Please wait...



Goto Application


Note: Text based on automatic Optical Character Recognition processes. Please use the PDF version for legal matters

[ EN ]




[0001 ] This application claims the benefit of co-pending U.S. Provisional Application No. 61/806,438 filed March 29, 2013.


[0002] The present invention relates to wide area networks and, more particularly, to a global wide area network infrastructure that incorporates aircraft and land-based nodes.


[0003] A wide area network (WAN) is a communications network that connects local computer networks into a larger working network that may cover both national and international locations. WANs often connect multiple smaller networks, such as local area networks (LANs) or metro area networks (MANs) or even end points. Current infrastructure for WANs includes microwave and radio frequency transceivers, land cables, undersea cables, and satellites. This infrastructure is required for companies to communicate across town, across the country and across the world.

[0004] Unfortunately, there are drawbacks with the current infrastructure options. Generally, the land- and undersea-based infrastructure for WANs is very expensive, hard to maintain, and has limited bandwidth. An undersea cable infrastructure is necessary to carry data traffic between different continents and

regions, and the cost to install these cables is so great that it is generally done with a consortium of service providers. The land-based cable infrastructure is also very expensive and constrained to routes with access.

[0005] Satellites have attempted to address the communication needs of the general population and those people without access to the land-based cable infrastructure. The use of satellites, however, has met with limited success. Satellite communication is unreliable partially because satellites do not have the ability to be upgraded and therefore cannot take advantage of new technologies that will increase bandwidth and decrease latency. Additionally, Satellites in even low-earth orbits are about 120 miles above land and thus the best satellite solutions have up to 500ms latency due to the large separation between the satellite and the land-based facility. The orbital height causes additional problems in that, because satellites use microwave frequencies that require line-of-sight transmission, the curvature of the earth means that the maximum distance for communication between a satellite and a land-based link is roughly 600 miles.

[0006] Submarine telecommunication cables also have been attempted to address communication needs. Submarine telecommunication cables present great economical and technical challenges, however. They have to support low

temperatures of less than 80-85 degree Celsius, as well as be exposed to strong oceanic currents, natural disasters (i.e. earthquakes), animal attacks, and damages caused by human interactions (i.e. anchors, fishing boats, etc.). Recent studies also have highlighted that undersea cables can have a negative impact on the environment.

[0007] Airborne aircraft convey flight status and related flight information using air-to-land communications, such as from the in-flight aircraft to air traffic control towers. Airborne aircraft communicate flight information to other airborne aircraft without using a land-based link by using radio of various frequencies for voice and data communication. Unfortunately, radio signal strength for aircraft-to-aircraft communication is only good for about 50 miles.

[0008] Due to the numerous drawbacks of current land, undersea, and satellite infrastructures for WANs, data and voice traffic between regions and sites is extremely expensive for companies. For example, in some regions like the Middle East, a company can have monthly recurring costs for a fairly small bandwidth of approximately USD $20,000 per month. With the growth of mobile interactions, the demand for high-quality voice and video transmissions, the diverse digital content proliferation and social interaction, and the transition to cloud computer services and solutions, there is a great need for cost-efficient and high-performing network infrastructures. Wide area network providers and developers continually seek ways to improve upon the existing infrastructure and ways to provide a reliable, high-performing, and cost-effective WAN infrastructure on a global level.


[0009] An infrastructure for wide area networks comprises airborne aircraft as nodes that wirelessly communicate with land-based nodes to form a mesh network for transmission of network packets in a vast wide-area network. Although the inflight aircraft are in constant motion, with the addition of certain hardware and software their whereabouts can be determined in real time enabling a live network with constantly changing nodes. Aircraft WAN components comprise receivers, transmitters, and repeaters. Land-based node WAN components comprise transmitters, receivers and computer system components necessary for operating

software to manage and route messages. Land-based nodes are in communication with data repositories that house data such as the location of each of the aircraft in flight, weather, speed and cost of packet hops, all updated in real time. By knowing this information in real time, software can determine possible packet routes in real time and select a desired route based on pre-defined criteria. These routes can be uploaded into each of the air- and land-based nodes for real time dynamic routing of network packets.

