GB2207517A

GB2207517A - Signal processing

Info

Publication number: GB2207517A
Application number: GB08816030A
Authority: GB
Inventors: Robert W Bales; Lynn T Olson
Original assignee: Tektronix Inc
Current assignee: Tektronix Inc
Priority date: 1987-07-27
Filing date: 1988-07-06
Publication date: 1989-02-01
Anticipated expiration: 2008-07-06
Also published as: GB8816030D0; NL8801828A; JPH0778511B2; GB2207517B; JPS6443769A

Abstract

In a spectrum analyser a signal is processed to determined a property e.g. amplitude of the signal within a selection sweep interval by employing a storage device (402A,402B) having a plurality of separately addressable storage locations, each defined by a first address word having n different values. The sweep interval is resolved (222) into m display slots and a range of possible values of the property is resolved (218) into n magnitude slices. The value of the selected property is determined (218) a plurality of times during each display slot. The number of the magnitude slice is employed as the first address word, and the contents of a storage location are incremented (408) each time that the storage location is addressed. The contents of the storage device are used to drive a display device (600), and the intensity with which a pixel is displayed (602) depends on the contents of a corresponding storage location. The display device (600) includes an mxn memory (608) into which stored values are written prior to display. <IMAGE>

Description

METHOD AND APPARATUS FOR SIGNAL PROCESSING This invention relates to a method and apparatus for signal processing. -.

Backqround of the Invention Spectrum analyzers that are currently available, such as the Tektronix 49X series of spectrum analyzers, provide a display in a rectangular Cartesian coordinate system of values of signal power plotted along a vertical axis and values of signal frequency plotted along a horizontal axis. In a typical spectrum analyzer, the display has 1,000 resolution points in the horizontal direction, each represent inc a frequency band, and 225 resolution points in the vertical direction. each representing a power band, for a total of 225,000 addressable display points. In the event that the display is provided on the screen of a cathode ray tube (CRT), the display is created by sweeping the electron beam across the screen of the CRT in a succession of vectors from one addressed display point to the next in the direction of increasing frequency.

Current spectrum analyzers have a real time mode of operation and a storage mode. In the storage mode, as few as 11 or as many as 22,000 sample values are generated for each of the horizontal resolution points. Each sample value represents the value of signal Power at a frequency in the frequency band represented by the particular horizontal resolution point. However such spectrun analyzers are limited to providing a display of only a single power value for a given frequency value. Tlierefore, in the storage mode the multiple power values are processed to yield a single value, e.g. the peak value or the mean value, and this single value is stored and is used to provide the display. Accordingly, the spectrum analyzer's memory has 1,000 memory locations, each of which is capable of storing at least 225 distinct sample values.The display that is provided by a spectrum analyzer orating in the storage mode is stable but does not convey information regarding the distribution of the power levels in a particular frequency band.

In the real-time mode, the display point currently being addressed represents the real-time value of signal power at a particular frequency.

For a given frequency band, several power values .may be displayed. The intensity of a particular display point depends upon the amount of time for which the input signal has the combination of power and frequency values represented by that point. If the input signal is time-varying, the display that is provided in the real-time mode conveys useful information regarding the distribution of power values in a particular frequency band. However, if the variations in signal power over time are substantial, the display may be disturbing to the user because it is not stable. Also, if the display refresh rate is low, the display is objectionable because it flickers in intensity.

Sun mary of the Invention In a preferred embodiment of the present invention, signal is processed to reveal information concerning a property of the signal within a selected sweep interval by employing a storage device having a plurality of separately addressable storage locations, the address of each storage location being defined by a first address word having n different values. The sweep interval is resolved intp m display slots and a range of possible values of the property is resolved into n magnitude slices. The value of the selected property is determined a plurality of times during each displayvslot. The number of the magnitude slice is employed as the first address word, and the contents of a storage location are incremented each time that the storage location is addressed.

Brief Description of the Drawings For a better understaning of the invention, and to show how the same may be carried into effect, reference will now be made, by way of example, to the accompanying drawinq, the single figure of which is a block diagram of signal processing apparatus embodying the present invention.

