WO2006014344A2

WO2006014344A2 - Wireless monitors for audience measurement and methods of determining a distance between a wireless monitor and an informaiton presenting device

Info

Publication number: WO2006014344A2
Application number: PCT/US2005/023478
Authority: WO
Inventors: Venugopal Srinivasan; Morris Lee
Original assignee: Nielsen Media Research, Inc.
Priority date: 2004-07-02
Filing date: 2005-06-29
Publication date: 2006-02-09
Also published as: WO2006014344A3

Abstract

Wireless monitors for audience measurement and methods of determining a distance between a wireless monitor (206) and an information presenting device (106) are disclosed. A method comprises obtaining first audio information via a first apparatus (204), obtaining second audio information via a second apparatus (206) wherein the second audio information is associated with the first audio information, determining a propagation delay based on the first audio information and the second audio information, and determining a distance between the information presenting device (106) and the second apparatus (206) based on the propagation delay.

Description

WIRELESS MONITORS FOR AUDIENCE MEASUREMENT AND

METHODS OF DETERMINING A DISTANCE BETWEEN A WIRELESS MONITOR AND AN INFORMATION PRESENTING

DEVICE

RELATED APPLICATION

[0001] This patent claims priority from U.S. Provisional Application Serial No. 60/585,116, entitled "Wireless Microphone as a People Meter for Audience Measurement" and filed on July 2, 2004. U.S. Provisional Application Serial No. 60/585,116 is hereby incorporated by reference in its entirety.

FIELD OF THE DISCLOSURE

[0002] This disclosure relates generally to audience measurement and, more particularly, to wireless monitors for audience measurement and methods for determining a distance between a wireless monitor and an information presenting device.

BACKGROUND

[0003] The ability to determine the composition of an audience in a panel household (i.e., a monitored household) at any instant of time is a desirable feature of a television audience measurement system. Some traditional methods to determine the identity and number of audience numbers enable each audience member of a monitored household to "log in" when television consumption begins and to "log out" when television consumption ends. Corresponding known metering systems include a metering device configured to periodically prompt the audience members to perform a "log in" function. Additionally, some known metering systems are configured to prompt audience members to respond via, for example, a remote control device when a channel change is detected by the metering device. Prompting may be accomplished via a display of one or more light-emitting diodes (LEDs) configured to flash at the appropriate times.

[0004] More recently, audience measurement systems which determine audience composition without requiring audience members to log in and log out via a remote control unit have been proposed, thereby reducing the intrusion to an audience member's viewing experience. For example, some prior-art systems employ a portable people meter ("PPM") which is a pager-sized device powered by a chargeable battery and configured to be worn or carried by an audience member. In general, a PPM may be configured to perform television monitoring using any number of known techniques including, for example, signature generation and/or code detection. A PPM configured to detect codes may be adapted to detect a television audio signal via a microphone and use a digital signal processor (DSP) to extract imperceptible watermarks embedded in the audio signal. The embedded imperceptible watermark is typically injected into the audio signal at the broadcast facility of a TV station and may contain unique station identification information and/or program identification information that may be collected and/or stored by the PPM. Each PPM in use may associate its collected data with the audience member wearing the particular PPM and the time/date when the data is collected. Thus, when the data is exported from the PPM to a central processor system, the collected data may be associated with a particular audience member to determine the viewing activities of that audience member. However, a drawback of such a PPM device configured to detect an embedded audio watermark is its sensitivity. For example, an audio signal emanating from a television located within a room adjacent to the room in which the PPM is located may trigger the PPM to detect a watermark within that audio. Detection of this watermark may result in erroneously crediting a television program as being viewed by the audience member wearing the PPM, even though the audience member was not within the viewing range of the particular television set producing the detected audio signal.

[0005] Additionally, systems exist which employ electronic tags ("tags") of various types to monitor the location of objects, including people. Applicable tags are typically passive devices powered with energy provided by high-energy radio frequency (RF) bursts emitted by strategically-located tag readers. For example, a passive tag absorbs energy from the high-energy RF burst and uses it to generate a responsive RF burst. The responsive RF burst may contain embedded data that includes identification information uniquely associated with that tag. However, while most tags provide a simple "yes/no" response to whether the tag is in the vicinity of a tag reader, the tags typically do not provide location information. Some recent tags do provide location information in the form of distance information. Such distance information corresponds to the distance between the tag and the tag reader in communication with the tag, but the determined distance is typically limited to an accuracy of only about one meter. Furthermore, these tags are expensive and may require the installation of an undesirable number of tag readers and/or RF transceivers within a household.

BRIEF DESCRIPTION OF THE DRAWINGS [0006] FIG. 1 is a plan view of an example monitored household. [0007] FIG. 2 is a detailed view of an example metering system that may be used to determine a distance between an audience member and a monitored information presenting device.

[0008] FIG. 3 is a block diagram of an example base unit for use in the example metering system of FIG. 2.

[0009] FIG. 4 is a block diagram of an example adaptive audio filter for use in the example metering system of FIG. 2. [0010] FIG. 5 is a graph illustrating example adaptive audio filter coefficients corresponding to the example audio filter of FIG. 4.

[0011] FIG. 6 is a block diagram illustrating an example manner of implementing the base unit of FIG. 3.

[0012] FIG. 7 is a block diagram illustrating an example manner of implementing a wireless monitor.

[0013] FIG. 8 is a flowchart representative of example machine readable instructions which may be executed by the processor of FIG. 7 to implement the example wireless monitor.

[0014] FIGS. 9A-9B are a flowchart representative of example machine readable instructions which may be executed by the processor of FIG. 6 to implement the example base unit of FIG. 3.

DETAILED DESCRIPTION

[0015] Although the following discloses example systems including, among other components, software executed on hardware, it should be noted that such systems are merely illustrative and should not be considered as limiting. For example, it is contemplated that any or all of these hardware and software components could be embodied exclusively in hardware, exclusively in software, exclusively in firmware, or in any combination of hardware, firmware and/or software. Accordingly, while the following describes example systems, persons of ordinary skill in the art will readily appreciate that the examples provided are not the only ways to implement such systems.