[0010] Possible data paths within the network are established in advance of packet transmission using known flight schedules to estimate the location of aircraft at given times and known land-based node positions. Each data path comprises a series of possible hops at a given time from a source to a final destination. Each data path is created by calculating a set of possible hops to a final destination and selecting a next hop according to prescribed criteria. The method of transmitting messages comprises updating the location of each airborne node in real time and transmitting a packet down a chosen data path to a final destination according to prescribed criteria. Packet transmissions are made wirelessly on these software-defined networks from aircraft-to-aircraft while they are in flight, enabling data to be transmitted across vast space nearly instantaneously. Thus combining airborne hops with traditional ground-based hops and implementing a software-defined network to update node and packet location in real time creates a new network infrastructure for transmitting data globally.


[0011 ] FIG. 1 is an illustration of the wide area network infrastructure of the present invention.

[0012] FIG. 2 is a schematic illustrating the components of the wide area network infrastructure of the present invention where two land-based nodes wirelessly communicate through a series of aircraft.

[0013] FIG. 3 is a flowchart illustrating the process of determining the best route for a network packet to travel using the wide area network infrastructure of the present invention.

[0014] FIGS. 4A-4C illustrate a simplified network of the present invention at three different times.


[0015] The present invention combines airborne nodes, land-based nodes, data repositories, and software-defined networks to transmit network packets. Data is updated in real time to enable the software-defined network to change, select, and manage routes in real time. As used herein, real time means as an event is happening or sufficiently immediate to it to render any variance irrelevant.

[0016] Fig. 1 (not to scale) shows one embodiment of the wide-area network infrastructure 10. Airborne nodes 12 are in communication with other airborne nodes 12, land-based nodes 16, a sea-borne node 17. Transmissions involving an airborne node or sea-borne node are wireless and therefore line-of sight:

Transmission between non-stationary nodes is indicated by arrow 26 and

transmission between stationary nodes is indicated by arrow 28. Transmissions between land-based nodes are well known in the art, and include wireless and wired communications.

[0017] As used herein, a node is any network-capable device whose location can be tracked in real time and whose hardware and software can be updated.

Nodes include airborne nodes, sea-based nodes, and land-based nodes.

[0018] As used herein, an airborne node is any device whose location is not usually stationary while in the network and can be tracked in real time, including airplanes, helicopters, gliders, balloons, drones and any other form of manned or unmanned vehicle. For convenience, ships at sea are also considered airborne nodes as used herein. An airborne node is also referred to herein as an aircraft. Preferably the airborne nodes are capable of landing, which enables their hardware and software to be updated, unlike that of satellites. In the preferred embodiment, the airborne nodes are airplanes that travel scheduled routes repeatedly, such as

commercial passenger and freight airplanes. Optionally aircraft may be added specifically as nodes to the network for the purpose of covering underserved areas. These fill-in nodes are not necessarily travelled nor scheduled repeatedly and may be moved dynamically based on the positioning and availability of other aircraft.

Airplanes have a line of sight that is not often obstructed by landmarks and physical barriers.

[0019] At any one point in time there are approximately 20,000 airplanes in the sky. The airplanes travel at approximately 36,000 feet and have roughly 250 miles of distance for line of sight to the horizon or 500 miles to another airplane also travelling at approximately 36,000 feet. Generally an airplane is within 500 miles of another aircraft and as disclosed herein a network mesh is constructed in real time between inflight aircraft, each of which becomes an airborne node. In one embodiment the present network is built using existing flight routes. As the number of airborne aircraft changes and their routes change, the network automatically changes too to accommodate the different number and geographical location of each node. Thus the present solution offers elastic scalability and expandability. For example, for a network packet to circle the earth utilizing the network described herein,

approximately 100 planes would be needed at minimum. Without the added signaling transaction time incurred using conventional land-based only routes, the network packet could circle the earth in under 500 ms and could travel from point to point in under 250 ms, far faster than WANs using solely land-based nodes and undersea cables.

