Receiver circuit

Receiver circuit technology field

[0001] The present disclosure relates to a receiver circuit that includes a receiver circuit that provides co-channel interference compensation (CIC).

Background technique

[0002] The in-band on-channel (IB0C) digital radio broadcasting standard for the FM frequency band is defined by the "in-band/on-channel digital radio broadcasting standard" document published by the National Radio System Committee (NRSC). This file is also the basis for the transmitted IB0C signals that can be received by HD RadioTM certified receivers. HD RadioTM transmission is also based on "HD RadioTM Air Interface Design Specification Layer 1 FM" file number SY_IDD_1011sG revision G on August 23, 2011.

Summary of the invention

[0003] According to a first aspect of the present disclosure, there is provided a receiver system including:

[0004] An input terminal configured to receive input signaling including multiple antenna signals, each antenna signal being received at a different antenna, wherein the multiple antenna signals each include information corresponding to a first frequency interval and a second frequency interval ;

[0005] A first AoA block configured to determine a first angle of arrival associated with a first frequency interval;

[0006] A second AoA block configured to determine a second angle of arrival associated with a second frequency interval;

[0007] A first weighting determination block configured to perform any one of the following operations based on the first angle of arrival and the second angle of arrival:

[0008] The first weighting value is set to a value for constructively combining the information of the first frequency interval corresponding to the plurality of antenna signals; or

[0009] setting the first weighting value to a value for destructively combining the information of the first frequency interval corresponding to the plurality of antenna signals;

[0010] a first weighting application block configured to apply a first weighting value to information corresponding to a first frequency interval in the plurality of antenna signals, so as to provide a weighted first frequency interval signal;

[0011] a second weighting determination block configured to set a second weighting value to a value for constructively combining information of a second frequency interval corresponding to a plurality of antenna signals;

[0012] a second weighting application block configured to apply a second weighting value to information corresponding to a second frequency interval in the plurality of antenna signals, so as to provide a weighted second frequency interval signal;

[0013] A signal combining block configured to combine the weighted first frequency interval signal and the weighted second frequency interval signal to provide output signaling.

[0014] In one or more embodiments, the first AoA block includes an outer first AoA block configured to determine an outer first angle of arrival associated with an outer subgroup of the first frequency interval . The first weight determination block may be configured to set the set first weight value based on the outer first angle of arrival and the second angle of arrival. The first weighting application block may be configured to apply the first weighting value to information corresponding to the entire first frequency interval among the plurality of antenna signals, so as to provide a weighted first frequency interval signal.

[0015] In one or more embodiments, the second AoA block includes an intermediate second AoA block that is configured

The second angle of arrival is determined based on the intermediate subgroup of the second frequency interval. The second weight application block may be configured to apply the second weight value to the information corresponding to the entire second frequency interval among the plurality of antenna signals, so as to provide a weighted second frequency interval signal.

[0016] In one or more embodiments, the first weight determination block is configured to set the first weight value based on whether the difference between the first angle of arrival and the second angle of arrival satisfies an external threshold.

[0017] In one or more embodiments, the first weight value includes a plurality of first antenna weight values, and each of the plurality of antenna signals has a first antenna weight value. The second weight value may include multiple second antenna weight values, and each of the multiple antenna signals has one second antenna weight value.

[0018] In one or more embodiments, the first weighting application block is configured to: multiply each of the first antenna weight values by a first frequency corresponding to an associated antenna signal of the plurality of antenna signals The interval information is used to provide a plurality of first antenna weighted signals; and the plurality of first antenna weighted signals are added together to provide a weighted first frequency interval signal.

[0019] In one or more embodiments, the second weighting application block is configured to: multiply each of the second antenna weight values by a second frequency corresponding to an associated antenna signal of the plurality of antenna signals Interval information so as to provide a plurality of second antenna weighted signals; and the plurality of second antenna weighted signals are added together to provide a weighted second frequency interval signal.

[0020] In one or more embodiments, the first weight determination block is configured to set the first weight value based on a second weight value for constructively combining information of the first frequency interval corresponding to the plurality of antenna signals A weighted value.

[0021] In one or more embodiments, the first weight determination block is configured to: determine the second weight value by modifying the second weight value based on the spatial covariance matrix of the information in the plurality of antenna signals corresponding to the first frequency interval Adjusting the second weighting value; and setting the first weighting value to the adjusted second weighting value.

[0022] In one or more embodiments, each of the multiple antenna signals further includes information corresponding to the third frequency interval. The receiver system may additionally include: a third AoA block configured to determine a third angle of arrival associated with the third frequency interval; a third weight determination block configured to be based on the third angle of arrival sum The second angle of arrival performs any one of the following operations: the third weighting value is set to a value for constructively combining the information of the third frequency interval corresponding to a plurality of antenna signals, or the third weighting value is set Is used to destructively combine the value of the information corresponding to the third frequency interval of the multiple antenna signals; a third weighting application block, the third weighting application block is configured to apply the third weighting value to the multiple antennas Information of the third frequency interval in the signal, so as to provide a weighted third frequency interval signal; and [0023] wherein the signal combination block is configured to combine the weighted first frequency interval signal, the weighted second frequency interval signal, and the weighted The third frequency interval signal in order to provide output signaling.

[0024] In one or more embodiments, the first AoA block includes: an internal first AoA block configured to determine an internal first angle of arrival associated with an internal subgroup of the first frequency interval; configured to determine The outer first AoA block of the outer first angle of arrival associated with the outer subgroup of the first frequency interval. The first weighting determination block may be configured to: determine the outer first angle difference as the difference between the outer first angle of arrival and the second angle of arrival; if the outer first angle difference is less than the outer threshold, then: The weighting value is set to a value used to constructively combine the information of the first frequency interval corresponding to the multiple antenna signals; the internal first angle difference is determined as the difference between the internal first arrival angle and the second arrival angle If the inner first angle difference is less than the predetermined ratio of the outer first angle difference, the first weighted value is set to a value for constructively combining the information of the first frequency interval corresponding to the multiple antenna signals; If the inner first angle difference is greater than the predetermined ratio of the outer first angle difference, the first weighted value is set as a value for destructively combining the information corresponding to the first frequency interval of the plurality of antenna signals

value.

[0025] In one or more embodiments, the first weighting application block is configured to apply the first weighting value to the information corresponding to the entire first frequency interval in the plurality of antenna signals, so as to provide the weighted first frequency Interval signal.

[0026] In one or more embodiments, the second weighting application block is configured to apply the second weighting value to the information corresponding to the entire second frequency interval in the plurality of antenna signals, so as to provide the weighted second frequency Interval signal.

[0027] In one or more embodiments, the third AoA block includes: an internal third AoA block configured to determine the internal third angle of arrival associated with the internal subgroup of the third frequency interval; configured to determine The outer third AoA block of the outer third angle of arrival associated with the outer subgroup of the third frequency interval. The third weighting determination block may be configured to: determine the outer third angle difference as the difference between the outer third angle of arrival and the second angle of arrival; if the outer third angle difference is less than the outer threshold, then: the third weighted value Set to the value used to constructively combine the information of the third frequency interval corresponding to the multiple antenna signals; determine the internal third angle difference as the difference between the internal third angle of arrival and the second angle of arrival; if the internal If the third triangle difference is smaller than the predetermined ratio of the outer third angle difference, the third weighting value is set to a value for constructively combining the information of the third frequency interval corresponding to the multiple antenna signals; if the inner third angle difference If the value is greater than the predetermined ratio of the external third angle difference, the third weighting value is set to a value for destructively combining the information of the third frequency interval corresponding to the plurality of antenna signals.