Detailed Description The signal processing apparatus illustrated in the drawing includes three major functional sections, namely a data acquisition section 200, a data accumulation section 400, and a display section 600. Other functional sections, such as a controller which controls operation of the three illustrated sections, are not shown because they are not necessary to an understanding of the invention.

The data acquisition section comprises a vertical information digitizer 218 and a horizontal information diqitizer 222, which receive and digitize respective signals from a signal source 100.

The data acquisition section also comprises a sampling clock generator 224 which provides a sampling clock signal for controlling operation of the digitizers 218 and 224.

The signal source 100 may be of conventional form and is specifically illustrated as a spectrum analyzer having an input terminal 102 at which it receives an input signal to be analyzed. The input signal is mixed in a mixer 104 with the output signal at a frequency fL1 of a sweeping local oscillator 106. The output signal of the mixer 104 is applied to a first intermediate frequency filter stage 108, which includes a fixed frequency bandpass filter, to produce an output signal at a frequency in the pass band of the filter. The power level of this signal depends on the power of the input signal at a frequency which depends on the instantaneous frequency of the output signal of the local oscillator 106.A second mixer 110 and a local oscillator 112 having a fixed output frequency are employed to provide an output signal at a frequency that is lower than that of the output signal of the first I.F. stage 108. The output signal from the second mixer 110 is fed to a second intermediate frequency filter and amplifier stage 114. The output of the second IF. stage is detected by a detector 116 and fed through amplifiers and video processing circuits (not shown) to the digitizer 218 of the data acquisition section.

The spectrum analyzer includes a ramp generator 120 which generates a constant amplitude repetitive ramp signal. The constant amplitude ramp signal is applied to a variable gain amplifier 124, and the variable amplitude ramp signal provided by the amplifier 124 is used to sweep the local oscillator 106. The gain of the amplifier 124 is adjustable under user control. The constant amplitude ramp signal is applied to the digitizer 222. Accordingly, the digital output of the digitizer 222 is representative of the instantaneous frequency of the output signal of the local oscillator 106, scaled in accordance with the gain of the amplifier 124. The output signal provided by the digitizer 218 represents signal power as a function of the frequency represented by the output signal of the digitizer 222.

The slope of the ramp signal provided by the ramp generator 120 is adjustable under the control of the user of the apparatus. Therefore, the frequency increment corresponding to consecutive power sample values is adjustable. The resolution of the digitizer 222 is such that in the time taken for the ramp signal to increase sufficiently for the LSB of the digital frequency signal to change, multiple power samples are taken. The number of power samples taken for each change in the LSB of the frequency signal depends on the slope of the ramp signal and is thus under user control. In the following description it will assumed that 100 power samples are taken for each change in the LSB of the frequency signal, but it will be understood that this value is given by way of example and not limitation.

The display section 600 includes a monitor which incorporates a CRT 602 and a drive circuit 606. The drive circuit 606 responds to a video.

signal in composite, e.g. NTSC, format by causing the electron beam of the CRT to execute a raster scan of the display surface of the CRT from left to right and top to bottom and by varying the beam current during the active interval of an active line in accordance with the amplitude of the video signal. The raster has 525 lines (of which 480 are active) each having a duration of 63.5 us (of which about 50 us is active).

The data accumulation section 400 receives data provided by the vertical digitizer 218, accumulates this data under control of the digital frequency signal provided by the horizontal digitizer, and provides the accumulatedsdata to the display section 600. The display section includes a display memory 608. The data received by the display section from the accumulation section is written into the display memory. Concurrently with writing data into the display memory, data stored in the display memory is read from the display memory and is converted to analog form by a digital-to-analog converter (DAC) 610. The analog output signal of the DAC 610 is combined with horizontal and vertical sync signals to provide the composite video signal that is applied to the drive circuit 606 of the monitor.

