A portable vehicle Ethernet physical layer testing method

Abstract

The present invention discloses a convenient vehicle-mounted Ethernet physical layer test method, including: constructing a test system, selecting a data acquisition module with a DA function, combining software to generate signal data, connecting a test terminal to the data acquisition module, and using the radio frequency signal introduced by the test fixture as the input of the data acquisition module; connecting the test system to a device under test, and setting two channels between the test system and the device under test, including a radio frequency signal channel and a control channel, setting the register of the Ethernet PHY of the device under test through the control channel, and transmitting the test signal through the radio frequency channel; the test system tests the device under test. The present invention utilizes high-performance modular hardware, combined with efficient and flexible software to replace traditional instruments, completes various test, measurement and automation applications, and solves the drawbacks of existing cabinet-type test systems.

Description

Technical Field

[0001] The present invention relates to the field of cross-vehicle Ethernet and signal processing technology, and in particular to a convenient vehicle Ethernet physical layer testing method. Background Art

[0002] Compared with common Ethernet, one unique feature of automotive Ethernet is the physical layer. The automotive 100M Ethernet technology has been standardized by IEEE and named 100BASE-T1, and the automotive Gigabit Ethernet IEEE1000BASE-T1 specification has also been released. They can support 100M / Gigabit speeds through a pair of unshielded twisted pair cables. This also has strict specification requirements for the physical layer testing of automotive Ethernet. This requirement is mainly reflected in the OPEN Alliance's automotive Ethernet ECU test standard TC8.

[0003] In TC8, physical layer testing includes IOP (Interoperability) testing and PMA (Physical Media Attachment) testing.

[0004] IOP testing includes tests on link-up time, signal quality SQI and harness diagnostics, while PMA testing includes checking transmitter output drop, transmitter timing jitter in MASTER mode, transmit clock frequency, transmitter power spectral density, MDI return loss, mode conversion (loss), MDI common mode transmission and transmitter distortion.

[0005] Physical layer testing, especially PMA testing, involves relatively complex signal processing. Existing testing systems generally use imported high-end instruments and develop corresponding test management software for integration.

[0006] Due to the particularity of the physical layer of automotive Ethernet and the requirements of the corresponding test cases, the test system constructed using existing test instruments is relatively large and is generally cabinet-type. Its components are as shown in the attached figure. Figure 1 shown.

[0007] The disadvantages of the existing test system are mainly in three aspects: Complex system and expensive price: The system contains multiple high-end instruments. Even when it can be imported, the price has always been relatively high. Inconvenient to move: Since the ready-made instruments are used for integration, the bulky size is inevitable. After the bulky size, the system almost becomes "fixed". This imposes certain restrictions on the use scenarios of the system. Troublesome maintenance and upgrade: The upgrade cost of imported high-end instruments and software systems is relatively high. And the maintenance cost after the warranty period is also very high, and sometimes the maintenance cycle is uncontrollable.

[0008] The first problem we encounter when we need to perform physical layer testing on equipment for automotive Ethernet-related projects is that purchasing / integrating test systems is too expensive, so we usually outsource the testing to a third party.

[0009] The third-party test system is also cabinet-based and inconvenient to move. We need to send the equipment to them for testing, and the number of test iterations is limited, which is not convenient for solving problems found during testing. In addition, if we rely on a certain test system, when the failure of the test case is caused by the test system itself or the third-party test method, it is not easy to analyze, which brings more variables to the problem solving.

[0010] Sometimes, it is not necessary to conduct a complete test of all items. For example, during the equipment development process, offline sampling, or when it has been installed on a real vehicle, it is hoped that some test items can be tested, but a fixed cabinet test system obviously cannot meet such requirements. If high-end instruments are used for integration, the above disadvantages are difficult to overcome. Summary of the invention

[0011] The purpose of the present invention is to overcome the defects of the prior art and provide a convenient vehicle-mounted Ethernet physical layer testing method, which utilizes high-performance modular hardware combined with efficient and flexible software to replace traditional instruments, completes various testing, measurement and automation applications, and solves the drawbacks of existing cabinet-type testing systems.

[0012] The object of the present invention is achieved in this way: a portable vehicle-mounted Ethernet physical layer testing method comprises the following steps:

[0013] Build a test system, select a data acquisition module with DA function, combine the software to generate signal data, connect the test terminal to the data acquisition module, and use the RF signal introduced by the test fixture as the input of the data acquisition module;

[0014] Connect the test system to the device under test. There are two channels between the test system and the device under test, including a radio frequency signal channel and a control channel. The register of the Ethernet PHY of the device under test is set through the control channel, and the test signal is transmitted through the radio frequency channel.