[0028] In one or more embodiments, the third weighting application block is configured to apply the third weighting value to the information corresponding to the entire third frequency interval in the plurality of antenna signals, so as to provide a weighted third frequency Interval signal.

[0029] A car radio receiver system including any receiver circuit disclosed herein or configured to perform any method disclosed herein may be provided.

[0030] An integrated circuit or electronic device including any circuit or system disclosed herein may be provided.

[0031] Although the present disclosure is susceptible to various modifications and alternative forms, its details have been shown in the drawings by way of example and will be described in detail. However, it should be understood that other embodiments beyond the specific embodiments described are also possible. It also covers all modifications, equivalents and alternative embodiments falling within the spirit and scope of the appended claims.

[0032] The above discussion is not intended to represent every example embodiment or every implementation within the scope of the current or future set of claims. The drawings and the detailed description also illustrate various example embodiments. Various example embodiments can be more fully understood by considering the following detailed description in conjunction with the accompanying drawings.

Description of the drawings

[0033] One or more embodiments will now be described with reference to the accompanying drawings only by way of example, in which:

[0034] FIG. 1 shows a simplified form of one type of IBOC signal;

[0035] FIGS. 2a and 2b show the IBOC signal of FIG. 1, and the first adjacent lower adjacent FM channel and the upper adjacent FM channel; [0036] FIG. 3 shows a first adjacent (FM) interference signal and Estimation of the periodogram power spectral density (PSD) of the (H) IB0C transmission of the first neighboring (FM) interference signal;

[0037] FIG. 4 shows a spectrum plot of an all-digital implementation of the IBOC signal;

[0038] FIG. 5 shows an example embodiment of a receiver;

[0039] FIG. 6 shows a detailed view of a method that can be performed by a receiver system, such as the receiver system in FIG. 5;

[0040] FIG. 7 shows another method of operating the receiver system; and

[0041] FIG. 8 shows a receiver system representing a generalized version of the processing of FIG. 7.

Detailed ways

[0042] The in-band on-channel (IB0C) digital radio broadcasting standard for the FM frequency band is defined by the "in-band/on-channel digital radio broadcasting standard" document published by the National Radio System Committee (NRSC). This file is also the basis for the transmitted IB0C signals that can be received by HD RadioTM certified receivers. HD RadioTM transmission is also based on "HD RadioTM Air Interface Design Specification Layer 1 FM" file number SY_IDD_1011sG revision G on August 23, 2011.

[0043] FIG. 1 shows a simplified form of one type of IBOC signal 100, which is a so-called "hybrid IBOC FM" signal and is denoted herein as "hybrid IBOC". The frequency is shown on the horizontal axis, where 0 Hz represents the carrier frequency. The vertical dimension of Figure 1 represents power.

[0044] The mixed signal 100 is a combination/addition of an analog FM signal 110 and a digital modulation signal 112, 114. The analog FM signal 110 occupies a bandwidth of 200 kHz, which is between -100 kHz and 100 kHz and represents a central part centered on the carrier frequency. The digital modulation signals 112, 114 occupy a bandwidth of approximately 200 kHz. However, the digitally modulated signal is divided into a lower sideband 112 and an upper sideband 114, each having a bandwidth of about 100 kHz. The lower sideband is positioned on the frequency spectrum at a distance of 100 kHz below the carrier frequency. The upper sideband 114 is located on the spectrum at a distance of 100 kHz higher than the carrier frequency. In this way, the lower sideband 112 is lower than the lowest frequency of the central part, and the upper sideband 114 is higher than the highest frequency of the central part. The digitally modulated signals 112, 114 can use Orthogonal Frequency Division Multiplexing (OFDM), where the number of sub-carriers can be changed according to the selected service/transmission mode. [0045] The total power of the digital modulation signals 112, 114 may be about one percent of the power of the analog host FM signal 110. The mixed IBOC signal 100 can therefore be regarded as a noisy FM signal.

[0046] The so-called "channel grid" defines the reserved channel bandwidth for analog FM signals. The channel bandwidth in Band II is adjusted to 200 kHz, as shown in the simulation center part 110 in FIG. 1. Therefore, the lower digital OFDM sideband 112 and the upper digital OFDM sideband 114 may correspond to the frequency ranges of the first adjacent lower adjacent FM channel and the upper adjacent FM channel. This will be described with additional reference to Figures 2a and 2b.

[0047] FIG. 2a shows the IBOC signal of FIG. 1, and the first adjacent lower adjacent FM channel 220a and the first adjacent upper adjacent FM channel 230a.

[0048] The primary sidebands 212, 214 of the IBOC signal occupy about 100 kHz of the lower adjacent channel 220a and the upper adjacent channel

About 100kHz of 230a. Due to the fact that the primary sidebands 212, 214 are outside the 200kHz grid, the primary sidebands 212, 214 are susceptible to adjacent channel interference, that is, first neighbor (FM) interference. Therefore, each of the first neighboring (FM) signals 220a, 230a transmitted by IBOC may introduce co-channel interference for the lower sideband 212 and the upper sideband 214 of the digital modulation.

[0049] Co-channel interference can be adjusted in power up to one hundred times stronger than the lower sideband 212 and upper sideband 214 of the digital modulation. In addition, the first neighboring interference signals 220a and 230a can both exist at the same time, so the lower sideband 212 and the upper sideband 214 are both distorted by adjacent FM transmission in this case.

[0050] FIG. 2b shows the IBOC signal of FIG. 1 and the first adjacent adjacent FM channels 220b and 230b. Compared with FIG. 2a, the power of the first adjacent lower adjacent FM channel 220b and the first adjacent upper adjacent FM channel 230b is less than the power of the corresponding digital modulation lower sideband 212 and upper sideband 214 that they interfere with.

[00511 FIG. 3 shows the periodogram power spectral density (PSD) estimation of IBOC transmission with a lower first neighboring (FM) interference signal and an upper first neighboring (FM) interference signal. The parts of the signal in Fig. 3 have been given the same reference numerals as the corresponding parts of the signal shown in Fig. 2a.

[0052] FIG. 3 shows that it is the lower first adjacent adjacent FM transmission 320 (light gray curve) and the upper first adjacent adjacent FM transmission (dark gray curve) 330 respectively causing (H) IBOC transmission (black curve) The lower sideband 312 and upper sideband 314 are severely distorted.

[0053] "In-band on the channel" (IB0C) radio transmission system can be used to simultaneously transmit digital radio broadcast signals and analog radio broadcast signals on the same frequency. As will be discussed below, there is also an all-digital version in which two digital signals are combined. The term (H) IB0C is used herein to refer to IB0C signals that can be in mixed or all-digital form.