[0016] The example systems, methods, and apparatus described herein may be used to implement an audience member metering system configured to determine whether audience members are sufficiently exposed to a media presentation to be counted as audience members. Specifically, the example systems, methods, and apparatus are configured to determine a distance between an audience member and a monitored information presenting device, such as a monitored television, radio, stereo, personal computer, etc. In this manner, a media exposure research system may more accurately credit media content as having been consumed by an audience member based on whether the audience member was sufficiently close to the information presenting device to adequately consume the media content. [0017] The example systems, methods, and apparatus described herein are configured to generate distance information indicating the proximity of audience members to information presenting devices so that subsequent analysis of the distance information can be used to determine whether a particular audience member was likely to have consumed programs presented by the information presenting device. For example, the distance information may indicate that the audience member was within the same room as a television, but so far away from the television that the audience member was not sufficiently exposed to the television to be considered as having consumed the television programs. Media research studies may use this information to more accurately credit television programs as having been consumed by audience members.

[0018] The example audience monitoring system 200 of FIG. 2 includes a wireless monitor 206 and a base unit 204. The wireless monitor 206 and base unit 204 are communicatively coupled and may be configured to execute a plurality of machine- executable programs, such as those represented by the flowcharts of FIGS. 8 and 9A-9B. In general, the example audience monitoring system 200 determines a distance between an audience member and a monitored information presenting device based on the propagation delay of an audio signal emanating from the monitored information presenting device and subsequently received by the wireless monitor 206 carried by the audience member. [0019] FIG. 1 is a plan view of an example monitored household 100. The example monitored household 100 includes a first room 102 and a second room 104. The first room 102 includes a television 106, an example base unit 204 and a first example wireless monitor 206 configured to be worn or carried by a first audience member 112. The second room 104 includes a second audience member 114 carrying or wearing a second wireless monitor 206. The example base unit 204 and the wireless monitors 206 form an example audience metering system 200 configured to determine, in this example, a first distance between the first wireless monitor 206 and the television 106 and a second distance between the second wireless monitor 206 and the television 106. The base unit 204 may determine, for example, in real-time or through post processing, whether either of the audience members 112 or 114 was sufficiently close to the television 106 to consume television program content presented by the television 106. More specifically, in the example scenario depicted in FIG. 1, the first distance determined by the base unit 204 may be used to ascertain that the first audience member 112 was located in the first room 102, the same room as the television 106, and, therefore, was sufficiently close to the television 106 to consume program content presented thereby. Furthermore, the second distance determined by the base unit 204 may be used to ascertain that the second audience member 114 was located in another room 104 and, therefore, was not sufficiently close to the television 106 to consume the program content. Given such distance information, the television program content can be properly credited as being consumed by the first audience member 112, but not by the second audience member 114.

[0020] FIG. 2 depicts the example audience metering system 200 of FIG. 1 in more detail. As discussed above, the audience metering system 200 may be used to determine a distance between an audience member, such as audience member 112 of FIG.1, and a monitored information presenting device 202, such as the television 106 of FIG. 1. The example audience metering system 200 includes an example base unit 204 and an example wireless monitor 206. The example base unit 204 is communicatively coupled to the information presenting device 202 and configured to detect and process audio signals from the information presenting device 202 corresponding to an audio portion of presented television program content. For example, if the information presenting device 202 is a television 202, the base unit 204 may be communicatively coupled to an audio out interface 208 of the television 202 via a cable 210 to obtain a line level audio signal from the television 202. Alternatively, the base unit 204 may be communicatively coupled to speakers 212 of the television 202 via wires 214 to obtain speaker level audio signals from the television 202. As a further alternative, a microphone (not shown) may be placed proximate to the speakers 212 to obtain an audio signal corresponding to the program presented by the television 202. [0021] In the illustrated example, the base unit 204 includes an antenna 216 to permit communication with the wireless monitor 206 via an antenna 218. Preferably, the base unit 204 and the wireless monitor 206 communicate wirelessly via the antennas 216 and 218 using frequency modulation (FM) in the 900 MHz or 2.4 GHz bands. Alternatively, the base unit 204 and the wireless monitor 206 may communicate wirelessly using any suitable wireless communication protocol, such as, for example, Bluetooth®, Wi-Fi® (i.e., 802.1 1), etc. [0022] The wireless monitor 206 is configured to process audio portions of program content presented via the speakers 212. In the illustrated example, the wireless monitor 206 includes a microphone 220 to receive an audio signal output via the speakers 212. The audio signal received by the wireless monitor 206 corresponds to the audio signal received by the base unit 204 via the cable 210 or the wires 214. As described in greater detail below, the example audience metering system 200 may obtain a first audio signal containing first audio information and corresponding to an audio content presentation at the base unit 204 co- located with the information presenting device 202, and a second audio signal containing second audio information and corresponding to the same audio content presentation at the wireless monitor 206. In other words, the second audio information and the first audio information correspond to the same audio content, but the second audio information is perceived at the location of the wireless monitor 206 and the first audio information is perceived at the location of the information presenting device 202. The wireless monitor 206 transmits its audio signal to the base unit 204 using, for example, an FM transmission as discussed above. The base unit 204 then determines a propagation delay between the first audio signal and the second audio signal. The propagation delay is associated with an amount of time required for the second audio signal to travel from the speakers 212 to the wireless monitor 206. The base unit 204 may then determine a distance between the information presenting device 202 and the wireless monitor 206 based on the propagation delay.

[0023] A block diagram illustrating an example base unit 204 is shown in FIG. 3. In the example base unit 204 of Fig. 3, audio data corresponding to an audio signal detected and transmitted by a wireless monitor 206 is received by a wireless receiver 504 located, for example, within the base unit 204. Preferably, the base unit 204 samples and digitally processes an analog audio signal received by the wireless receiver 504 to produce a wireless monitor audio input 506. Alternatively, in the case of reception of a digital audio signal, the base unit 204 may produce the wireless monitor audio input 506 by reading digital audio samples output by the wireless receiver 504. The base unit 204 also similarly produces a reference audio input 508 based on an audio signal from an audio interface 510 corresponding to, for example, a television line audio output sampled and converted to a digital representation by an AID converter (not shown). As discussed above in connection with FIG. 2, the reference audio input 508 may be obtained through a line output of the television, wires connected to the television speaker circuitry, and/or a microphone placed in proximity to the television speaker or speakers.

