon, T-Mobile and the like), participation within a mesh network, and the like.

Figure 7 illustrates the bi-directional nature of virtualization of the embodiments of this invention. The first type of virtualization is virtualization from the perspective of the core network. Figure 7 shows two wireless communication networks 710, 720 comprising multi-RAT 722 nodes. Both communication networks Wireless 710, 720 are coupled in communication to the 730 cloud. Although Figure 7 shows two separate connections to the computing cloud 730, the communication network 710 could be coupled in communication to the communication network 720, which in turn could provide the communicative coupling for both networks to the computer cloud 730 in one alternative embodiment. Also, although Figure 7 shows seven multi-RAT nodes, the number of nodes could be as small as two and may not have an upper limit.

For illustration purposes, Figure 7 includes two central networks 740, 750. By way of example, one of the central networks 740 could be an AT&T network, and one of the central networks 750 could be a Verizon network. The virtualization methods and devices described in this document create a dual or bi-directional virtualization as follows. From the perspective of the AT&T 740 core network, you see its connection to the 730 cloud as a connection to a single base station. Similarly, from the perspective of the Verizon 750 core network, it sees its connection to the 730 cloud as a single connection to a single base station. None of the central networks 740, 750 is aware of the other.

In-network communication data has been summarized as described in more detail in US Patent Application No. 13/889631 filed on May 8, 2013 and entitled "Heterogeneous Mesh Network and a Multi-RAT Node Used Therein. " As such, the data is agnostic with regard to frequency, protocol, duplexing scheme and media access technology. This summary data is not incorrectly multiplexed or encoded with header information to indicate to the EPC or the core network that multiple different radio technologies are being serviced. The signaling that takes place between the computing cloud 730 and the central network 740, 750 is the same type of signaling that would occur between a base station and the central network 740, 750 in some embodiments. In alternative embodiments, the signaling between the computing cloud 730 and the central network 740, 750 could be similar to the type of signaling that occurs between a base station and the central network 740, 750, if, for example, we were using a proxy S1 / S1 'and custom patented extensions between multi-RAT nodes and the S1 / S1' proxy.

The second type of virtualization is from the perspective of an individual multi-RAT node 722. The multi-RAT node 722, which is part of a mesh network, is coupled in communication to the computing cloud 730. The nature of a network of Mesh allows dynamic adjustments to be made within the network. Cloud 730 monitors these dynamic adjustments by continuously evaluating environmental conditions and operational parameters to be able to make dynamic changes within the network. Some of these changes could be, for example, a change from an LTE access channel to a blank space access channel so that the multi-RAT 722 node improves its efficiency. In this situation, the multi-RAT node 722 could be instructed by the cloud 730 to change its access radius to a blank space radio. Cloud computing 730 could instruct the multi-RAT node 730 as to how often to change within a blank space sector. Before receiving this instruction, multi-RAT node 722 may have ignored the availability of a white space network, or what frequencies within the white space network were available. In this way, the networks to which the cloud computing 730 is connected are virtualized from the perspective of any multi-RAT node 722. In addition, unlike the virtualization methods of the prior art, the frequencies within the network of spaces Blank do not need to be dedicated or assigned exclusively to a virtualized resource consumer.

In embodiments of alternative methods of the present invention, computer-implemented methods have been created to virtualize a network, so that the central network believes that the virtualized network is a single base station, when in reality the virtualized network is Many base stations. In some ways, this is the inverse of typical virtualization: the multi-tenure virtualization version. The computer implemented steps of the present invention could occur within a memory device stored in the computing cloud 340, 530, 630 or 730 or in a distributed manner.

Figure 8 shows these stages. In the first stage, the computing cloud 340, 530, 630 or 730 groups 810 the radio resources of the multi-RAT nodes within a wireless communication network. To make a network, there must be more than one single multi-RAT node actively participating within the network. In some embodiments, multi-RAT nodes could have the same radio resources; and in other embodiments, its radio resources could be varied. For example, in a wireless communication network, there could be four multi-RAT nodes, each with a Wi-Fi radio and an LTE radio. The grouped resources would therefore be four Wi-Fi radios and four LTE radios. In a different embodiment that has four multi-RAT nodes, the grouped resources could be: two TV blank radios; three LTE radios, two Wi-Fi radios and a 3G radio. In any of these embodiments, the pooled resources can be used for access or return.