[0015] The test system tests the device under test, including tests when the transmitter output drops, tests when the transmitter power spectrum density is low, and tests when the transmission is distorted.

[0016] As a further limitation of the present invention, the bandwidth of the data acquisition module is above 333 MHz, can support dual-channel synchronous acquisition, the sampling frequency is above 512 MHz, and the sampling accuracy is 10-bit conversion accuracy.

[0017] As a further limitation of the present invention, the test system has a control function, including control of the test system itself and control of the device under test; the control of itself includes setting of the data acquisition module and setting of the test fixture; the control of the device under test includes control of attribute switching of Ethernet PHY and control of test mode switching of the device under test.

[0018] As a further limitation of the present invention, the test when the transmitter output drops specifically includes: according to the test requirements, the test system sets the working parameters of the Ethernet PHY of the device under test, sets the working parameters of the data acquisition module, the test system performs data acquisition and data area screening, finds the initial peak voltage of the rising edge in the waveform, and determines whether the peak voltage is within a reasonable range;

[0019] If yes, the test qualification is judged according to the rules, the test results are recorded, and the working parameters of the Ethernet PHY of the device under test are restored; if no, an abnormality is prompted.

[0020] As a further limitation of the present invention, the test system performs data acquisition and data area screening, finds the initial peak voltage of the rising edge in the waveform, and determines whether the peak voltage is within a reasonable range, specifically including: the data acquisition module performs time domain data pre-selection, finds the maximum value, and determines whether the peak voltage corresponding to the maximum value is within a reasonable range;

[0021] If yes, extract the value at 500ns after the initial peak, calculate the drop between the peak voltage and the device output voltage, and determine whether it is within the allowable range. If it is within the allowable range, the test passes this round; if it is not within the allowable range, mark the failure type;

[0022] If not, the failure type is marked; and finally the marked failure type is recorded.

[0023] As a further limitation of the present invention, the test of the power spectrum density of the transmitting end specifically includes: according to the test requirements, the test system sets the working parameters of the Ethernet PHY of the device under test, sets the working parameters of the data acquisition module, the test system performs data acquisition and data area screening, selects time domain data of suitable length for time-frequency transformation, and determines whether the power spectrum density is within a reasonable range;

[0024] If yes, temporarily store the power spectrum density data, perform average processing after a certain number of test iterations, make a test qualification judgment according to the rules, and record the result; if no, indicate an abnormality.

[0025] As a further limitation of the present invention, the test system performs data acquisition and data area screening, selects time domain data of suitable length for time-frequency transformation, and determines whether the power spectral density is within a reasonable range, specifically including: the data acquisition module selects time domain data according to resolution requirements, selects a suitable window function for fast Fourier transform, records the spectrum data within the required range, and then determines whether the number of iterations meets a certain number; if so, averages all iterative spectrum data by frequency to obtain average spectrum data, calculates upper and lower limit spectrum data according to test rules, compares them with the average spectrum data, determines whether they are within the allowable range, and then marks the data to see whether they have passed the test, and records the average spectrum data within the required range.

[0026] As a further limitation of the present invention, the test during transmission distortion specifically includes: according to the test requirements, the test system sets the working parameters of the Ethernet PHY of the device under test, sets the working parameters of the data acquisition module, generates a sine wave for interference, controls the data acquisition module to acquire the bus data waveform, and determines whether the effective acquisition bit number of the acquisition card is less than the set bit number. If so, low-pass filtering is performed on the captured continuous acquisition data, and then the time domain data of a unit is selected for peak distortion measurement. Finally, after the number of iterations is met, the test system records the test results and restores the working parameters of the Ethernet PHY on the device under test.

[0027] As a further limitation of the present invention, the low-pass filtering of the captured continuously collected data and then selecting a unit of time domain data for peak distortion measurement specifically includes: calculating the peak distortion, i.e., the minimum mean square error, based on the ideal reference data, and after a certain number of iterations, judging all the peak distortion records, when there is a peak distortion greater than a set value, marking the test as failed, if not, marking the test as passed, and outputting all the peak distortion records and the judgment results.

[0028] The present invention adopts the above technical solution, and compared with the prior art, the beneficial effects are: the data acquisition module is combined with software to replace a variety of test instruments, which not only reduces the complexity of the system structure, but also greatly reduces the cost. The signal generation and synchronization functions make the external form of the test system simpler. The volume of the test system is greatly reduced, which is easy to carry and adaptable to more application occasions. When the software function is improved, the function of the test system is also expandable; the external form of the test system is simple, and daily maintenance becomes easier. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 The composition of the existing test system.

