Dynamic real-time TV white space awareness

Abstract

This invention describes a cloud-based architecture that orchestrates the detection and dissemination of highly-dynamic, real-time, and fine-grained TV white space information to improve spectrum information used for White Space Devices (WSDs). Wasted spectrum opportunities were first identified, both temporal and spatial, due to the current approach of white spaces detection. This invention introduces a next generation of geo-location databases capable of tracking the state of the relatively static TV transmitters and the highly dynamic TV receivers. A quantitative evaluation of the potential gain in white space availability for large scale deployments of this novel architecture demonstrated significant improvement in the availability of white spaces.

Claims

1. A method, implemented as software running on a computer, for determining free white space opportunities in a broadcast spectrum comprising the steps of: a. consulting a geo-location database containing information regarding TV transmitters; b. defining and identifying a potential cell for use as free white space cell based on data in said geo-location database; c. collecting spectrum sensing data from one or more white space devices located within said identified potential cell; d. collecting ambient sensing data from within said identified potential cell, said ambient sensing data including visual and acoustic fingerprints of a TV receiver, and comparing said visual and acoustic fingerprints to known fingerprints of broadcast TV programs to determine the channel to which said TV receiver is tuned; and e. declaring said identified potential cell to be a free white space cell for a particular channel when it is determined with a pre-determined confidence level that TV receivers within said identified potential cell are not receiving a broadcast TV signal for said channel.

2. The method of claim 1 wherein said step (a) further comprises obtaining static information about TV transmitters including location, transmission power, antenna height and channels being transmitted.

3. The method of claim 1 wherein said spectrum sensing data consists of information contributed by enabled white space devices regarding sensed interference on each channel.

4. The method of claim 3 wherein said information contributed by white space devices is pushed by said white space device to said computer.

5. The method of claim 3 wherein said information contributed by white space devices is pulled from said white space device by said computer.

6. The method of claim 1 wherein said step of collecting ambient sensing data includes collecting data from sensors on smartphones located within said identified potential white space cell.

7. The method of claim 5 wherein said step of collecting ambient sensing data includes collecting data from embedded sensors located within said identified potential white space cell.

8. The method of claim 1 wherein said ambient sensing data includes data collected from cameras, microphones and accelerometers.

9. The method of claim 8 wherein said ambient sensing data is used to detect TV viewer behavior, including determining whether said viewer is sitting, using the remote or texting about TV shows.

10. The method of claim 1 further comprising the step of constructing a watching profile for a viewer based upon the collected ambient sensing data.

11. The method of claim 10 wherein said watching profile can be used to predict when a particular TV receiver may be On or Off and to which channel said TV receiver may be tuned.

12. A system for determining free white space opportunities in a broadcast spectrum comprising: a computer, said computer having access to the internet and running software performing the functions of: a. extracting data from a geo-location database containing information regarding TV transmitters; b. defining and identifying a potential cell for use as free white space cell based on data in said geo-location database; c. collecting spectrum sensing data from one or more white space devices located within said identified potential cell; d. collecting ambient sensing data from within said identified potential cell, said ambient sensing data including visual and acoustic fingerprints of a TV receiver, and comparing said visual and acoustic fingerprints to known fingerprints of broadcast TV programs to determine the channel to which said TV receiver is tuned; and e. declaring said identified potential cell to be a free white space cell for a particular channel when it is determined with a pre-determined confidence level that TV receivers within said identified potential cell are not receiving a broadcast TV signal for said channel.

13. The system of claim 12 wherein said spectrum sensing data consists of information contributed by enabled white space devices within said identified potential white space cell regarding sensed interference on each channel.

14. The system of claim 13 wherein said ambient sensing data includes data from embedded sensors located within said identified potential white space cell comprising visual and acoustic fingerprints of a program being displayed by a nearby TV receiver.

15. The system of claim 14 wherein said visual and acoustic fingerprints can be compared to known visual and acoustic fingerprints of programs being broadcast by said TV transmitter to determine the channel to which said TV receiver is tuned.

16. The system of claim 13 wherein said ambient sensing data is collected from sensors of a smartphone located within said potential white space cell.

