Apparatus and method for eliminating RF interference in wired data communication links

Abstract

This invention relates to an apparatus and method for eliminating RF interference coupled to a wired data communication link. The apparatus includes: an input terminal for providing an input signal that forms a digital representation of a signal received through the wired data communication link; an adaptive feedforward digital filter for filtering the input signal to provide a feedforward filtered input signal; an adaptive feedback digital filter for filtering a superimposed signal to provide a feedback filtered output signal, the superimposed signal representing the sum of the feedforward filtered input signal and the feedback filtered output signal; and a regulator for adjusting the adaptive feedforward digital filter and the adaptive feedback digital filter according to an error signal representing the difference between the input signal and the superimposed signal.

Description

Technical Field

[0001] This invention relates to an apparatus and method for eliminating radio frequency (RF) interference coupled into wired data communication links. Specifically, this invention relates to instantaneous narrow band interference (NBI) cancellation techniques used in automotive, industrial, and consumer electronics environments. Background Technology

[0002] like Figure 1 As shown, wired high-speed serial communication PHYs (physical interfaces) 110 and 120 exchange data via cable 130, the maximum length of which is defined by the corresponding standard. The length of cable 130 varies from tens of centimeters to approximately 100 meters. Cable 130 has limited coupling attenuation, depending on the cable type and quality. When these cables are placed in an adverse environment with an electromagnetic (EM) field 131, cable 130 acts as an antenna and picks up RF interference from the surrounding environment. As long as the RF interference 131 falls within the frequency band of the operating frequencies of PHYs 110 and 120, the receiver front-ends 111 and 121 treat the RF interference 131 as a (usable) signal. This phenomenon increases the bit error rate (BER), thereby reducing receiver performance and increasing the likelihood of link disconnection. This is a real problem, and it is even more severe in automotive environments because, due to safety requirements, sudden link disconnection is not permissible. RF interference 131 adds an additional burden to the complexity of the PHY, as it necessitates strict specification of the analog front-end and digital signal processing blocks. Summary of the Invention

[0003] The purpose of this invention is to provide a concept for eliminating or at least reducing RF interference at the PHY to solve the problems described above.

[0004] Specifically, the object of the present invention is to provide a concept for instantaneous detection and elimination of such RF interference without significantly increasing PHY complexity and power consumption.

[0005] The basic idea of ​​this invention is to detect and eliminate such RF interference "on the fly," that is, to perform such detection and elimination while the process of change is still in progress. RF interference detection can be performed by a time-domain correlator. Elimination can be based on prediction and history. If necessary, bootstrapping can be used to speed up detection. Different bootstrapping methods are proposed in this invention, namely: (a) FFT-based bootstrapping, for example, by using a 32-point FFT, the complexity is only 1 / 142 of the complexity of a 2048-point FFT used for comparison; (b) counter-based bootstrapping; and (c) digital phase-locked loop (DPLL)-based bootstrapping.

[0006] The novel concept proposed in this invention provides an efficient algorithm with adaptive feedforward and feedback schemes, as well as a power-efficient and silicon-region-efficient implementation.

[0007] By using this novel interference detection and cancellation method, the following advantages can be achieved: reduced requirements for more stringent receiver analog front-end specifications, improved signal-to-noise ratio (SNR) performance, less complex digital signal processing modules, and maintenance of consistently reliable and robust communication links with low power consumption. Adding blocks for interference detection and cancellation does not increase PHY complexity or power consumption. Digital signal processing algorithms can be used to detect and cancel interference while reducing hardware workload.

[0008] The concepts described in this invention can be applied to automotive applications using wired serial data communication links (e.g., Ethernet cables defined by 1000BASE-T1, 10GBASE-T1, and the future 25GBASE-T1). As future autonomous vehicles require increasing bandwidth to reliably transmit sensor data to the central processing unit, the techniques according to this invention can be advantageously applied to improve data transmission efficiency.

[0009] Besides automobiles, the technologies described in this article can also be applied to industrial and automation applications as well as consumer electronics.

[0010] To describe the invention in detail, the following terms, abbreviations, and symbols are used:

[0011] PHY physical layer device or physical interface

[0012] NBI narrowband interference

[0013] DPLL digital phase-locked loop

[0014] Fast Fourier Transform (FFT)

[0015] EM electromagnetic (field)

[0016] RF radio frequency

[0017] BER (Bit Error Rate)

[0018] SNR signal-to-noise ratio

[0019] FFE feedforward equalizer

[0020] DFE Decision Feedback Equalizer

[0021] FIR Finite Impulse Response (Filter)

[0022] ADC analog-to-digital converter

[0023] LPF low-pass filter

[0024] PAM pulse amplitude modulation

[0025] VCO voltage-controlled oscillator

[0026] BER (Bit Error Rate)

[0027] According to a first aspect, the present invention relates to an apparatus for eliminating radio frequency interference coupled to a wired data communication link, the apparatus comprising: an input terminal for providing an input signal forming a digital representation of a signal received through the wired data communication link; an adaptive feedforward digital filter for filtering the input signal to provide a feedforward filtered input signal; an adaptive feedback digital filter for filtering a superimposed signal to provide a feedback filtered output signal, the superimposed signal representing the sum of the feedforward filtered input signal and the feedback filtered output signal; and a regulator for adjusting the adaptive feedforward digital filter and the adaptive feedback digital filter according to an error signal representing the difference between the input signal and the superimposed signal.

[0028] This device can advantageously detect and eliminate RF interference on the fly without significantly increasing PHY complexity and power consumption. The "on-the-fly" detection and elimination of RF interference has the technical advantage of being able to perform this detection and elimination while the process of change (e.g., receiving sensor data) is still in progress. No interruption of data transmission is required.

[0029] In an exemplary implementation of the device, the regulator is used to adjust the adaptive feedforward digital filter and the adaptive feedback digital filter according to the bootstrap configuration of the feedforward digital filter and the feedback digital filter.