[0030] Figure 2 The test system of the present invention constitutes embodiment 1.

[0031] Figure 3 The test system of the present invention constitutes embodiment 2.

[0032] Figure 4 Embodiment 1 of the schematic diagram of the connection between the test system and the device under test of the present invention.

[0033] Figure 5 Embodiment 2 of the schematic diagram of the connection between the test system and the device under test of the present invention.

[0034] Figure 6 Embodiment 3 of the schematic diagram of the connection between the test system and the device under test of the present invention.

[0035] Figure 7 A connection diagram of a prior art test system when checking transmitter distortion.

[0036] Figure 8 A block diagram of the test process of the test system of the present invention when checking the output drop of the transmitter.

[0037] Fig. 9 The algorithm flow chart of the test system of the present invention when checking the transmitter output drop.

[0038] Fig.10 A block diagram of the test process of the test system of the present invention when checking the power spectrum density of the transmitting end.

[0039] Fig.11 The algorithm flow chart of the test system of the present invention when checking the power spectrum density of the transmitting end.

[0040] Fig.12 A block diagram of the test process of the test system of the present invention when checking transmission distortion.

[0041] Fig.13 Flowchart of the algorithm of the test system of the present invention when checking the transmission distortion measurement. DETAILED DESCRIPTION

[0042] A convenient vehicle-mounted Ethernet physical layer testing method comprises the following steps:

[0043] Build a test system, select a data acquisition module with DA function, combine the software to generate signal data, connect the test terminal to the data acquisition module, and use the RF signal introduced by the test fixture as the input of the data acquisition module;

[0044] like Figure 2 In the first embodiment shown, the test terminal adopts a high-performance laptop computer, which is connected to the data acquisition module via a USB 3.0 interface. The input of the data acquisition module is the radio frequency signal introduced by the fixture.

[0045] like Figure 3In the second embodiment shown, the test terminal adopts a high-performance industrial computer, and a data acquisition module is inserted into the internal PCIe interface. The input of the data acquisition module is also the radio frequency signal introduced through the fixture.

[0046] The external RF signal is collected by the ultra-high-speed data acquisition module and processed by the software; the test system formed is much smaller in size than the cabinet test system and is a portable system, especially when a laptop is used.

[0047] Connect the test system to the device under test. There are two channels between the test system and the device under test, including a radio frequency signal channel and a control channel. The register of the Ethernet PHY of the device under test is set through the control channel, and the test signal is transmitted through the radio frequency channel.

[0048] like Figure 4 As shown in the figure, there are two channels between the test system and the device under test (DUT): an RF signal channel transmitted by a pair of unshielded twisted pair (UTP) of 100 / 1000BASE-T1, and a control channel using a serial port or other interface. This connection is suitable for checking the transmitter output drop, the timing jitter of the transmitter in the MASTER mode, the power spectrum density of the transmitter, the MDI return loss, and the mode conversion (loss).

[0049] like Figure 5 As shown in the figure, there are two channels between the test system and the device under test (DUT): RF clock signal (TX_TCLK) channel and control channel. In addition, 100 / 1000BASE-T1 UTP transmission is used between the device under test and the test partner (LP). LP can use a verified and qualified device with 100 / 1000BASE-T1 interface. This connection is suitable for checking the transmission clock frequency and the timing jitter of the transmitter in SLAVE mode.

[0050] like Figure 6 As shown in the figure, there are two channels between the test system and the device under test (DUT): 100 / 1000BASE-T1 RF signal channel and control channel. This connection is suitable for checking transmitter distortion. When measuring transmission distortion using this method, no independent frequency divider and interference generator are required. The test system is greatly optimized and the connection of the entire test is more concise.

[0051] For ease of illustration and comparison, Figure 7 The connection diagram of a conventional test system for checking transmitter distortion is given. The TX_TCLK of the device under test (DUT) is output to a frequency divider, which divides two 10MHz clock signals, one for the interference signal generator and the other for the digital storage oscilloscope DSO as a synchronization clock.

[0052] From the connection of the above test system, we can know that under the new test system architecture, the signals that need to be collected are: the RF signal transmitted on the bus; the TX_CLK clock signal; for 100BASE-T1, the baud rate of the signal on the bus is TX_CLK is also For 1000BASE-T1, the baud rate of the signal on the bus is 750MHz. TX_CLK is also 750MHz.