[0020] The concentration of airplanes is greatest over the geographical areas that have the largest population, which also has the greatest amount of data transmission requirements. This high concentration of airplanes means that the

bandwidth of the present network can be concentrated in the areas that require the most bandwidth, thus the highest ability to avoid network congestion. The network built using the aircraft avoids physical barriers such as mountains and seas, resulting in the shortest direct data paths.

[0021 ] Each airborne node is equipped with certain hardware to transmit data from airborne node to other airborne node and from airborne node to land-based nodes. The transmission hardware in the planes comprises receivers and

transmitters operating in a band that enables a high rate of data transfer, or bandwidth, with signal strength sufficient for reliable communications between airborne aircraft. In a preferred embodiment, the receivers and transmitters operate in the microwave band, and more preferably in the Ka band covering the frequencies of about 26.5 - 40 GHz. Aircraft may also be equipped with antennae to aid

transmissions. Airborne WAN equipment is preferably powered by battery, but may also be powered by sun, wind, or an aircraft's engine.

[0022] The receivers and transmitters are mounted inside or outside the aircraft at locations for best transmission, including the sides, nose, tail, top or bottom. In some embodiments the receivers and transmitters are directional, the direction controlled by the data repository which provides real-time information about the location of the next hop, as explained below. In some embodiments there are multiple receivers and transmitters, operating at the same or different frequencies.

[0023] Aircraft are also equipped with routers to examine each packet of data as it is received, determine where to send it to its next node. The routers are preferably installed inside the aircraft for protection from the elements, and are preferably housed in the passenger cabinet or aircraft wiring closet, or alternatively in bin space under the passenger cabin. Optionally, aircraft may be equipped with

repeaters to receive data from, and forward data to, other WAN components.

Repeaters can also retransmit a packet at a higher power so that the signal can cover longer distances or in a different direction to overcome obstacles such as storms.

[0024] Airborne nodes may be equipped with location identification

components 15 that are able to identify and track the past and present location of each airborne node 12. Devices for determining the location of an aircraft are known in the art and include satellite tracking, radar, and global positioning systems (GPS). Airborne nodes may be equipped with computer system components 36 needed for processing data and instructions. Optionally, aircraft may also be equipped with sensors for collecting weather information.

[0025] Each land-based node is equipped with certain hardware to transmit data to airborne nodes, to other land-based nodes, and to the existing network infrastructure. The land-based receivers and transmitters receive and forward data to the airborne nodes within line of sight. The land based nodes can also comprise endpoints.

[0026] The WAN transmission hardware in the land-based nodes comprises receivers and transmitters operating in a band that communicates with the airborne nodes. In a preferred embodiment, the land-based receivers and transmitters operate in the microwave band, and more preferably in the Ka band. If communicating wirelessly, land-based nodes may also be equipped with antennae to aid

transmissions. Land-based node WAN equipment is preferably powered by mains, but may also be powered by battery, sun, wind, or other energy source.

[0027] Land-based nodes communicate with other land-based nodes by wired and wireless methods known in the art. Land-based nodes 16 are in communication with one or more data repositories 30, as discussed below. The data repositories may be physically within a land-based node, or physically separate. Land-based nodes are also equipped with routers and, optionally, repeaters 14 to receive data from, and forward data to, other WAN components.

[0028] The land-based nodes are also in communication with computer systems 34 necessary for operating software 32 that includes software defined network (SDN) management components to dynamically update network packet routes and locations, as discussed below.