Description

DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 summarizes three different scenarios in which white spaces opportunities were lost.

(2) FIG. 2 illustrates an overview of the system's operation scenario.

(3) FIG. 3 illustrates the system's operation flow.

DETAILED DESCRIPTION OF THE INVENTION

(4) The identified wasted spectrum opportunities requires alterations in conventional geo-location databases to evolve from storing and processing the relatively static information regarding TV transmitters (e.g. location, transmission power, antenna height and channel), to collect, process and store the highly dynamic, real-time information regarding TV receivers reported by equipped WSDs. To address this evolution, the scalability and reliability of cloud storage and processing is exploited. A new architecture is presented to generate real-time, dynamic geo-location databases via a cloud-based system that is responsible for processing and aggregating sensory information on TV receivers. This information is different in: 1) nature (e.g. spectrum and ambient sensory information); and 2) reliability metrics (e.g. credibility of certified geo-location databases compared to contributing sensors). This architecture aims to address the expected bandwidth demand on white space networks by tracking every available spectrum opportunity in real-time.

(5) System Architecture

(6) Overview and Operation.

(7) FIG. 2 gives an overview of the invention. Contributing devices 100, equipped with localization mechanisms, submit location tagged spectrum sensing data (from WSDs) or ambient sensing data (from standard cell phones and embedded sensors) to the cloud infrastructure of the system. The system divides the map into fine grained cells 120 and calculates white space availability for each cell separately based on the collected sensory information. Cells may be of any desirable size, ranging from 1 cm.sup.2 to a single room to an entire district. In practice, cell size is dependent on several factors, including, for example, system complexity and density of devices. It is also possible to have a hierarchy of cells, where a large cell can be composed of smaller cells.

(8) A cell is considered a “free cell” 140 with respect to a certain channel in two cases. The first case occurs when there is no TV reception of the channel within the cell (i.e., no TV signal within the particular cell or all TV sets in the cell turned off). The second case occurs when the cell has no TV sets that are tuned to the channel (i.e., TV set may be tuned to different channels). The system supports both push and pull modes for obtaining sensory information. The two modes enable the system to enhance its view of the spectrum, through either continuous pushes or on demand pulls from available contributors. It should be noted that the system will default to traditional geo-location databases when there is no sensory information available for a particular cell.

(9) The multi-layered operation flow, shown in FIG. 3, is divided based on the different types of available white space opportunities. These opportunities vary in terms of persistence and reliability. For each White Spaces Information Query 210, the system aims to obtain the most persistent and reliable availability information.

(10) The first layer 220 consults conventional geo-location databases to identify white space opportunities based only on the parameters of licensed TV transmitters. This layer presents white space information with the highest confidence and persistence.

(11) The second layer 230 checks contributed spectrum sensing readings which lie within the white space network's coverage area given its transmission power. If there is no contributed spectrum information found, the system will pull real-time spectrum information on demand. This layer can increase the number of available white spaces or help the network enhance its transmission parameters based on the sensed interference on each channel. The white space information obtained from this layer is highly persistent as the view of the spectrum is not likely to change fast. However, this sensory information needs to be confirmed by more than one contributing device to establish its credibility.

(12) The third layer 240 detects the presence and the state of TV sets within the white spaces network's coverage area to seek more white space opportunities (i.e. a TV transmitter is operational with no TV sets tuned to its channel). This type of white space information is highly volatile because TV viewers switch channels randomly. Moreover, to increase the confidence in such information more than one contributor needs to support the same sensor based decision.

(13) It is envisioned that as TV sets become more “aware” and are equipped with the capability to connect to the internet, that individual TV sets will be able to report their status directly to the cloud, thereby providing more reliable information on which cells may be free.

(14) Note that, in the event that conventional geo-location databases are unable to be modified to contain spectrum sensing data from WSDs and ambient sensing data from standard cell phones and embedded sensors, this secondary information may be kept in one or more separate databases.