[0053] 1) Parameter requirement design of data acquisition module:

[0054] a. Number of sampling channels: The data acquisition module can be single-channel or multi-channel. 100 / 1000BASE-T1 transmits differential signals. If the data acquisition module is single-channel, a device (balun) that converts the differential end to the single end needs to be added during the transmission of the RF signal. The differential end is a balanced end, and the single-ended side is an unbalanced end, so the balun is called a balanced-to-unbalanced device, represented by balun. The present invention selects a dual-channel acquisition module, which can save the balun device.

[0055] b. Bandwidth selection: The 5x rule is often used when selecting an oscilloscope. Not only for bandwidth, but also when it comes to fast edge signal rise time testing, the 5x rule is also used to select an oscilloscope based on the rise time. This solution uses a data acquisition module with software to replace oscilloscopes and other instruments, so the 5x rule is also used when selecting an acquisition module, that is, the bandwidth of the acquisition module should be selected to be 5 times or more of the signal frequency. This is proposed to ensure the amplitude test accuracy of the signal.

[0056] The analog channel of the data acquisition module has a frequency response of a low-pass filter, and the bandwidth refers to the 3dB cutoff frequency of the low-pass filter. If a sine wave signal with the same frequency as the calibration bandwidth of the acquisition module is tested, the voltage amplitude test result will drop to 0.707 times the actual voltage value. If expressed in logarithm, the measured amplitude will be reduced by 3dB.

[0057] If the analog channel of the acquisition module is equivalent to an RC low-pass filter, for simplicity, only the first-order RC low-pass filter is considered, then a formula can be derived:

[0058]

[0059] Through calculation, the following table gives the amplitude test accuracy of the acquisition module at several common frequencies:

[0060]

[0061] It can be seen that when the measured signal frequency is 1 / 5 of the acquisition module bandwidth, the amplitude test accuracy can reach 98%.

[0062] Obviously, the larger the bandwidth of the data acquisition module is relative to the signal frequency, the higher the amplitude test accuracy. However, the requirements for the data acquisition module are also higher. In general, it is recommended that the bandwidth of the data acquisition module should not be less than 5 times the maximum frequency of the measured signal. Combined with the signal characteristics on the bus and the specific conditions of each test item, the bandwidth parameters that the data acquisition module should have can be obtained.

[0063] According to the above analysis of the measured signal, when performing time domain test analysis on 100BASE-T1 signal, in order to ensure the amplitude test accuracy, the acquisition module bandwidth is required to be above 333MHz; 1000BASE-T1 requires the acquisition module bandwidth to be above 3750MHz. In practical applications, considering the cost factor, 500MHz and 2GHz are generally selected respectively.

[0064] c. Selection of sampling frequency

[0065] Shannon's sampling theorem is well known: the sampling frequency must be at least twice the highest frequency of the signal of interest. Half of the sampling frequency is called the Nyquist frequency. If there are no frequency components in the signal that are higher than the Nyquist frequency, there is no aliasing. However, it is difficult to guarantee this for signals in the real world, so anti-aliasing filtering is generally performed before sampling. Since the anti-aliasing filter has a filtering steepness, there is still a possibility of aliasing in some areas above the filter cutoff frequency (Nyquist frequency), so when selecting the sampling frequency, it is generally selected to be more than 2.56 times the highest frequency of the signal of interest.

[0066] In the PMA test of the automotive Ethernet physical layer, a higher sampling rate is generally selected, so that the requirements for the anti-aliasing filter can be reduced and the amplitude accuracy in the frequency domain can be higher.

[0067] In the multiple test cases of 100BASE-T1 PMA test, the highest frequency of the signal of interest is 200MHz, so the sampling rate is required to be above 512MHz. Generally, 1GHz can be selected.

[0068] d. Selection of sampling accuracy: Common sampling bits (or vertical resolution) are: 8bit, 10bit, 12bit, 14bit, 16bit. An important concept is involved here: quantization noise. Quantization noise corresponds to quantization signal-to-noise ratio. For every 1bit increase in the number of quantization bits, SNRq will increase by 6.02dB. In multiple test cases of the automotive Ethernet physical layer PMA test, there are requirements for signal-to-noise ratio. For example, when checking transmitter distortion, a higher quantization signal-to-noise ratio is required. If the vertical resolution of the acquisition module is less than 10 bits, a low-pass filter must be used in the post-processing process.

[0069] The higher the sampling bit number of the data acquisition module, the higher the quantization signal-to-noise ratio. However, from the previous analysis, it can be seen that the sampling rate requirement of the data acquisition module is also very high. For the ultra-high-speed acquisition module, increasing the sampling bit number means higher cost. The present invention selects an ultra-high-speed acquisition module with a sampling bit number of 10bit / 12bit.