[0024] The example base unit 204 includes an adaptive audio filter 512 which is configured to subtract the reference audio input 508, corresponding to an audio signal emitted by the information presenting device 202, from the wireless monitor audio input 506, corresponding to the audio signal detected at the location of the wireless monitor 206, with the goal of yielding a null output by suitably modifying the reference audio input 508. Generally, a perfect subtraction will not be possible because the wireless monitor may also detect other ambient audio, including, for example, human speech. As discussed in greater detail below, the adaptive audio filter 512 includes filter coefficients (or weights) which operate upon the reference audio input 508 to achieve the desired subtraction. As further discussed in detail below, analysis of the audio filter coefficients by the filter coefficient analyzer 516 yields information about the time (propagation) delay between the wireless monitor and reference versions of the audio signal applied as inputs 506 and 508, respectively, to filter 512. Finally, the distance determiner 520 is configured to convert the propagation delay determined by the filter coefficient analyzer 516 into a distance metric representative of the distance between the wireless monitor 206 providing the wireless monitor audio input 506 and the information presenting device providing the reference audio input 508. In the case of digital audio transmission by the wireless monitor 206, the distance determiner 520 may also be configured to adjust the propagation delay by one or more parameters representative of the delay incurred by digitally processing the audio within the wireless monitor 206.

[0025] The audio filter 512 processes input audio samples from inputs 506 and 508 preferably sampled at a rate of 16 kHz, although other sampling rates, such as 8 kHz, could be used. The audio signal received by the wireless monitor and ultimately corresponding to the wireless monitor audio input 506 is delayed relative to the line audio signal providing the reference audio input 508 due to the propagation delay of the sound waves emanating from the speakers of the information presenting device 202 and arriving at the wireless monitor 206. Furthermore, multiple sound wave paths may exist due to reflections from walls and possibly other objects in the room. Additionally, the acoustic wave associated with the audio signal received by the wireless monitor 206 and corresponding to the wireless monitor audio input 506 is typically attenuated in amplitude relative to the line audio signal providing the reference audio input 508. Therefore, in order to subtract the line audio signal (i.e., reference audio input 508) from the wireless monitor audio signal (i.e., wireless monitor audio input 506) with the goal of matching the two inputs to yield a null output, the audio filter 512 adapts to suitably delay and attenuate the line audio signal (i.e., reference audio input 508). An example adaptive audio filter 512 is shown in greater detail in FIG. 4. [0026] The example adaptive audio filter 512 of FIG. 4 is a finite impulse response (FIR) filter implemented via a tap delay line 604 having adaptive weights 608. The example adaptive audio filter 512 processes input line audio samples applied to the reference audio input 612 through the delay line 604, which is depicted as a set of M-I shift-registers 616- 624. Mathematically, the most recent sample of the reference audio input 612 processed by the adaptive audio filter 512 is represented by x[k] , whereas the earliest sample processed by

the adaptive audio filter 512 is represented by x[k - (M - 1)] . The audio samples applied to

the reference audio input 612 are scaled by corresponding tap weights 628-640, where the

weight w₀ scales the most recent audio input sample x[k] , while the weight w_M__x scales the

earliest audio input sample x[k - (M - 1)] . The filter output 644 of the adaptive audio filter

512, denoted by y[k] , is computed by adding the weighted input audio samples, which may

be expressed mathematically as:

[0027] Preferably, the filter weights 628-640 are initialized to values of zero prior to being adaptively updated by the weight adapter 648, although any suitable set of initialization values may also be used. The weight adapter 648 updates the filter weights 628-640 based on an error signal 652, denoted by e[k] . The error signal 652 (e[k]) is determined by

subtracting audio samples applied to the wireless monitor audio input 656, denoted by d[k]

and corresponding to remote audio samples transmitted by a wireless monitor 206 and received by the wireless receiver 504 of FIG. 3, from the filter output 644 (}>[£]). As a result

of the configuration of FIG. 4, the weight adapter 648 operates to adapt the filter weights 628-640 to minimize the error signal 652 (e[&]), thereby resulting in a filter output 644

( y[k]) that approximates the wireless monitor audio input 656 (d[k] ). More specifically, the

weight adapter 648 updates the filter weights 628-640 based on the error signal 652 (e[k])

according to the following equation:

^w _m,k+\ = ^w _m,k + μ e[k]x[k - m], m = 0,...,M - \ ,

where the index k denotes sample time such that filter weight updates occur at each sample instant in time. The variable μ corresponds to a learning factor which is usually set to a low

value, such as 0.05 in the example of FIG. 4, to avoid having the filter weights 628-640 become unstable while still allowing the filter weights 628-640 to converge to steady-state values in a reasonable length of time. A person of ordinary skill in the art will appreciate that the adaptive audio filter 512 gradually converges to a filter solution (i.e., a set of filter weights 628-640) that achieves the least mean squared (LMS) error between the wireless monitor audio input 656 and the television reference audio input 612. (See, for example, Haykin, Adaptive Filter Theory, 2^nd Ed., Englewood Cliffs, NJ: Prentice-Hall (1991) and Widrow and Steams, Adaptive Signal Processing, Englewood Cliffs, NJ: Prentice-Hall (1985), both of which are hereby incorporated by reference in their entirety). [0028] In a typical implementation using a 16 kHz sampling rate, the number of filter weights (or taps) 628-640 is preferably M = 400 . This number of filter weights 628-640 supports a maximum time (propagation) delay of 25 milliseconds (400/16 kHz = 25 milliseconds) between the wireless monitor audio input 656 and the reference audio input 612. Furthermore, the filter weights 628-640 typically adapt to relatively stationery, steady- state values in under one second, and yield an error signal 652 containing virtually no audio content corresponding to the monitored program.

