Patent Grant
7116943
The present invention is directed to radio communication devices, and more particularly to technology used in a radio communication device to classify or identify signals in a radio frequency band.
In certain radio communication environments, it would be desirable to know whether and what types of other signals or devices are active. For example, an unlicensed radio frequency band is, by its nature, free to be used by any device that emits signals within certain power levels in that part of the allocated spectrum. It is possible that many devices may share the unlicensed frequency band at the same time, potentially causing interference with each other. Under these circumstances, it would be useful to identify or classify signals detected in the frequency band in order to know, for example, whether a device should take certain actions to avoid interfering with other devices operating in the frequency band.
Briefly, a system and method are provided for classifying signals occurring in a frequency band. One or more characteristics of one or more signals in the frequency band are detected using any suitable technology, such as a device that can generate characteristics of signal pulses detected in the frequency band, wherein the signal is associated with the presence of energy having a spectral (center frequency, frequency bandwidth, etc.) characteristics and timing (pulse time duration, time between pulses, etc.) characteristics. Data pertaining to the signal pulses is accumulated over time. The accumulated signal data is compared against reference data associated with known signals to classify the one or more signals in the frequency band based on the comparison. The accumulated data may include one or more characteristics selected from the group consisting of: pulse center frequency, pulse bandwidth, pulse duration, time between pulses and number of different active pulses, and wherein the reference data associated with each of a plurality of known signals comprises one or more characteristics selected from the group consisting of: pulse center frequency, pulse bandwidth, pulse duration and time between pulses. The accumulated signal data is compared against the reference data, and depending on the degree of match with reference data, a signal can be classified. The reference data may include pulse timing signatures for known signals that are compared against the accumulated data. Further, raw samples of the signals can be taken and examined to classify the signal. Further still, classification may involve iterating through a plurality of different signal characteristics to detect signals that meet one or more of the plurality of different signal characteristics.
The above and other objects and advantages will become readily apparent when reference is made to the following description taken in conjunction with the accompanying drawings.
The communication device 1000 has a radio transceiver 200 coupled to one or more antennas 100. An RF interface 300 is coupled to the radio transceiver 200. A spectrum analysis engine (SAGE) 400 is coupled to the radio transceiver 200. The SAGE 400 is hardware peripheral that generates real-time spectrum activity information. The output of the SAGE 400 is the raw information used in the classification process. The SAGE 400 will be generally described hereinafter with reference to
Though not specifically shown, the SAGE 400 may reside in a baseband section of a communication device in a VLSI implementation. A processor 600 executes a classification software program, called the classification engine 500, stored in a processor readable memory 620. Information used by the classification engine 500 to classify signals may be stored locally in a device fingerprint database 610 (also referred to as a profile or reference database). The concept of a fingerprint will be described hereinafter. In addition, new and updated device fingerprints may be downloaded to the communication device 1000 from another communication device 2060 that stores a more comprehensive and updated database 2070 of fingerprint definitions. The processor 600 executing the classification engine 500 may be an on-chip processor in the baseband section, or may be a host processor external to the baseband section, but in the communication device 1000 itself. Alternatively, the classification engine 500 may be executed on another device entirely separate from communication device 1000, in which case communication device 1000 would transmit spectrum information generated by the SAGE 400 via a wireless or wired link to the other device, such as to the server 2055 where the classification engine would be executed. For example, the classification engine could be executed on a server computer that communicates by wire or wirelessly to communication device 1000. Alternatively, if the communication device 1000 is a WLAN station, it may transmit the raw classification information to WLAN device 2045 which may be an access point with greater processing power and capable of overseeing communications in the WLAN. Data output by the SAGE 400 is stored in a memory 620 that the processor 600 accesses when performing the classification operations. The memory 620 may contain the fingerprint database 610 as well.
The communication device 1000 samples the RF spectrum in which it is operating via the radio transceiver 200. The radio transceiver 200 may downconvert the entire frequency band of interest for a given time interval or a portion of it at a given time interval. In addition, the radio transceiver 200 may scan to different portions of the frequency band to ultimately gather information about the entire frequency band. The SAGE 400 receives and processes the RF information from the radio transceiver 200 to output first level spectrum information, described hereinafter. The classification engine 500 processes the first level spectrum information with the use of device fingerprints to output information characterizing the activity in the RF spectrum and which classifieds/identifies devices operating in the spectrum. When a signal is “classified,” data that describes the spectral parameters of the signal are determined sufficient to know generally what type of signal it is. The parameters may include the bandwidth, center frequency, frequency hopping rate (if it hops), pulse duration, time between pulses, etc. A signal that is classified can also be “identified” such as by the brand of the chipset that transmits its signal, or perhaps so specific as to identify the brand and model of the device that transmits the signal.
In an unlicensed band, it is inevitable that two or more of these devices will be transmitting at the same time. There is, therefore, a high likelihood that they will interfere with each other. When interference occurs, a signal from one device to another may not be received properly, causing the sending device to retransmit (and therefore reduce throughput), or possibly entirely destroying the communication link between two communication devices. Therefore, being able to classify or identify signals is an important prerequisite to intelligently managing the use of a shared frequency band. Once a signal type is known, actions in other devices operating in the frequency band can be tailored appropriately.
With reference to
As described in that application, the SA 410 generates data representing a real-time spectrogram of a bandwidth of RF spectrum, such as, for example, up to 100 MHz using a Fast Fourier Transform (FFT) process. As such, the SA 410 may be used to monitor all activity in a frequency band, such as the 2.4 GHz or 5 GHz bands. As shown in
Internal to the SA 410 are a lowpass filter (LPF), a linear-to-log converter, a decimator and a statistics block. The LPF performs a unity-gain, single-pole lowpass filtering operation on the power values of the signal at each FFT frequency. Using Pfft(k) to denote the power value of signal at FFT frequency f(k), the lowpass filter output Plpf(k) is updated once per FFT period as follows: Plpf(k,t)=α1·Plpf(k,t)+(1−α1)·Plpf(k,t−1), 1≦k≦256, where α1 is a parameter specifying the LPF bandwidth. The linear-to-log block at the output of the FFT computes the decibel value PdB(k)=10*log(|Plpf
The signal detector 420 comprises a peak detector 422 and one or more configurable pulse detectors 424 coupled to the peak detector. The processor 600 (or another processor coupled to the processor 600, not shown, or another application program) configures the one or more pulse detectors to detect signal pulses that fall within specified ranges of bandwidth, power, center frequency, duration, etc., to detect signal pulses of certain types of signals.
More specifically, the peak detector 422 detects a peak as a set of FFT points in contiguous FFT frequency bins, each above a configured minimum power level. Once per FFT interval, the peak detector 422 outputs data describing those frequency bins that had a FFT value above a peak threshold and which frequency bin of a contiguous set of frequency bins has a maximum value for that set. In addition, the peak detector 422 passes a power vs. frequency bin data field for each FFT interval. This can be represented by the pseudo code (where k is the frequency bin index):
The peak detector 422 outputs the bandwidth, center frequency and power for each detected peak.