[0070] e. Choice of interface type: The data acquisition module must meet the above parameter design requirements. In addition, it should be noted that the amount of data generated by ultra-high-speed acquisition is very large, and there are high requirements for the transmission rate between the acquisition module and the computer. Therefore, the following two options are recommended for the interface between the acquisition module and the computer: USB3.0 (for laptops), PCIe (for industrial computers). If it is only for the 100BASE-T1 test system, the USB2.0 interface can also be used according to the specific situation of the acquisition module.

[0071] 2) Design of signal generation function: In the present invention, an extended and optimized option is to enable the ultra-high-speed data acquisition module to have a signal generation function. In the Ethernet physical layer test, there are several test cases that require auxiliary test signals. In the conventional Ethernet physical layer test system, a general AWG (arbitrary waveform generator) is used, such as Keysight's 33612A waveform generator. But in fact, the test only requires a few waveforms: 11.111Mhz, 5.4Vpp sine wave interference signal (for transmission distortion measurement), frequency preset sweep signal (for MDI return loss and mode conversion loss measurement), amplitude controllable Gaussian noise (for IOP signal quality SQI measurement, PMA test is not used); so in the scheme of the present invention, an ultra-high-speed data acquisition module with DA function is selected, combined with software to generate signal data, so that the test system has a signal generation function. This can reduce the instruments and equipment required for the test system and further simplify the complexity of the test system. The present invention adopts this method to ensure the accuracy of the measured value when TX_TCLK cannot be drawn when performing transmission distortion measurement. When performing transmit distortion measurement, the IEEE 802.3bw specification requires that the test equipment used for transmit distortion measurement must be synchronized with the TX_TCLK of the DUT, so that the internal sampling clock of each device shares a common clock source and is phase-locked. Conventional test systems use a frequency divider to generate a 10MHz clock from the TX_CLK of the device under test DUT to provide it to the interference generator and the test system. The interference generator injects the generated RF interference interference signal into the test system (signal channel). However, sometimes the TX_TCLK of the device under test DUT may not be exposed or brought out of the probe pin due to design and form factors, and TX_TCLK is inaccessible. At this time, the conventional test system is equivalent to performing distortion testing when the test equipment is not synchronized. Because there will be a phase offset between the clock sources of each test equipment, the measured distortion value will be worse than when the TX_TCLK of the DUT is available.

[0072] In the present invention, an ultra-high-speed data acquisition module with DA function is selected, and signal data is generated in combination with software. In the data acquisition module, the signal acquisition and signal generation use the same clock, and the test system can also extract information from the collected signal (calculate the peak position of the collected signal, etc.) for the correction of the phase offset value of the generated signal, thereby ensuring the accuracy of the measured value when TX_CLK cannot be derived. For 100BASE-T1, when measuring the transmission distortion, the interference signal is a 5.4Vpp 11.111MHz sine wave. It should be noted that the DA output amplitude of the acquisition module is not so large, and a broadband amplifier is required to amplify the DA output signal.

[0073] 3) Signal processing and analysis: The present invention is implemented by using a high-performance acquisition module combined with efficient and flexible software. Signal processing and analysis are important parts of the test system. The signal processing of the test system includes the processing of time domain data and the processing of frequency domain data.

[0074] The data processing in the time domain is relatively simple, such as extreme value search, delay calculation, and judgment on whether the transmitter output drop is within the range. Depending on the specific system construction and test case requirements, there is an optional low-pass filtering requirement.

[0075] There are many methods for time-frequency conversion. Considering the spectrum test requirements in automotive Ethernet PMA testing, the short-time fast Fourier transform algorithm FFT is selected.

[0076] Before performing the FFT transformation, the selection of the window function is involved. For the selection of the window function, the properties of the analyzed signal and the processing requirements should be considered. If only the main lobe frequency is required to be accurately read without considering the amplitude accuracy, a rectangular window with a relatively narrow main lobe width that is easy to distinguish can be selected, such as measuring the natural frequency of the object; if a narrowband signal is analyzed and there is strong interference noise, a window function with a small sidelobe amplitude should be selected, such as a Hanning window, a triangular window, etc.; for functions that decay exponentially over time, an exponential window can be used to improve the signal-to-noise ratio. Taking all factors into consideration, the Hanning window is selected in the present invention.

[0077] In multiple test cases of PMA testing, further processing of spectrum data is involved, such as average calculation, reference spectrum data generation, comparison, etc. Taking the power spectrum density PSD check as an example, it is necessary to generate upper and lower limit spectrum data according to the PSD mask definition defined in the specification for qualification comparison.