[0029] Returning to FIG. 3, the filter coefficient analyzer 516 is configured to analyze the distribution of filter coefficients (weights) 628-640 at a particular instant in time (i.e.,

w_{m k} , m = 0, ... , M - 1 , for a given k) to determine the tap index n corresponding to a filter

weight w_{n k} such that the absolute value of the weight w_{n k} is greater than the absolute values

of all the other weights. The filter weight w_{n k} having the maximum absolute value is

assumed to correspond to the most direct "line-of-sight" path between the television speakers and the wireless monitor because the direct line-of-sight path is subjected to the least attenuation and, thus, is the strongest audio input signal to the microphone relative to other paths further attenuated by reflections from walls and/or various objects in the room. Referring to the filter structure of FIG. 6, persons having ordinary skill in the art will appreciate that the filter weight having the largest absolute value of all the filter weights 628- 640 corresponds to the time delay that the reference audio input 612 must be subjected to in order to become sufficiently aligned with the wireless monitor audio input 656 to minimize the error signal 652. Therefore, accounting for any processing delays, the time delay associated with the filter weight having the largest absolute value corresponds to the propagation delay between reference audio input 508 of FIG. 5 and the wireless monitor audio input 506.

[0030] The distance determiner 520 uses the propagation delay determined by the filter coefficient analyzer 516 to determine the distance between the wireless monitor 206 responsible for the wireless monitor audio input 506 and the monitored information presenting device 202 responsible for the reference audio input 508. The distance determiner 520 may determine this distance (d) using the expression:

d " = V sound '

where n is the index of the filter weight having the maximum absolute value, f_s is the audio

sampling rate (e.g., 16 kHz) and V_sound is the velocity of sound, which has a nominal value of

340.29 m/s at sea level.

[0031] Thus, returning to FIG. 2, the distance between the monitored information presenting device 202 and a user/wearer of the wireless monitor 206 may be determined by analyzing the filter coefficients (weights) of the adaptive audio filter 512 to identify the index n corresponding to a direct line-of-sight between the monitored information presenting device 202 and the wireless monitor 206. As mentioned above, wireless monitor users located greater than a predetermined distance from the monitored information presenting device 202 are automatically excluded from the audience viewing the information presenting device 202, whereas wireless monitor users located within the predetermined distance are automatically included as members of the audience viewing the information presenting device 202. As a result of this capability to automatically identify (e.g., include and/or exclude) audience members, prompting of audience members and/or manual logging-in by audience members may be eliminated or significantly reduced. [0032] Returning to FIGS. 3 and 4, to prevent erroneous distance determination, the base unit 204 may also measure the power levels of various internal signals to ensure they meet predetermined thresholds. For example, when considerable ambient noise is present in a room, the error signal 652 usually contains significant energy. Detection of this significant energy may allow the measurement to be aborted rather than resulting in an erroneous distance determination.

[0033] The filter coefficient analyzer 516 may also be configured to ensure reliability by reporting a measurement only when 1 ) the filter weight having the maximum absolute value has a significantly higher absolute value than the other weights in its neighborhood, and 2) the other weights exhibit a decay characteristic relative to the filter weight having the maximum absolute value. For illustrative purposes, an example filter weight distribution 700 is shown in Fig. 5. For convenience, only the first 200 of the 400 taps are shown. In the example of FIG. 5, the filter weight having the maximum absolute value is filter weight 702 having a tap index of 112. The tap index of 112 corresponds to a distance of 2.38 m (112 / 16000 * 340.29 = 2.38 m). Furthermore, the maximum filter weight magnitude, 0.033 in the example of FIG. 5, indicates the amount of attenuation that has to be applied to the reference audio input 508 to match the level of the wireless monitor audio input 506. To further ensure reliable measurements, processing may be aborted if the ratio of the power levels of these two audio signals (or, similarly, the maximum filter weight magnitude) falls below a predetermined threshold.

[0034] Returning to FIG. 2, privacy may be a concern in audience measurement systems employing wireless monitoring. To address privacy concerns, the wireless monitor 206 may be configured to operate in a burst mode such that the wireless monitor 206 transmits audio to the base unit 204 for bursts as short as three seconds and then remains idle for several minutes following this transmission. An adaptive filter implemented by the base unit 204, such as the adaptive audio filter 512 of FIG, 6, may be configured to require only approximately one second to converge to steady-state and yield a distance value. Such filter convergence and burst mode of operation is appropriate for audience metering applications because it is generally accepted that people location information obtained in intervals exceeding two minutes is usually adequate.

[0035] As illustrated in FIG. 1 , when more than one member of a household is in the audience, it is necessary to provide each member with a separate wireless monitor 206. It is also necessary to uniquely identify the transmission from each wireless monitor 206. A simple method of implementing such identification is to use multiple frequencies for transmission. Alternatively, the burst mode transmission may be modified to include a one second tone modulation of an RF carrier. The frequency of this tone (which precedes the actual audio transmission) identifies the wearer of the wireless monitor 206 and may be easily detected at the base unit 204. Furthermore, when several audience members are present, the wireless monitors 206 may be configured to operate in the burst mode such that the time interval between bursts of the same wireless monitor has a random duration. These random bursts between transmissions allow collisions to be avoided just as in modern network communication systems.

[0036] FIG. 6 is a block diagram of an example manner of implementing the base unit 204 of FIG. 2. The example base unit 204 of FIG. 6 includes a processor 302, a memory 304, a local communications interface 306, a remote communication interface 308, an audio interface 310, and an input interface 314, all of which are communicatively coupled as shown in FIG. 6. The processor 302 may be any suitable processor capable of interpreting and executing machine readable instructions, such as, for example, any general purpose processor or digital signal processor (DSP). Although one processor 302 is shown, the base unit 204 of FIG. 3 may include two or more processors. For example, the base unit 204 may include a general purpose processor and a DSP communicatively coupled to the general purpose processor. In such a configuration, the general purpose processor may be configured to manage system management and housekeeping operations and the DSP may be configured to perform audio signal processing operations (e.g., audio filter operations, audio signal identification operations, etc.).