[0030] The bootstrap configuration of the feedforward and feedback digital filters is the initial configuration of the filter taps of the two filters in order to improve adaptive convergence.

[0031] However, this initial configuration is continuously updated to increase convergence not only during startup but also in the case of time-varying channels.

[0032] In an exemplary implementation of the device, the bootstrap configuration of the feedforward digital filter and the feedback digital filter is based on the signal characteristics of the input signal.

[0033] This offers the advantage that applying a bootstrap configuration can detect RF interference more quickly. Different bootstrap methods can be used, as described below.

[0034] In an exemplary implementation of the device, the bootstrap configuration of the feedforward digital filter and the feedback digital filter is based on the bandwidth of the input signal.

[0035] This offers the advantage of pre-configuring filter taps based on the bandwidth of the input signal, taking that bandwidth information into account. This means that filter taps for bandwidth portions present in the reference input signal can be prioritized over filter taps for bandwidth portions not present in the reference input signal.

[0036] In an exemplary implementation of the device, the device includes a fast Fourier transform (FFT) circuit that provides the spectrum of the input signal based on the FFT of the input signal.

[0037] This offers the advantage that by using this FFT, the frequency information of the input signal can be rapidly calculated, resulting in accurate and efficient bootstrapping configuration, thus enabling rapid convergence of the adaptive process.

[0038] In one example, the FFT circuit can perform a 32-point FFT. This 32-point FFT can be computed efficiently with low complexity.

[0039] In an exemplary implementation of the device, the device includes a detection circuit for detecting the bandwidth of the input signal by comparing the spectrum of the input signal with a threshold.

[0040] This offers the advantage that the threshold detection can be easily computed with low complexity.

[0041] In an exemplary implementation of the device, the device includes a correlator for providing autocorrelation of an input signal, wherein the bootstrap configuration of the feedforward digital filter and the feedback digital filter is based on the autocorrelation of the input signal.

[0042] This provides an advantage that frequency information of particularly narrowband interference can be estimated efficiently by using this autocorrelation.

[0043] The device can advantageously utilize the symmetry of the autocorrelation function and the fact that the periodic time function also includes a periodic autocorrelation function to provide a suitable bootstrapping configuration.

[0044] In one example, the correlator could be an adaptive correlator, specifically a 1-tap adaptive correlator. Such a 1-tap adaptive correlator can be implemented with low computational complexity.

[0045] In an exemplary implementation of the device, the device includes: a zero-crossing detector for detecting zero-crossings of the autocorrelation of the input signal; and a counter for determining the distance between the zero-crossings of the autocorrelation of the input signal, wherein the bootstrap configuration of the feedforward digital filter and the feedback digital filter is based on the distance between the zero-crossings of the autocorrelation of the input signal.

[0046] This provides an advantage that these distances between zeros indicate one or more frequency components of the input signal, i.e., frequency information of the input signal, which can be advantageously used to determine the bootstrap configuration of the filter.

[0047] In an exemplary implementation of the device, the device includes a frequency conversion table for converting the distance between zero-crossings of the autocorrelation of the input signal into frequency values, wherein the bootstrap configuration of the feedforward digital filter and the adaptive feedback digital filter is based on the frequency values.

[0048] This offers the advantage that, by using such a frequency conversion table, the distance between the zero-crossings of the autocorrelation of the input signal can be easily and efficiently converted into frequency values, which can be used to determine the bootstrap configuration of the feedforward digital filter and the adaptive feedback digital filter.

[0049] In an exemplary implementation of the device, the device includes a low-pass filter for low-pass filtering a frequency value, wherein the bootstrap configuration of the feedforward digital filter and the feedback digital filter is based on the low-pass filtered frequency value.

[0050] This offers the advantage that the low-pass filtered frequency component is the average frequency component of the input signal, with less fluctuation than the unfiltered frequency value. The low-pass filter can be adjusted to account for the history of the input signal.

[0051] In an exemplary implementation of the device, the device includes a digital phase-locked loop (PLL) for determining the frequency of autocorrelation of an input signal, wherein the bootstrap configuration of a feedforward digital filter and a feedback digital filter is based on the frequency of autocorrelation of the input signal.

[0052] This provides an advantage that DPLL can efficiently determine the frequency components of the input signal to provide accurate bootstrap configurations for feedforward and feedback filters.

[0053] In an exemplary implementation of this device, radio frequency interference is narrowband interference relative to the bandwidth of the input signal. For example, as Figure 8 As shown and described below, RF interference can be in the range of several MHz, while the bandwidth of the input signal can be in the range of approximately 400 MHz.

[0054] This provides an advantage that narrowband interference can be easily detected by using the frequency estimation methods described in this invention, such as FFT-based estimation, correlation-based estimation, counter-based estimation, or DPLL-based estimation described in this invention.

[0055] In an exemplary implementation of the device, the device includes: an equalizer for equalizing the channel transfer function of a wired data communication link, the equalizer including a feedforward equalizer for receiving superimposed signals, a decision feedback equalizer, and a decision device for providing an estimate of the symbols transmitted through the wired data communication link.

[0056] This offers the advantage that the equalizer can effectively eliminate ISI interference and provide accurate estimates of transmitted symbols, for example, the following regarding Figure 2 As described.

[0057] In an exemplary implementation of the device, the input signal includes a pulse-amplitude modulated (PAM) user signal.

[0058] This offers the advantage that such a device can be efficiently applied to automotive applications that utilize PAM-based user signals.

[0059] In an exemplary implementation of the device, the wired data communication link includes a shielded twisted-pair cable.

[0060] This offers the advantage that shielded twisted-pair cables are less sensitive to the coupling of RF interference caused by shielding. Therefore, the detection and elimination of RF interference can be performed with less complexity than in the case of unshielded cables.

[0061] In an exemplary implementation of the device, the wired data communication link includes an Ethernet cable, specifically an automotive Ethernet cable.

[0062] This device can be advantageously applied in automotive applications, such as in autonomous vehicles, to reliably transmit sensor data to a central processing unit using Ethernet cables defined in standards such as 1000BASE-T1, 10GBASE-T1 and the future 25GBASE-T1.