[0029] The land-based nodes are in communication with one or more data repositories, which as used herein means a collection of databases and metadata about those databases. The data repository aggregates the information that is used to build the dynamic routing tables for the devices that make up the inventive network. The databases include static data and changing data; empirical data and forecast data; and data derived from analysis of other data in the repository. The databases comprise data about airborne nodes such as current location data, including longitude, latitude, altitude and speed; scheduled flight plan route and departure/arrival information for aircraft; and connectivity attributes of each airplane. The databases comprise data about land-based nodes such as their fixed locations; power outages, planned and actual; and cost per hop between nodes by carrier. The data bases include data about transmission, including number of retries; number of packets with errors; number of packets that were discarded, lost, or retransmitted; average and maximum jitter; average latency and maximum latency; cost by carrier; and time-of-day. The data bases include data about users or with a particular category of user, such as residential users, business users, military users. The databases also include data types common to both types of nodes, such as network address information, for example the IP address, preferably IPv6; current and

forecast weather around and between each node; and even censorship rules for a country being serviced. For example, for data being transmitted from an aircraft to Iran is censored based on the laws of that country. The censorship rules are stored in a data repository and any data transmitted to the land-based receivers in Iran are routed through the censor engine and either filtered at the source or in route. Each transmission could be filtered, or just the first packet before the data path is established to ensure compliance to rules governing data transmitted to or from a region or country.

[0030] Preferably each data repository has high availability, whether the ready access to data is due to inherent architecture, redundancy or other technology.

[0031] With such large datasets, the preferred embodiment uses big data processing technology, such as massively parallel software running on tens, hundreds, or even thousands of servers to ensure access to the data in real time. For example, in one embodiment 500 petabytes of data is stored in one or more of Hadoop™, Apache Cassandra™, HBase™, and MongoDB™ databases on solid-state and standard Winchester Technology™ drives, available commercially.

Standard OLTP and in-memory databases can also be used for real time data processing. The data is managed using different database and NoSQL technologies running on the latest X86 processors, over a large bandwidth network. This data can be slow changing and fairly static such as planned flight departure and arrival times, city locations, flight paths, etc. Other data is dynamic, such as weather events, flight delays, route changes, and equipment failures, Other data is time-sensitive data such as packet drops from specific nodes, latency times of packet deliveries, etc. The data ingestion can be from many different sources, into different types of data repositories based on the use case. The data is used to provide the latest network maps to the

SDN controller, based on desired criteria including shortest path, data redundancy requirements, retransmit times etc.

[0032] The data is encrypted in transit, preferably using Internet Protocol Security to authenticate and encrypt each packet of a communication session. IPv6 is preferred, and brings the added benefit of avoiding IP address exhaustion looming with IPv4, although IPv4 will suffice.

[0033] The increased number of potential nodes relative to a conventional land-based network vastly increases the number of potential paths the packets can take to their final destination. Historically paths were determined between nodes that were at essentially fixed locations, and therefore known far in advance. The next hop was determined by the physical device that held the packet, e.g. a router, based on a limited set of information. Knowing the location of the fixed endpoints made the construction of routing tables relatively straightforward, even for large data sets. However, such autonomous systems do not allow a node to move without changing the node's identity on the network nor the node's relative position to other nodes.

[0034] In contrast, the present system uses moving nodes such that the location of the endpoints of each hop are not fixed and are therefore not exactly known far in advance. That is, a node will likely not be at the same exact physical location as it was when the packet was sent, nor in the same position relative to other nodes. Creating and maintaining continuous communication over networks of moving nodes is managed by a software-defined network ("SDN") which decouples the system that makes decisions about where a packet is sent (the control plane) from the underlying systems that forward the packet to the selected destination (the data plane or physical layer). SDN uses controllers and CPUs on the ground to determine the routing maps and specific connectivity for each packet. These SDN components

can access all the relevant factors for desired packet path with the CPU power to run the algorithms to determine the next best hop, in cooperation with the data

repositories. The SDN transmits this information to the data plane of individual routers, gateways and switches about aircraft using protocols such as OpenFlow™. In this way the airborne router doesn't have to have comprehensive network information. Instead, each data plane device needs only the information that details how to get the data packets to the next hop.