(15) Of note is that the information collected over time from multiple devices can determine the set of channels typically viewed at a particular TV receiver, ultimately increasing the persistence of the spectrum availability data but reducing the granularity of channel availability. In addition, the TV viewer's watching profile can be computed and stored in the cloud, which can then be used in the spectrum availability estimation.

(16) White space opportunities, obtained from each layer, are associated with a validity period to improve the system's reliability. The WSD is required to re-check with the database for the white space's availability after the validity period expires. This period depends on both the persistence level and confidence associated with each reading. These levels are calculated based on the number of contributors conforming on the same reading as well as the source of the sensed information. Upon the reception of the system's response to its query, the WSD can determine the channels to transmit on, based on its required quality of service and the validly period of each available white space.

(17) WSD Spectrum Sensing.

(18) In one example of a novel contribution model for the new database, signal readings are collected from contributing spectrum sensing enabled WSDs. This architecture supports obtaining either a partial or full view of the TV bands spectrum from contributing devices along with their accurate location information. Spectrum sensing enabled devices as specified by the FCC are either sensing only devices, that depend solely on spectrum sensing to detect available channels, or, as the FCC encourages, mixed devices that use spectrum sensing with geo-location databases. Another source of spectrum information includes dedicated sensing infrastructures that could be deployed to enhance spectrum utilization in spectrum starved areas. Several cooperative spectrum sensing algorithms can be used to make the best use of collected spectrum sensory information.

(19) FCC regulations for sensing only devices require initial sensing of 30 seconds to select the channel to be used. After that, these devices are required to ensure that the channel is vacant every 60 seconds. While these regulations ensure protection of TV spectrum incumbents, they require a mobile device to regularly check to ensure the consistency of the spectrum view.

(20) In contrast, the described system's interest lies in determining the spectrum view for each statically determined cell and the cell's area is smaller than the approved sensing only device coverage area. Therefore, the spectrum sensing requirements for this system are more relaxed.

(21) TV Set Awareness Using Standard Mobile Devices.

(22) In another example, sensory information collected by standard mobile devices (cell phones, laptops, tablets, etc) can be used as unconventional spectrum sensors to deduce whether a TV set is available within an environment or not. If a TV set is detected, regular checks are performed to detect the channel to which the TV is currently tuned. Sensors like cameras, microphones, accelerometers, etc. can be used to detect both TV set behavior (e.g. visual and acoustic fingerprints) and TV viewer behavior (e.g. sitting, using the remote and texting about TV shows). This information could be used to identify the presence of TV sets and whether they are ON or OFF. Moreover, using online streaming sites and channel guides, the channel currently playing could be detected using only the acoustic fingerprint of the TV set.

(23) Smart infrastructures (e.g. smart homes) can also be used for inferring information as they are equipped with sensors designated for different functionalities required to improve the quality of people's life. These functionalities include TV controls which can be directly used to update the system. Another form of sensors are the new generation of Internet-enabled smart TV sets that can be used to update the system's database in real-time with information on the channels being viewed. These sources of information could reliably tell whether there is a TV that is ON in a certain area of interest and the channel to which that TV is tuned.

(24) By accumulating and aggregating detected channel information for each TV set, collected from different sources with different confidence levels, and correlating them with time, high confidence TV set detection decisions can be reached. The system then becomes capable of constructing or estimating a watching profile for a viewer to predict when a particular TV may be in use and to which channel the TV may be tuned. This stochastic behavior can further be leveraged to enhance the system's decisions and confidence estimation.

(25) Results and Discussion

(26) Simulations were conducted for Miami City, Fla. and New York County in New York, to illustrate the potential gain of spectrum availability in white spaces in urban areas. According to the United States Census Bureau, New York County, N.Y. has 732,204 households within an area of 22:83 mi.sup.2 and Miami City has 149,077 households within an area of 35 mi.sup.2. In this simulation, these households were distributed uniformly over the two areas, assigning one TV set for each household. 21% of the TV sets were randomly picked as operational to account for the statistics showing that the average American watches TV about 5.2 hours a day. Then, 10% of the TV sets were selected to be tuned to a broadcast channel. One of the broadcast TV channels in the designated area was assigned to each of the TV sets (27 channels in New York County and 26 channels in Miami City).