[0063] Furthermore, this device can be advantageously used to improve data communication, for example, as defined in the IEEE 802.3 standard for wired Ethernet.

[0064] Besides automobiles, this device can also be advantageously applied to industrial and automation applications as well as consumer electronics.

[0065] According to a second aspect, the present invention relates to a method for eliminating radio frequency interference coupled to a wired data communication link, the method comprising: providing an input signal that forms a digital representation of a signal received through the wired data communication link; filtering the input signal using an adaptive feedforward digital filter to provide a feedforward filtered input signal; filtering a superimposed signal using an adaptive feedback digital filter to provide a feedback filtered output signal, the superimposed signal representing the sum of the feedforward filtered input signal and the feedback filtered output signal; and adjusting the adaptive feedforward digital filter and the adaptive feedback digital filter according to an error signal representing the difference between the input signal and the superimposed signal.

[0066] This method can advantageously detect and eliminate RF interference on the fly without significantly increasing PHY complexity and power consumption. The "on-the-fly" detection and elimination of RF interference has the technical advantage of being able to be performed while the process affecting it (e.g., receiving or transmitting sensor data) is still in progress. No interruption of data transmission is required.

[0067] This method for eliminating radio frequency interference coupled into wired data communication links offers the same advantages as the corresponding device of the first aspect described above.

[0068] According to a third aspect, the present invention relates to a computer program product comprising computer-executable code or computer-executable instructions, which, when executed, cause at least one computer to perform the method according to the second aspect.

[0069] The computer program product may include a non-transient readable storage medium storing program code for use by a processor, the program code including instructions for performing the methods or computation blocks described below.

[0070] Computer program products can run on a computer, for example, by using... Figure 1 The wired base serial data communication link 130 shown runs on a processor or controller of a communication system. For example, a computer program product can run on the communication system, which includes: processing circuitry, such as a processor for processing and generating data (e.g., the program code described above); and a transceiver, including, for example, a transmitter and a receiver, such as... Figure 1 The PHY shown is used to communicate with other components of the communication system (e.g., Figure 1 The other PHY shown exchanges data; non-transient memory is used to store data (e.g., the program code described above).

[0071] Using this computer program product improves data transmission efficiency by eliminating RF interference. Attached Figure Description

[0072] Other embodiments of the present invention will be described in conjunction with the following drawings.

[0073] Figure 1 A schematic diagram of a wired serial data communication link 130 is shown, in which an EM field couples RF interference into the data communication link.

[0074] Figure 2 A block diagram of an apparatus 200 for eliminating RF interference according to a first example is shown.

[0075] Figure 3 A block diagram of an apparatus 300 for eliminating RF interference according to a second example is shown.

[0076] Figure 4 A block diagram of an apparatus 400 for eliminating RF interference according to a third example is shown.

[0077] Figure 5 A block diagram of an apparatus 500 for eliminating RF interference according to the fourth example is shown.

[0078] Figure 6 A block diagram of an exemplary digital phase-locked loop (DPLL) that can be used in device 500 according to the fourth example is shown.

[0079] Figure 7 A schematic diagram of a method 700 for eliminating RF interference according to the present invention is shown.

[0080] Figure 8 Performance figures 800a and 800b of the apparatus for eliminating RF interference according to the present invention are shown. Detailed Implementation

[0081] In the following detailed description, reference is made to the accompanying drawings, which form a part of this specification, illustrating specific aspects of the invention that can be practiced. It should be understood that other aspects may be utilized and structural or logical changes may be made without departing from the scope of the invention. Therefore, the following detailed description is not intended to be limiting, and the scope of the invention is defined by the appended claims.

[0082] It should be understood that the notes relating to the described method also apply to the device or system corresponding to performing the method, and vice versa. For example, if specific method steps are described, the corresponding device may include units that perform the described method steps, even if such units are not explicitly described or shown in the accompanying drawings. Furthermore, it should be understood that features of the various exemplary aspects described herein can be combined with each other unless otherwise explicitly stated.

[0083] The devices and methods described herein can be used to receive data via wired serial data communication links according to IEEE 802.3 and other standards for wired Ethernet. Early Ethernet used coaxial cable as the shared medium, while newer Ethernet variants, in conjunction with switches, use twisted-pair and fiber optic links. Ethernet standards include several cabling and signaling variants of the OSI physical layer (PHY) used with Ethernet.

[0084] Figure 1 A schematic diagram of a wired serial data communication link 130 is shown, in which an EM field couples RF interference into the data communication link. Two wired high-speed serial communication PHYs 110 and 120 exchange data through the wired serial data communication link 130. The length of cable 130 can range from a fraction of a meter to approximately 100 meters. Cable 130 has limited coupling attenuation, depending on the type and quality of the cable. Cable 130 is located in an unfavorable environment with an electromagnetic (EM) field 131. In this environment, cable 130 can act as an antenna for the EM field, picking up RF interference. As long as the RF interference 131 falls within the frequency band of the operating frequencies of PHYs 110 and 120, the receiver front-ends 111 and 121 will treat the RF interference 131 (e.g., in the form of narrowband interference 131) as a (usable) signal. This phenomenon increases the bit error rate (BER), thereby reducing receiver performance and increasing the likelihood of link disconnection.

[0085] Figure 2 A block diagram of an apparatus 200 for eliminating RF interference according to a first example is shown. Figure 1 As shown, RF interference 131 is coupled into the wired data communication link 130.

[0086] The device 200 includes an input terminal for providing an input signal 210. The input signal 210 forms a digital representation of a signal 201 received via a wired data communication link 130. An analog-to-digital converter 202 can be used to provide the digital input signal 210 based on the analog signal 201 received via the wired data communication link 130.

[0087] The device 200 includes an adaptive feedforward digital filter 220 for filtering the input signal 210 to provide a feedforward filtered input signal 221.