The writing of data to the display memory from the accumulation section and the reading of data from the display memory 608 are managed by a state machine 603. The state machine receives the sampling clock signal provided by the sampling clock generator 220. The state machine counts the sampling clock pulses and generates a pixel clock pulse each time the count reaches a predetermined number and then resets to 0. The predetermined number is selected such that the period of the pixel clock signal is 106 ns, corresponding to a frequency of about 9.43 MHz. Each pixel clock period is divided into a read interval, during which data is read from the display memory, and a write interval, during which the display memory is available to receive data written from the accumulation section.The pixel clock pulses are counted in a sequence from 0 to 600 and the pixel clock count is then reset to 0 and a horizontal reset pulse is generated. The period of the horizontal reset signal is 63.5 us, which is equal to the horizontal line time of an NTSC signal. The horizontal reset pulses are counted in a sequence from 0 to 525 and the horizontal reset count is then reset to 0 and a vertical reset pulse is generated. The horizontal and vertical reset pulses are applied to a sync generator 614 to control generation of the horizontal and vertical sync pulses.

The display memory 608 has 500 columns and 480 rows of memory locations. Each memory location is able to store a digital word. (The term "word" is used herein to describe a binary digital quantity having multiple bits without limitation to any particular number of bits.) The memory 608 is accessed by application of a row address word and a column address word in conjunction with a read or write enable signal. On a write access, the row address word is provided by the state machine 603 and the column address word is the digital frequency signal providedd by the digitizer 222. On a read access, the row address word is the horizontal reset count provided by the state machine and the column address word is the pixel clock count provided by the state machine.There is no necessary relationship between the location addressed on a write access and the location addressed on the next read access. A multiplexer 612 is used to select the column address word provided by the digitizer 222 or the state machine 603 depending on whether the memory 608 is in the read or write state.

During successive read accesses, the memory locations of the display memory 608 are scanned a row at a time, synchronously with the scanning of the CRT's display surface. Even-numbered rows are scanned during a first field interval and oddnumbered rows are scanned during the next field interval, in order to provide an interlaced signal.

A column address in the range from 501 to 600 or a row address in the range from 480 to 525 is outside the address space of the memory 608, and therefore does not result in selection of a memory location in the memory 608. The ranges of values for the column and row addresses that are outside the address space of the display memory allow time for beam retrace of the CRT. For each address that defines a location within the address space of the display memory, the contents of the addressed location are read from the memory. Therefore the CRT provides a visual representation of the contents of the display memory 608 such that the intensity with which a pixel is illuminated depends on the value of the number stored at the corresponding location of the memory 608.For addresses that are outside the address space of the memory 608, the CRT's electron beam is blanked.

The accumulation section 400 of the spectrum analyzer is connected between the acquisition section 200 and the display section 600 and comprises a pair of buffers 402A and 402B. The buffers are placed in a load state in alternating fashion.

Switching of the buffers into the load state is controlled by the least significant bit of the frequency signal provided by the digitizer 222. When the LSB of the frequency signal is 0, the buffer 402A is in the load state, and the buffer 402B is in the load state when the LSB of the frequency signal is 1. When a buffer is in the load state, data representing the distribution of power values in the signal frequency slot represented by the frequency signal is loaded into that buffer. When one buffer enters the load state, the other enters a transfer state. In the transfer state, the data that was loaded into the buffer when it was in the load state is transferred from the buffer to the display section and the buffer is cleared.Only 480 write accesses are required to transfer the contents of a buffer to the display memory and clear the buffer, and 480 consecutive write access intervals occupy about 51 us. If the sweep rate of the spectrum analyzer is low, each buffer remains in the transfer state for more than 51 ps. When a buffer first enters the transfer state, the transfer and clear operations are carried out during the first 480 write access intervals, and the buffer then enters a wait state until it is switched to the load state.

Each buffer 402 has 480 memory locations, and each memory location is able to store a digital word of, e.g., 12 bits. Address signals are applied to the buffers 402A and 402B by way of respective multiplexers 404A and 404B.

When the buffer 402A, for example, is in the load state (the LSB of the frequency signal is 0), the multiplexer 404A selects the digital power signal provided by the digitizer 218 as the address signal. In the case of the example in which the frequency signal remains constant for 100 periods of the sampling clock generator, 100 address words, each representing a sampled power value, are provided successively to the buffer 402A by way of the multiplexer 404A. In response to each address word, the number stored at the addressed memory location is read from the buffer 402A and is applied through a multiplexer 406 to an adder 408.