[0054] The signal of FIG. 3 can be considered as having three non-overlapping frequency intervals: a lower frequency interval between -300kHz and -100kHz, an intermediate frequency interval between -100kHz and +100kHz, and a + Upper frequency range between 100kHz and +300kHz. The lower frequency interval and the upper frequency interval are examples of the first frequency interval and the third frequency interval. They may include some interference (for example from adjacent adjacent FM channels 320, 330) and a part of the desired digital sideband signals 312, 314 . The intermediate frequency interval is an example of the second frequency interval, which may include the desired FM signal 310 (or a digital equivalent signal, as will be discussed below).

[0055] FIG. 4 shows a spectrum plot of an all-digital implementation of the IBOC signal. For the all-digital IBOC FM signal, the analog FM signal is replaced by (secondary) digital modulation signals 420b, 422b. In the all-digital mode, the bandwidth of the primary digital sidebands 420a, 422a is fully extended with the lower power secondary sidebands.

[0056] The all-digital IBOC signal has a bandwidth of about 400 kHz. In the same manner as described above with reference to Figure 2a, approximately 100 kHz of each of the lower and upper adjacent channels (that is, frequencies outside the 200 kHz "channel grid") are occupied. The lower digital sideband of the all-digital IBOC signal is shown as 420 and the upper digital sideband is shown as 422. Each digital sideband has primary sections 420a, 422a and secondary sections 420b, 422b. In Figure 4, the number of extended frequency partitions (E) is fixed to 4. In a mixed mode system, as shown in Figures 1 and 2, for example, the number of extended frequency partitions can be: 0, 1, 2, or 4, depending on the service mode being transmitted.

[0057] FIG. 5 shows an example embodiment of a receiver system 500.

[0058] The receiver system 500 includes an input terminal 502 that receives input signaling 504. The input signaling 504 includes multiple antenna signals, and each antenna signal is received at a different antenna. Using multiple antenna signals from corresponding multiple antennas may be a convenient way to enable the angle of arrival of the received signal to be determined. Each of the plurality of antenna signals includes information corresponding to at least the first frequency interval and the second frequency interval. The first frequency interval may include the first adjacent interference signal and the digital sideband signal of the desired IBOC signal. The second frequency interval may include the center portion of the desired IBOC signal. As will be understood from the following, the received antenna signal can be in the time domain, but still includes information corresponding to these frequency intervals. [0059] Advantageously, the receiver system 500 can determine:

[0060] Referring to the angle of arrival of the signal in the first frequency interval, the first frequency interval potentially includes interference signals (for example, the upper frequency interval (+100kHz to +300kHz) shown in FIG. 3) or the lower frequency interval (_300kHz to +300kHz). -100kHz)); and [00611 The angle of arrival of the signal in the second frequency interval, which includes the central part of the desired IBOC signal (for example, the intermediate frequency interval shown in Figure 3 (-100kHz to + 100kHz)).

[0062] The system 500 may then use the angle of arrival to determine whether the signals in the two frequency bins may originate from different sources, and therefore determine whether they should be considered as interfering with each other. Based on this determination, weighting coefficients/values can be set so that the signals from multiple antennas are combined with each other in a way that can produce improved output signaling.

[0063] The first AoA block 506 determines the first angle of arrival 508 associated with the first frequency interval. Example implementations of this function are discussed below, and may involve determining the covariance matrix between the portions of the antenna signal corresponding to the first frequency interval, where the antenna signal is received at different (spaced apart) antennas. Optionally, as will be discussed below, the first AoA block 506 may determine the first angle of arrival 508 based on a sub-group of the first frequency interval (for example, an outer sub-group of the first frequency interval), since the sub-group should also be located in Outside the frequency range of the digital sideband of the desired IBOC signal in the first frequency interval, this subgroup can be considered as a good representation of any first neighboring interference signal.

[0064] The second AoA block 510 determines a second angle of arrival 512 associated with the second frequency interval. The second angle of arrival 512 may be determined in a similar manner to the first angle of arrival 508. Optionally, as will be discussed below, the second AoA block 510 may determine the second angle of arrival 512 based on a sub-group of the second frequency interval (for example, the middle sub-group of the second frequency interval), since the sub-group and the second frequency The ends of the interval are separated and therefore unlikely to include interference from the first neighboring interference signal, so this subgroup can be considered a good representation of the desired signal.

[0065] The first weight determination block 514 may set the first weight value 516 based on the first angle of arrival 508 and the second angle of arrival 512. The first weight determination block 514 may set the first weight value 516 as a value for constructively combining the information of the first frequency interval corresponding to the plurality of antenna signals, or may set the first weight value 516 to Therefore, the value of the information corresponding to the first frequency interval of the plurality of antenna signals is destructively combined.

[0066] For example, if the difference between the first angle of arrival 508 and the second angle of arrival 512 exceeds a threshold value (which may be referred to as an external threshold value and a value such as 10 degrees may be adopted), the first weighting value 516 is set with For destructive combination, it is based on that the signal in the first frequency interval originates from a different location and is therefore assumed to be an interfering signal. Such destructive combination can reduce the influence of the interference signal on the digital sideband signal located in the first frequency interval, and thus improve the ability of the subsequent processing of the decoded digital sideband signal.

[0067] If the difference between the first angle of arrival 508 and the second angle of arrival 512 is less than the threshold, the first weighting value 516 is set for constructive combination based on the signal in the first frequency interval derived from the Those signals in the two frequency intervals are in the same direction, and therefore the signals cannot be separated from each other because they do not have spatial separation. Since there is no spatial separation between the signals (or small enough), the "conventional" CIC algorithm should be applied (as can be the case without beamforming). If the first neighboring interference signal is strong, the "conventional" CIC algorithm is best /Best practice. Therefore, the combination is applied to obtain the "strongest" signal. In addition, if the AoA is similar/same, if the first adjacent interference source does not exist, the first adjacent interference signal (the first frequency interval) can also be the residual of the desired signal (the second frequency interval), that is, the leakage, in this case The lower combination can also be used to prevent "self-zeroing". [0068] As will be discussed below, the example implementation that generates the first weight value 516 may also use a covariance matrix, and therefore one of the first weight determination block 514 and the first AoA block 506 may be reused for another block Some of the processing performed. In a manner similar to the first AoA block 506, the first weight determination block 514 may set the first weight value 516 based on the subgroup of the first frequency interval and the subgroup of the second frequency interval.

[0069] The first weighting application block 518 applies the first weighting value 516 to the information corresponding to the first frequency interval in the plurality of antenna signals to provide a weighted first frequency interval signal 520. In this way, parts of the multiple antenna signals corresponding to the first frequency interval are combined constructively or destructively into a single weighted first frequency interval signal 520.

[0070] The first weighting application block 518 may apply the first weighting value 516 to the entire frequency range of the first frequency interval, even if the arrival angles 508, 512 and/or the first weighting value 516 are calculated using a subset of the frequency interval . This is why the input signaling is shown in FIG. 5 as a direct input to the first weighting application block 518.

[0071] The second weight determination block 522 sets the second weight value 524 as a value for constructively combining the information of the second frequency interval corresponding to the plurality of antenna signals. It is assumed that the second frequency interval corresponds to a part of the channel grid reserved for the signal in question. Therefore, constructively combining these signals can improve the SNR or SINR in the second frequency interval. Furthermore, the second weight determination block 522 may set the second weight value 524 based on the subgroup of the second frequency interval.

