HELP SAVE MESH NETWORKING —This application from General Electric seeks to patent a system for extending the range of communication for low rate wireless mesh networks.

Help narrow US patent applications before they become patents.... HERE!

QUESTION: Have you seen anything that was published before June 27, 2011 that discusses:

  1. A low-rate wireless mesh network that uses several directional antennas (communication nodes including: directional wSIM, any directional wireless sensor interface) to extend the range of wireless communication between devices; OR

  2. A method of assembling such a network; OR

  3. A sensor and controller connected to such a mesh network.

If so, please submit evidence of prior art as an answer to this question. Only one piece of prior art per answer below. We welcome multiple answers from the same individual.

EXTRA CREDIT --- A reference to anything that meets all of the criteria above AND ALSO uses one or more omnidirectional antennas. A reference to anything that meets all of the criteria above AND ALSO uses a memory device connected to a processor. A reference to anything that meets all of the criteria above AND ALSO uses two adjacent directional antennas with opposite polarization.


Summary: [Translated from Legalish into English] A system for increasing the range of communication between devices in low-rate wireless mesh networks by using multiple directional antennas (for example directional wSIM or any directional wireless sensor interface device).

Publication Number: US 2012/0326927

Assignee: General Electric

Prior Art Date: June 27, 2011

Open for Challenge at USPTO: open until June 27, 2013

Claim 1 requires each and every element below:

A wireless mesh network comprising:

  1. At least one data collection device; and

  2. A plurality of directional antennas operatively coupled to each other and to said at least one data collection device,

  3. Wherein said wireless mesh network is configured to transmit information with a data rate of approximately 250 Kilobits per second (Kbits/sec) or less, and at a frequency of approximately 2.4 Gigahertz (GHz) or less.

In plain English this could mean:

  1. At least one device that collects data input from any kind of sensor or user input interface (including, for example, a microcontroller programmed to collect, store and process input from sensors, user interfaces such as touch screens or keyboards, or GPS); AND

  2. Multiple mesh nodes using directional radio antennas configured in such a way that they are in communication with each other and with the device described in (1); AND

  3. The mesh network is configured to transmit information with a low-data rate, using either the Zigbee® specification, the WirelessHART™ standard, or any low-rate specification, standard or protocol based on IEEE standard 802.15.4™ [a data rate of approximately 250 Kilobits per second (250Kbits/sec) or less, at a frequency of approximately 2.4 Gigahertz (GHz) or less].

Claim 15 requires each and every element below:

A monitoring system comprising:

  1. At least one sensor measurement device;

  2. At least one data collection device; and

  3. A wireless mesh network coupled to said at least one data collection device and said at least one sensor measurement device, said wireless mesh network comprising a plurality of directional antennas operatively coupled to each other and to said at least one data collection device,

  4. Wherein, said wireless mesh network is configured to transmit information with a data rate of approximate 250 Kilobits per second (Kbits/sec) or less, and at a frequency of approximately 2.4 GigaHertz (GHz) or less.

In plain English this could mean:

  1. At least one sensor

  2. At least one device that collects data input from any kind of sensor or user input interface (including, for example, a microcontroller programmed to collect, store and process input from sensors, or user interfaces such as touch screens or keyboards, or GPS); AND

  3. Multiple mesh nodes using directional antennas configured such that the nodes are in communication with each other, the sensor, and the data collection device.

  4. The mesh network is configured to transmit information with a low-data rate, using either the Zigbee® specification, the WirelessHART™ standard, or any low-rate specification, standard or protocol based on IEEE standard 802.15.4™ [a data rate of approximately 250 Kilobits per second (250Kbits/sec) or less, at a frequency of approximately 2.4 Gigahertz (GHz) or less].

You’re probably aware of ten pieces of art that meet these criteria already... Separately, the applicant is claiming:

• A method and system using all of the elements described above and including multiple omnidirectional antennas;

• A data collection device composed of any memory device coupled to a processor, where the memory device includes programmed instructions to the processor to transmit data through the network;

• An arrangement of antennas where the first directional antenna is positioned within range of a first omnidirectional antenna; and a second directional antenna positioned within range of a second omnidirectional antenna.

What is good prior art? Please see our FAQ

Want to help? Please vote or comment on submissions below. We welcome you to post prior art you know about on other questionable US patent applications.

18 Answers 18


This page about the OLPC OX-1 laptop[1], dating from 2008, describes a mesh network with directional antennas. See point #11 under the heading "WiFi Power Save mode". To quote:

"11. While I am at it, directional antennae for the outermost nodes could dramatically enhance both range and power consumption. Item 9 above suggests that a high proportion of homes that would be "just out of range" could become "just within range" if they are able to point a directional antenna inwards. (Nodes closer to the school portal cannot use directional antenna because they need to relay traffic in other directions). "

This quote talks about nodes relaying traffic, and the rest of the page, and other pages on the OLPC XO-1 laptop [2], make it abundantly clear that the directional antennas are being discussed in the context of a mesh network.