A pulse detector 424 calculates relative thresholds based on configuration information, and checks whether a peak exceeds the relative thresholds. If a peak exceeds the relative threshold, it defines the peak as a pulse candidate. Once a pulse candidate is found, the pulse detector compares the identified pulse candidate with a pulse definition such as ranges for power, center frequency, bandwidth and duration (defined by the pulse detector configuration information). After matching a pulse candidate with a defined pulse associated with the configuration information, the pulse detector declares that a pulse has been detected and outputs pulse event data (power, center frequency, bandwidth, duration and start time) associated with the detected pulse
If a pulse detector detects a pulse that meets the configured criteria, it outputs signal pulse event data for that pulse, including one or more of center frequency, bandwidth, duration, time between pulses and power. The spectrum analyzer 410 outputs duty cycle statistics such as the percentage of time of energy at each frequency in a frequency band and the average power and maximum power at each frequency (for example, at each of the 256 FFT frequency bins processed by the spectrum analyzer 410). A pulse detector 424 in the signal detector 420 can also be configured by the processor 600 to trigger the snapshot buffer to store raw analog-to-digital (ADC) samples of the received signal when a pulse of a particular type is detected. Other devices or processes may be used to generate raw spectrum information useful by a signal classification process.
Consequently, the first level spectrum information may include one or more of:
With reference to the flow chart of
Steps 3030 through 3050 depict the various ways that the accumulated signal pulse data may be processed to classify a signal. The classification engine 500 uses device fingerprint definitions from a fingerprint database to compare with accumulated signal pulse data. A fingerprint definition includes signal descriptive or other information that is used to identify a device or a class of devices from the signal pulse data. A fingerprint definition may include:
In step 3030, the classification engine 500 compares the accumulated signal pulse data with reference data of known signals in the fingerprint database 610 and tries to classify the pulse(s). In step 3040, pulse timing signatures are for signal classification. Pulse timing signatures templates of known signals are compared against pulse timing signatures derived from the accumulated signal pulse events. Examples of how pulse timing signatures are used is described in more detail hereinafter in conjunction with
In step 3050, additional algorithms or iterative tests can be performed that are designed to classify otherwise hard to match pulse types. For example, in some cases, fingerprinting may be enhanced by detecting the leading sync-word of a pulse. Raw ADC samples of the frequency spectrum are analyzed for matches with sync-word formats or patterns in the fingerprint database. An algorithm running on a standard microprocessor can classify most sync-words expected to be detected in the unlicensed frequency band.
In step 3060, an identification/classification alert is generated for each signal that either classifies the signal or specifically identifies it. In addition, the power, duty cycle and center frequency (channel) information for each detected and/or identified pulse (referred to as general spectrum utilization descriptions) is output, as well as information generated by the spectrum analyzer (SA statistics) and the signal detector in the SAGE 400. An identification/classification alert may contain center frequency information (when relevant), a signal identification/classification (described above), a probability indicator, as a well as power and duty cycle information for the signal. The signal identification/classification information may indicate whether the signal is a microwave oven, frequency hopping signal (Bluetooth™ SCO or Bluetooth™ ACL, for example), cordless telephone, IEEE 802.11 signal, IEEE 802.15.3 device, or one of various radar types.
The order of the steps for the flowchart shown in
Information used to build or compile a fingerprint definition is obtained from one or more of the following sources:
Examples of the pulse event data that is generated for exemplary pulses shown in
Though not listed above, also included in the information for each pulse is the start time of a pulse, thereby enabling computation of the time between consecutive pulses detected by a pulse detector.
The pulse event data for pulses 7–14 are very similar, with the exception of the center frequency. For example, pulses 7–14 may have a pulse bandwidth of 1 MHz, a pulse duration of 350 microsec, whereas the center frequency will vary across nearly all of the 2400 MHz to 2483 MHz frequency band. The SDID for pulses 7–14 is 2, since pulse detector 2 is configured to detect these types of pulses, for example.
Turning to
Similarly, there is a bandwidth histogram that tracks a given bandwidth observed for signal pulses. For example, if the entire bandwidth of the frequency band is 80 MHz, then there may be 80 1 MHz slots that are used to count the amount of time a pulse is at a particular bandwidth. Again, after a certain amount of time, each bandwidth count is divided by a total amount of time that the pulses were being transmitted in order generate a percentage value. For example, each byte bwPercent[N] represents the percentage of pulses having a bandwidth of N.
The pulse duration and time between pulses (also called pulse gap) observed for a signal pulse may be tracked with separate histograms that have generally the same form as shown in
Each of the bins represents the percentage of occurrences of gaps or durations within the range of that bin. The percentage values range from 0 to 100, with accuracy with ½ percentage point. For example, an unsigned value from 0 to 200 can be used to represent the 0 to 100 percentage value.
Another data that can be generated and used for signal classification and other purposes is data that tracks the number of different simultaneous transmissions in the frequency band during a period of time. The active transmission histogram is an example of such data. It shows the percentage or time that one to 4 or more different signal transmissions were in progress during a sampling interval, based on the number of different pulses tracked by the 4 pulse detectors. It should be understood that 4 is only an example of the number of different pulses that could be simultaneously tracked. This type of data is very useful in determining when two or more signals of the same or different type are simultaneously active in the frequency band.
The reference signal pulse data against which the accumulated signal pulse data is compared, is similar in structure to the accumulated signal pulse data. The active transmission data is useful to classify signals that are simultaneously occurring in the frequency band; it is not generally part of the reference data for a particular signal type, but is useful intelligence to process the signal data.
Pulse time signatures can provide a distinctive representation of a device or class of devices. They are useful to classify signals that have very rigorous timing attributes.
A similar analysis may be performed on the pulse data against pulse signature information for a Bluetooth™ SCO signal in which activity consists of two bursts of energy (pulses) very close in time. Energy associated with a first pulse may occur at one frequency in the band, and energy associated with a second pulse may occur at another frequency in the band, separated from the first pulse by a time interval that recurs on a consistent basis. In fact, the Bluetooth™ SCO signal shown in
Pulse timing signatures of known signals are compared against the accumulated data (typically over relatively short periods of time) to determine if there is a match within certain predetermined and adjustable tolerances. The visual paradigm is as if sliding a pulse timing template of a known signal along the accumulated pulse data of an unknown signal to determine if there is a sufficient match.
The pulse timing signature analysis for a frequency hopping signal is slightly different if the spectrum information is derived from sampling of only a portion of the frequency band, rather than the entire band that the signal may hop in. For example, while a frequency hopping signal is just as likely to occur anywhere in a frequency band, such as the 2.4 GHz band, if data for only a 20 MHz portion of the band were provided as input to the classification process, then the signal pulse data would show a relatively smaller percentage of pulses from the frequency hopping signal. The pulse timing signature analysis would be adjusted accordingly.
Classifying a pulse using timing signature templates is particularly useful when more than one device is transmitting in the frequency band. Pulse timing signature information for a signal can be represented by data describing the characteristics of a pulse, such as pulse duration, time between pulses, etc. This information can then compared against similar pulse timing signature information to determine whether there is a match.
Another way to iteratively search for different signal characteristics is to iteratively operate the pulse detectors in the SAGE with different parameters, with the goal of eventually detecting signal pulses that match the configured parameters of a pulse detector. The matching signal pulse characteristics are accumulated over time, and then run through the signal classification processes.
When several signals are being classified simultaneously, it may be useful to loosen the match requirements initially when processing data, and then once signals have been classified with a minimum confidence level, the match requirements can be tightened to ensure, over time, that the initial signal classifications holds over time.
An interface 630, such as a Cardbus, universal serial bus (USB), mini-PCI, etc., interfaces the output of the SAGE 400 to a host device 5000. There may be an optional embedded processor 605 to perform local processing, an Ethernet block 640 to interface to a wired network, FLASH memory 650 and SDRAM 660. There are also an optional lower MAC (LMAC) logic block 670 associated with a particular communication protocol or standard (“protocol X”) and a modem 680 associated with protocol X. Protocol X may be any communication protocol that operates in the frequency band, such as an IEEE 802.11x protocol. Multiple protocols may be supported by the device. Many of the blocks may be integrated into a gate array ASIC. The larger block around the radio(s) and other components is meant to indicate that the spectrum sensor device may be implemented in a NIC form factor for PCI or mini-PCI deployment. Alternatively, many of these components may be implemented directly on a processor/CPU motherboard. The embedded processor 605 may execute software programs to perform one or more of the processes described hereinafter, including the classification engine.