In the adder, the number is incremented by a selected amount and is written back through the multiplexer 406 to the same memory cation. Thus, when the buffer 402A is switched to the transfer state by a change in the LSB of the frequency signal from 0 to 1, the contents of the buffer 402A represent the distribution of power in the input signal within the frequency slot represented by the digital frequency signal during the time for which the buffer was in the load state. When the buffer 402A is in the transfer state, the contents of the buffer are transferred to one of the columns of the display memory.

The amount by which the adder 408 increments a number received from a buffer 402 before writing the number back-to the buffer is adjustable. In particular, at high sweep speeds the amount would be more than at low sweep speeds in order to compensate at least partially for reduction in display brightness that normally occurs at high sweep speeds.

The transfer of data from the accumulation section 400 to the display memory takes place during a succession of write access intervals defined by the pixel clock signal. Thus, during a write access interval which occurs while the buffer 402A is in the transfer state, one of the columns of the display memory 608 is selected in accordance with the digital frequency signal provided by the digitizer 222 and one of the memory locations in that column is selected in accordance with the row address word provided by the state machine. The row address word also selects one of the locations of the buffer 402A. The data value that was accumulated in the selected location of the buffer 402A while it was in the load state is transferred to the selected memory location of the display memory by applying a write enable signal to the display memory 608 and a read enable signal to the buffer 402A.In this manner, the data value stored at the addressed location of the buffer 402A is read through the multiplexer 406 and written into the addressed location of the display memory. When the data value stored at the addressed location of the buffer 402A has been transferred to the display menory 608, a write enable signal is applied to the buffer 402A and the multiplexer 406 selects a buffer 410 which contains all zeros. The contents of the buffer 410 are written into the addressed location, thus clearing that location to zero. This transfer/clear sequence is carried out for all 480 locations of the buffer 402A.

When the contents of the buffer 402A have been transferred to the selected column of the display memory, the buffer 402A switches to the wait state.

When the LSB of the frequency signal changes to 0, the' buffer 402A switches to the load state once more. The buffer 402B switches to the transfer state and the contents of the buffer 402B are transferred to the next column of the display memory. On successive switches of the buffers to the transfer state, the columns of the display memory are selected in increasing order of the column address word. Since the display memory has 500 columns, each of the buffers 402 is selected 250 times in order to complete the acquisition phase and, in'the case of the example discussed above, the contents of the memory 608 represent the values of 50,000 samples.

When all the columns of the display memory have been selected by the digital frequency signal, the column address word wraps around to the beginning of the display memory. The addressing of the display memory and the buffers is controlled so that the contents of the buffer 402A are always transferred to the first column of the display memory. The rate at which the contents of the display memory are replaced depends on the rate at which the ramp generator 120 sweeps the local oscillator 106. However, the sweep rate is not related to the refresh rate of the CRT. The frequency span of the analyzer can be increased without changing the sweep rate by reducing the number of samples that are taken between successive switches in the states of the buffers 402, i.e.

reducing the slope of the ramp signal.

The display that is provided by the CRT 602 corresponds to that which is provided by a conventional spectrum analyzer when operated in the real-time mode but the display is stable and does not flicker if the refresh rate of the CRT 602 is sufficiently high, e.g. 60 HZ.

It will be appreciated that the present invention is not restricted to the particular embodiment that has been described and illustrated, and that variations may be made therein without departing from the scope of the invention as defined in the appended claims and equivalents thereof. For example, the invention is not limited to use of a spectrum analyzer as the signal source and may be applied to processing other signals having a relatively low sweep rate. Also, the invention is not restricted to the particular configuration of the signal processing apparatus that has been described with reference to the drawing. In particular, in order to permit the acquisition of signals having higher sweep rates, the contents of the display memory may be incremented directly instead of through use of the accumulation section.

In order to reduce the rate at which memory accesses must be made, in the read portion of the read/write cycle multiple data values, e.g. four or eight, may be read from successive locations in a row of the display memory and loaded into a shift register. The data values are read out serially from the shift register and used to create the desired video signal. If the frequency at which read accesses are made is reduced, the duration of the write access interval may be increased to allow the contents of multiple locations in a buffer to be transferred to a column of the display memory during a single write access interval. It is desirable that the accumulation section include a device for detecting when the MSB of a word stored in a location of one of the buffers turns on, in order to avoid wrap-around of words stored in the buffers. The invention is not restricted to generation of a video signal which is compatible with standard broadcast formats. In particular, many computer monitors employ a progressively-scanned, non-interlaced raster. If it were desired to generate a video signal for driving such a monitor, the rows of the display memory would be scanned sequentially and each field would be scanned in the same manner.