[0037] The memory 304 may be used to store collected audio signal data (i.e., collected audio information), program instructions (e.g., software, firmware, etc.), program data (e.g., variables, calculation results, etc.), and/or any other data or information used by the base unit 204. For example, after acquiring an audio signal, the processor 302 may store a quantized and digitized version of the audio signal as audio information in the memory 304. The memory 304 may be implemented using any combination of suitable volatile and/or non¬ volatile memory, including, for example, a random access memory (RAM), a read-only memory (ROM), a flash memory device, a hard drive, an optical storage medium, etc. Additionally, the memory 304 may be any removable or non-removable storage medium. Although one memory device is shown, the example base unit 204 may be implemented using any number of memory devices.

[0038] The local communications interface 306 may be configured to allow the base unit 204 to communicate wirelessly via an antenna 316 with one or more wireless monitors 206. The local communications interface 306 may be used to receive an audio signal transmitted by the wireless monitor 206 and corresponding to an audio signal output by, for example, the information presenting device 202 of FIG. 2 and detected by the wireless monitor 206. Preferably, local communications interface 306 is implemented using an FM receiver or transceiver configured to operate, for example, in the 900 MHz or 2.4 GHz bands. Alternatively, the local communications interface 306 may be implemented using any type of suitable digital wireless receiver or transceiver such as, for example, a Bluetooth® transceiver, an 802.11 (i.e., Wi-Fi®) transceiver, a cellular communications transceiver, an optical communications transceiver, etc. As appropriate, in some implementations a wired interface may be used to implement the local communications interface 306. In such a scenario, a user (e.g., an audience member) may communicatively couple a wireless monitor 206 to the base unit 204 via a wired interface (e.g., a coaxial cable) to communicate information from the wireless monitor 206 to the base unit 204.

[0039] The remote communication interface 308 may be used to communicate information between the base unit 204 and, for example, a home processing system and/or a central facility (not shown) via a network 318. The remote communication interface 308 may be implemented using any suitable wired or wireless communication technology, including, for example, a telephone modem, a DSL modem, a cable modem, a cellular communication circuit, an Ethernet communication circuit, an 802.11 communication circuit, etc. The remote communication interface 308 may be used to communicate audio data and/or analysis results to the home processing system and/or the central facility via the network 318.

[0040] The audio interface 310 may be used to obtain audio signals from an information presenting device 202, such as a television, radio, stereo, personal computer, etc. For example, the base unit 204 may be communicatively coupled to the information presenting device 202 as described above in connection with FIG. 2 via the audio interface 310. The audio interface 310 may be implemented using a digital interface, such as an optical interface, a Sony-Philips Digital Interface Format (SPDIF), an analog interface, etc. The audio interface 310 may include the physical connector, such as an optical coupling, an electro¬ mechanical coupling, etc., to communicatively couple the base unit 204 to, for example, an audio out interface 208 of the information presenting device 202. Additionally or alternatively, the audio interface 310 may be configured to couple directly to the speaker circuit of the information presenting device 202, for example, via the wires 214 of FIG. 2. The audio interface 310 may also include, either additionally or alternatively, a microphone or other suitable audio sensor configured to be placed proximate to a speaker or speakers of the information presenting device 202 to receive an audio signal or signals output thereby. [0041] The base unit 204 may also include an optional input interface 314 to, for example, facilitate configuration of the base unit 204. The input interface 314 may be implemented using a digital interface, such as an RS-232 interface, a USB interface, an IR interface, etc., configured to allow the base unit 204 to be configured by a separate computer or other configuring device (not shown). Additionally or alternatively, the input interface 314 may also include a user input device, such as a keyboard, touch screen, etc., and corresponding display device, such as an LCD display, etc., to allow the base unit 204 to be configured without requiring a separate computer or other configuring device. The input interface 314 may also provide an audio interface and/or other general purpose interfaces to support one or more test/debug modes of operation, wired communication with one or more wireless monitors, etc.

[0042] FIG. 7 is a block diagram on example manner of implementing the wireless monitor 206 of FIG. 2. The example wireless monitor 206 includes a processor 402, a memory 404, a local communications interface 406, and an audio sensor (e.g., preferably a microphone) 408, all of which are communicatively coupled as shown in FIG. 4. The processor 402 may be any suitable processor capable of interpreting and executing machine readable instructions, such as, preferably, a micro-controller capable of controlling, for example, the local communications interface 406 and/or the audio sensor 408. Alternatively, the processor 402 may be any general purpose processor or digital signal processor (DSP). Although one processor 402 is shown in FIG. 7, the wireless monitor 206 may include two or more processors. For example, in an alternative implementation the wireless monitor 206 may include a general purpose processor and a DSP communicatively coupled to the general purpose processor. In such a configuration, the general purpose processor may be configured to manage system management and housekeeping operations and the DSP may be configured to perform audio signal processing operations (e.g., audio filter operations, audio signal identification operations, etc.).

[0043] In a preferred implementation, the memory 404 may be used to store instructions to be executed by the processor 402 to, for example, control the local communications interface 406 and/or the audio sensor 408. Additionally, the memory 404 may store configuration parameters to allow the local communications interface 406 and/or the audio sensor 408 to be calibrated and/or properly initialized by the processor 402. In an alternative implementation, the memory 404 may also be used to store collected audio signal samples and other audio information for use by the wireless monitor 206. For example, in the alternative implementation, after acquiring an audio signal, the processor 402 may store a quantized and digitized version of the audio signal as audio information in the memory 404. Additionally, the memory 404 may store one or more parameters representative of the delay incurred by digitally processing the acquired audio signal. This processing delay may be determined, for example, through a calibration procedure that compares an input analog audio signal with the processed audio signal ultimately transmitted by the wireless monitor 206. In any case, the memory 404 may be implemented using any suitable combination of volatile and/or non- volatile memory, including, for example, a random access memory (RAM), a read-only memory (ROM), a flash memory device, etc. Although one memory device is shown, the example wireless monitor 206 may be implemented using any number of memory devices.