The host device 5000 may be a computer (e.g., PC) having a processor 5002 and memory 5004 to process the spectrum activity information supplied by the spectrum sensor via a wired network connection, USB connection, or even a wireless connection (such as an 802.11x wireless network connection). The memory 5004 may store software to enable the host processor 5002 to execute processes based on the output of the SAGE 400 (including the classification engine), as further described hereinafter. A display monitor 5010 may be coupled to the host device 5000. The host device 5000 may be a desktop or notebook personal computer or personal digital assistant, or any other computer device local to or remote from the spectrum sensor. The memory 5004 in the host device may also store driver software for the host device, such as drivers for operating systems such as Windows operating systems (Windows® XP, Windows® CE, etc.). Either the embedded processor 620 or the host processor 5002 may perform the signal classification processes described herein.
Still another variation is to implement the functions of the SAGE 400 in software on the host processor 5002. The output of an ADC of any one or more device(s) operating in the frequency band (particularly those devices having a wideband capable radio receiver) can be supplied to a host processor where the SAGE and other functions described herein are performed entirely in software, such as the classification engine, etc. For example, the output of the ADC 240 may be coupled across any one of the interfaces shown in
The measurement engine 6100 collects and aggregates output from the SAGE 400 and normalizes the data into meaningful data units for further processing. Specifically, the measurement engine 6100 accumulates statistics for time intervals of output data from the SAGE 400 to track, with respect to each of a plurality of frequency bins that span the frequency band, average power, maximum power and duty cycle. In addition, the measurement engine 6100 accumulates pulse events for signal pulses output by the SAGE 400 that fit the configured criteria. Each pulse event may include data for power level, center frequency, bandwidth, start time, duration and termination time. The measurement engine 5100 may build the histograms of signal pulse data that is useful for signal classification, referred to above. Finally, the measurement engine 6100 accumulates raw received signal data (from the snapshot buffer of the SAGE 400) useful for location measurement in response to commands from higher levels in the architecture. The measurement engine 6100 may maintain short-term storage of spectrum activity information. Furthermore, the measurement engine 6100 may aggregate statistics related to performance of a wireless network operating in the radio frequency band, such as an IEEE 802.11 WLAN. In response to requests from other software programs or systems (via the network spectrum interface described hereinafter), the measurement engine 6100 responds with one or more of several types of data generated by processing the data output by the SAGE 400.
The classification engine 500 may compare data supplied to it by the measurement engine 6100 against a database of information of known signals or signal type, as described at length above. The signal classification database may be updated for new devices that use the frequency band. The output of the classification engine 5120 includes classifiers of signals detected in the frequency band. As described above, a classification output may specify the signal type, for example, “cordless phone”, “frequency hopper device”, “frequency hopper cordless phone”, “microwave oven”, “802.11x WLAN device”, etc. In addition, the classification engine may output information describing one or more of the center frequency, bandwidth, power, pulse duration, etc. of the classified signal, which is easily obtained directly from the signal detector output of the SAGE. This may particularly useful for a classified signal that is determined to interfere with operation of other devices in the frequency band.
The location engine 6130 computes the physical location of devices operating in the frequency band. One example of a location measurement technique involves using snapshot buffer data collected by the measurement engine 5100 to perform time difference of arrival measurements at known locations of a signal transmitted by the device to be located and another reference signal to determine a location of a variety of devices (such as interferers) operating in the region of the frequency band. Sometimes simply moving an interferer to a different location can resolve transmission problems that another device or network of devices may be experiencing. The location engine 6130 may coordinate measurements obtained from multiple locations in the network. An example of a location engine is disclosed in commonly assigned co-pending U.S. Application No. 60/319,737, filed Nov. 27, 2002, entitled “System and Method for Locating Wireless Devices in an Unsynchronized Wireless Network,” the entirety of which is incorporated herein by reference.
Many other techniques to determine the location of wireless radio communication devices are known in the art and may be used as well. When an interference condition in the frequency band is detected, the spectrum expert 6140 may command the location engine 6130 to locate the source of the interferer. The output of the location engine 6130 may include position information, power level, device type and/or device (MAC) address. In addition, the security application 6220 may command the location engine 6130 to locate a rogue device that may present a possible security problem.
The spectrum expert 6140 is a process that optimizes operation of devices operating in the frequency band, given knowledge about the activity in the frequency band obtained by the measurement and classification engines. For example, the spectrum expert 6140 processes data from the SAGE 400 and optionally statistics from a particular wireless network operating in the frequency band, such as an IEEE 802.11x network, in order to make recommendations to adjust parameters of a device, or to automatically perform those adjustments in a device. The spectrum expert 6140 may be a software program that is executed, for example, by a network management station. Parameters that can be adjusted (manually or automatically) based on output of the spectrum expert 6140 include frequency channel, transmit power, fragmentation threshold, RTS/CTS, transmit data rate, CCA threshold, interference avoidance, etc. Example of interference mitigation techniques are described in commonly assigned and co-pending U.S. application Ser. No. 10/248,434, filed Jan. 20, 2003, and entitled “Systems and Methods for Interference Mitigation with Respect to Periodic Interferers in Short-Range Wireless Applications,” the entirety of which is incorporated herein by reference. The spectrum expert 6140 may operate on triggers for alert conditions in the frequency band, such as detection of a signal that interferes with the operation of a device or network of devices operating in the frequency band, to automatically report an alert, and/or adjust a parameter in a device in response thereto. For example, the spectrum expert 6140 may operate to control or suggest controls for a single WLAN AP.
The NSI 6150 may be transport independent (e.g., supports Sockets, SNMP, RMON, etc.) and parses spectrum information into sub-sections for session and radio control, measurement, events (classification), location and protocol specific enhanced statistics and controls. End user on-demand commands to check the spectrum knowledge or activity information at a particular device may be received from an application residing above the NSI 6150 and translated into a request for a particular process below the NSI 6150 to supply the requested information.
The higher level application services 6200 may include software and systems that perform broader analysis of activity in a frequency band, such as the network expert 6210 that, for example, manages multiple WLANs, a WLAN associated with one or more wireless LANs, etc. These applications may call upon the services of any one or more of the software processes shown on the other side of the NSI 6150. For example, there may also be security services 6220, location services 6230, user interfaces 6240, and systems integrations 6250 to integrate the lower level processes with other applications, such as a network management application 6260. The network management application 6260 may be executed by a network management station (e.g., server 2055) that is located in a central monitoring or control center (telephone service provider, cable Internet service provider, etc.) coupled to the sensor devices, APs, etc., as well as the devices which it controls (e.g., APs) via a wide area network (WAN) connection, e.g., the Internet, a dedicated high speed wired connection, or other longer distance wired or wireless connection.