Claims

1. A method of processing a signal to reveal information concerning variation in the value of a selected property of the signal as a function of the value of an independent variable, said method comprising providing at least a first storage device having a plurality of separately addressable storage locations, the address of each storage location being defined by a first address word, resolving a range of possible values of the independent variable into m slots and resolving a range of possible values of the selected property of the signal into n slices, selecting one of the m slots in the range of possible values of the independent variable, determining the value of the selected property for at least one value of the independent variable within said one slot, identifying the slice in which said value of the selected property lies, addressing said first storage device utilizing an identification value of the slice as said first address word, and incrementing the contents of that storage location of said first storage device which is defind by said first address word.

2. A method according to claim 1, further comprising: providing a second storage device having a plurality of separately addressable storage locations, the address of each storage location being defined by said first address word, selecting a second of the m slots in the range of possible values of the independent variable, determining the value of the selected property for at least one value of the independent variable within said second slot, identifying the slice in which said value of the selected property lies, addressing said second storage device utilizing an identification value of the slice as said first address word, and incrementing the contents of that storage location of said second storage device which is defined by said first address word.

3. A method of processing a signal to reveal information concerning variation in the value of a selected property of the signal as a function of the value of an independent variable, said method comprising: (a) providing first and second storage devices each having a plurality of separately addressable storage locations, the address of each storage location of each storage device being defined by a first address word, (b) resolving a range of possible values of the independent variable into m slots and resolving a range of possible values of the selected property of the signal into n slices, (c) (i) -selecting a first of the m slots in the range of possible values of the independent variable, (ii) determining the value of the selected property for at least one value of the independent variable within said first slot, (iii) identifying the slice in which said value of the selected property lies, (iv) addressing said first storage device utilizing an identification value of the slice as said first ad dress word, and (v) incrementing the contents of that storage location of said first storage device which is defined by said first address word, (d) (i) selecting a second of the m slots in the range of possible values of the independent variable, (ii) determining the value of the selected property for at least one value of the independent variable within said second slot, (iii) identifying the slice in which said value of the selected property lies, (iv) addressing said second storage device utilizing an identification value of the slice as said first ad dress word, and (v) incrementing the contents of that storage location of said second storage device which is defined by said first address word, and (e) repeating steps (c) and (d) in alternating fashion until all m slots in the range of values of the independent variable have been selected.

4. A method according to claim 3, comprising: (f) providing a memory device having a multiplicity of separately addressable memory locations organized as m columns, where m is an even number, and n rows, the address of each memory location being defined jointly by a column address word and a row address word, (g) during the ith occurrence of step (c), where i is an integer in the range from 2 to m/2, transferring the contents of the second storage device to the (2i-2)th column of the memory device, and (h) during the ith occurrence of step (d), transferring the contents of the first storage device to the (2i-l)th column of the memory device.

5. A method according to claim 4, comprising employing the contents of the memory device to drive a display device having a multiplicity of display locations corresponding respectively to the memory locations of the memory device, and wherein the appearance of a display location depends on the contents of the corresponding memory location.

6. A method according to claim 4, comprising employing the contents of the memory device to drive a display device having a multiplicity of display locations corresponding respectively to the memory locations of the memory device, each display location being stimulated to a selected brightness level which depends on the contents of the corresponding memory location.

7. A method of processing a signal to reveal information concerning variation in the value of a selected property of the signal as a function of the value of an independent variable, said method comprising: (a) providing memory means having at least m x n separately addressable memory locations, the address of each of the m x n memory locations being defined jointly by a first address word having a value in the range from 1 to m and a second address word having a value in the range from 1 to n, (b) resolving a range of possible values of the independent variable into m slots and resolving a range of possible values of the selected property of the signal into n slices, (c) determinins the value of the selected property for a plurality of values of the independent variable within jth slot, where j is an integer in the range from 1 to m, (d) identifyin the slice in which each value of the selected property lies for said plurality of values of the independent variable, and (e) writing into the memory location defined by a first address word having the value j and a second address word having the value k, the number of times the kth slice is identified in step (d!.