[0078] 4) Test process control: Each test case defines the test prerequisites, test settings, test steps, pass criteria, and test iteration times. The control function of the test system is divided into two aspects: control of the test system itself and control of the device under test (DUT).

[0079] The control of itself includes the setting of the data acquisition module and the setting of the test fixture, etc. The setting function of the test fixture is optional. When the test fixture does not have the controlled function, manual adjustment is required.

[0080] The control of DUT includes the control of Ethernet PHY attribute switching (Master / Slave Change Method) and the control of DUT test mode switching.

[0081] The switching of test modes is a major part of the control. When testing, the TEST_MODE register of the Ethernet PHY of the DUT needs to be set to match the execution of different test items.

[0082] 100BASE-T1 includes test modes 1, 2, 4, and 5. Test mode 1 is used to check transmitter output drops. In test mode 1, the PHY sends a '+1' symbol for 600 nanoseconds, then sends a '-1' symbol for 600 nanoseconds, and this sequence is repeated continuously. Test mode 5 is used to test the transmitted power spectral density (PSD) mask. In test mode 5, the Ethernet PHY transmits a random sequence of PAM-3 symbols.

[0083] 5) The test system also has other functions including test management and some auxiliary functions. Management functions include test project management and test result management. Test items are managed according to the PMA test case description in the TC8 standard, including test prerequisites, test parameter requirements, test support, test pass criteria, etc. Test result management includes the management of test time, location, personnel, test data (some cases include graphic data), test judgment results, etc. Test result management can also include the automatic generation of test reports. Auxiliary functions include the self-test function of the test system, calibration function, and selected test item judgment function. Auxiliary functions provide certain support for the smooth progress of the test.

[0084] In the method of the present invention, since the system is for 100BASE-T1, from the perspective of cost, the data acquisition module selects the digitizer FCFR-USB9912 as the ultra-high-speed acquisition module;

[0085] FCFR-USB9912 supports dual-channel synchronous 1GHz acquisition and 300MHz 3dB bandwidth, but the sampling accuracy is only 8bit, so low-pass filtering is performed using software during post-acquisition processing.

[0086] The interface between FCFR-USB9912 and the host is USB2.0, and it can meet the data transmission requirements between the host and the FCFR-USB9912.

[0087] The advantage of using FCFR-USB9912 is that it has analog output function, 4GSa buffer and high output analog bandwidth, which provides good support for the signal transmission function of this system.

[0088] The quality of the test fixture has a great impact on the test results. The test fixture used should meet the restrictions on fixture self-conversion in the "IEEE 100BASE-T1 Physical Media Attachment Test Suite V1" to avoid large test errors.

[0089] A necessary accessory for testing is a high-bandwidth probe. The requirements for fixtures and probes are basically the same as those for cabinet test systems.

[0090] The test system of the present invention needs to be equipped with a high-bandwidth preamplifier. There are several test items that require the signal to be amplified before acquisition in order to achieve a certain test accuracy. Independent instruments and equipment generally have signal preamplifiers integrated inside. The test system of the present invention uses ultra-high-speed acquisition modules and software to replace independent instruments and equipment, so an additional preamplifier is required. Combined with the index requirements of the test case, a low-noise, high-bandwidth preamplifier is selected.

[0091] During transmitter testing, a "Short Automotive Cable" matches the DUT to the necessary test fixtures for each test setup. This cable should be limited in length to reduce losses between the DUT transmitter and the oscilloscope front end.

[0092] The software is developed using VisualStudio201X and SQLite is used to manage test projects and test results.

[0093] When the present invention works, Figure 8 As shown, the test when the transmitter output drops specifically includes: according to the test requirements, the test system sets the working parameters of the Ethernet PHY of the device under test, sets the working parameters of the data acquisition module, and the test system performs data acquisition and data area screening, finds the initial peak voltage of the rising edge in the waveform, and determines whether the peak voltage is within a reasonable range;

[0094] If yes, the test qualification is judged according to the rules, and the test results are recorded until the test iteration coefficient (≥10) is met, and the working parameters of the Ethernet PHY of the device under test are restored; if no, an abnormality is prompted. For each iteration, it is necessary to judge whether the rule requirements are met based on the waveform data. The test can only pass if all iterations meet the rule requirements.