Spectrum Activity Information and Accessing it Using the NSI
The measurement engine 6100, classification engine 500, location engine 6130 and spectrum expert 6140 generate information that maybe used by software programs or systems that access it through the NSI. The software or systems above the NSI may access the data generated by the software residing below the NSI using session control messages. The NSI 6150 may be embodied by instructions stored on a computer/processor readable medium and executed by the processor (server 1055 or network management station 1090) that executes the one or more application program or systems. For example, this processor would execute instructions for an NSI “client” function that generates the request and configurations for spectrum analysis functions and receives the resulting data for the application program. The processor(s) that execute(s) the measurement engine, classification engine, location engine and/or spectrum expert will execute instructions stored on an associated computer/processor readable medium (shown in
It should be further understood that the classification engine, location engine and spectrum expert can be viewed as a client to the measurement engine and would generate requests to, and receive data from, the measurement engine similar to the manner in which an application program would interact with the measurement engine. Further still, the spectrum expert can be viewed as a client to the classification engine and location engine and request analysis services of those engines.
The NSI 6150 may be transport independent (e.g., supports Sockets, SNMP, RMON, etc.) and may be designed for implementation in a wired or wireless format, such as by TCP/IP traffic from an 802.11 AP to a PC which is running software designed to accept the traffic for further analysis and processing. The TCP/IP traffic (or traffic using some other network protocol) could also be carried by a PCI bus inside a laptop PC, provided the PC has built-in 802.11 technology, or an 802.11 NIC. If the source of the spectrum information data stream is a TCP/IP connection, the application program would implement a socket, and access the correct port, to read the data stream. A sample of typical code for this purpose is shown below. (The sample is in Java, and shows client-side code.) Once the port connection to the data stream is established, the use of the data stream is determined by the network management software itself.
The class DataInputStream has methods such as read. The class DataOutputStream allows one to write Java primitive data types; one of its methods is writeBytes. These methods can be used to read data from, and write data to, the NSI 6150.
If the transport of the data stream occurs over other low-level media, other methods are used to access the data stream. For example, if the data is carried over a PC's PCI bus, a PCI device driver will typically provide access to the data.
The information provided by the NSI to an application program corresponds to data generated by the measurement engine 6100 (through the SAGE), classification engine 500, location engine 6130, and/or the spectrum expert 6140.
In acting as the API, the NSI has a first group of messages that identify (and initiate) the spectrum analysis function (also called a service or test) to be performed and provide configuration information for the function. These are called session control messages and are sent by the application program to the NSI. There is a second group of messages, called informational messages, that are sent by the NSI (after the requested spectrum analysis functions are performed) to the application program containing the test data of interest.
Most of the spectrum analysis functions (i.e., tests) have various configuration parameters, which are sent via session control messages, and which determine specific details of the test. For example, in monitoring the spectrum, session control messages tell the NSI how wide the bandwidth should be (narrowband or wideband), and the center frequency of the bandwidth being monitored. In many cases, detailed test configuration parameters for a spectrum analysis function can be omitted from the session control messages. In those cases, the NSI uses default settings.
Examples of spectrum analysis functions that the measurement engine 6100 (in conjunction with the services of the SAGE 400) may perform, and the resulting data that is returned, include:
Spectrum Analyzer Power vs. Frequency Data. This data describes the total power in the spectrum as a function of frequency, over a given bandwidth.
Spectrum Analyzer Statistics Data. This data provides a statistical analysis of the data in RF power vs. frequency measurements.
Pulse Event Data—This data describes characteristics on individual RF pulses detected by the SAGE 400. The characteristics for (and thus the types of pulses) detected by the SAGE 400 can be configured.
Pulse Histogram Data. This data describes the distribution of pulses per unit of time, in terms of the percentage of pulses distributed among different frequencies, energy levels, and bandwidths.
Snapshot Data. This data contain portions of raw digital data of the RF spectrum captured by the snapshot buffer of the SAGE 400. The data can help identify the location of devices, and can also be used to extract identifier information which can determine the brand of certain devices operating in the frequency band, for example. Snapshot data may also be useful for signal classification.
The classification engine 500 may perform spectrum analysis functions to determine and classify the types of signals occurring in the frequency band, and together with optional recommendation or descriptive information that may be provided by the classification engine 500 or the spectrum expert 6140, the resulting data that is returned are called spectrum event data, which describe specific events, such as detecting a particular signal type as going active or inactive in the frequency band. The spectrum expert 6140, as well as the network expert 6210 and other applications or processes may use the output of the classification engine 500.
There are numerous ways to format the NSI messages to provide the desired API functionality in connection with the spectrum analysis functions. The following are examples of message formats that are provided for the sake of completeness, but it should be understood that other API message formats may be used to provide the same type of interface between an application program and spectrum analysis functions pertaining to activity in a frequency band where signals of multiple types may be simultaneously occurring.
A common message header may be used by both session control messages and information messages. The common header, called the sm1StdHdr_t header, comes at the very beginning of all messages and provides certain general identifying information for the message. An example of the general format of the common header is explained in the table below.
Informational messages are started with two headers: the common header (sm1StdHdr_t), followed by the Info Header (sm1InfoHdr_t). The sm1InfoHdr_t header provides specific identifying parameters for information messages:
A summary of all the messages that may be sent via the NSI is contained in the table below. The numeric values in the table below correspond to the values that are used in the msgType sub-field of the sm1StdHrd_t field.
Examples of informational messages, which as suggested above, are NSI formatted versions of the output of the measurement engine 6100 and classification engine 500, and optionally the spectrum expert 6140, are described.
Spectrum Analyzer Power vs. Frequency Data
The SAGE 400 will analyze a frequency band centered at a frequency which may be controlled. Moreover, the bandwidth of the frequency band analyzed may be controlled. For example, a portion, such as 20 MHz (narrowband mode), of an entire frequency band may be analyzed, or substantially an entire frequency band may be analyzed, such as 100 MHz (wideband mode). The selected frequency band, is divided into a plurality of frequency “bins” (e.g., 256 bins), or adjacent frequency sub-bands. For each bin, and for each sample time interval, a report is made from the output of the SAGE 400 on the power detected within that bin as measured in dBm. The measurement engine 6100 supplies the configuration parameters to the SAGE drivers 6000 and accumulates the output of the SAGE 400 (
An example of the structure of the spectrum analyzer power vs. frequency data is as follows.
In the second standard header, the msgType is 46 to identify the message as an informational message, and the sessType is 10 (SM_L1_SESS_SAPF) to identify that data results from a session that is a spectrum analyzer power vs. frequency test.
The field below is the standard information header for spectrum analyzer power vs. frequency data.
This field sm1SapfMsgHdr_t below describes the frequency spectrum that is being monitored. While this message provides the center frequency and the width of the bins, it may not provide the total bandwidth being measured. This can be calculated (low end=frqCenterkHz−128*binSize, high end=frqCenterkHz+128*binSize. The radio receiver being used to monitor the bandwidth need not actually span the full bandwidth. As a result, some of the frequency bins at either end of the spectrum will typically show zero (0) RF power.
For a single snapshot of the RF spectrum at a moment in time, the sapfListEntries field explained below contains the information of primary interest, namely, the power level in dBm for each of the frequency bins.
The frequency range corresponding to bin “N”, where N goes from 0 to 255, is given by:
LowFrequency[N]=sm1SapfMsgHdr_t.frqCenterkHz+(N−128)*sm1SapfMsgHdr_t.binSizekHz
HighFrequency[N]=sm1SapfMsgHdr_t.frqCenterkHz+(N−127)*sm1SapfMsgHdr_t.binSizekHz
Spectrum Analyzer Statistics Data
The spectrum analyzer statistics data/messages provide a statistical analysis of the data in the frequency spectrum.
A single message is built from a specified number of FFT cycles, where a single FFT cycle represents an, e.g., 256 frequency bin output of the FFT. For example, 40,000 successive FFTs of the RF spectrum, taken over a total time of 1/10 of a second, are used to construct the statistics for a single message.