In addition, the XO-1 allows a the mesh network to be rate limited to any desirable value, so it satisfies the criteria of 250kbit/s or less [3]. The referenced document talks about "connection rate limiting ". The OX-1's mesh network was operating at 2.4GHz [4]. The XO-1 also includes sensors, such as a microphone and video camera. The XO-1 distance measurement activity shows that it is a mesh connected measurement device [5]. The chat and record activities also use the mesh network to connect sensors and measuring devices between nodes [6,7].


[1] http://wiki.laptop.org/index.php?title=Talk:XO-1/Software_specification&oldid=130524

[2] http://wiki.laptop.org/go/Mesh_Network_Details

[3] http://wiki.laptop.org/go/OLPC_Bitfrost#P_NET:_network_policy_protection

[4] http://laptop.org/en/laptop/hardware/specs.shtml

[5] http://wiki.laptop.org/go/Acoustic_Tape_Measure

[6] http://wiki.laptop.org/go/Chat

[7] http://wiki.laptop.org/go/Record#Collaboration

| improve this answer | |
  • 2
    Excellent! The first thing I thought of when I saw the title was the XO laptop =) – user1675549 Jul 25 '13 at 10:35

Back in 2004-2005, I worked for a company (now defunct) called MeshTel, part of the Idealab family of companies in Pasadena, CA. We were granted US patent 7,626,966 (Ad hoc wireless communication system).

The system operated in the 900MHz ISM band at a fairly low datarate, definitely 1Mbit or below, but I cannot recall whether we were below 250Kbit. To extend range (or conserve power) our system was designed to automatically (ad hoc) use intervening nodes in the network.

During development, we frequently connected our system to a computer, which contained a data collection device to maintain logs of sensor data (mainly radio metadata) to improve our algoritms and design.

Although our system used omnidirectional antennas, it seems obvious to me (and would have in 2005) that fixed-position nodes could benefit from directional antennas. Our system was intended for mobile nodes.

| improve this answer | |

Waterloo Wireless was a local group from around 2002 in Waterloo, Ontario, Canada set up to implement a mesh network. They held a couple of antenna workshops; I do remember discussing plans for, and creating directional antennas and omni-directional antennas, but I'm not sure if they discussed multiple antennas. At any rate, the technology was 'obvious to a practitioner skilled in the art'.

Waterloo Wireless no longer exists, but I'm still in contact with some of the participants. There is a link to their site through Archive.org: http://web.archive.org/web/20020928173111/http://www.waterloowireless.org/

http://web.archive.org/web/20021012085158/http://www.waterloowireless.org/otherwireless.html lists many other organizations and projects for wireless and mesh networks in Canada.

--Bob Jonkman


| improve this answer | |

In wlan slovenija, open wireless network of Slovenia, we are using sensors attached to routers for quite some time now. But we are using WiFi to communicate between our network devices in the mesh network. We use multiple types of antennas. And we do use tools to help us deploy such networks. I am not sure if all this is of any help, but still I am listing here some references.

Already in 2010 we were organizing the workshop on how to attach sensors to WRT54GL routers and prepared some instructions for that. We have data collected from this sensor node since 2010. Data is collected on a central server over the mesh network through our node database system, nodewacher. It regularly connects to each node and downloads sensor data. This same system is also used when deploying new nodes to ease configuration of such nodes. So the node is a WiFi mesh node running Linux, which has on it some scripts which publish sensor data over HTTP, and this data is then collected by nodewatcher over the network into a central location and nicely drawn on graphs. In 2010.

In 2009 we also deployed a node powered by solar panel which was then sending its power measurements to our central server to draw the graphs.

Of course our network could run also on speeds of 250kbit/s or less, but we are running it on a higher speed. Why exactly running at slower speed is innovative to be patentable is beyond me. :-)

| improve this answer | |

In terms of establishing that the claims are "obvious to a skilled practioner" and "anticipated by the existing art" (does that apply in the US patent system?), see the about page for the Serval Project which is essentially unchanged since late 2010:


There we talk about, among other things:

"The second is a permanent system for remote areas that requires no infrastructure and creates a mesh-based phone network between Wi-Fi enabled mobile phones, and eventually specially designed mobile phones that can operate on other unlicensed frequencies, called Batphone. The two systems can also be combined."