In the next stage of this virtualization method, the computing cloud 340, 530, 630 or 730 maintains 820 a connection to a central network or an ascending device. The computing cloud 340, 530, 630 or 730 could establish a connection with a central network or upstream device and then maintain that connection. It could maintain multiple connections to multiple central networks or upstream devices as well.

The computing cloud 340, 530, 630 or 730 also in this virtualization method manages 830 the grouped radio resources, so that the central network or the upstream network device believes that it has a single connection to a base station. The core network or upstream devices are unaware of how many radio resources are being managed 830 by the cloud computing, the fact that the radio resources have been grouped, etc. In the past, radio networks, operator policies, service agreements between service providers and their customers, different operating parameters, and the like, have resulted in the partition and segregation of radio resources. The virtualization embodiments described herein address these drawbacks.

One of the goals of the cloud computing 340, 530, 630 or 730, which is behind the curtain of virtualization, is the efficient management of all of its grouped radio resources. This management can come into play in many different communication life cycles that occur within a communication network. There is the establishment of a call or data connection. There is continuous maintenance of the current call or data session. There is a transfer decision, pagination requests, return requirements. There are environmental conditions as discussed above that continually affect the performance of the network. There are equipment failures that affect performance. Cloud computing 340, 530, 630 or 730 takes these factors into consideration when managing 830 the grouped radio resources.

In terms of radio resource management, this management could be done in several different ways. In one embodiment, a complete virtualization embodiment, multiple operators can fully share all available radio resources. Radio resources are not reserved for any operator. Any subscriber of any operator can use all available radio resources at any time. In a second embodiment, the divided virtualization, the computer cloud 340, 530, 630 or 730 could store in a memory a list of dedicated radio resources reserved for each operator. If the radio resources dedicated to a particular operator are fully utilized, a network connection or a connection request from a subscriber belonging to that operator will be denied.

In a third embodiment, virtualization of equitable access, computer cloud 340, 530, 630 or 730 could store in a memory a list of dedicated radio resources reserved for specific operators and a set of unreserved radio resources. In this embodiment, if the radio resources for a particular operator were fully utilized, the computing cloud 340, 530, 630 or 730 could provide a radio resource for that operator of the group of unreserved radio resources. In a fourth embodiment, the unbalanced virtualization, the computing cloud 340, 530, 630 or 730 could provide radio resources dedicated to a subset of operators, while providing a group of radio resources not reserved for it or a different subset of operators, or for all operators.

In additional embodiments, the computing cloud 340, 530, 630 or 730 could manage 830 grouped radio resources by assigning a radio bearer in light of a quality of service value, a use value of radio bearer resources, or other data priority metrics such as a data priority value, a type of data, contractual relationships with network operators, or a service level agreement.

The call admission control can determine whether a call request will be accepted or rejected based on the virtualized resources available by operator to guarantee the quality of the requested service. If a call is accepted, the quality of service of existing calls can be maintained. The call admission control may use the virtualization embodiments described herein to admit or reject new calls.

Additional embodiments of the systems described herein could include layered virtualization. In these embodiments, the basic architecture shown in Figure 3 could be repeated N times, creating a total of N cloud computing servers 340, 530, 630 or 730. These N cloud computing servers 340, 530, 630 or 730 could include modules stored in the internal memory that virtualize the total number of cloud computing servers, which makes it appear that there is only one point of contact with the EPC, which seems Be an eNodeB or a base station. An advantage of these embodiments of the system is that, from the perspective of the central network, there is only one connection that requires the management of metrics such as transfers, paging, QoS measurements, billing and the like. In addition, the use of multiple clouds, in different networks, creates virtual redundancy. The resulting efficiencies include frequency reuse, maximization of network resources across multiple radio technologies, a significant reduction in core management overhead and the like.

The embodiments described herein reduce the costs of the system for transfers because the computing cloud 340, 530, 630 or 730 optimizes the transfers between the grouped radio technologies. Cloud computing 340, 530, 630 or 730 tracks all transfers for each of the multi-RAT nodes within your network. When a UE moves from the coverage area of ​​one multi-RAT node to another, the computing cloud 340, 530, 630 or 730 absorbs mobility on the transfer side by sending all packets destined for that UE to the new multi-RAT node which is serving the EU. These packages could include anchor data. Specifically, computer cloud 340, 530, 630 or 730 replaces the traditional function of the MME with respect to signaling between an MME and a traditional base station.