An example of the overall structure of the spectrum analyzer statistics data is:
This message header sm1SaStatsMsgHdr_t field contains parameters which describe the sampling process, examples of which are below.
There are, for example, 256 consecutive statsBins, each with four sub-fields as shown in the table below. Each statsBin, with its four subfields, contains the statistical data for a particular bandwidth. To calculate the width of each frequency bin, the following formula may be used:
binWidth=sm1SaStatsMsgHdr_t. bwkhz/256
The lower and upper bandwidth for each bin is giving by the following formulas:
LowBandwidth[N]=sm1SaStatsMsgHdr_t. centerFreqkHz+((N−128)*binWidth)
HighBandwidth[N]=sm1SaStatsMsgHdr_t. centerFreqkHz+((N−127)*binWidth)
There are ten consecutive activeBins which record “peak” activity. The bins may be viewed as being indexed consecutively, from 0 to 9. For each bin, the value in the bin should be interpreted as follows. In the Nth bin, if the value in the bin is X, then for (X/2) % of the time, there were N peaks in the RF spectrum during the sampling period, except for the special case below for the 10th bin, called bin 9.
As described above in conjunction with the SAGE 400, peaks are spikes, or very brief energy bursts in the RF spectrum. If a burst persists for a certain period of time (e.g., approximately 2.5 μsec), the SAGE 400 will detect the peak, and the peak will be included in the statistics described in this subsection. Such brief peaks are generally not included in pulse data or pulse statistics. Also as described above, if a series of consecutive peaks are seen over a continuous time period, all at the same frequency, this series—once it reaches some minimum time threshold—it will be counted as a pulse.
The exact minimum duration of a pulse, for testing purposes, is configurable by the application program, but a typical time may be 100 μsec. Since the SAGE 400 can detect RF events as brief as 2.5 μsec, a typical pulse would need to persist through at least 40 FFTs before being acknowledged as being a pulse.
Pulse Event Data
A signal pulse is a sustained emission of RF energy in a specific bandwidth starting at a specific time. The SAGE 400 detects pulses in the radio frequency band that satisfy certain configurable characteristics (e.g., ranges) for bandwidth, center frequency, duration and time between pulses (also referred to as “pulse gap”). When the SAGE 400 detects a pulse that has these characteristics, it outputs pulse event data for the pulse including:
Start Time—Measured from when the SAGE first begins detecting pulses.
Duration—The lifetime of the pulse.
Center Frequency—The center frequency of the pulse.
Bandwidth—How wide the pulse is.
Power—Average power in dBm.
The overall structure of a pulse event (PEVT) data/message is shown in the table below.
This information header field is the standard information header for pulse event messages.
There may be one or many pulse events in the message. Each instance of the classPevts field below, describes the properties of one pulse.
Pulse Histogram Data
While it is possible to access information about individual pulses, it may also be useful to work with the statistical information about pulses detected and occurring in the frequency band over time. That information is provided by pulse histogram data. The pulse histograms track distributions of: duration of the pulses (the percentage of pulses with short, medium, and long durations); gaps in time between the pulses (the percentage of pulses with short time gaps between them, medium time gaps, and long time gaps); bandwidth of pulses; frequency of pulses; and power of pulses.
The overall structure of the pulse histogram data is shown in the following table.
This PhistMsgHdr field describes the frequency spectrum which is being monitored, and some other parameters of the overall sampling process.
The pulse duration histogram fields contain a series of bytes. Each of the data bytes, or bins—in sequence—indicates the percentage (multiplied by two) of pulses that fall into a given range of durations. The table below categorizes data into smallBins, mediumBins, and largeBins and are only examples of how to track pulse duration.
The first bin (bin 0) contains the percentage (×2) of pulses that were between 0 μsec and 9 μsec. The second bin (bin 1) contains the percentage, multiplied by 2, of pulses that were between 10 μsec and 19 μsec in duration. Each of these “bins” is 10 μsec wide. This continues up to the 20th bin (bin 19), whose value is the percentage, multiplied times 2, of pulses that were between 190 and 199 μsec in length.
The next twenty-six bins are similar, except they are wider; specifically, they are 50 μsec wide. Bin 20 has a value which indicates the percentage (×2) of pulses that were between 200 μsec and 249 μsec in length. Again, there are twenty-six bins which are 50 μsec wide. Bin number 45 has a value which indicates the percentage (times 2) of pulses that were between 1450 μsec and 1499 μsec in length.
The final set of 27 bins each indicate the percentage (×2) of pulses that are wider still, specifically 500 μsec wide. Bin number 46 includes pulses whose duration was between 1500 μsec and 1999 μsec in length. Bin 72 includes pulses whose duration was between 14499 and 14999 μsec.
Pulse Duration Histogram Bins
The pulse gap histogram indicates the percentage (multiplied by two) of gaps between pulses, where the duration of the gap falls within a given time range. The bins do not reflect when the gaps occurred; they reflect how long the gaps were. Gaps are measured between the start of one pulse and the start of the next. This is because the start of a pulse tends to be sharply delineated, while a pulse may trail off more gradually. For example, assume there were a total of twenty gaps between pulses. Of these twenty, only two gaps had a duration between 10 μsec and 19 μsec. The first gap, which lasted 12 μsec, occurred at time 15.324 seconds. The second gap, which lasted 15 μsec, occurred at time 200.758 seconds. Both gaps are recorded in the second bin (numbered as bin 1). Since the two gaps reflect 10% of all recorded gaps, the value in the second bin (bin 1) will be 2×10%=20 (since all percentages are multiplied by two).
Pulse Gap Histogram Bins
For the pulse bandwidth histogram, each data bin reflects a progressively wider bandwidth. For example, if the first bin represents pulses from 0 to 9.999 kHz in width, then the second bin represents pulses from 10 kHz to 19.999 kHz, the third bin pulses from 20 kHz to 29.999 kHz in width, etc. The value stored in the bin is the percentage (×2) of the pulses that had a bandwidth somewhere within the indicated range. For example, assume the size of each bin is 80 kHz. Suppose also that the SAGE 400 detected 1000 pulses and there are 256 frequency bins. The pulses had a bandwidth between 0 and 20,480 kHz. As another example, assume the SAGE 400 detects 65 pulses, each of which had a bandwidth somewhere between 400 and 480 kHz. Then, 6.5% of the pulses fall within the sixth bandwidth range, so the 6th bin (bin number 5) will have a value of 2×6.5%=13.
The bandwidth bins may have exactly the same width. For example, if the first bin is 80 kHz wide (and includes data for pulses with bandwidths from 0 to 79.999 kHz), then all successive bins will be 80 kHz wide. The second bin includes pulses from 80 kHz to 159.999 kHz; and the 256th bin—still 80 kHz wide—includes pulses with bandwidths from 20,400 kHz to 20,479.999 kHz.
Pulse Bandwidth Histogram Bins
For the pulse center frequency histogram, each data bin reflects a range of frequencies. The value stored in the bin is the percentage, multiplied times two, of the pulses whose center frequency fell within the indicated range of frequencies.
All frequency bins may be exactly the same width. However, in general, the lowest bin (byte number 0) does not start with the frequency 0 Hz. Recall that the pulse histogram message header (PhistMsgHdr_t) has a sub-field histCenterFreqkHz, which is measure in kHz. This field defines the center frequency for the pulse center frequency histogram.