[0088] The device 200 includes an adaptive feedback digital filter 230 for filtering the superimposed signal 222 to provide a feedback filtered output signal 231. The superimposed signal 222 represents the sum 240 of the feedforward filtered input signal 221 and the feedback filtered output signal 231.

[0089] The device 200 includes a regulator 250 for adjusting an adaptive feedforward digital filter 220 and an adaptive feedback digital filter 230 based on an error signal 223. The error signal 223 represents the difference 241 between the input signal 210 and the superimposed signal 222.

[0090] For example, the adaptive feedforward digital filter 220 and the adaptive feedback digital filter 230 can be implemented using finite impulse response (FIR) digital filters. For instance, both filters 220 and 230 can have one filter tap. Alternatively, the two filters 220 and 230 can have two, three, four, five, six, seven, eight, nine, ten, or any other number of filter taps. In one example, the two filters 220 and 230 can have the same number of filter taps. Alternatively, the number of filter taps can be different. For example, the adaptive feedback digital filter 230 can have more filter taps than the adaptive feedforward digital filter 220, such as one, two, three, four, five, six, seven, eight, nine, ten, or any other value more. Alternatively, the filter taps of the adaptive feedback digital filter 230 may be fewer than those of the adaptive feedforward digital filter 220, for example, by 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, or any other value.

[0091] Alternatively, the adaptive feedforward digital filter 220 and the adaptive feedback digital filter 230 can be implemented using infinite impulse response (IIR) digital filters. Alternatively, one of the two filters 220 and 230 can be implemented using an FIR digital filter, while the other can be implemented using an IIR digital filter. The adaptive feedforward digital filter 220 and the adaptive feedback digital filter 230 can be implemented on a processor or controller using hardware, firmware, and / or software, for example, in conjunction with the coefficient adaptive engine 250.

[0092] The regulator 250 can be implemented by a coefficient adaptive engine (e.g., a processor or controller for implementing an adaptive algorithm). The adaptive algorithm can be a gradient algorithm, such as a least squares algorithm, such as the least mean squares (LMS) algorithm or the recursive least squares (RLS) algorithm.

[0093] The regulator 250 can be used to adjust the adaptive feedforward digital filter 220 and the adaptive feedback digital filter 230 according to the bootstrap configuration of the feedforward digital filter 220 and the feedback digital filter 230.

[0094] The bootstrap configuration of the feedforward digital filter 220 and the feedback digital filter 230 is the initial configuration of the filter taps of the two filters in order to improve adaptive convergence.

[0095] However, this initial configuration can be continuously updated to increase convergence not only during startup but also in the case of time-varying channels.

[0096] The bootstrap configuration of the feedforward digital filter 220 and the feedback digital filter 230 can be based on the signal characteristics of the input signal 210, such as the bandwidth of the input signal 210.

[0097] Radio frequency interference 131 can be narrowband interference relative to the bandwidth of the input signal 210. In some examples, narrowband interference may include a single frequency.

[0098] The apparatus 200 may include an equalizer 260 for equalizing the channel transfer function of the wired data communication link 130. The equalizer 260 may include a feedforward equalizer 265 that receives the superimposed signal 222, a decision feedback equalizer 263, and a decision device 262 for providing an estimate of the symbol 264 transmitted through the wired data communication link 130.

[0099] The equalizer 260 can be used to reduce inter-symbol interference and support the recovery of transmitted symbols.

[0100] The feedforward equalizer 265, implemented as a linear equalizer, such as an FIR filter, can be placed in series with the channel and can be used to generate an estimate of the channel's back-transfer function. If the transmitted symbols are real-valued, the feedforward equalizer 265 may include a real-valued FIR filter; or if the symbols are complex-valued (e.g., in the case of a QAM system), the feedforward equalizer 265 may include a complex-valued FIR filter. For example, their coefficients can be updated using the least mean squares (LMS) algorithm or any other suitable gradient algorithm.

[0101] For channels with severe amplitude distortion, the performance of a standalone linear equalizer is not very good. In fact, since the equalizer reproduces the channel's inverse transfer function to compensate for strong attenuation in certain frequency bands, it can produce strong gain in the same frequency band. This means that not only is the signal amplified in these bands, but any noise present is also amplified. This problem can be solved by using a decision feedback equalizer (DFE) 263. The decision feedback equalizer (DFE) can be implemented as a filter that uses the feedback of the detected symbols to generate an estimate of the channel output. The DFE 263 is fed with the detected symbols 264 provided by the decision device 262 and generates an output that is combined (261) with the output of the feedforward equalizer 265. The combination (261) can be additive or subtractive. As with the feedforward equalizer 265, the DFE 263 can include a real-valued FIR filter if the transmitted symbols are real-valued, or a complex-valued FIR filter if the symbols are complex-valued.

[0102] Since DFE 263 can only estimate the back cursor, it can be used in conjunction with feedforward equalizer 265.

[0103] During steady-state operation, DFE 263 includes an estimate of the channel's impulse response or an estimate of the channel's convolution with feedforward equalizer 265. Because DFE 263 can replicate the channel output, and the DFE output is combined with the input signal, DFE 263 can compensate for severe amplitude distortion without adding noise in high-distortion frequency bands.

[0104] Similar to the feedforward equalizer 265, the DFE coefficients can be updated using a gradient algorithm (such as the LMS algorithm). This gradient algorithm can also be implemented by the processor or controller that implements the regulator 250.

[0105] When using feedforward equalizer 265 and DFE 263 simultaneously, adaptive algorithms for these two equalizers can be designed accordingly to utilize the characteristics of the two equalizers 265 and 263 and avoid equalization conflicts.

[0106] The input signal 210 may include a pulse-amplitude modulated (PAM) user signal.

[0107] The PAM user signal is a modulated signal, in which the message information is encoded as the amplitude of a series of signal pulses. This is an analog pulse modulation scheme, where the amplitude of the carrier pulse train varies according to the sampled values ​​of the message signal. Demodulation is performed by detecting the amplitude level of the carrier in each individual cycle.