Multi-RAT nodes 612, 614, 616 of the embodiments described herein are coordinated with the computing cloud 630 to carry out UE transfers in a manner that is invisible to the MME. These same benefits can be obtained in networks that have been scaled to include large numbers of multi-RAT nodes or multiple computing clouds. By taking the signaling to the MME outside the transfer scenario, the data traffic to the MME, and therefore through the central network, is greatly reduced. The cloud 630 has a global view of the load experienced by each multi-RAT node 612, 614, 616 to which it is coupled in communication. This global view allows the cloud 630 to achieve an optimal network load by selectively downloading the UEs from the edge of the cell to neighboring cells by means of intelligent redirection, as well as planning of the cell radius. In addition, the 630 cloud computing can use TWAG activation to release the LTE context in the MME and in a multi-RAT node 612, 614, 616. In these embodiments, the e Pc does not have to be informed every time a transfer occurs, thereby greatly reducing the operational costs of the network.

The virtualization embodiments described in this document also reduce the operational costs of the network in terms of paging. Currently, when a UE is in idle mode and the central network must reach the UE, for example, to connect a call, the central network sends a page to the UE. The central network generally sends a paging message to the base station that previously serves the UE and perhaps to other base stations within the same tracking area as the last active base station in the UE. If the UE cannot be located near its last known location, the central network will begin to broaden its search for the EU to send paging messages to an increasingly wide audience of base stations.

These virtualization methods and systems lend themselves to a more precise way to locate a UE. In the virtualized embodiments described herein, the computing cloud 340, 530, 630 or 730 has more information about any particular UE than the central network or an upstream network device it may have at any given time. The computer cloud 340, 530, 630 or 730, because it is an aggregator node that is between a multi-RAT mesh network and the central network, has access to information related to the activities of any UE in other networks, such as Wi-Fi networks. If, for example, a particular UE was in INACTIVE mode on an LTE network, but was downloading content from a local server over a Wi-Fi connection, the cloud computing 340, 530, 630 or 730 would know when a page will arrive for that UE, where it was located, and what multi-RAT node should be paginated. In this example, cloud computing 340, 530, 630 or 730 could include a TWAG, which would allow you to use Wi-Fi connection knowledge to more quickly locate a given UE. Similarly, if an UE was driving and was using its GPS for navigation purposes, the IT cloud 340, 530, 630 or 730 would know where that UE was at any given time. This type of heuristic paging could further reduce the tax imposed on the core network because the inherent uncertainty of the core network to locate a UE in idle mode is reduced.

In the embodiments described above, an SDN controller could be coupled in communication to the cloud computing server 340, 530, 630 or 730. In these embodiments, the SDN controller could insert the SDN policies into the virtualized network. The SDN controller could optionally be coupled in communication to an external SDN controller, acting, as well as an SDN gateway for the virtualized network. In this embodiment, the SDN gateway could be implemented in proxy mode, passing all control information to SDN controlled devices, or in hybrid mode, implementing some of its own policies on the control information it receives. In these embodiments of SDN, the SDN controller or controllers would also be virtualized from the perspective of the core network.

The above analysis discloses and describes merely exemplary embodiments of the present invention. In further embodiments, the methods described herein can be stored in a computer-readable medium, such as a computer memory storage, a compact disc (CD), flash drive, optical drive or the like. In addition, the computer-readable medium could be distributed through memory storage devices within multiple servers, multi-RAT nodes, controllers, cloud computing components, and the like. As will be understood by those skilled in the art, the present invention can be carried out in other specific ways without departing from the scope of the appended claims. For example, the wireless network topology can also be applied to wired networks, optical networks and the like. Various components may be added, removed or replaced in the devices described herein, with those having the same or similar functionality. Several steps, as described in the figures and the specification can be added or removed from the processes described herein, and the steps described can be performed in an alternative order, consistent with the scope of the appended claims. Accordingly, the disclosure of the present invention is intended to be illustrative, but not limiting the scope of the appended claims. The disclosure, including any easily discernible variant of the teachings herein, defines, in part, the scope of the terminology of the preceding claims.