The following formulae give the actual frequency range covered by each bin of this histogram, indicating both the low frequency and the high frequency of the range. The number N is the bin number, where bin numbers are counted from freqBins 0 to freqBins 255:
Low Frequ. (bin N)=histCenterFreqkHz−(128 *binSizekHz)+(N*binSizekHz)
High Frequ. (bin N)=histCenterFreqkHz−(128*binSizekHz)+((N+1)*binSizekHz))
Suppose the size of each bin, in kHz, is 100 kHz, and that the bandwidth is 2.4 GHz. Frequencies are actually being monitored in the range from 2,387,200 kHz to 2,412,800 kHz. Suppose also that SAGE 400 detected 1000 pulses, and 80 pulses with center frequencies in the range from 2,387,600 kHz to 2,387,699 kHz. Then 8% of the pulses fall within the fifth bandwidth range, so bin 4 will have a value of 2×8%=16.
The field structure for the pulse center frequency histogram is indicated in the table below.
Pulse Center Frequency Histogram Bins
For the pulse power histogram, each bin reflects a certain power range, measured in dBm. The value of each bin reflects the percentage (×2) of those pulses whose power level fell within the indicated range.
Pulse Power Histogram Bins
Snapshot Data
Snapshot data, unlike other data provided by the NSI, is not based on data analysis by the SAGE or software. Rather, this data provide raw data from the ADC which precedes the SAGE and that converts the received signal analog signal to digital data.
The raw ADC data may be expressed in n-bit I/Q format, where ‘n’ is indicated by ‘bitsPerSample’. The snapshot samples can be used for location measurements, or for detailed pulse classification (such as identifying the exact model of a device). The size of the sample data contained in ‘snapshotSamples’ is typically 8 K bytes. The overall structure of the message is shown in the following table.
An example of a snapshot message smSnapshotMsg_t field is defined below.
Spectrum Event Data (e.g., Monitoring Activity of Signals)
The msgType for spectrum event data is 46 and the sessType is 14 (SM_L1_SESS_EVENT). A format for the smEventMsg_t spectrum event message field is described in the table below.
Examples of the manner in which spectrum event messages may be displayed are shown in
Software and systems communicate requests to the NSI for data from the services on the other side of the NSI using the session control messages referred to above. An example of the format of the session control messages is as follows. There is a standard header followed by information elements. An information element is a data structure with several parts, as described in the following table:
Typical information elements provide data such as the SAGE configuration data, radio configuration data, and service specific data (e.g., pulse data, spectrum data, etc.). Examples of NSI information elements are provided in the table below:
There is an advantage to using information elements in NSI session control messages. The format of session control messages can be modified or expanded over time, as technology is further developed, while requiring no revisions to existing software or systems that use the NSI. In other words, enhancements to the messages do not break legacy code.
In traditional software design, the network management software would be coded with the expectation of specific data structures for each of the session control messages. Any time the session control messages were changed or enhanced, changes would be required in the code for the network management software, and the code would need to be recompiled.
With session control messages, however, this is no longer necessary. Session control messages are processed as follows.
1. The requesting software or system reads the message header, and determines what kind of message it is receiving.
2. Software developers know what kinds of information elements will follow the header field based on a specification document. Design decisions are made to determine what kinds of actions the software or system will take in response to those information elements.
3. In the code itself, after reading the header field, the software loops through information elements which follow. Only for information elements of interest—which can by flagged by the infoElementType field in each information element—the software takes appropriate action.
Additional information elements may be added to some of the session control messages. However, during the “looping” process the requesting software ignores any information elements which are not of interest to it, so the additional information elements in the control messages do not require any changes in the software code. Of course, it may be desirable to upgrade a software program to take advantage of additional types of information; but again, until that new software is in place, existing software continues to function.
This benefit works in both directions. For example, in sending messages to the NSI, the software program can send an information element which fine-tunes the behavior of the SAGE. Typically, however, SAGE's default operating modes are satisfactory, and there is no need to make changes. Rather than having to send an information element containing redundant, default configuration data for SAGE, this information element can simply be omitted.
A handshaking type protocol may be used to setup, initiate and terminate a session between the application and the NSI. There are numerous techniques known in the art to provide this function. For example, all tests are started by sending a sm1StdHdr_t field. Additional, optional information elements may follow. The NSI responds with messages indicating that the test has started successfully; that it was rejected; or that the test is pending (the test is queued behind other requests for the same service). The four possible session control reply messages are Started, Pending, Rejected, and Stop.
All Start Messages may have the following structure:
1. A required sm1StdHdr_t field with a msgType value of SESS_START_REQ (40), and a value for sessType to indicate the test to be performed. This field may come first. For example, to start a pulse event test, the sessType value of 12 is used, to start a pulse histogram test, a sessType value of 13 is used, to start a spectrum analyzer power vs. frequency test, a sessType value of is used, etc.
2. An optional common session configuration information element. This configures parameters which are of interest for all the possible tests, described below.
3. For the Pulse Event test only, an optional information element to configure the pulse detectors.
4. Optional information elements to configure the SAGE and the radio.
5. An optional, vendor-specific information element, typically (but not necessarily) related to further configurations to the radio.
6. An optional session-type specific information element, with configuration information for the particular test (PEVT, PHIST, SAPF, etc.).
The general/common session configuration element IE_Session_CFG is optional when starting tests, i.e., with SESS_START_REQ. If it is not sent, the default values are used.
The radio is configured to a starting bandwidth (either 2.4 GHz or one of the 5 GHz bands, for example) before the NSI can begin any testing. Similarly, before many pulse test services can be run, at least one (if not more) of SAGE's four pulse detectors need to be configured at least once. These services include Pulse Events, Pulse Histograms, Snapshot Data, and Spectrum Analyzer Power vs. Frequency (but only if this test is to be triggered by pulse events). Once the pulse detectors are configured, they can be left in their initial configuration for subsequent tests, although the application program can reconfigure them.
The radio configuration element IE_Radio_CFG is described in the table below. It is used to fine-tune the performance of the radio. If the information element is not sent as part of the message, the radio is configured to the default values.
The SAGE configuration information element IE_SAGE_CFG is optional. It fine-tunes the performance of the SAGE 400. If the information element is not sent as part of the message, the SAGE 400 is configured to the default values. An example of the SAGE configuration element is set forth below.
The IE_VENDOR_CFG information element contains vendor specific configuration information. Typically this is a configuration that is specific to the particular radio in use.
The NSI provides a pulse detector configuration element (IE_PD_CFG) which is used to configure the pulse detectors. This element must be used the first time the pulse detectors are configured. It is also used if and when the pulse detectors are reconfigured (which may be infrequent). The optional pulse events test configuration element (IE_PEVT_CFG) are shown in the table below. If this configuration element is not sent, the default values are used for the test.
Configuring the pulse detectors involves selecting which pulse detector(s) to use for a test. It also involves providing parameters which indicate the kind of signal pulse (for example, ranges for signal power, pulse duration, pulse center frequency, etc.) will, in fact, be interpreted as being a pulse. There are a variety of options when dealing with pulse detectors:
Use the existing pulse detector configuration for the service.
Allocate a currently unused detector.
Reconfigure an existing pulse detector.
Release a pulse detector so that other sessions may use it.
Whether configuring a pulse detector before using it for the first time, or reconfiguring the detector, the header field will first be sent with a particular msgType. This will be followed by the pulse detector configuration element, IE_PD_CFG, described in the table below. (Other information elements may be included in the message as well.) Pulse detectors are selected using PD_ID sub-field values from 0 to 3. These do not correspond to physical pulse detectors; rather, they are a logical reference to a pulse detector that is used by that transport connection supporting the sessions.