[0108] The PAM user signal can be modulated according to Ethernet communication standards. The PAM user signal can include a PAM-3 signal according to the 100BASE-T4 or BroadR-Reach Ethernet standard. The PAM user signal can include a PAM-5 signal according to 1000BASE-T Gigabit Ethernet. The PAM user signal can include a PAM-16 signal according to 10GBASE-T 10 Gigabit Ethernet, which uses a version of Tomlinson-Harashima Precoded (THP) with 16 discrete levels (PAM-16) of pulse amplitude modulation, in a two-dimensional checkerboard encoding called DSQ128. The PAM user signal can include a PAM-4 signal according to some copper Ethernet variants of 25 Gigabit Ethernet, or 100 Gigabit Ethernet, or 200 Gigabit Ethernet.

[0109] Device 200 is able to eliminate RF interference from all of the input signals and cable types described above.

[0110] In one example, wired data communication link 130 may include shielded twisted-pair cable.

[0111] In one example, wired data communication link 130 may include an Ethernet cable, specifically an automotive Ethernet cable.

[0112] Ethernet is a set of wired computer networking technologies commonly used in local area networks (LANs), metropolitan area networks (MANs), and wide area networks (WANs). Ethernet was first standardized as IEEE 802.3 in 1983. Since then, Ethernet has been continuously improved to support higher bit rates, more nodes, and longer link distances, while retaining many backward compatibility features.

[0113] Device 200 is able to eliminate RF interference coupled to the various Ethernet cables described above.

[0114] Figure 2The apparatus 200 shown can be based on a digital signal processing (DSP) algorithm that can run on a processor or controller implementing the regulator 250 and digital filters 220, 230 as described above. As described above, the taps of the adaptive feedforward filter 220 and feedback filter 230 can depend on the nature of the channel loss and bandwidth. The error signal 223 can be used to identify the coefficients for detecting and correcting RF interference. Therefore, the apparatus 200 achieves instantaneous detection and correction of RF interference.

[0115] Figure 3 A block diagram of an apparatus 300 for eliminating RF interference according to a second example is shown.

[0116] Device 300 corresponds to the above regarding Figure 2 The device 200 is described, but has some additional functions as described below.

[0117] This additional function relates to a specific evaluation of the input signal 210 to provide a bootstrap configuration to the regulator 250, thereby increasing the convergence of the coefficient adaptive engine 250.

[0118] Device 300 includes a Fast Fourier Transform (FFT) circuit 310 for providing the spectrum 311 of the input signal 210 based on the FFT of the input signal 210. This spectrum 311 of the input signal 210 can be evaluated by regulator 250 to increase convergence. For example, coefficient adaptive engine 250 can use the frequency information 311 of the input signal 210 to weight adaptive coefficients based on this frequency information 311. If the input signal 210 is distorted due to narrowband interference 131 (as mentioned above...), Figure 1 If the distortion is due to (as described), or even due to single-frequency interference, the coefficient adaptive engine 250 can adjust its adaptive method to improve convergence within the frequency range indicated by the frequency information 311. For example, a bootstrap configuration corresponding to the filter coefficients of the narrowband interference 131 or the single-frequency interference can be applied. When using such a bootstrap configuration of the filter taps, the convergence of the regulator 250 can be greatly increased.

[0119] This bootstrapping configuration of the filter taps can be applied not only to the initial configuration, but also to time-varying channel variations, and even time-varying RF interference 131.

[0120] In one example, FFT circuit 310 can perform a 32-point FFT. In other examples, FFT circuit 310 can perform 4, 8, 16, 64, 128, 256-point FFTs or more.

[0121] The device 300 may also include a detection circuit ( Figure 3 (Not shown in the image), the detection circuit is used to detect the bandwidth of the input signal 210 by comparing the spectrum 311 of the input signal 210 with a threshold. The regulator 250 can use this bandwidth information detected by the detection circuit to give greater weight to filter taps that are related to the bandwidth of the input signal 210, rather than giving greater weight to filter taps that are unrelated to the bandwidth of the input signal 210.

[0122] The threshold can be an adaptive threshold based on the spectral power of the input signal 210, etc. The threshold can be used to provide a sufficient distance to the noise spectrum of the input signal 210, such that only frequencies of the input signal 210 that have a predetermined or configurable distance from the noise spectrum of the input signal are detected.

[0123] The performance of RF detection and cancellation can vary depending on the bandwidth of the input signal 210, which may be a baseband signal. For example, a PHY system with a wide bandwidth and dependent on the end application may be slow. However, the performance of RF detection and cancellation can be greatly improved by using a device 300 utilizing a bootstrap configuration as described above. Specifically, as described above, performance can be improved by using a small “32-point FFT” block 310 for bootstrap. This means that a simple FFT can be used to initialize the coefficients of filters 220 and 230.

[0124] Figure 4 A block diagram of an apparatus 400 for eliminating RF interference according to a third example is shown.

[0125] Device 400 corresponds to the above regarding Figure 2 The device 200 is described, but has some additional functions as described below.

[0126] This additional function relates to a specific evaluation of the input signal 210 to provide a bootstrap configuration to the regulator 250, thereby increasing the convergence of the coefficient adaptive engine 250.

[0127] As described below, replacing the above regarding Figure 3 The fast Fourier transform (FFT) circuit 310 described is used to provide frequency information 311 of the input signal 210. The device 400 uses another mechanism to provide this frequency information of the input signal 210 to the regulator 250.

[0128] Device 400 includes a correlator 410 for providing an autocorrelation 412 of the input signal 210. The bootstrap configuration of the feedforward digital filter 220 and the feedback digital filter 230 can be based on the autocorrelation 412 of the input signal 210. For narrowband interference 131 (as mentioned above...) Figure 1(As described), even with single-frequency interference, the autocorrelation 412 of the input signal 210 can depend on the frequency of the input signal. This frequency information can be used by the coefficient adaptive engine 250 to apply appropriate bootstrapping configurations to the filter taps of the feedforward filter 220 and the feedback filter 230.