[0035] The data repository and computer software on land-based and airborne nodes cooperate to dynamically update the potential data paths and their locations as the airborne nodes change positions The land-based system delivers a complete model in real time of the network's end-to-end topology and distribution of aircrafts. The airplanes' routers will also have the future routes of all aircraft in the network, so even if the router is out of communication for a period and doesn't get regular updates, the routers will update once back in communication and load the correct routing information to connect to the correct current destinations. Through the SDN, the route of any network packet can be changed and managed in real time as shown in FIG. 3.

The SDN in cooperation with the big data repository uses algorithms to enable redundancy of network packet transmission and validation. The aircraft routers also forward latency, packet routing efficiency and other real-time information back to the data repository for further tuning of the routing tables. Additionally, the SDN in cooperation with the big data repository uses algorithms that would provide the least-cost forwarding of the packets and determine the best routes based on time, weather, local laws and other factors. By using an SDN, the management and all other operational aspects are centralized, and there is a high level of security controls. [0036] This infrastructure comprises an established set of data paths that packets can take, known in advance of transmission. Each data path comprises a set of established nodes, also known in advance of transmission. However, while the expected location of each of the nodes is known based on known aircraft routes and position of stationary land nodes, the exact location of the airborne nodes at a given time is not known in advance due to the variances of weather, airport delays, and other factors that change randomly or unsystematically, and in real time. The infrastructure uses real-time data to determine the location exactly, prior to packet transmission, so that the proper data path can be chosen and the connection made. The mesh network comprises many of these data paths.

[0037] In a simplified example, assume that a message is being transmitted from New York City to Mumbai on the present network. Fig. 4A-C shows planes in the network in flight; the direction and duration of future travel is indicated by the associated arrow. Planes that have landed and that are no longer active in the network at the given time are indicated in parentheses. As used herein, a hop is one portion of the path between the source and the destination, from a first node to a second node. For simplification of this example, intermediate hops between the nodes necessary to bridge the communication distance are not shown.

[0038] In advance of a packet being sent, the infrastructure knows from data previously loaded into the data repository that planes A-G are supposed to be airborne at Time 1. See Fig. 4A. At Time 2, planes A, B, C, F, G and K are supposed to be airborne, but planes D and E have landed. See Fig. 4B. At Time 3, only planes A and K are supposed to be airborne, and planes B, C, D, F, E and G have landed. See Fig. 4C.

[0039] For a message to be sent from New York City to Mumbai at Time 1 , the message could take the following data path N1 : New York City to plane C to plane D to plane E to plane G to Mumbai. In shorthand:

N1 = NYC - C - D - E - G - Mumbai

[0040] At Time 2, however, that data path using those nodes would not work because the nodes have moved and no longer form a data path from New York City to Mumbai. For example, plane C is no longer within communication distance from New York City, as shown in Fig. 4B. Furthermore, planes D and E have landed and therefore are not active in the mesh network. However, plane K is now airborne over New York City and can thus serve as the endpoint of the hop from NYC to a first airborne node. Thus, for a message to be sent from New York City to Mumbai at Time 2, the message could take the following data path N2: New York City to plane K to plane C to plane F to a land-based node in Moscow to another land-based node in Mumbai. In shorthand:

N2 = NYC - K - C - F - Moscow - Mumbai

[0041] At Time 3, the message would take the following data path N3: New York City to plane A to plane K to a land-based node in Vienna to a land-based node in Moscow to a land-based node in Mumbai. . In shorthand:

N3 = NYC - A - K - Vienna - Moscow - Mumbai

[0042] While the potential data paths are known in advance, because the airborne modes are movable, the system has to receive data in real time about the actual location of the airborne nodes in order to be able to determine whether the known data path will work. If real-time data indicates that one of the nodes will be out of place at a time necessary for message forwarding, the system will choose an alternate data path for the message. For example, if at Time 1 plane D has mechanical trouble and is grounded and therefore not active in the network at the expected time, the system is updated in real time and the path N1 is taken out of the potential data paths for a New York City-to-Mumbai message. Due to the number of nodes that could be substituted in various permutations to form other data paths, the message can still be transmitted in the infrastructure at Time 1. For example, the message could instead take the path:

[0043] N4 = NYC - C - London - Vienna - Moscow - G - Mumbai

[0044] For data being transmitted from New York to Mumbai through nodes a-z, all the hops would be determined proactively, before transmission is initiated. The data path changes in real time, and the physical layer turns off the routes to a node as the node becomes unavailable.