[0072] The second weighting application block 526 applies the second weighting value 524 to the information corresponding to the second frequency interval in the plurality of antenna signals to provide a weighted second frequency interval signal 528. In this way, the portions of the multiple antenna signals corresponding to the second frequency interval are combined constructively into a single weighted second frequency interval signal 528.

[0073] The second weighting application block 526 may apply the second weighting value 524 to the entire frequency range of the second frequency interval,

Even if the second angle of arrival 512 and/or the second weighting value 524 are calculated using a subgroup of the frequency interval.

[0074] The receiver system 500 also includes a signal combining block 532 configured to combine the weighted first frequency interval signal 520 and the weighted second frequency interval signal 528 to provide output signaling 534. This combination can preserve the distance in the frequency domain between the first frequency interval and the second frequency interval (even if the signal added together is a time domain signal), so that the weighted first frequency interval signal 520 and the weighted second frequency interval signal 520 The frequency components of the signal 528 do not overlap with each other in the output signaling 534.

[0075] In this example, each of the plurality of antenna signals further includes information corresponding to the third frequency interval. The third frequency interval may be another first adjacent interference signal; that is, a first adjacent interference signal that is opposite to the first adjacent interference signal included in the first frequency interval. Therefore, if the first frequency interval corresponds to the frequency range of -300kHz to -100kHz, the third frequency interval corresponds to the frequency range of +100kHz to +300kHz, and vice versa.

[0076] The receiver system 500 includes blocks for processing the third frequency interval, which are similar to those described above with reference to the first frequency interval, as will be discussed below.

[0077] The third AoA block 536 determines the third angle of arrival 538 associated with the third frequency interval. The third weight determination block 540 sets the third weight value 542 based on the third angle of arrival 538 and the second angle of arrival 512. As above, the third weight determination block 540 sets the third weight value 542 as a value for constructively combining the information of the third frequency interval corresponding to a plurality of antenna signals, or sets the third weight value as a value for The value of the information corresponding to the third frequency interval of the plurality of antenna signals is destructively combined. The third weight application block 546 applies the third weight value 542 to the information corresponding to the third frequency interval among the plurality of antenna signals to provide a weighted third frequency interval signal 548.

[0078] In this example, the signal combining block 532 may combine the weighted first frequency interval signal 520, the weighted second frequency interval signal 528, and the weighted third frequency interval signal 548 to provide output signaling 534.

[0079] FIG. 6 shows a detailed view of a method 600 that may be performed by a receiver system, such as the receiver system in FIG. This method can be considered as providing selective beamforming with double zero control.

[0080] The improved reception of the transmitted (H)IBOC signal can be obtained by electronically controlled antenna radiation patterns of multiple (two) antennas, such as a uniform linear (antenna) array (ULA) composed of two isotropic antennas. . Manipulating the radiation pattern by using the composite baseband signal and phase and amplitude estimates can be referred to as electronically steered composite baseband beamforming. This type of beamforming also allows "zero control" to suppress (or remove) interfering signals. For example, this may allow suppression of the first neighbor (FM) interference signal transmitted by IBOC (in hybrid mode and in all-digital mode). In addition, this type of beamforming allows so-called co-channel interference cancellation (CIC), because the first neighbor (FM) interference signal is a co-channel interference signal of the digital modulation sideband transmitted by IBOC. However, if there is not enough "spatial separation" between the interfering signal and the desired signal, the suppression of the first neighboring interfering signal can also result in suppression of the received desired signal; this is called "self-zeroing". This self-zeroing should be avoided by a suitable self-zeroing prevention program, examples of which are described below.

[0081] One or more of the examples described below may prevent self-zeroing (or reduce its likelihood) based on the steering vector of the received signal. The steering vector represents the angle of arrival (AoA) of the received signal. To calculate the steering vector, we propose to use a part of about 100kHz wide (100kHz frequency interval) to obtain training for the desired IBOC signal, lower first neighbor (FM) interference signal, and upper first neighbor (FM) interference signal Signal or alternative signal (representation). These replacement signals will then be used to calculate the steering vectors for the two first neighboring interference signals and the desired signal. After obtaining the steering vector, we derive the one-to-one relationship between the angle of arrival (AoA) and the steering vector. We then propose AoA to determine the spatial separation of signals. Finally, if there is a lack of spatial separation, we propose constructive combinations of application signals. This entire program can be called the "self-zeroing prevention" algorithm for IBOC transmission.

[0082] The beamforming process has the goal of removing or at least reducing the two first neighboring interference signals by using multiple antennas (in some cases only two antennas due to low complexity and cost reasons). It should be noted that the maximum ratio combination (MRC) cannot remove or reduce the co-channel interference caused by the first neighbor (FM) interference signal, because MRC only optimally combines the total signal energy, that is, includes the first neighbor Signal energy. Since the expected (H) IB0C signal and the two first neighboring (FM) interference signals originate from different locations, this is normal for FM transmission in Band II. The inventors found that the beamforming method can be used in the spatial domain. Try to remove the first nearby (FM) interference signal. Now, the task of the beamforming algorithm (electronically controlled composite baseband) is to separate the three different signals by their spatially different information ("spatial signature").

[0083] However, if there is a lack of spatial separation, the received desired signal may also be suppressed, which is called "self-zeroing" and is an undesirable effect caused by beamforming. To prevent self-zeroing, we propose to calculate the steering vector of the received signal in this article. The steering vector represents the angle of arrival (AoA) of the received signal, as we will show later. [0084] The method of FIG. 6 starts at step 652, where for each of the multiple antenna signals, a time-discontinuous baseband signal (for example, with a sampling frequency of 650 kHz) may be selected and down-converted so that the signal from each antenna All information in the frequency domain/time domain from 325kHz to +325kHz of each signal can be captured. In step 654, N data samples may be collected. It should be noted that although the signals shown in FIGS. 1 to 4 illustrate the sampling frequency, the N data samples in FIG. 6 are time samples rather than frequency samples. N may be a high enough number so that the training signal effectively represents the interference signal and the desired signal. [0085] For example, about 2000 samples (1872 samples in one example) can be used to form a training signal for a frequency of 650 kHz. Those skilled in the art will understand that the length of each sample will depend on the sample frequency-for example, in terms of 650kHz, 1.65 microseconds (ys), so that 2000 samples for 650kHz can result in at most about 3 milliseconds (ms) The extension Time. It should be noted that the analog-to-digital conversion of the received signal can be performed as part of the IBOC reproduction process; therefore, no additional processing is required to reproduce the samples other than writing the samples. This is because the samples are just a digital representation of the received signal that is repeatedly processed (divided) into N time samples. Therefore, sampling can be continued so that new samples are constantly available. Therefore, step 654 may be repeated multiple times before moving to the steps in paths 656 to 664 as discussed below. Thus, the most recent N data samples can be maintained, so that when new samples are obtained, earlier samples can be discarded. In some embodiments, the samples may be updated at a frequency, rather than continuously updated, which frequency depends on the speed of movement of the antenna array or the interference level of the desired signal.