[0129] For example, the autocorrelation function of an infinitely long sinusoidal signal is a cosine signal. Device 400 can utilize the symmetry of the autocorrelation function and the fact that periodic time functions also include periodic autocorrelation functions to provide a suitable bootstrapping configuration for filters 220 and 230. It can also be observed that the maximum value of the autocorrelation function occurs at position 0. This maximum value appears in each period of the autocorrelation function. Without shifting, the function is most similar to itself. Due to the periodicity of the function, it is the same at the beginning of each period. Periodic signals (e.g., including...) Figure 1 This characteristic of the autocorrelation function of the input signal 210 of the narrowband interference 131 shown can be advantageously used to provide frequency information to the coefficient adaptive engine (i.e., regulator 250).

[0130] The correlator 410 can be implemented as an adaptive correlator, specifically a 1-tap adaptive correlator. Alternatively, the correlator 410 can be implemented as a 2-tap correlator, a 3-tap correlator, a 4-tap correlator, a 5-tap correlator, a 6-tap correlator, a 7-tap correlator, an 8-tap correlator, a 9-tap correlator, a 10-tap correlator, or a correlator with more than one tap.

[0131] The device 400 may further include a zero-crossing detector 420 for detecting zero-crossings 421 of the autocorrelation 412 of the input signal 210. The device 400 may also include a counter 430 for determining distances 431 between the zero-crossings 421 of the autocorrelation 412 of the input signal 210. The bootstrap configuration of the feedforward digital filter 220 and the feedback digital filter 230 may be based on the distances 421 between the zero-crossings 421 of the autocorrelation 412 of the input signal 210. As described above, these distances 431 between the zero-crossings 421 provide indications of one or more frequency components of the input signal 120, i.e., frequency information of the input signal 210.

[0132] For example, counter 430 can be reset after a corresponding zero crossing is detected to accurately count the distance 431 between two zero crossings 421.

[0133] The device 400 may further include a frequency conversion table 440 for converting the distance 431 between the zero-crossings 421 of the autocorrelation 412 of the input signal 210 into frequency values ​​441. The bootstrap configuration of the feedforward digital filter 220 and the adaptive feedback digital filter 230 may be based on these frequency values ​​441.

[0134] The frequency conversion table 440 can be pre-configured or adapted based on knowledge of the input signal 210 and / or narrowband interference 131.

[0135] The device 400 may include a low-pass filter 450 for low-pass filtering the frequency value 441. The bootstrap configuration of the feedforward digital filter 220 and the feedback digital filter 230 may be based on the low-pass filtered frequency value 451. By using the low-pass filter 450, the frequency information can be stable and free from fluctuations. This results in a more efficient bootstrap configuration.

[0136] The following example illustrates how to calculate the frequency 441 (442) based on the counter output 431. The following variables can be used:

[0137] Tp: Precision of counter 430, in seconds;

[0138] N: The counter value between two zero crossings.

[0139] Time can be defined as Time (T) = Tp * N (in seconds).

[0140] The frequency can be determined as 1 / time (in Hertz).

[0141] Frequency calculation 442 can be implemented using a lookup table with discrete intervals. The following time-frequency correspondence can be used as an example:

[0142] T1 <T<=T2 F1

[0143] T2 <T<=T3 F2,

[0144] Among them, T1 <T2<T3。

[0145] This means that if the time value T is between T1 and T2, the frequency value F1 is calculated; if the time value T is between T2 and T3, the frequency value F2 is calculated, where T1... <T2<T3。

[0146] The performance of RF detection and cancellation can vary depending on the bandwidth of the input signal 210, which may be a baseband signal. For example, a PHY system with a wide bandwidth and dependent on the end application may be slow. However, the performance of RF detection and cancellation can be significantly improved by using a device 400 utilizing a bootstrap configuration as described above. Specifically, as Figure 4 As shown and as described above, performance can be improved by using a 1-tap correlator with a counter for bootstrapping. This means that a simple correlator 410 can be used to initialize the coefficients of filters 220 and 230.

[0147] Figure 5 A block diagram of an apparatus 500 for eliminating RF interference according to the fourth example is shown.

[0148] Device 500 corresponds to the above regarding Figure 4 The device 400 is described, but has some different functions for determining the frequency information of the input signal 210 as described below.

[0149] This additional function relates to a specific evaluation of the relevant information 413 determined by correlator 410, in order to provide a bootstrap configuration to regulator 250, thereby increasing the convergence of coefficient adaptive engine 250.

[0150] As described below, replacing the above regarding Figure 4 The zero-crossing detector 420, counter 430, frequency calculation 442, and low-pass filter 450 described herein, the device 500 uses a digital phase lock loop (DPLL) 510 to provide frequency information 511 of the input signal 210 to the regulator 250.

[0151] The device 500 includes a digital phase-locked loop 510 for determining the frequency 511 of the autocorrelation 410 of the input signal 210. The bootstrap configuration of the feedforward digital filter 220 and the feedback digital filter 230 can be based on the frequency 511 of the autocorrelation 410 of the input signal 210 determined by the DPLL 510.

[0152] DPLL 510 can be described as follows: Figure 6 Implemented as described.

[0153] The performance of RF detection and cancellation can vary depending on the bandwidth of the input signal 210, which may be a baseband signal. For example, a PHY system with a wide bandwidth and dependent on the end application may be slow. However, the performance of RF detection and cancellation can be significantly improved by using a device 500 utilizing a bootstrap configuration as described above. Specifically, as Figure 5 As shown, and as described above, or as... Figure 6 As shown, performance can be improved by using a DPLL 510 for bootstrapping. This means that a simple DPLL 510 can be used to initialize the coefficients of filters 220 and 230.

[0154] Figure 6 A block diagram of an exemplary digital phase-locked loop (DPLL) that can be used in the device 500 according to the fourth example is shown.

[0155] The DPLL 510 includes a phase detector 520 for receiving the autocorrelation signal 412 of the input signal 210. The output of the phase detector 520 is passed to the PI controller 530.