[0044] The local communications interface 406 may be configured to allow the wireless monitor 206 to communicate wirelessly via an antenna 410 with the base unit 204 of FIG. 2. For example, the local communications interface 406 may be used to transmit an audio signal corresponding to audio content output by the television 202 of FIG. 2 and detected by the wireless monitor 206. Preferably, the local communications interface 406 is configured to transmit an analog audio signal using, for example, frequency modulation (FM) in the 900 MHz or 2.4 GHz bands. Alternatively, the local communications interface 406 may be implemented using any type of suitable digital wireless transmitter or transceiver such as, for example, a Bluetooth® transceiver, an 802.11 (i.e., Wi-Fi®) transceiver, a cellular communications transceiver, an optical communications transceiver, etc. In the latter alternative implementation, the wireless monitor 206 may communicate the audio data to the base unit 204 in real-time, at predetermined intervals, when a predetermined amount of audio data has been stored in the memory 404, or any other timing mechanism, and also communicate one or more parameters representative of the delay incurred in digitally processing the audio signal. As appropriate, in some implementations a wired interface may be used to implement the local communications interface 406. In such a scenario, a user (e.g., an audience member) may communicatively couple the wireless monitor 206 to the base unit 204 via a wired interface (e.g., a coaxial cable) to communicate information from the wireless monitor 206 to the base unit 204.

[0045] Flowcharts representative of example machine readable instructions which may be executed by the example wireless monitor 206 of FIG. 7 and/or the example base unit 204 of FIG. 6 are shown in FIGS. 8 and 9A-9B, respectively. In these examples, the processes represented by the flowcharts may be implemented by sets of machine readable instructions that may comprise one or more programs for execution by a processor, such as the processor 302 or the processor 402 shown in the examples of FIGS. 6-7. The one or more programs may be embodied in software stored on a tangible medium such as a CD-ROM, a floppy disk, a hard drive, a DVD, or a memory 304, 404 associated with the processor 302, 402, but persons of ordinary skill in the art will readily appreciate that the program or programs and/or portions thereof could alternatively be executed by a device other than the processor 302, 402 and/or embodied in firmware or dedicated hardware in a well-known manner. For example, any or all of the wireless monitor 206, the base unit 204, the audio filter 512, the filter coefficient analyzer 516, the distance determiner 520 and/or the weight adapter 848 could be implemented by any combination of software, hardware, and/or firmware. Further, although the example programs are described with reference to the flowcharts illustrated in FIGS. 8 and 9A-9B, persons of ordinary skill in the art will readily appreciate that many other methods of implementing the example methods and apparatus described herein may alternatively be used. For example, with reference to the flowcharts illustrated in FIGS. 8 and 9A-9B, the order of execution of the blocks may be changed, and/or some of the blocks described may be changed, eliminated, combined and/or subdivided into multiple blocks.

[0046] A flowchart illustrating an example program 800 which may be executed by the processor 402 of FIG. 7 to implement the example wireless monitor 206 is shown in FIG. 8. The example program 800 begins at block 804 at which the wireless monitor 206 is first enabled, for example, by pressing an activation button on the wireless monitor 206, removing the wireless monitor 206 from a charging cradle, etc. Control passes to block 808 at which the program 800 initializes the operating state of the wireless monitor 206, for example, by setting hardware registers controlling the memory 404, the local communication interface 406 and/or the audio sensor 408 to predetermined initial values, etc. After initialization at block 808 completes, control passes to block 812 at which the wireless monitor 206 enters a predetermined sleep cycle to conserve power and, thus, extend the operating time of the wireless monitor 206. Once the sleep cycle is programmed at block 812, the wireless monitor 206 enters a low-power sleep/hibernation mode of operation. [0047] At the end of the sleep cycle, control passes to block 816 based on, for example, triggering of a hardware interrupt based on a low-power clock/counter operating during the low power sleep/hibernation mode. At block 816, the wireless monitor 206 enters a "wake- up" cycle to enable further processing. Control then passes to block 820 at which the local communication interface 406 is opened to enable wireless transmission by the wireless monitor 206. Next, control passes to block 824 at which the wireless monitor 206 commences transmitting the audio signal detected by the audio sensor 408. For example, and as discussed above, the wireless monitor 206 may transmit the audio signal for reception by a base unit 204 to allow the base unit 204 to determine the propagation delay and, thus, the distance between the wireless monitor 206 and the monitored information presenting device 202 Control then passes to block 832.

[0048] At block 832, the wireless monitor 206 determines whether the wake-up cycle is complete or whether transmitting of the audio signal detected by the audio sensor 408 should continue. If at block 832 the wireless monitor 206 determines that the wake-up cycle is not complete, control returns to block 824 to continue transmission of the audio signal. However, if at block 832 the wireless monitor 206 determines that the wake-up cycle is complete, control then passes to block 840 at which the local communication interface 406 is closed/disabled. Finally, control returns to block 812 and blocks subsequent thereto to cause the wireless monitor 206 to re-enter the low-power sleep/hibernation mode until it is time to wake-up for further processing.

[0049] A flowchart illustrating an example program 900 which may be executed by the processor 302 of FIG. 6 to implement the example base unit 204 of FIG. 3 is shown in FIGS. 9A-9B. The example program 900 may, for example, be executed indefinitely in an iterative fashion in the context of a broader control program. The example program 900 begins at block 904 of FIG. 9A at which the processor 302 initializes the audio filter 512 of the base unit 204. For example, and as discussed above, the audio filter 512 is configured to process incoming audio samples corresponding to the wireless monitor audio input 506 and the reference audio input 508 to effectively determine propagation delay and attenuation values between the two audio inputs. In the case of the example adaptive audio filter 512 of FIG. 4, at block 904 the program 900 may initialize the filter weights 628-640 to predetermined values, such as initializing all weights to zero.

[0050] After the audio filter 512 is initialized at block 904, control passes to block 908 at which the processor 302 enables the wireless receiver 504 (or, e.g., opens the local communication interface 306 of FIG. 6) to enable wireless signal reception by the base unit 204. Next, control passes to block 912 at which the wireless receiver 504 commences reception of a wireless transmission from a wireless monitor 206. For example, and as discussed above, the wireless monitor 206 transmits detected audio to the base unit 204 for processing to determine the distance between the wireless monitor 206 and the information presenting device 202 monitored by the base unit 204.