[0086] In addition, due in part to the increased complexity and processing power required, the new range of the beamformer may not be calculated for each updated sample collected - therefore, it may be updated more frequently than the range Samples are collected so that the samples can be obtained on demand when it is determined that it is time to calculate a new range. As mentioned above, if no samples are collected when the range is to be calculated (for example, when the frequency is tuned for the first time), the delay will be approximately the time required to collect N samples, which can be expressed as Calculate from t-(N*length_sampi_e) Acquire samples, and the beamformer does not continue until N samples are collected.

[0087] In addition, the interval between recalculations of the beamformer in one embodiment may depend on factors such as the speed of the receiver (for example, assuming that the receiver is on a motor vehicle) - theoretically, if the receiver is moving fast , In extreme cases, the beamformer can be recalculated for every N+1 samples (for example, the first calculation can be run on 0-2000 samples, and the second calculation can be run on 1-2000 samples). Therefore, the update rate will be the same as the sampling rate, but this situation will require significant processing power.

[0088] The method includes a first estimation path 656, a second estimation path 658, and a third estimation path 660 for determining weighted values of the first frequency interval, the second frequency interval, and the third frequency interval, respectively. The method also includes three correction paths:

The first correction path 657, the second correction path 659, and the third correction path 661. These correction paths are used to combine the associated frequency intervals of multiple antenna signals by using the weighting value determined by the estimated path. As will be described below, in this example the estimation path deals with a subset of the associated frequency intervals, while the correction path combines the entire frequency interval.

[0089] As shown in the first estimation path 656, for each antenna signal, the representation of the first neighboring interference signal calculated from N samples may be at a frequency of about -250kHz (between -200kHz and -300kHz) Get 666 in the interval. There may be minimal interference in this frequency range, for example, from the lower digital sideband of the desired IBOC signal. That is, the first estimation path 656 may involve processing of the outer subgroup (between -200 kHz and -300 kHz) of the first frequency interval (between -100 kHz and -300 kHz).

[0090] In the second estimation path 658, for each antenna signal, for example, the representation of the desired signal calculated from N samples again may be about 0 Hz (0 kHz) (between -50 kHz and +50 kHz). Get 668 in the frequency range. There may be relatively little interference in this frequency range, for example, from the lower first neighboring interference signal and the upper first neighboring interference signal. In other words, the second estimation path 658 may involve the intermediate subgroup (between -100kHz and +100kHz) of the second frequency interval (between -100kHz and +100kHz).^-Between 50kHz and +50kHz).

[0091] In the third estimation path 660, the representation of the upper first neighboring interference signal calculated from N samples can be obtained 670 in a frequency interval of about +250kHz (between +200kHz and +300kHz), where There is minimal interference, for example, from the upper digital sideband of the desired IBOC signal. That is, the third estimation path 660 may involve processing of the external subgroup (between +200 kHz and +300 kHz) of the third frequency interval (between +100 kHz and 300 kHz).

[0092] Therefore, the representation of the lower first neighboring interference signal and the representation of the upper first neighboring interference signal may be centered in step 666 and step 670, so that the signal is offset to a direct current (DC) bias. It should be noted that for the second estimation path 658, the antenna signal does not need to be offset to a DC offset, because the second frequency interval is already centered at 0 Hz. Therefore, the second estimation path 658 can continue from step 654 to step 668, or in an alternative embodiment step 668 can wait for step 672 and step 674 to run in parallel.

[0093] In steps 672, 668, and 674, each of the frequency-shifted lower first neighboring interference signal, the desired signal, and the frequency-shifted upper first neighboring interference signal representation may use a 50kHz low-pass filter (LPF ) Filter at about OHz (for example from _50kHz to 50kHz), where the finite impulse response (FIR) is 24 taps. In one embodiment, a low-pass filter is combined with a band-pass filter to shift each signal to 0 (steps 666 and 670), and half of the signal is filtered out to ensure that the cleanest signal is possible (steps 672, 668 and 674). Therefore, these three different frequency bands of 100 kHz (they are a subset of the first frequency interval, the second frequency interval, and the third frequency interval) can be referred to as training signals or replacement signals.

[0094] Steering vector calculation

[0095] The calculation of the steering vector received by IBOC will be introduced in this chapter and represented by the following sub-steps in Figure 6:

[0096] Participate in generating the first frequency interval matrix 676,

[0097] Participate in generating the second frequency interval matrix 678,

[0098] Participate in generating the third frequency interval matrix 680, and

[0099] The steering vector and AoA are determined by reference to the minimization of mean square error (MMSE) 607.

[0100] The calculation of the steering vector is based on the (sample-based) spatial covariance matrix described below. As will be discussed below, weighting values are generated based on the spatial covariance matrix by applying the "Maximization of SINR Standard" algorithm, and these optimal weighting values are applied to the first frequency interval, the second frequency interval, and the third frequency interval. The MMSE standard is used to obtain AoA from the spatial covariance matrix, and AoA is used for "veto", that is, the combination process.

[0101] In addition, as discussed above, by using about 100kH_ZPart of the wide (100kHz frequency interval) obtains each of the (sample-based) spatial covariance matrices to obtain the desired IBOC signal, the lower first neighbor (FM) interference signal, and the upper first neighbor (FM) ) Training signal or substitute signal (representation) of interference signal. In the rest of this chapter, we will introduce and explain the calculation of steering vectors in detail.

[0102] The received lower first neighboring (FM) interference signal, the received desired signal and the received upper first neighboring (FM) interference signal substitute are processed by the so-called "self-zeroing prevention" algorithm . This self-zeroing prevention algorithm calculates the composite number representing the estimated composite number of the received lower first neighboring interference signal and the upper first neighboring interference signal and the steering vector of the received desired signal. As we will show later, the steering vector of each of the received signals has a one-to-one relationship with their angle of arrival (AoA). Therefore, the steering vector contains the necessary spatial information to compare signals in the spatial domain. In addition, we treat these comparison criteria as under the heading "Angle of Arrival (AoA) Comparison". As we will show, by solving the eigenvalue problem by means of the "principal component analysis" (pea) method, the so-called "estimation and packing" technique, the calculation of the steering vector can be extremely fast. In our case, the statistical pea method uses the solution of the eigenvalue problem as an orthogonal variation. By solving this eigenvalue problem, a set of observations of possibly related variables, that is, the observations for our spatial covariance matrix, are transformed into a set of linearly uncorrelated (orthogonal) variables (called principal components) Value. In our case, these components are the eigenvectors of the spatial covariance matrix, that is, the singular value decomposition (SVD) of the spatial covariance matrix. In addition, through the pea party Method, the eigenvector with the largest eigenvalue is the first principal component. We will show later that this principal component is proportional to the required steering vector. In addition, the specific version of the pea method used in this example, that is, the SVD of the (2x2) spatial covariance matrix can be applied extremely quickly (this can be considered instantaneous). Therefore, the proposed composite baseband "self-zeroing prevention" algorithm is fast, that is, it has low latency. In fact, the delay is only determined by the observations (samples) required to calculate the spatial covariance matrix. It should be noted that for the determination of the weight value as discussed above, we have calculated the 2x2 covariance matrix for the desired signal vector 2 received by the two antennas, and also applied the pea method as described above to calculate the optimal beamforming Weighted value. We describe a similar procedure below to obtain the steering vector, which is then applied to prevent self-zeroing.