[0045] The system calculates a set of hops between nodes using data in the data repository that meet some or all of prescribed criteria. In one embodiment, the system loads a table of available paths to the final destination from a routing database as well as empirical statistical and real-time quality, speed, and cost criteria. Criteria may include for example the number of retries; number of packets with errors; number of packets that were discarded, lost, or retransmitted; average and maximum

jitter; average latency and maximum latency; cost by carrier; and time-of-day. Criteria may be associated with a particular user or with a particular category of user, such as residential users, business users, military users, etc. Some users may require the highest quality service regardless of cost; whereas other may require the absolute lowest-cost routes regardless, within reason, of service quality. Some users may require highest quality during a specific time of day which, if the desired time of day is during peak demand, may not be the lowest cost.

[0046] Example 1 - Initial Routes

[0047] As part of the normal power up of the electronics on an airplane, an airplane - as a soon-to-be airborne node - would contact a land-based node, which in turn would contact the data repository. All of the receivers and transmitters would be on line, and the land-based SDN router would forward the current and anticipated route information from the data repository to the airplane. The data repository would provide up-to-date information about the planned time of departure for all flights, eventual destination, and any weather issues already loaded. The exact location of the aircraft could be set by latency to other aircraft that it is connected to, and also GPS and other land-based receivers. Thus, the initial data paths would be set up even before the aircraft takes off, but until airborne there would be little or no packet traffic going through the airplane. A set of possible packet routes is calculated from the data in the data repository, including the plurality of possible hops from this airplane, the soon-to-be airborne node, and all available second nodes. The airplane is now ready to enter the mesh network as a node.

[0048] Ideally, once a data path is selected it will remain as the desired path until the packet reaches its final destination. Transit time is so fast that the minute changes in node location during the transmit time do not usually affect the continuity of hops. However the system can compensate for such minute changes by changing the data path during transmit, as well as if a node becomes disconnected

unexpectedly. For example, upon receipt of a packet the airplane's airborne router selects other nodes from the plurality of possible hops to a desired destination, according to prescribed criteria. The data path to the endpoint is updated in the data repository for each possible destination of the packets at the second node, and packets are forwarded to the best next hop at the second node, which becomes the first node in the next hop. Again a plurality of possible hops is continually calculated from the current node to the second node of the next hop. If the next node is airborne, too, that airplane's airborne router selects a second node from the plurality of possible hops according to prescribed criteria. The packet is transmitted from the first node to the second node of the second hop which has the current information about the best path to the final destination. This determines the next hop for the inflight packet. The process is repeated until the packet reaches the final destination.

[0049] Example 2. Data repository getting weather data and adjusting routes.

[0050] In another example of the flexible network, while the second airplane in the previous example is in flight, the weather service forwards information about a high altitude cloud that could reduce the distance that the network signals could travel. Data from other aircraft in the area could be used to validate that this indeed is happening. The routing information for the aircraft or land based nodes would be adjusted based on the weather information in real time, causing the plurality of potential packet routes to change accordingly. Packets, for example, might be

forwarded to another aircraft or ground station to mitigate any data loss due to the weather.

[0051] Example 3 - Loss of Connectivity

[0052] Planes will not have permanent connectivity to the rest of the network. So, for the node to know how when to reconnect and to which node, the routes of the future nodes are also loaded. For example, a portion of the network is implemented using commercial passenger aircraft traveling from Dubai to South Africa. At least one of the aircraft in the network also carries passengers over water to the Maldives. In doing so, it is beyond maximum aircraft-to-aircraft communication for about three hours so it doesn't have connectivity during that time. Once the Maldives flight gets to a location where it can connect to other aircraft, it has to have mapping to the other aircraft so that it can know where to direct its packet transmission to reconnect. These aircraft are different from the ones that it was communicating with when it lost connectivity. So, the present WAN would use the N+1 network map to determine the location of the new aircraft to contact after a period of no connectivity.