[0051] After the base unit 204 receives a transmission burst from the wireless monitor 206, control passes to block 916 at which the processor 302 processes the received wireless monitor audio burst through the audio filter 512. In the case of the example adaptive audio filter 512 of FIG. 4, at block 916 the processor 302 samples the received wireless monitor audio burst and applies the resulting audio sample to the wireless monitor audio input 656. Control then proceeds to block 920 at which the program 900 receives and processes the next audio sample from the audio interface 510 through the audio filter 512. In the case of the example adaptive audio filter 512 of FIG. 4, at block 920 the processor 302 applies the audio sample to the reference audio input 612 of the adaptive audio filter 512. After the audio filter 512 processes the audio samples, control passes to block 924 at which the processor 302 adapts the audio filter 512. In the case of the example adaptive audio filter 512 of FIG. 4, at block 924 the weight adapter 648 updates the filter weights 628-640 according to the mathematical expressions presented above. Control then proceeds to block 928 at which the processor 302 checks for convergence of the audio filter 512. In the case of the example adaptive audio filter 512 of FIG. 4, at block 928 the processor 302 checks whether the filter weights 628-640 have converged to steady-state values. Persons of ordinary skill in the art will appreciate that there are many procedures which may be used at block 928 to check for filter convergence. For example, the mean squared error (MSE) between successive filter weight updates can be used to determine the amount of change exhibited by the filter weights 628-640 after adaptation. Mathematically, based on the preceding discussion, the expression for the filter weight MSE is given by:

Thus, at block 928, if the filter weight MSE is less than a predetermined threshold, the processor 302 may determine that the filter weights 628-640 have converged to steady-state values. Persons of ordinary skill in the art will also appreciate that the processing at block 928 may be performed after each audio sample is processed through the audio filter 512, after a group of audio samples have been processed, and/or after processing of the entire received wireless monitor audio burst is complete. In any case, after processing at block 928 completes, control proceeds to block 932.

[0052] At block 932, the processor 302 determines whether the entire received wireless monitor audio burst has been processed. If at block 932 the processor 302 determines that the entire received audio burst has not been processed, control returns to block 916 and blocks subsequent thereto to further process the received wireless monitor audio burst. However, if at block 932 the processor 302 determines that the entire received audio burst has been processed, control then proceeds to block 936 of FIG. 9B at which the processor 302 determines whether the audio filter 512 has converged based on the processing at block 928 of FIG. 9A. If at block 936 the processor 302 determines that the audio filter 512 has converged to steady-state, control proceeds to block 940 at which the processor 302 checks the signal quality of the received wireless monitor audio burst. For example, and as discussed above in the context of the adaptive audio filter 512 of FIG. 4, the processor 302 may examine the energy of the error signal computed as the difference between the wireless monitor audio input 656 and the filter output 644 to determine whether the received wireless monitor audio burst is satisfactory and not dominated by noise. Control then proceeds to block 944 at which the processor 302 uses the result of processing at block 940 to determine whether the quality of the received wireless monitor audio burst is satisfactory. If at block 944 the processor 302 determines that the quality is satisfactory, control then passes to block 948.

[0053] At block 948, the filter coefficient analyzer 516 determines the audio filter weight having the largest magnitude (i.e., absolute value). Control then proceeds to block 952 at which the filter coefficient analyzer 516 determines whether the filter peak determined at block 948 is satisfactory. For example, and as discussed above, the filter coefficient analyzer 516 may determine whether 1) the filter peak has a significantly higher absolute value than the other weights in its neighborhood, and 2) the other weights exhibit a decay characteristic relative to the filter peak. If at block 952 the filter coefficient analyzer 516 determines that the filter peak is satisfactory, control proceeds to block 956 at which the distance determiner 520 determines the distance between the information presenting device 202 monitored by the base unit 204 and the wireless monitor 206 responsible for the received wireless monitor audio burst. For example, and as discussed above, the distance determiner 520 may scale the tap index corresponding to the filter peak determined at block 948 by the sampling frequency and speed of sound to determine the distance between the monitored information presenting device 202 and the wireless monitor 206. After determining the distance at block 956, control passes to block 960 at which the base unit 204 reports the determined distance, for example, to a central facility via the remote communication interface 308 and network 318 of FIG. 6. The reported distance may be used to include or exclude the audience member corresponding to the wireless monitor 206 from the viewing audience of the information presenting device 202 monitored by the base unit 204. After processing at block 960 completes, the program 900 terminates and may be re-initiated to allow processing of the next wireless monitor audio burst.

[0054] Returning for completeness of discussion to block 936, if at block 936 the processor 302 determines that the audio filter 512 has not converged, the processor 302 may choose to abort the distance measurement and terminate the program 900. Similarly, the processor 302 may also abort the measurement and terminate the program 900 if at block 944 the processor 302 determines that the audio signal quality is not satisfactory or if at block 952 the processor 302 determines that the filter peak is not satisfactory. As discussed above, aborting the distance measurement under these circumstances reduces the possibility of erroneous distance reporting.

[0055] From the foregoing, persons of ordinary skill in the art will appreciate that audio based audience measurement systems which measure the time required for audio emanating from a monitored information presenting device to arrive at a wireless monitor worn by an audience member have been disclosed. This time delay multiplied by the velocity of sound yields distance information corresponding to the distance between the monitored information presenting device and the audience member wearing the wireless monitor. By applying a threshold to the distance measurement, individuals in proximity to the monitored television as specified by the threshold distance are identified as audience members, while individuals located outside the threshold distance are not identified as audience members (i.e., are excluded from the audience). [0056] An example audience measurement system disclosed herein includes a base unit 204 co-located with a monitored information presenting device. The base unit 204 includes several sub-systems that perform various tasks, such as, for example, using inaudible audio watermarks embedded in the monitored audio to determine the channel and/or program being presented by the monitored information presenting device and/or collecting/generating signatures that uniquely represent the channel and/or program being presented for comparison to reference signatures for the purpose of identifying the presented channel and/or program. Additionally, the base unit 204 has audio inputs that receive a line audio signal directly from the monitored information presenting device. However, in cases in which the monitored information presenting device does not have an audio line output, probes/wires could be attached to the speaker circuitry or a microphone(s) could be placed proximate to the speaker(s) of the information presenting device. To support legacy systems, the front panel of the base unit 204 may also contain a display having an array of light emitting diodes (LEDS). One or more of these LEDS may be flashed to prompt a viewer to signal his or her presence by operating the appropriate button on, for example, a remote control pad. When the viewer responds, the flashing terminates.