[0103] The calculation standard of the steering vector used in the self-zeroing prevention algorithm is the minimization of mean square error (MMSE), that is, the known Wiener-Hopf standard. The solution of the Wiener-Hopf criterion can provide a steering vector under some assumptions, which we will introduce later in this chapter. The Wiener-Hopf MMSE standard takes the first partial derivative of the mean square error (MSE) with respect to its weight vector, that is, the gradient of the MSE, sets this derived result to 0 and solves the equation. We will show this procedure as an example in the remainder of this chapter, namely, the Wiener-Hopf standard to calculate the steering vector X of the transmitted desired signal. [0104] The desired signal received by multiple (two) antennas is given by the following equation:

[0105] s = x+n = ax+ri Equation 1

[0106] Among them, three is a number of two) antenna weird) 0 average composite white Gaussian noise signal vector with noise variance of each vector component. In addition, we assume that the received desired signal 5 is a "point source signal" with a steering vector 5. Now by applying the weighting vector w to the received desired signal 2, we can obtain the (noisy) estimate X of the transmitted desired signal, which is given by:

[0107] f twoTM (w^h,a) x + / equation 2

[01081 where (·)¹¹Is the Hermitian transpose, that is, the application of complex conjugate (·)' and transpose (·)⁷Calculate both. We now define the estimated MSE of t for the desired signal transmitted by:

[0109] MSE (w)-E{(.v-square stalk style 3

[01W] Where E{ is called the statistical average value. An additional evaluation of Equation 3 yields:

+ iV

Eiss¹¹}

VOMWMM ./

[0111]

[0112] i\-,·-vd iv + vvA\, vv Equation 4 where <is the variance of the transmitted signal X, Vy = is the spatial cross-correlation vector, and

I: Efs\s^H} Is the spatial covariance matrix.

[0113] Review the Wiener-Hopf standard to calculate the gradient of MSE and set the result to zero. Therefore, the gradient of the MSE given by Equation 4 with respect to iv^wbecome:

[0115]

[0116]

[0114]

<img file="CN107370499A_D0001.tif" />

And by setting the gradient to 0, Eq. 5, we get-~~* = Eq. 6

[0117] In fact, the accepted Wiener-Hopf solution is given.

[0118] By assuming that the received desired signal 2 is a noisy point source signal with a steering vector 3, that is, 4=¥+11, we can rewrite the spatial cross-correlation vector as:

[0119]

<img file="CN107370499A_D0002.tif" />

<img file="CN107370499A_D0003.tif" />

Equation 7

[0120] It is a "scaled version" of steering vector 3 (with the variance of the transmitted desired signal X). Combining its equation 6 to give:^[θ121]No ssiE*: d package seven fppt vibration program 8

[0122] The result is that we actually need to solve the eigenvalue problem. The solution to this eigenvalue problem provides the best weight vector. In addition, this optimal weight vector represents the steering vector of the transmitted desired signal X and can therefore be given by:^[0123] Equation 9

[0¹²⁴] Where P{·} is an operator based on the principal eigenvector of the pea returned matrix, which is proportional to the steering vector 3 and. [0125] In the case of a two-antenna uniform linear array (ULA), the spatial covariance matrix R_ssIt is a 2x2 matrix, wherein calculating the expected received signal Laid eigenvectors 2 eigenfunctions only "straight through" quadratic function requires solution. This quadratic function is given by the following formula:

[0126]

[0127]

[0128] λ'- tr{A} Ά + dvtiAj = 0 = λ_}·.

ΙΦΜ soil v tr^z{A}-4 riethM Equation 10 where Α to trpM is the trace of matrix A, and det{A} is the determinant of matrix A.

Now the first (largest) principal eigenvector Wopt representing the steering vector 3 can be based on Cayley Hamilton's theorem by using the solution matrix (Α_λ₂Ι) column is found, where λ₂Is the smallest eigenvalue outside the quadratic function of Equation 10.

Solving this program means that in fact the main eigenvector w_QptIt is proportional to the steering vector 3 of the desired signal received. The obtained solution wopt of the eigenvalue problem will be used to calculate the received desired signal ^JAoA. In addition, the similar result i can be similar to that used for the received lower first neighbor (FM) interference signal with a spatial covariance matrix of 7 and the received upper first neighbor (FM) interference signal with a spatial covariance matrix. Derivation of the way.

[0129] Recall that the process used to obtain these spatial covariance matrices is based on the substitution signals (ie subgroups of associated frequency intervals) mentioned earlier.

[0130] Finally, for practical considerations, the approximation of the spatial covariance matrix is used. For this approximation, the infinite-length statistical average operator E{ ·} will be replaced by the finite running length sum average, which obtains the sample-based spatial covariance matrix and is given by the following formula:

[0131] ί¥ For the next interference,

<img file="CN107370499A_D0004.tif" />

[0132]

[0133] ι For the desired signal, two 1 (/frd-Γ1X1) is not for the upper interference, equation 11 mt

[0134] where TIM is the stream of sample vectors used for the received lower first neighbor (FM) interference signal substitute, is the stream of sample vectors used for the received substitute of the desired signal, and 0«! is a stream of sample vectors for the substitute of the received upper first neighbor (FM) interference signal, and is provided as the output of steps 672, 668, and 674 shown in FIG. 6.

[0135] These sample-based spatial covariance matrices are private_H As step 676 shown in FIG. 6,

The output of 678, 680 is provided and used to calculate the angle of arrival and also to calculate the weighted value.

[0136] Now, the first weighting value, the second weighting value, and the third weighting value representing the steering vector calculated by the sample-based spatial covariance matrix can then be written as:

[0137] Xun,_pU = TM For lower interference,

[0138] = S For the desired signal,

[0139] Η»ί_；.；._;Ί-ί - also/for upper interference, equation 12

[0140] The eigenvector is calculated by solving the quadratic characteristic function of each of the three replacement signals in three different frequency intervals. As a result, we obtain three steering vectors. We use these three steering vectors to calculate the received interference signal and the received desired signal's angle of arrival (AoA). Then compare these AoAs to determine if there is a lack of spatial separation. This AoA comparison procedure will be described in the next chapter.

[0141] Angle of Arrival (AoA) Comparison

[0142] In this section we will introduce comparison criteria to prevent self-zeroing. This self-zeroing prevention decision algorithm is based on the availability of steering vectors, which is calculated, for example, as explained above under the heading "Steering Vector Calculation". The criterion

It is actually quite "simple" and relies on the angle of arrival (AoA) information obtained from the steering vector. More precisely, the determination is based on the difference between the received desired signal and the received AoA of the lower first neighboring interference signal, and also based on the received desired signal and the received upper first neighboring interference signal The difference between AoA. In the rest of this chapter, we will explain the comparison and the corresponding criteria.