The field bwThreshDbm takes a signed dBm value that helps determine which RF signals will be counted as pulses. A pulse is defined by a series of time-contiguous, and bandwidth continuous “peaks”, or brief spikes, which determine the overall bandwidth of the pulse (thus the reference to “bandwidth threshold”). A “peak floor” is established to determine which spikes of radio energy qualify as a valid “peak”. Energy spikes below this “peak floor” do not qualify, whereas those above the “peak floor” do qualify. The bwThreshDbm parameter determines the “peak floor” based on whether ‘bwThreshDbm’ is positive or negative:
If bwThreshDbm is negative (ex: −65 dBm), then the peak floor is the same as the value of bwThreshDbm.
If bwThreshDbm is positive (ex: 24 dBm), then the peak floor is determined dynamically based on the current noise floor:
peak floor dBm=noise floor dBm+bwThreshDbm.
The noise floor based mechanism (bwThreshDbm is positive) is used almost exclusively because it responds well to changes in the radio spectrum environment.
There may be pre-defined pulse detection configurations, shown in the table below, to detect certain types of signal pulses.
This following short pulse profile is suitable for detecting short pulse frequency hoppers, such as Bluetooth™ headsets and many cordless phones.
The following long pulse profile is suitable for detecting long pulses output by Microwave Ovens and television transmissions (infant monitors, surveillance cameras, X-10 cameras, etc.).
Before running a pulse histogram test for the first time, the pulse detectors need to be configured. This is done by first running a pulse event test, described above. A session control message is sent containing a header field with a sessType value of ‘13’. That is followed by the optional information elements, as shown in the table below detailing the optional pulse histogram test configuration element (IE_PHIST_CFG). If it is not sent, the default values (shown in the table) are used.
The spectrum analyzer power vs. frequency test is started by sending a session control message containing a header field with a sessType value of ‘10’; that is followed by the optional information elements, as shown below.
The spectrum analyzer statistics test is started by send a session control message containing a header field with a sessType value of ‘11’. That is followed by the optional information elements, as described below.
The field pwrThreshDbm takes a signed dBm value that helps determine the minimum power level for the “duty cycle” and the “peak count.” The pwrThreshDbm parameter determines the “floor”, or minimum energy level for these measurements, based on whether pwrThreshDbm is positive or negative:
If pwrThreshDbm is negative (e.g.,: −65 dBm), then the floor is the same as the value of pwrThreshDbm.
If pwrThreshDbm is positive (e.g.,: 24 dBm), then the floor is determined dynamically based on the current noise floor: power floor dBm=noise floor dBm+pwrThreshDbm. A noise floor based mechanism (pwrThreshDbm is positive) is used almost exclusively because it responds well to changes in the radio spectrum environment.
The spectrum event data test is started by sending a message containing a header field with a sessType value of ‘14’.
The snapshot message test is started by sending a message containing a header field with a sessType value of ‘17’, followed by the optional configuration elements. The optional snapshot message configuration element (IE_SNAP_CFG) follows. If it is not sent, default values are used for the test.
By specifying which pulse detector is used to trigger the snapshot capture, it is possible to control which types of signal pulses are detected to trigger a raw ADC data capture.
The NSI may reply to test start messages to inform the requesting software application of the status of the test, and the ability of the underlying applications to deliver data for the requested tests. It is also possible to stop a test that has been requested. The table below summarizes the session control status messages which may be sent via the NSI.
An example of how the NSI can be used to configure and obtain data from a SAGE pulse detector is shown in
Referring to
The GUI application may be written in Java® and may use sockets over TCP, for example, to communicate with the spectrum activity information associated with a particular radio communication device. The GUI application software loads a PE.ini file at initialization that contains all the configuration related information like hostname and the port number. Once the communication is established the application will spawn and thread which will wait on the port to detect spectrum activity information messages coming from the source device. As information comes through the socket it is processed and displayed to the various components that are detecting these messages. The message dispatcher dispatches the processed messages to appropriate display panels. All the messages coming through the socket will also be stored in a log file located in a directory specified by the user in the PE.ini against the key PE_LOGS. The GUI application is fed by data from the measurement engine and the classification engine (and optionally the spectrum expert or other applications) referred to above in conjunction with
The GUI consists of several sub-parts:
Fault Management. Provides a means to detect, receive and provide fault information. The fault information describes the cause of the fault.
Configuration Management. Provides a means to configure the spectrum components. A spectrum advisor provides configuration related information and guides the user through the configuration process.
Performance Management. Monitors traffic of a communication protocol, such as an IEEE 802.11 network, and collects statistical information indicative of spectrum utilization and displays them.
Event Management. Provides a means to monitor various spectrum events and display this information in the form of graphs and histograms.
A graphs panel consists of the graphs or plots on the right of the screen and plot type on the left tree view. When the tree view is opened and any plot type is clicked, the corresponding plot will be added and displayed on the right side. Any plot on the right side of the screen can be removed by clicking on the close icon on the plot. As soon as the “Start” button is hit and data is available on the socket the spectrum analyzer plots will be plotted. If the “Stop” button is pressed the plotting action is disabled and the spectrum analyzer plots will no longer be updated with incoming data. The spectrum activity information is displayed on the spectrum analyzer graphs, pulse histograms and pulse plots.
The spectrum analyzer graph in
The Number of Peaks chart shows the percentage of time that there are “N” peaks in the RF spectrum. For example, if the “0” bar is hovering around 50%, then 50% of the time there are no peaks at all. If the “1” bar is hovering at around 20%, then 20% of the time there is just 1 peak in the RF spectrum. If the “2” bar hovers at 5%, then 5% of the time SAGE is detecting 2 peaks in the RF spectrum. (The “9” bar is a special case: If the “9” bar is hovering at, say, 3%, then 3% of the time SAGE is seeing 9 or more peaks in the RF spectrum.
Center Frequency shows the distribution of the central frequencies of the pulses. The graph spans a bandwidth of 100 MHz. The actual central frequency is determined by combining the central frequency shown on the graph with the overall RF center frequency (2.4 GHz). Also, both ends of the graph are typically flat, since the actual bandwidth captured by the radio is 83 MHz.
Bandwidth shows the distribution of the bandwidths of the pulses.
Pulse Duration shows the distribution of the duration of the pulses. For example, a peak at around 200 μsec indicates that many of the pulses persist for about 200 μsec.
Pulse Gap shows the distribution of the gap times. A peak at about 1500 μsec indicates that many of the pulses are separated in time by gaps that are about 1500 μsec long.
Pulse Power indicates the distribution of the power of the pulses.
Pulse Count indicates, on a logarithmic scale, the number of pulse events counted per sample interval. Colors may be used indicate that the number of pulses poses little risk, some risk, or significant risk, for example, to a particular type of communications occurring in the radio frequency band, such as 802.11 communications.
By clicking on the “Event Log” button on the spectrum management console window in
The Alert Level, ranging from Low to High to Severe, indicates how much interference the event may cause for 802.11 communications.
The Type of event includes, “Interferer” (for example, a signal that may interfere with IEEE 802.11 communications), “Information” and “Error”.
A specific Message describing the event.
The Date & Time of the event. This is the date and time is filled in by the application (i.e., the Event Log software), based on the computer's internal clock.
A Time Stamp in seconds and microseconds, indicating the time when the event occurred, counting from when testing first began. This data is provided by the measurement engine (from the SAGE).
The ID indicates the device type, and a table below provides a partial list of IDs.