[0053] There are several advantages of the present invention over existing WAN infrastructure technology. For example, the WAN infrastructure 10 can provide unlimited bandwidth and worldwide coverage. It can dramatically reduce the cost of transmitting data between locations. Also, it can reduce the time it takes for a building or location to achieve WAN connectivity from days to hours or minutes. The system also provides the ability to manage the congestion in the overall network. For example, specific types of traffic can be routed based on latency

requirements. Other traffic can be rerouted based on managing the congestion between different points in the network. The system provides real time proactive

management to effectively utilize bandwidth, optimize profit and establish an overall class of service for the traffic

[0054] The capital investment to set up, modify, upgrade, and operate the system of the present invention is also much less than with existing WAN technology. Existing technology stacks can be incorporated into the WAN infrastructure 10, and the WAN infrastructure 10 easily can be upgraded and modified at any time. Network edges can be distributed with lower power requirements. Additionally with this WAN infrastructure 10 the costs of training people to deploy it are significantly reduced.

[0055] WAN infrastructure 10 provides a higher level of availability and reliability than currently available WAN infrastructures. Additionally, it allows central management, wider network distribution, and prediction of network and resource demand. Logical processing and algorithms are optimally removed from the network edges and globally centralized. Mapping is centralized and reduced to consolidated cores, global definitions are decoupled from the physical components and interfaces of the infrastructure, and extended dynamic any-location computing allocation is enabled. Further, the present invention is resilient from failures of network edges components and offers infinite routing and balance load capabilities.

[0056] The WAN infrastructure of the present invention can be used for consolidating data centers and pooling current static partitions of components. Also, the WAN infrastructure can be used for combining virtual computing and virtual networking to optimize resource allocations. The WAN infrastructure can be an Infrastructure as a Service (laaS) solution and can be used to develop large internet applications and to deploy global definitions for identity and policy management. It also can be used to distribute application firewalls and to redirect suspicious traffic to higher-level IDS/IPS controls and DLP security systems. The WAN infrastructure can

be used to connect buildings together in the same metropolitan area and to increase the bandwidth and lower the cost of transporting data between regions. Additionally, it can be used for in-flight videoconferencing and VOIP services, real-time GIS and ground imagery data provisioning, and underwater tracking of submarines and undersea wildlife.

[0057] The WAN infrastructure can be used to route data from any two nodes, whether both are airborne or land-based, or one airborne and one land-based. Each airborne node may also possess additional capacity for data storage and computing power to provide network capabilities between aircraft independent of the network, without needing to communicate with a land-based node. In this way a network of solely or primarily airborne nodes may be established, for example to be used as a network communication service between aircraft or as a buffer for aircraft


[0058] In addition to the specific embodiments described above, the network routing discussed here provides an application framework that is independent of any present (or future) networking protocol architecture. The methods described herein may be implemented in a combination of digital electronic circuitry and software residing in a programmable processor (for example, a special-purpose processor, or a general-purpose processor in a computer) in combination with storage media. Suitable storage media for tangibly embodying computer program instructions and data include all forms of non-volatile memory, and include by way of example, semiconductor memory devices; ROM and flash memory devices; magnetic disks such as internal hard disks and removable disks; magneto-optical disks; optical disks such as compact disks (CDs), digital video disks (DVDs), and other computer-

readable media. Any of the foregoing may be supplemented by, or incorporated in, a specially-designed ASIC.

[0059] While there has been illustrated and described what is at present considered to be the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made and equivalents may be substituted for elements thereof without departing from the true scope of the invention disclosed, but that the invention will include all embodiments falling within the scope of the claims.