[0143] Assuming that the received desired signal is a single point source RF signal s (1) with plane wave propagation, the received desired RF signal at the ULA of the two elements is given by the following equation:

[0144] le /2 η-·

<img file="CN107370499A_D0005.tif" />

Equation 13 rtshw, Λ'ten? ι; When J kisses_λ

[0145] where fc is the RF carrier frequency (for example, FM band 88&lt;fc&lt;108MHz), where% is the initial phase, = let is the wavelength, where c is the speed of light, and d is the distance between the two antenna elements of the ULA (for example &amp;, x(t) is the desired RF signal, ϋ is the RF noise signal, and A is the AoA of the transmitted desired signal. Now, s is evaluated as the composite baseband signal s, where and combined with Equation 1, for the steering The vector gets:

[0146] α-[^;rrs_m;/7,| ™ [_Πί | Equation 14

[0147] The AoA of the transmitted desired signal can now be obtained from the steering vector component. Assuming that the steering vector (or estimate) is available, this results in:

[0148] φ_χ = SHI ¹ I one; one j equation 15

[0149] where ln (·) represents the use of natural logarithms,: Η represents the use of the imaginary part of a complex number, and Sin-1 (·) is the arc sine.

[0150] In another embodiment, the representation of AoA can be determined by using the natural logarithm and the imaginary part of the complex number a1, that is, the arc sine as shown in Equation 15 is not calculated.

[0151] Through Equation 15, we obtain the direct relationship between the AoA of the desired signal transmitted and the steering vector i. It should be noted that AoA (only) is calculated from the spatial covariance matrix, which is already available from steps 676, 678, 680, and is also used to obtain the first weighted value through the "SINR maximization" criterion described below , The second weighting value and the third weighting value (which can also be referred to as beamforming weights). Therefore, AoA calculations can be a fairly "complexity-friendly" extension of the beamforming algorithm used to determine beamforming weights. In addition, it is advantageous that the AoA is "consistent" with the calculated beamforming weight, and therefore the "matching determination criterion" is introduced through the proposed method.

[0152] Calculation of Weighted Value

[0153] The following description relates to an implementation of how the maximization of the SINR criterion in block 651 can determine a weighting value (which may also be referred to as a weighting coefficient or beamforming weight). It should be noted that if no "veto" occurs, these are beamforming weights for the three frequency intervals.

[0154] Through the composite digital baseband beamforming with double zero control, the weighting coefficient can be generated by maximizing the SINR.

Proceed as follows. It should be noted that the estimation criterion is the maximization of SINR, which calculates the optimal weight by using the first derivative of the signal-to-interference plus noise ratio (SINR); the result of this derivative can be set to 0, which can solve the equation. SINR can be expressed as SINRyun (Equation 16), where w is the weight and (·)η is the Hermitian transpose, that is, the complex conjugate and transpose operation

W^llKj_uW - both, and R_ss Will ^En'=i(s[n]-s^w[n]), R_jn -~Ση=ι ({ί[^η] + η [η]}-{i[nl + 1}^Η) (Equation 17) is the sample covariance matrix (approximation of the covariance matrix on a finite number of N samples), and_n[n] means 0 mean and variance σ for each of the components² = _ΝϋThe composite Gaussian noise vector. In the example case, the covariance matrix R_inMay not be available, however, for IBOC transmission, the representative signal can be used to obtain the sample covariance matrix R_jnApproximation (and also used for the sample covariance matrix R_in),produce:

[0155] R_in «R_n Δ ~SS=: iQl'nJ-!^H[n]) Down interference

[0156] R_ss R_g§ Δ Ji·s^H[nj) Expected signal (equation is)

[0157] R_jnRighteousness

[0158] where i[n] is the stream of samples for the substitute of the next adjacent interference signal, correction n] is the stream of samples for the substitute of the desired signal, and ί[n] is A stream of samples used for the first neighboring interference signal substitute. [0159] The IBOC transmission with two first neighboring interference signals (and, in some embodiments, compound Gaussian noise) can be expressed as three spatially different and independent signals (ie, the lower first signal with independent compound Gaussian noise). 1 Neighboring interference signal, the desired signal, and the first adjacent interference signal). Therefore, SINR can be expressed as SINR_H1B0G ¥ Ε ί'7ηη/ 7'S' ~ Look at 1||1,|||) to earn , Na{Peng face . Hupeng number, that is, the weight vector can be obtained by maximizing SINR, E_optThree argmax^.iSlNR} = argrnax_wEquation 20).

(equation

[0160] For the received IBOC transmission, such as the transmission 100 shown in FIG. 1, the interference signals 220a, 230a and the desired signal 210 shown in FIGS. 2a and 2b can be divided into different frequency intervals, so that for each Frequency range, SINR will be optimized to obtain weights, so that

[0161] w_opU argmax^. Lower interference

[0162] Also,_pt，_sSome desired signals (equation²1), fwP RxjeWi)

[0163] w_optJ arg max^. upper interference

[0164] It can be regarded as the initial step of calculating the weighting coefficients in the three subbands. In order to solve the problem of maximization, the composite gradient of the SINR in each interval can be obtained relative to the composite weight and the result set to 0, which obtains, for example, for the first neighboring interference signal &amp; {(wi'RssWOCwf^nWi)·¹}-0 (Equation 22),

Where is the expression used to adopt the compound gradient. Application of partial differentiation can lead to (w^RijWi)²Riiwi (w-ⁱRssw_i) = 0 (Equation 23), which can be rewritten as an expression

RssWiCwi'RiiWi)¹ = (W^RnwJ-^RiiWi^R^Wi) R_ssWi = _{R w R w}People (Equation 24), where \zhu 1 two Sheng (Equation 25) can be defined as ~ WjRnWj ¹¹—^{1 1} i+n W| RiiWj SINR of the frequency interval of the first adjacent interference signal. This can be rewritten as:

[0165] R_ssw; = RiiWjAi (Rh^ss)^-AjW; (Equation 26), which provides the eigenvalue problem with the best weight for maximizing SINR for its solution: W_opt,i = P{Rh^1rss.) (Equation 27), where P{ ·} is an operator that returns the principal eigenvector of the PCA-based matrix.

[0166] Through the ULA of the two antennas, the sample covariance matrix can be a 2×2 matrix, and the eigenfunction for calculating the eigenvector is a quadratic function expressed as: X²-tr{AU+det{A} = 0 4 λυ = Badu Myou (Equation 28), where Alan is the trace of matrix A, and det{A} is the determinant of matrix A.

[0167] Similar results can be derived in a similar manner for the desired signal and the first adjacent interference signal. However, the sample covariance matrix R is independent of the zero mean compound Gaussian noise variable_nnIt can be assumed to be a diagonal matrix, where the noise variance on the main diagonal is σ². It should be noted that the desired signal may not have an interfering signal (which is one reason for the training signal). By using the inversion of the noise matrix, there will only be values on the main diagonal, so that the desired signal will be scaled, but the eigenvectors will not be changed. Therefore, the inverted sample covariance matrix can also be expressed as a diagonal matrix and

[0168] R; i*R_ss &lt;x R_ss W_opt,_s = PiR^Rss) P{R_SS} (Equation 29). Finally, the optimal weights for the lower first adjacent interference signal and the upper first adjacent interference signal and the desired signal can be expressed as [0169] w_opt i = P{RH^lRs_S} Lower interference

[0170] ，_pt,_s = p{_Rss}The desired signal (Equation 30)

[0171] W_Opt,j = upper interference

[0172] where the eigenvector is calculated by solving the "simple" quadratic characteristic function of each of the three frequency intervals.