(27) WSDs were distributed over the two areas to measure the potential increase in the number of available channels. Then, the amount of free white spaces on which each WSD can operate without violating the FCC's protection criteria for the TV sets was measured. The protection criteria for co-channel transmission was selected to be 23 db SNR and −33 db SNR for adjacent channel transmissions. The Okumura-Hata model for urban areas was applied to identify the separation needed between the WSD and the TV set to maintain the minimum field strength of 41 dbu for TV service at the TV set, which is specified by the FCC. The simulation was made for the following transmission powers: 1 mW used in experiments conducted for local area white space networks in. 5 and 10 mW account for possible transmission powers that could be used to increase white spaces availability and range. 40 mW which is the maximum transmission power specified by the FCC for WSDs working in within a TV stations protected service area broadcasting on adjacent channel. 100 mW which is the maximum allowed transmission power by the FCC for portable WSDs.

(28) The awareness of the geo-location database of these different transmission powers presents another enhancement that enables the detection of white spaces relative to the WSD's transmission power. This results in avoiding the assumption of current geo-location databases that the WSDs can work only on two levels of power (i.e. 40 mW and 100 mW). Table 2 summarizes the parameters for each of the different transmission obtained using the Okumura-Hata model for urban areas.

(29) TABLE-US-00002 TABLE 2 Parameters used in the simulations for TV band devices with transmission powers 1, 5, 10, 40, and 100 mW in terms of the maximum coverage distance, the minimum distance between the TV band device, and the TV set to avoid interference. Min. separation distance Min. separation distance Cover- for adjacent channel for co-channel Power age transmission transmission  1 mW  59 m   9 m 182 m  5 mW  86 m 13.2 m 265 m 10 mW 101 m 15.5 m 310 m 40 mW 140 m 22.4 m 430 m 100 mW  173 m 26.4 m 533 m

(30) The results of the simulation are summarized in Table 3. There is a significant difference between the two cities due to the difference in population densities. Comparing the spectrum availability of the new system to conventional geo-location databases (Table 1) for Miami City, devices working with any transmission power obtain at least 12.2 extra channels instead of no channels in case of 100 mW for conventional geo-location databases. Similarly, in New York County, despite the high density of TV sets, almost 20% of devices working with 100 mW transmission power obtain 1.27 channels on average to work on instead of not being able to work at all when using the conventional geo-location databases.

(31) TABLE-US-00003 TABLE 3 Simulation of potential gain in white spaces for WSDs working in Miami City, Florida and New York County, New York with transmission powers of 1, 5, 10, 40, and 100 mW given information on TV sets locations and the channels they are currently tuned to. Percent of WSDs Average number of Percent of WSDs Average number of gaining more channels in channels gained in gaining more channels in channels gained in Power New York County New York County Miami Miami  1 mW  100% 9.65 100% 23.4  5 mW .sup. 99% 4 100% 21 10 mW .sup. 92% 2.7 100% 19.6 40 mW 49.8% 1.53 100% 15.5 100 mW  19.9% 1.27 100% 12.2

(32) As the transmission power is lowered to increase the bandwidth available for enterprise local area networks, the number of available channels increases, reaching 23.4 in Miami and 9.65 in New York County for all devices working with 1 mW (i.e. having a coverage of 59 m). This highlights the significant gains that can be achieved through a city scale deployment of the invention described herein.

(33) In conclusion, this invention can detect available white space opportunities through a multi-layered architecture. This architecture uses unconventional spectrum sensing, supported by conventional spectrum sensing and geo-location databases to detect spectrum opportunities that are diverse in reliability and persistence. This system exploits the cloud's communication, computational and storage capabilities to infer available white spaces. This approach was shown to provide significant enhancement to current conventional approaches (i.e. geo-location databases) in city wide deployments.

(34) The present invention has been described in accordance with several examples, which are intended to be illustrative in all aspects rather than restrictive. Thus, the present invention is capable of many variations in detailed implementation, which may be derived from the description contained herein by a person of ordinary skill in the art. The scope of the invention is captured in the claims below.