[0095] like Fig. 9 As shown, the test system performs data acquisition and data area screening, finds the initial peak voltage of the rising edge in the waveform, and determines whether the peak voltage is within a reasonable range. Specifically, the data acquisition module performs time domain data pre-selection, finds the maximum value, and determines whether the peak voltage corresponding to the maximum value is within a reasonable range;

[0096] If yes, extract the value at 500ns after the initial peak, calculate the drop between the peak voltage and the device output voltage, and determine whether it is within the allowable range. If it is within the allowable range, this round of testing passes; if not, mark the failure type; finally, record the marked failure type and record the data collected within 1ms.

[0097] like Fig.10As shown, the test of the power spectrum density of the transmitting end specifically includes: according to the test requirements, the test system sets the working parameters of the Ethernet PHY of the device under test, sets the working parameters of the data acquisition module, the test system performs data acquisition and data area screening, selects time domain data of suitable length for time-frequency transformation, and determines whether the power spectrum density is within a reasonable range;

[0098] If yes, temporarily store the power spectral density PSD data, pass a certain number of test iterations (≥50), then perform average processing, make a test qualification judgment according to the rules, and record the result; if no, an abnormality is prompted.

[0099] like Fig.11 As shown, the test system performs data acquisition and data area screening, selects time domain data of suitable length for time-frequency transformation, and determines whether the power spectrum density is within a reasonable range. Specifically, the data acquisition module selects time domain data according to the resolution requirement, selects a suitable window function for fast Fourier transform, and records the spectrum data within the required range. Each iteration can obtain spectrum data once, and then determine whether the number of iterations meets a certain number (≥50); if so, all iterative spectrum data are averaged by frequency to obtain average spectrum data, and the upper and lower limit spectrum data are calculated according to the test rules, and compared with the average spectrum data to determine whether it is within the allowable range, and then the data is marked to see whether it has passed the test, and the average spectrum data within the required range is recorded.

[0100] like Fig.12 As shown, the test during transmission distortion specifically includes: according to the test requirements, the test system sets the working parameters of the Ethernet PHY of the device under test, sets the working parameters of the data acquisition module, generates an 11.111MHz sine wave for interference, controls the data acquisition module to collect bus data waveforms of more than 2ms, and determines whether the effective acquisition bit number of the acquisition card is less than the set bit number (<10). If so, low-pass filtering is performed on the captured 2 milliseconds of continuous acquisition data, and then a unit (15ns) of time domain data is selected for peak distortion measurement. Finally, after the number of iterations is met (≥10), the test system records the test results and restores the working parameters of the Ethernet PHY on the device under test.

[0101] like Fig.13As shown, the captured 2 milliseconds of continuous acquisition data is low-pass filtered, and then a unit (15ns) of time domain data is selected for peak distortion measurement. Specifically, the peak distortion is calculated based on the ideal reference data, that is, the minimum mean square error. After a certain number of iterations (≥10), all peak distortion records are judged. When there is a peak distortion greater than 15mV, the test is marked as unpassed. If not, the test is marked as passed, and all peak distortion records and judgment results are output. In this test, the peak distortion is measured by capturing the test mode waveform supported by the Ethernet PHY chip and finding the minimum mean square error. The peak distortion refers to the difference between the possible maximum value of inter-symbol interference and the useful signal sample value. The useful signal sample value is the ideal reference data in the flow chart.

[0102] The present invention adopts virtual instrument technology, uses high-performance acquisition modules combined with efficient and flexible software to simplify the composition of the test system to achieve the purpose of convenient testing. This scheme describes the composition of the test system and the connection during the test, conducts an in-depth analysis of the measured signal and the auxiliary test signal, and then designs the parameters of the main module of the test system, the ultra-high-speed acquisition module, and designs the acquisition module function that can further simplify the system.

[0103] The present invention is not limited to the above-mentioned embodiments. On the basis of the technical solution disclosed in the present invention, technicians in this field can make some substitutions and deformations to some technical features therein according to the disclosed technical content without creative labor, and these substitutions and deformations are all within the protection scope of the present invention.

Claims

1. A portable vehicle-mounted Ethernet physical layer testing method, characterized in that: The following steps are involved: Build a test system, select a data acquisition module with DA function, combine the software to generate signal data, connect the test terminal to the data acquisition module, and use the RF signal introduced by the test fixture as the input of the data acquisition module; The data acquisition module selected the digitizer FCFR-USB9912 as the ultra-high-speed acquisition module; the software was used to perform low-pass filtering in the post-acquisition processing; the software was developed using VisualStudio201X, and SQLite was used to manage the test items and test results; Connect the test system to the device under test. There are two channels between the test system and the device under test, including a radio frequency signal channel and a control channel. The register of the Ethernet PHY of the device under test is set through the control channel, and the test signal is transmitted through the radio frequency signal channel. The test system tests the device under test, including tests when the transmitter output drops, tests when the transmitter power spectrum density is low, and tests when the transmission is distorted.