For example, a display value of 7 is the same as ([011][1]), meaning a Bluetooth Headset was turned on. 8 ([100][0]) means an Infant Monitor was just turned off. An additional field may be provided to indicate (and display) a measure of match confidence of the classification.
The Source ID identifies the target source. This parameter is only significant when more than one source (Access Point or STA) is feeding data to the application program.
More detailed information is displayed about a particular event by clicking on an event row which will open up a dialog. This dialog contains detailed information about the event in the form of a text area containing a description of the event and a text area containing details of the event. Examples of detailed event dialogs are shown in
In sum, a method is provided for classifying signals occurring in a frequency band, comprising accumulating data pertaining to one or more characteristics of radio frequency energy received in the frequency band over time; comparing the accumulated data against reference data associated with known signals; and classifying one or more signals occurring the frequency band based on the step of comparing. In addition, a system is provided for classifying signals occurring in a frequency band, comprising: at least one spectrum sensor device that receives radio frequency energy in the radio frequency band and outputs data pertaining to one or more characteristics of radio frequency energy; and a processor that accumulates the data output by the spectrum sensor device and compares the accumulated data against reference data to classify the one or more signals in the frequency band.
A processor readable medium is provided that is encoded with instructions that, when executed by a processor, cause the processor to perform steps of: comparing data associated with one or more characteristics of radio frequency energy received in a frequency band against reference data associated with known signals; and classifying one or more signals in the frequency band based on the comparison.
A software system is provided that processes data representing activity in a radio frequency band where signals of multiple types may occur, comprising: a first process for accumulating data associated with activity in the radio frequency band; and a second process that classifies types of signals occurring in the radio frequency band based on data from the first process.
Finally, a device is provided that receives radio frequency energy in a radio frequency band and processes signals representative thereof, comprising: a radio receiver that receives radio frequency energy in a radio frequency band where signals of multiple types may be occurring; a spectrum analyzer that computes power values for radio frequency energy received in at least part of the radio frequency band for a time interval; a signal detector coupled to the spectrum analyzer that detects signal pulses of radio frequency energy that meet one or more characteristics; and a processor coupled to receive output of the spectrum analyzer and the signal detector, wherein the processor is programmed to accumulate data pertaining to signal pulses detected by the signal detector; and compare accumulated data against reference data associated with known signals to classify one or more signals occurring in the frequency band.
The above description is intended by way of example only.
This application claims priority to each of the following U.S. Provisional Applications, all of which are incorporated herein by reference: Application No. 60/374,365, filed Apr. 22, 2002. Application No. 60/374,363, filed Apr. 22, 2002. Application No. 60/380,891, filed May 16, 2002. Application No. 60/380,890, filed May 16, 2002. Application No. 60/319,435, filed Jul. 30, 2002. Application No. 60/319,542, filed Sep. 11, 2002. Application No. 60/319,714, filed Nov. 20, 2002. Application No. 60/453,385, filed Mar. 10, 2003. Application No. 60/320,008, filed Mar. 14, 2003. This application is a continuation-in-part of U.S. application Ser. No. 10/246,364, filed Sep. 18, 2002, now U.S. Pat. No. 6,850,735, the entirety of which is incorporated herein by reference.
| Number | Name | Date | Kind |
|---|---|---|---|
| 3609553 | Frazier et al. | Sep 1971 | A |
| 3812291 | Brodes et al. | May 1974 | A |
| 4166980 | Apostolos et al. | Sep 1979 | A |
| 4227255 | Carrick et al. | Oct 1980 | A |
| 4271523 | Gable | Jun 1981 | A |
| 4501020 | Wakeman | Feb 1985 | A |
| 4597107 | Ready et al. | Jun 1986 | A |
| 4845707 | Isaacson et al. | Jul 1989 | A |
| 4947338 | Vistica | Aug 1990 | A |
| 4979211 | Benvenuto et al. | Dec 1990 | A |
| 5005210 | Ferrell | Apr 1991 | A |
| 5144642 | Weinberg et al. | Sep 1992 | A |
| 5210820 | Kenyon | May 1993 | A |
| 5353346 | Cox et al. | Oct 1994 | A |
| 5432862 | Hirsch | Jul 1995 | A |
| 5608727 | Perreault et al. | Mar 1997 | A |
| 5651030 | Wong et al. | Jul 1997 | A |
| 5687163 | Fox et al. | Nov 1997 | A |
| 5832038 | Carsello | Nov 1998 | A |
| 5905949 | Hawkes et al. | May 1999 | A |
| 5912922 | Koszarsky et al. | Jun 1999 | A |
| 6084919 | Kleider et al. | Jul 2000 | A |
| 6105015 | Nguyen et al. | Aug 2000 | A |
| 6198779 | Taubenheim et al. | Mar 2001 | B1 |
| 6226680 | Boucher et al. | May 2001 | B1 |
| 6240282 | Kleider et al. | May 2001 | B1 |
| 6240372 | Gross et al. | May 2001 | B1 |
| 6385434 | Chuprun et al. | May 2002 | B1 |
| 6463294 | Holma et al. | Oct 2002 | B1 |
| 6466614 | Smith | Oct 2002 | B1 |
| 6519541 | Bitton | Feb 2003 | B1 |
| 6570991 | Scheirer et al. | May 2003 | B1 |
| 6629151 | Bahl | Sep 2003 | B1 |
| 6697013 | McFarland et al. | Feb 2004 | B1 |
| 6708146 | Sewall et al. | Mar 2004 | B1 |
| 6714605 | Sugar et al. | Mar 2004 | B1 |
| 6718297 | Pride et al. | Apr 2004 | B1 |
| 6850735 | Sugar et al. | Feb 2005 | B1 |
| 6978145 | Takaba et al. | Dec 2005 | B1 |
| 20010019543 | Mueckenheim et al. | Sep 2001 | A1 |
| 20020080951 | Tanrikulu | Jun 2002 | A1 |
| 20020086641 | Howard | Jul 2002 | A1 |
| 20020137485 | Nilsson et al. | Sep 2002 | A1 |
| 20020142744 | Okanoue et al. | Oct 2002 | A1 |
| 20020154614 | Jagger et al. | Oct 2002 | A1 |
| 20020155811 | Prismantas et al. | Oct 2002 | A1 |
| 20020173272 | Liang et al. | Nov 2002 | A1 |
| 20030040277 | Deats | Feb 2003 | A1 |
| 20030050012 | Black et al. | Mar 2003 | A1 |
| 20030123420 | Sherlock | Jul 2003 | A1 |
| 20040023674 | Miller | Feb 2004 | A1 |
| 20040219885 | Sugar et al. | Nov 2004 | A1 |
| 20050032479 | Miller et al. | Feb 2005 | A1 |
| 20050075863 | Jiang et al. | Apr 2005 | A1 |
| Number | Date | Country |
|---|---|---|
| 2260336 | Aug 2000 | CA |
| 2298316 | Aug 2000 | CA |
| 2000-022712 | Jan 2000 | JP |
| Number | Date | Country | |
|---|---|---|---|
| 20030224741 A1 | Dec 2003 | US |
| Number | Date | Country | |
|---|---|---|---|
| 60320008 | Mar 2003 | US | |
| 60453385 | Mar 2003 | US | |
| 60319714 | Nov 2002 | US | |
| 60319542 | Sep 2002 | US | |
| 60319435 | Jul 2002 | US | |
| 60380890 | May 2002 | US | |
| 60380891 | May 2002 | US | |
| 60374363 | Apr 2002 | US | |
| 60374365 | Apr 2002 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10246364 | Sep 2002 | US |
| Child | 10420362 | US |