[0156] The PI controller 530 employs a feedback control loop mechanism. The PI controller 530 performs continuous calculations. Error value It is calculated as the difference between the desired setpoint (SP) and the measured process variable (PV), and corrections are applied based on proportional and integral terms (denoted as P and I, respectively). Term P is proportional to the current value of the SP-PV error. The term I interprets past values ​​of the SP-PV error, and these past values ​​are integrated over time to obtain term I.

[0157] The output of the PI controller 530 is passed to a low-pass filter 540 that determines a voltage 541. This voltage 541 drives a voltage-controlled oscillator (VCO) 550, which determines frequency and phase information 511 based on the self-excited oscillator frequency (Fvco) 551 and the initial phase (Pvco) 552. The frequency and phase information 511 is then passed through a frequency divider 560 and further provided to a phase detector 520.

[0158] The frequency and phase information 511 provided by the DPLL 510 represents information about the frequency components of the input signal 210, and can be advantageously used to provide bootstrapping configurations for the filter taps of the adaptive feedforward filter 220 and the adaptive feedback filter 230, for example, as described above regarding... Figure 5 As described.

[0159] The following example describes how to operate the DPLL 510 and determine frequency and phase information 511, including phase and frequency, according to the following relationship:

[0160] Phase = Pvco + 2 π V Kvco Ts;

[0161] Frequency = 2 π Fvco t+ phase.

[0162] The following variables are used in this example:

[0163] Pvco(552): represents the initial phase of VCO(550);

[0164] V: Represents phase error;

[0165] Kvco: represents the gain constant of VCO (550) (Hz / V);

[0166] Ts: Indicates the sampling period;

[0167] Fvco (551): Represents the frequency of the self-excited oscillator, in Hz;

[0168] t: represents instantaneous time;

[0169] Phase: Represents the steady-state phase output;

[0170] Freq: Indicates the steady-state frequency output.

[0171] Figure 7 A schematic diagram of a method 700 for eliminating RF interference according to the present invention is shown.

[0172] Method 700 includes providing (701) an input signal 210, which forms a digital representation of a signal received via a wired data communication link 130, for example, as described above regarding... Figures 2 to 5 As described.

[0173] Method 700 includes filtering the input signal 210 (702) via an adaptive feedforward digital filter 220 to provide a feedforward filtered input signal 221, for example, as described above regarding Figures 2 to 5 As described.

[0174] Method 700 includes filtering the superimposed signal 222 by an adaptive feedback digital filter 230 (703) to provide a feedback filtered output signal 231, the superimposed signal 222 representing the sum 240 of the feedforward filtered input signal 221 and the feedback filtered output signal 231, for example, as described above regarding Figures 2 to 5 As described.

[0175] Method 700 includes adjusting (704) an adaptive feedforward digital filter 220 and an adaptive feedback digital filter 230 according to an error signal 223, where the error signal 223 represents the difference 241 between the input signal 210 and the superimposed signal 222, for example, as described above regarding... Figures 2 to 5 As described.

[0176] Figure 8 Performance figures 800a and 800b of the apparatus for eliminating RF interference according to the present invention are shown.

[0177] In Figure 800a on the left, Figure 801 represents the above regarding... Figures 2 to 7 The input signal 210 described is subjected to, for example Figure 1The effect of narrowband interference 131 is shown. Figure 802 represents the output signal of a device (e.g., device 100) according to the invention. This output signal corresponds to the input signal 210 in which narrowband interference 131 has been eliminated.

[0178] In this example, a 100mVpp interference is applied to the input signal 210. Figure 8 The following use cases were considered:

[0179] –1000BASE-T1 PHY automotive standard with PAM3;

[0180] – The insertion loss at the Nyquist frequency is relatively high, approximately 20 dB;

[0181] -1Vpp emitter voltage, no random noise.

[0182] Figure 800b on the right shows the eye diagram for each sample voltage. It can be observed that after a certain number of samples—in this example of about 40,000 samples—the eye diagram clearly shows three symbols that can be easily detected by applying some threshold. These three symbols correspond to the original symbols of PAM3 modulation.

[0183] Therefore, performance figures 800a and 800b illustrate the advantages of the technology described in this invention, namely: "instantaneous" detection and correction of RF interference; readjustment when frequency and phase changes occur, as well as low power consumption and low implementation complexity, resulting in a smaller silicon area.

[0184] Readjustment refers to the ability to readjust the frequency estimate when the frequency and phase change slightly. It eliminates the need for recalculating the frequency using FFT or other frequency estimation methods as described above.

[0185] This invention also supports computer program products comprising computer-executable code or computer-executable instructions, which, when executed, cause at least one computer to perform the execution and computation steps described herein, specifically the methods and processes described above. The computer program product may include a readable, non-transitory storage medium storing program code for computer use. The program code can perform the processing and computation steps described herein, specifically the methods and processes described above.

[0186] While a particular feature or aspect of the invention may have been disclosed in combination with only one of several implementations, such feature or aspect may be combined with one or more features or aspects of other implementations whenever any given application or particular application requires or is advantageous to any given application or particular application. Furthermore, with regard to the terms “comprising,” “having,” “having,” or synonyms thereof as used in the detailed description or claims, these terms are intended to be included in a manner similar to the term “comprising.” Similarly, the terms “exemplarily” and “for example” mean only examples and not best or optimal. The terms “coupled,” “connected,” and their synonyms may be used. It should be understood that these terms may be used to indicate that two elements cooperate or interact with each other, whether the two elements are in direct physical contact or electrical contact, or whether the two elements are not in direct contact with each other.

[0187] While specific aspects have been illustrated and described herein, those skilled in the art will understand that various alternatives and / or equivalent implementations may be used instead of the specific aspects shown and described without departing from the scope of the invention. This application is intended to cover any modifications or variations to the specific aspects discussed herein.

[0188] Although the elements in the claims are stated in a specific order by their corresponding labels, the elements are not necessarily limited to that specific order of implementation unless the claims imply a specific order for implementing some or all of these elements.