2. A portable vehicle-mounted Ethernet physical layer testing method according to claim 1, characterized in that: The bandwidth of the data acquisition module is above 333 MHz, and can support dual-channel synchronous acquisition. The sampling frequency is above 512 MHz, and the sampling accuracy is 10-bit conversion accuracy.

3. A portable vehicle-mounted Ethernet physical layer testing method according to claim 1, characterized in that: The test system has control functions, including control of the test system itself and control of the device under test; the control of the test system itself includes setting of the data acquisition module and setting of the test fixture; the control of the device under test includes control of attribute switching of Ethernet PHY and control of test mode switching of the device under test.

4. A portable vehicle-mounted Ethernet physical layer testing method according to claim 1, characterized in that: The test when the transmitter output drops specifically includes: according to the test requirements, the test system sets the working parameters of the PHY of the device under test, sets the working parameters of the data acquisition module, the test system performs data acquisition and data area screening, finds the initial peak voltage of the rising edge in the waveform, and determines whether the peak voltage is within a reasonable range; If yes, the test qualification is judged according to the rules, the test results are recorded, and the working parameters of the Ethernet PHY of the device under test are restored; if no, an abnormality is prompted.

5. A portable vehicle-mounted Ethernet physical layer testing method according to claim 4, characterized in that: The test system performs data acquisition and data area screening, finds the initial peak voltage of the rising edge in the waveform, and determines whether the peak voltage is within a reasonable range. Specifically, the data acquisition module performs time domain data pre-selection, finds the maximum value, and determines whether the peak voltage corresponding to the maximum value is within a reasonable range; If yes, extract the value at 500ns after the initial peak, calculate the drop between the peak voltage and the device output voltage, and determine whether it is within the allowable range. If it is within the allowable range, the test passes this round; if it is not within the allowable range, mark the failure type; If not, the failure type is marked; and finally the marked failure type is recorded.

6. A portable vehicle-mounted Ethernet physical layer testing method according to claim 1, characterized in that: The test of the power spectrum density of the transmitting end specifically includes: according to the test requirements, the test system sets the working parameters of the Ethernet PHY of the device under test, sets the working parameters of the data acquisition module, the test system performs data acquisition and data area screening, selects the time domain data of suitable length for time-frequency conversion, and determines whether the power spectrum density is within a reasonable range; If yes, temporarily store the power spectrum density data, perform average processing after a certain number of test iterations, make a test qualification judgment according to the rules, and record the result; if no, indicate an abnormality.

7. A portable vehicle-mounted Ethernet physical layer testing method according to claim 6, characterized in that: The test system performs data acquisition and data area screening, selects time domain data of suitable length for time-frequency transformation, and determines whether the power spectrum density is within a reasonable range, specifically including: the data acquisition module selects time domain data according to the resolution requirement, selects a suitable window function for fast Fourier transform, records the spectrum data within the resolution requirement range, and then determines whether the number of iterations meets a certain number; if so, averages all iterative spectrum data by frequency to obtain average spectrum data, calculates the upper and lower limit spectrum data according to the test rules, compares them with the average spectrum data, determines whether they are within the allowable range, and then marks the data to see whether it has passed the test, and records the average spectrum data within the required range.

8. A portable vehicle-mounted Ethernet physical layer testing method according to claim 1, characterized in that: The test during transmission distortion specifically includes: according to the test requirements, the test system sets the working parameters of the Ethernet PHY of the device under test, sets the working parameters of the data acquisition module, generates a sine wave for interference, controls the data acquisition module to acquire the bus data waveform, determines whether the effective acquisition bit number of the acquisition card is less than the set bit number, and if so, performs low-pass filtering on the captured continuous acquisition data, and then selects the time domain data of a unit for peak distortion measurement, and finally, after the number of iterations is met, the test system records the test results and restores the working parameters of the Ethernet PHY on the device under test.

9. A portable vehicle-mounted Ethernet physical layer testing method according to claim 8, characterized in that: The method of performing low-pass filtering on the captured continuously collected data and then selecting a unit of time domain data for peak distortion measurement specifically includes: calculating the peak distortion, i.e., the minimum mean square error, based on the ideal reference data, and judging all the peak distortion records after a certain number of iterations. When there is a peak distortion greater than a set value, the test is marked as failed; if not, the test is marked as passed, and all the peak distortion records and the judgment results are output.

Images