This summarises our plans to combine various radios on different bands, and connect together different kinds of mesh devices.

I also recall our early discussions for air-dropped units that were to have been large egg-shaped devices with multiple directional antennae that would use low-bit-rate long range links to find each other, and then turn their antennae to point to each other. We drew some diagrams at the time, but I am not sure that I can find them.

Meanwhile, looking through what material we do still have for anything else.

| improve this answer | |

It's australia, but I don't care...


| improve this answer | |


Metricom created the ricochet wireless network, mesh networks, quite a few things related to this patent.

| improve this answer | |
  • yes, I believe that Metricom was one of the first to deploy Wireless Mesh Network Technology. The company was founded in 1985. The network provided the equivalent of a wireless modem, initially it operated at 9.6k baud but was later upgraded to 128k baud. see: en.wikipedia.org/wiki/Metricom – codeslinger Oct 30 '13 at 17:47

I actually have both references and work experience with Wireless Mesh. I last worked with them in 2007 so my access to their data is limited to commercial information.

3D-P Tech meets all the following:

  1. Variable Rate mesh network utilizing Motorola Cards inside a ruggedized box that is attached to an open source computer that stores and passes information on to a central monitoring unit. Current tech is up to 300 mbps max but is able to transmit much lower as needed. Also has GPS connectivity.

  2. Multiple methods for assembling the network provided including basic theory.

  3. Base unit is attached mounted to a vehicle with omni directional antenna that gives information including GPS location and information about the vehicle. This is relayed over the mesh network and relayed back to a central monitoring station that often goes through a backhaul system which is point to point.

I have included the links to diagrams and info at the bottom but here is something really important coming off their network diagram information.

3D-P Intelligent EndpointsTM 3D-P Intelligent Endpoints are ruggedized networking and open computing platforms designed for on-machine use. With multiple wired and wireless interface options, the Intelligent Endpoint offers the flexibility to meet your needs.

Motorola Access Points Motorola’s MOTOMESH and 802.11n access points provide optimal in-pit coverage. Motorola’s AP 7181 is a high performance, multi-radio device featuring a maximum data rate of 300 Mbps and Motorola’s unique ADEPT (Advanced Element Panel Technology) antenna system.

Cambium Networks Backhaul Cambium is the gold standard for point-to-point and point- to-multipoint radios and has deployed millions of units around the globe. With solutions capable of 300Mbps and a range of up to 120 miles, Cambium products offer carrier- grade reliability and unmatched performance for network backhaul and distribution.

Note that the endpoints offer multiple wired and wireless options that go through the whole spectrum of devices and endpoints available.

Main Website: http://www.3D-P.com

Extra Credit: System uses Multiple Omni Directional Antennas that vary in size based off wattage. System also uses multiple directional Dishes on the Backhaul that use polarity dependent upon noise in the mine. (I personally had to set the polarity on two TrangoBand Wireless Devices from a central point in a mine and one was +1 the other -1 so they wouldn't interfere with each other) The endpoints have a processor built in that runs off a Linux Debian Kernel that stores information of the device it is connected to.

| improve this answer | |

I found

A. Nasipuri, K. Li; A directionality based location discovery scheme for wireless sensor networks, Proceedings of the 1st ACM international workshop on Wireless sensor networks and applications, 2002

In this work, we present a localization technique by which the sensor nodes determine their position with respect to a set of fixed beacon nodes that are capable of covering the entire network area by powerful directional wireless trans- missions.

There is

  • wireless sensor network
  • several directional antennae
  • 2002

Unfortunately the directional antennae are not connected to a single data collection device. But they arguably are "operatively coupled" since they together achieve the localisation function.

| improve this answer | |

Oh boy this is vague but still worth something. An NYT article from June 12th 2011 that mentions antennas included in the Commotion wireless project sponsored by New America foundation and the US Department of State.

"The group’s suitcase project will rely on a version of “mesh network” technology, which can transform devices like cellphones or personal computers to create an invisible wireless web without a centralized hub. In other words, a voice, picture or e-mail message could hop directly between the modified wireless devices — each one acting as a mini cell “tower” and phone — and bypass the official network...the suitcase would include small wireless antennas, which could increase the area of coverage; a laptop to administer the system; thumb drives and CDs to spread the software to more devices and encrypt the communications; and other components like Ethernet cables."

Commotion wireless press page. https://commotionwireless.net/press

| improve this answer | |

Search Wireless Mesh Networks on Google Scholar, or any online journal repository and you'll have academic research dating back over a decade


| improve this answer | |

Somebody already mentioned Commotion Wireless, but I thought that linking to their first substantial source code commit on 10 December 2010 at Github would provide corroborating evidence that they'd begun work on this project prior to the date on the patent filing.

| improve this answer | |

In 2007, the WiLDNet group at Berkeley published a paper using Wi-Fi with directional antennas (multiple antennas per site) to create a long-distance (extended range) wireless network. They call it "high-throughput" but in reality it essentially meets your low rate specs.