[0057] However, the use of a wireless monitor as disclosed herein greatly simplifies the tasks to be performed by an audience member. For example, the audience member may wear the wireless monitor 206 which detects an audio signal whenever the viewer is in the vicinity of the monitored information presenting device. This automatic detection of the emitted audio signal, and resulting automatic determination of the distance between the audience member and monitored information presenting device, eliminates or at least greatly reduces the need for the audience member to signal his or her presence manually to the base unit 204 of the audience metering system. [0058] Although certain example methods, apparatus, and articles of manufacture have been described herein, the scope of coverage of this patent is not limited thereto. On the contrary, this patent covers all methods, apparatus and articles of manufacture fairly falling within the scope of the appended claims either literally or under the doctrine of equivalents.

Claims

What is claimed is:

1. A method of determining a distance between an audience member and an information presenting device, the method comprising: obtaining, via a first apparatus, first audio information; obtaining, via a second apparatus, second audio information, wherein the second audio information is associated with the first audio information; determining a propagation delay based on the first audio information and the second audio information; and determining a distance between the information presenting device and the second apparatus based on the propagation delay.

2. A method as defined in claim 1 wherein the first apparatus obtains the first audio information from at least one of a speaker or an audio line out interface of the information presenting device.

3. A method as defined in claim 1 wherein the second audio information corresponds to audio broadcast from a speaker.

4. A method as defined in claim 1 wherein the second audio information includes at least a portion of the first audio information.

5. A method as defined in claim 1 wherein determining the propagation delay comprises processing the first and second audio information using an adaptive filter.

6. A method as defined in claim 1 wherein the adaptive filter is a finite impulse response filter.

7. A method as defined in claim 1 wherein determining the propagation delay comprises delaying and attenuating the first audio information.

8. A method as defined in claim 1 wherein determining the propagation delay comprises: generating a plurality of coefficients; selecting at least one of the plurality of the coefficients; and determining the propagation delay based on the at least one of the plurality of coefficients.

9. A method as defined in claim 8 wherein the at least one of the plurality of the coefficients is associated with an amount of time by which the second audio information is time shifted from the first audio information.

10. A method as defined in claim 8 wherein determining the propagation delay further comprises: identifying an index value associated with the at least one of the plurality of the coefficients; multiplying the index value by a velocity to generate a product value; and dividing the product value by a sampling frequency to determine the propagation delay, wherein the sampling frequency is associated with a sampling rate of at least one of the first or the second audio information.

11. A method as defined in claim 10 wherein the velocity corresponds to a speed of sound.

12. A method as defined in claim 1 further comprising communicating the second audio information from the second apparatus to the first apparatus via a wireless transmission.

13. A first apparatus for determining a distance between an information presenting device and a second apparatus, the first apparatus comprising: a filter configured to receive first audio information from the information presenting device and second audio information from the second apparatus and to generate a plurality of filter coefficients based on the first and second audio information; a filter coefficient analyzer communicatively coupled to the filter and configured to identify at least one of the plurality of filter coefficients; and a distance determiner communicatively coupled to the filter coefficient analyzer and configured to determine the distance between the information presenting device and the second apparatus based on the at least one of the plurality of filter coefficients.

14. A first apparatus as defined in claim 13 further comprising a wireless communication receiver communicatively coupled to the filter, and configured to receive the second audio information via a wireless transmission from the second apparatus and to communicate the second audio information to the filter.

15. A first apparatus as defined in claim 13 further comprising an audio interface communicatively coupled to the filter and configured to receive the first audio information from one of an audio line out interface of the information presenting device, a speaker interface or a microphone.

16. A first apparatus as defined in claim 13 wherein the filter is an adaptive filter configured to generate the plurality of filter coefficients based on the first and second audio information.

17. A first apparatus as defined in claim 13 wherein the filter is a finite impulse response filter.

18. A first apparatus as defined in claim 13 wherein the first audio information is an audio portion of a television program.

19. A first apparatus as defined in claim 13 wherein at least a portion of the second audio information corresponds to an audio portion of a television program that is broadcast via a speaker and detected via a microphone communicatively coupled to the second apparatus.

20. A first apparatus as defined in claim 19 wherein the information presenting device is at least one of a television, a radio, a stereo or a computer.

21. A system for determining a distance over which audio information travels, the system comprising: a first apparatus configured to be communicatively coupled to an audio interface of an information presenting device and to obtain first audio information from the information presenting device; and a second apparatus configured to be communicatively coupled to the first apparatus and configured to obtain second audio information output by the information presenting device, wherein the second apparatus is configured to communicate the second audio information to the first apparatus, and wherein the first apparatus is configured to determine a propagation delay based on the first audio information and the second audio information and to determine a distance over which the second audio information traveled based on the propagation delay.

22. A system as defined in claim 21 wherein the first apparatus is communicatively coupled to an audio line out of the information presenting device, a speaker interface, or a microphone.

23. A system as defined in claim 21 wherein the second apparatus is a wireless portable communication device.

24. A system as defined in claim 21 wherein the distance over which the second audio information traveled is associated with a distance between a person and the information presenting device.

25. A system as defined in claim 21 wherein the first apparatus includes an adaptive filter configured to determine the propagation delay based on the first audio information and the second audio information.

26. A system as defined in claim 21 wherein the first apparatus includes a wireless receiver and the second apparatus includes a wireless transmitter, and wherein the second apparatus communicates the second audio information to the first apparatus using a wireless communication protocol.