8. A method according to claim 7, wherein the memory means comprise a first memory device having n memory locations and a second memory device having m x n memory locations, the address of each memory location of the first storage device being defined by an address word having a value in the range from 1 to n, and step (e) comprises:: (e)(i) for a value of the independent variable within the jth slot for which the value of the selected property is identified in step (d) by the th slice, addressing the first memory device utilizing an address word having the value q and incrementing the contents of the addressed memory location, (e) (ii) carrying out step (e) (i) for each of the plurality of values of the independent variable within the jth slot, (e) (iii) transferring the contents of that memory location of the first memory device which is defined by an address word having the value k to that memory location of the second address device which is defined by a first address word having the value j and a second address word having the value k, and (e)(iv) carrying out step (e)(iii) for all values of k in the range from 1 to n.

9. Apparatus for processing a signal to reveal information concerning variation in the value of a selected property of the signal as a function of the value of an independent variable, said apparatus comprising: a storage device having a plurality of separately addressable storage locations, the address of each storage location being defined by a first address word, first means for resolving a range of possible values of the independent variable into m slots and resolving a range of possible values of the selected property of the signal into n slices, and for identifying the slice in which at least one value of the selected property lies for one of the slots in the range of possible values of the independent variable, second means for'addressing the storage device utilizing an identification value of the slice as the first address word, and third means for incrementing the contents of the storage location that is defined by the first address word.

10. Apparatus according to claim 9, wherein the first means comprise a first analog-to-digital converter for resolving the range of possible value of the independent variable into slots and providing a digital output signal which is representative of the slot in which the value of the independent variable lies, and a second analog-to-digital converter for resolving the range of possible values of the selected property into n slices and providing a digital output signal which is reDresentative of the slice in which the value of the selected property lies.

11. Apparatus for processing a signal to reveal information concerning variation in the value of a selected property of the signal as a function of the value of an independent variable, said apparatus comprising: first and second storage devices each having a plurality of separately addressable storage locations, the address of each storage loca tion-of each storage device being defined by a first address word, first means for resolving a range of possible values of the independent variable into m slots and resolving a range of possible values of the selected property of the signal into n slices, and for identifying the slice in which at least one value of the -selected property lies for first and second slots in the range of possible values of the independent variable;; second means for addressing the first storagedevice utilizing an identification value of the slice in which the value of the selected property lies for the first slot as the first address word and incrementing the contents of that storage location of the first storage device which is defined by the first address word, and for addressing the second storage device utilizing an indentification value of the slice in which the value of the selected property lies for the second slot as the first address word and incrementing the contents of that storage location of the second storage device which is defined by the first address word.

12. Apparatus according to claim 11, further comprising a memory device having a multiplicity of separately addressable memory locations organized as m columns and n rows, and means for placing each storage device selectively in a load state, in which the contents of at least one storage location of the storage device is incremented, and a transfer state, in which the contents of the storage device are transferred to one column of the memory device.

13. Apparatus according to claim 12, in which each storage device has n storage locations and the second means are operative to address a storage device which is in the load state utilizing an identification value of the slice as an address word.

14. Apparatus according to claim 13, wherein the third means comprise means for reading the data value stored at an addressed storage location of a storage device which is in the load state, incrementing the data value, and writing the incremented data value to said addressed storage location.

15. Apparatus for processing a signal to reveal information concerning variation in'the value of a selected property of the signal as a function of the value of an independent variable, said apparatus comprising: memory means having at least m x n separately addressable memory locations, the address of each of the m x n memory locations being defined jointly by a first address word having a value in the range from 1 to m and a second address word having a value in the range from 1 to n, resolving means for resolving a range of possible values of the independent variable into m slots and resolving a range of possible values of the selected property of the signal into n slices, and, for each of a plurality of values of the independent variable within a jth slot, identifying the slice in which the value of the selected property lies, and writing means for writing into the memory location defined by a first address word having the value j and a second address word having the value k, the number of times the kth slice is identified while the value of the independent variable lies within the jth slot.