In 2009, UCLA published a paper that used a mesh sensor networks to achieve long-range links. http://www.gps.caltech.edu/~clay/PDF/Martin_Lukac_Paper2.pdf

This 2004 survey may also prove useful: http://www.sciencedirect.com/science/article/pii/S1389128604003457

| improve this answer | |

An additional piece of prior art: Project Byzantium (a working group of HacDC) started work on a live distribution of Linux called Byzantium Linux on 26 February 2011 which seems to fit some of the criteria of this patent:

  • A method of assembling a network to extend the range of wireless communication between devices. This project had a web-based control panel (initial commit dated 2011-05-31) and now uses an automatic configuration daemon (initial commit dated 2012-11-06, which automates the node assembly and configuration process of the control panel).
  • At least one device that collects data input from any kind of sensor or user input interface. Byzantium Linux allows users sitting at console to use the mesh for communication in addition to providing network access for mobile devices.

Here is the first substantial commit to the source code repository on 26 February 2011. The first development sprint was documented and published on 15 March 2011, but actually took place over the weekend of 25 February 2011 when two mesh networks running two pre-existing mesh routing protocols (Babel and BATMAN-adv) were constructed and tested with the usual network troubleshooting tools (tcpdump, ping, traceroute) and functional tests (SSH from node to node, HTTP from node to node).

Full disclaimer: I am one of the core developers of this project.

| improve this answer | |

This may not help, either, but the mesh routing protocol OLSR (Open Link State Routing) was published as RFC 3626 by the IETF in October of 2003. The first substantial commit to olsrd's source code repository was on 2004-09-07.

| improve this answer | |

One more mesh routing protocol, which I think fits the description of "programmed instructions to the processor to transmit data through the network": Babel.

While the Babel mesh routing protocol wasn't introduced as an RFC to the IETF until April of 2011, the first release of the codebase (v0.1) was published on 2007-08-22 and the first substantial commit to their source code repository was on 2007-04-19.

| improve this answer | |

Some work I did on patent #6,584,080 Wireless burstable communications repeater granted in 2003 covered most of this territory, around 2.4GHz and targeting variable (but usually higher) speeds. Directional plus omnidirectional antenna mixes to extend the network outward into remote spots, putting a processor and data collection device in there, this stuff was barely state of the art in 1999 when that was filed. The sensors collecting data were actually parts of the network QoS system. I pointed out in there that you can run the concept at multiple speeds, only specifically mentioning the popular at the time 1 and 2MB/s versions though.

| improve this answer | |


A cellular communications system is provided have both satellite nodes and surface nodes for providing mobile cellular communications services for a plurality of mobile user units. The surface and satellite nodes are fully integrated by a network controller for providing service over large areas. In the alternative, each user unit includes link control means permitting the user unit to designate the mode of communications, either satellite node communications or ground node communications. In addition, multiple beam, relatively high gain antennas are disposed in the satellite nodes to establish satellite cells having enough gain in the satellite part of the system such that a user unit need only comprise a small, mobile handset with a non-directional antenna for communications with both ground nodes and satellite nodes.

1998 Also, look at the diagram at the above link.

The linked patent has:

-- multiple omni-directional and directional antennae in the same system.

-- a sensor (voice microphone) on a cellular communications device (cellphone)

-- the mobile cellphone converts the analog signal from the microphone into digital data, and processes this data using a processor. [This is how the digital cellphone system works].

-- the cellphone collects multiple analog signals from the microphone into batches of data before transmitting them to the network. This means it utilises short term storage using a memory device coupled to a processor. [This is how the digital cellphone system works].

-- The processor includes programmed instructions to transmit data through the cellphone company's network.[This is how the digital cellphone system works].

-- It complies with the speed and frequency requirements: "configured to transmit information with a data rate of approximately 250 Kilobits per second (Kbits/sec) or less, and at a frequency of approximately 2.4 Gigahertz (GHz) or less." Older cellphones transmit their data using data rates around 12Kbits/sec to 24Kbits/sec. This complies with the "..or less" part of the "250 Kilobits per second" bandwidth requirement. Even newer voice+data cellphones usually don't exceed 250 Kbits/sec. Cellphone frequencies are typically 800 MHz, 900 Mhz, 1200 MHz, 2100 MHz. All of these comply with the "..or less" part of the "2.4 Gigahertz" frequency requirement.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.