[0189] Based on the above insights, many alternatives, modifications, and variations will be apparent to those skilled in the art. Of course, those skilled in the art will readily understand that numerous other applications of the invention exist besides those described herein. Although the invention has been described with reference to one or more specific embodiments, those skilled in the art will understand that many changes can be made to the invention without departing from its scope. Therefore, it should be understood that the invention can be practiced in ways other than those specifically described herein, as long as it remains within the scope of the appended claims and their equivalents.

Claims

1. An apparatus (200) for eliminating radio frequency interference (131) coupled to a wired data communication link (130), characterized in that, The device (200) includes: An input terminal is provided to provide an input signal (210) that forms a digital representation of a signal received through the wired data communication link (130); An adaptive feedforward digital filter (220) is used to filter the input signal (210) to provide a feedforward filtered input signal (221). An adaptive feedback digital filter (230) is used to filter a superimposed signal (222) to provide a feedback filter output signal (231), the superimposed signal (222) representing the sum (240) of the feedforward filter input signal (221) and the feedback filter output signal (231). A regulator (250) is used to adjust the adaptive feedforward digital filter (220) and the adaptive feedback digital filter (230) according to an error signal (223), the error signal (223) representing the difference (241) between the input signal (210) and the superimposed signal (222).

2. The apparatus (200) according to claim 1, characterized in that, The regulator (250) is used to adjust the adaptive feedforward digital filter (220) and the adaptive feedback digital filter (230) according to the bootstrap configuration of the feedforward digital filter (220) and the feedback digital filter (230).

3. The apparatus (200) according to claim 2, characterized in that, The bootstrap configuration of the feedforward digital filter (220) and the feedback digital filter (230) is based on the signal characteristics of the input signal (210).

4. The apparatus (200) according to claim 2 or 3, characterized in that, The bootstrap configuration of the feedforward digital filter (220) and the feedback digital filter (230) is based on the bandwidth of the input signal (210).

5. The apparatus (200) according to claim 4, characterized in that, include: A Fast Fourier Transform (FFT) circuit (310) is provided to provide the spectrum (311) of the input signal (210) based on the FFT of the input signal (210).

6. The apparatus (200) according to claim 5, characterized in that, include: A detection circuit is used to detect the bandwidth of the input signal (210) by comparing the spectrum (311) of the input signal (210) with a threshold.

7. The apparatus (200) according to claim 4, characterized in that, include: A correlator (410) is provided to provide the autocorrelation (412) of the input signal (210). The bootstrap configuration of the feedforward digital filter (220) and the feedback digital filter (230) is based on the autocorrelation (412) of the input signal (210).

8. The apparatus (200) according to claim 7, characterized in that, include: Zero-crossing detector (420) is used to detect the zero-crossing (421) of the autocorrelation (412) of the input signal (210). A counter (430) is used to determine the distance between the zero crossings (421) of the autocorrelation (412) of the input signal (210). The bootstrap configuration of the feedforward digital filter (220) and the feedback digital filter (230) is based on the distance between the zero-crossing (421) of the autocorrelation (412) of the input signal (210).

9. The apparatus (200) according to claim 8, characterized in that, include: A frequency conversion table (440) is used to convert the distance between the zero crossings (421) of the autocorrelation (412) of the input signal (210) into a frequency value (441). The bootstrap configuration of the feedforward digital filter (220) and the adaptive feedback digital filter (230) is based on the frequency value (441).

10. The apparatus (200) according to claim 9, characterized in that, include: A low-pass filter (450) is used to perform low-pass filtering on the frequency value (441). The bootstrap configuration of the feedforward digital filter (220) and the feedback digital filter (230) is based on the low-pass filter frequency value (451).

11. The apparatus (200) according to claim 7, characterized in that, include: A digital phase-locked loop (510) is used to determine the frequency (511) of the autocorrelation (412) of the input signal (210). The bootstrap configuration of the feedforward digital filter (220) and the feedback digital filter (230) is based on the frequency (511) of the autocorrelation (412) of the input signal (210).

12. The apparatus (200) according to claim 1, characterized in that, The radio frequency interference (131) is narrowband interference relative to the bandwidth of the input signal (210).

13. The apparatus (200) according to claim 1, characterized in that, include: Equalizer (260), the equalizer (260) is used to equalize the channel transfer function of the wired data communication link (130); The equalizer (260) includes a feedforward equalizer (265), a decision feedback equalizer (263), and a decision device (262); the feedforward equalizer (265) is used to receive the superimposed signal (222), and the decision device (262) is used to estimate the symbols (264) that provide the transmission of the wired data communication link (130).

14. The apparatus (200) according to claim 1, characterized in that, The input signal (210) includes a pulse amplitude modulation (PAM) user signal.

15. The apparatus (200) according to claim 1, characterized in that, The wired data communication link (130) includes a shielded twisted-pair cable.

16. The apparatus (200) according to claim 1, characterized in that, The wired data communication link (130) includes an automotive Ethernet cable.

17. A method (700) for eliminating radio frequency interference (131) coupled to a wired data communication link (130), characterized in that, The method (700) is performed by the apparatus of any one of claims 1 to 16; The method (700) includes: An input signal (210) is acquired, which forms a digital representation of a signal received through the wired data communication link (130); The input signal (210) is filtered (702) by an adaptive feedforward digital filter (220) to provide a feedforward filtered input signal (221). The superimposed signal (222) is filtered (703) by an adaptive feedback digital filter (230) to provide a feedback filter output signal (231), the superimposed signal (222) representing the sum (240) of the feedforward filter input signal (221) and the feedback filter output signal (231). The adaptive feedforward digital filter (220) and the adaptive feedback digital filter (230) are adjusted (704) according to the error signal (223), wherein the error signal (223) represents the difference (241) between the input signal (210) and the superimposed signal (222).

18. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed on a processor, implement the method of claim 17.

19. A computer program product, characterized in that, Includes a program that, when run on a processor, implements the method of claim 17.

Images