[0173] Generate logical indicator based on AoA

[0174] Now, by reviewing the steering vector calculation discussed above under the heading "Steering vector calculation", we can obtain the received AoA of the first neighboring interference signal, the received AoA of the desired signal, and the received signal respectively. The AoA (for ULA of two antennas) of the first adjacent interference signal obtained is as follows:

[0175] （_PiIs sin-i (¾¹.¹Xi: Dagger B). Among them, for the next interference,

[0176] (p_{s sin}-1 where the social installation|two'j for the desired signal, equation 31

[0177] φ. sin-¹ Where 2 [to 1 for the upper interference, ¹ \ ^:ΤΓ· / JLJ /.Λ 19

[0178] Next, we can apply our comparison criteria by using these AoAs. Therefore, we propose to use the obtained AoA to identify whether the received lower first neighboring interference signal and upper first neighboring interference signal are spatially separated from the expected signal. If these signals are too close to the desired signal, self-zeroing can occur due to lack of spatial separation. The criterion for determining whether there is a lack of spatial separation between the received lower first adjacent interference signal and the received desired signal is given by the following formula: ί Κ <=> \ψί - φ.~\ <handsome_is

[⁰¹⁷⁹] ^Lis equation Mangmi(4. ϊψι i

[0180] And the criterion for determining whether there is a lack of spatial separation between the received upper first neighboring interference signal and the received desired signal is given by the following formula: j * -φ^\&lt;

[0181] = j equation 33 \φ;-q?₅|> A0_/5.

[0182] where {Δ φ _is, Δ is an external threshold, which can be considered as a suitable value for determining whether two signals are sufficiently close to each other in space (for example, 10° for both the first neighboring interference signal and the second neighboring interference signal) variable. These values can be obtained through other spatial modeling methods such as "on-site testing" or (6) IBOC transmission scenarios. Logical indicator [L_1S, L,_S] It is output by step 607 in Fig. 6 and used to decide which procedure to follow to prevent self-zeroing. That is, whether to set a weighting value for subsequent constructive or destructive combinations.

[0183] Setting the weighting value used to prevent self-zeroing

[0184] In this example, the step of setting the weighting value is shown as six sub-steps, which together set the first weighting value (Wopt, i), the second weighting value (Wopt, s), and the first weighting value (Wopt, s). Three weighted value (Woptj). These sub-steps are:

[0185] Participate in generating the first frequency interval matrix 676,

[0186] Participate in generating the second frequency interval matrix 678,

[0187] Participate in generating the third frequency interval matrix 680,

[0188] Determine the weighting value used to maximize the SINR 651 based on the three frequency interval matrices 676, 678, 680,

[0189] Determine the weighting value used to constructively combine each of the three frequency intervals 653, and

[0190] The reference is based on the logical indicator [L_1S, L,_S]Choose whether 655 is output from step 651 or

653 The weighted value obtained for each of the three frequency intervals.

[0191] The steps of generating the frequency interval matrices 676, 678, 680 and determining the weighting value for maximizing the SINR 651 have been discussed in detail above. As pointed out above, the steps of generating the frequency interval matrices 676, 678, 680 are common to both this section and the "steering vector calculation" section.

[0192] The step of determining the weighting value for maximizing the SINR 651 can be summarized as: adding the first weighting value (_Wcipt, i) is set to a value for destructively combining the information of the first frequency interval corresponding to the multiple antenna signals; the second weighting value is set to be used to constructively combine the first frequency interval corresponding to the multiple antenna signals The value of the information of the second frequency interval; and the third weighting value (Wopt.j) is set to the value used to destructively combine the information of the third frequency interval corresponding to multiple antenna signals. For more details, please refer to Above. Each of these (transitional) groups of weighted values is provided as an input to the selection step 655.

[0193] The step of determining a weighting value for constructively combining each of the three frequency intervals 653 may involve casting the first weighting value (Wopt, i) and the third weighting value (Wopt, j) to be the same The value of the same value in step 651 (determined for the most

The weighted value of the maximum SINR) is set for the second weighted value (ffopt.s). This can be considered as a rejection of the canceled first weighting value and second weighting value determined in step 651.

[0194] In steps 676, 678, and 680, N covariance matrices (Rii, Rjj) representing the sample signal of the first neighboring interference and N covariance matrices R representing the expected signal of the sample are generated._ss. It should be noted that in this example, since there are two antennas in the array, the matrix will be a 2x2 matrix; the main diagonal will correspond to the information about the interfering signal or the desired signal, and the secondary diagonal will correspond to the information about the interfering signal Or information about the spatial correlation of the desired signal (depending on which matrix is considered).

[0195] Once the matrices 676, 678, and 680 are generated, the step of determining the weighted value for maximizing the SINR 651 calculates weighted values, which are based on the N covariance matrices and representative samples of the first neighboring interference representative sample signal, respectively The inversion of the N covariance matrices of the desired signal to maximize the signal-to-interference plus noise ratio (SINR), see above. In this way, the first weighted value and the third weighted value are determined for the subsequent destructive combination, and the second weighted value is determined for the subsequent constructive combination.

[0196] The covariance matrix can be used as a measure of the signal power of the desired signal and the interference signal. The beamformer maximizes the signal used for the SINR, the maximum power of the desired signal, and the minimum contribution of interference plus noise, as this will result in a determination of directivity; this condition will correspond to the originating direction of the signal. The "estimate and stuff" technique can be used almost immediately to calculate the weights by solving the eigenvalue problem using the "principal component analysis" (PCA) method described above. Therefore, the composite baseband beamforming algorithm is fast, that is, it has low latency.

[0197] The step of selecting 655 whether to output the weighted value from step 651 or 653 is based on the logical indicator from step 607 [L_ls, L,_s]. Logical indicator [L_ls, L,_s] Indicates whether there is a lack of spatial separation, and the method uses this information to determine what action to perform to prevent self-zeroing.

[0198] That is, if L_lsIf true, the first weight value (for constructive combination) generated by step 653 is output to be applied to the first frequency interval. If L_lsIf false, the first weighted value (used for destructive combination) generated by step 651 is output to be applied to the first frequency interval. Similarly, if L,_sIf true, the third weight value (for constructive combination) generated by step 653 is output to be applied to the third frequency interval. If L,_sIf false, the third weighted value (used for destructive combination) generated by step 651 is output to be applied to the third frequency interval. This process can be thought of as preventing self-zeroing and will be described in more detail below.

[0199] For the received desired signal, the beamformer applying the weighted value for maximizing the SINR generated in step 651 can calculate the weighted value as the optimal beamforming weight to perform the received signal at ULA. Constructive addition of the desired signal, that is, it "constructively combines" the desired output signal of multiple (two) antennas. However, the method of FIG. 6 can determine whether there is a lack of spatial separation between the interference source and the desired signal, that is, whether the AoA of the desired signal and the AoA of one or both of the interfering signals have similar values, Makes them very close in space. If there is a lack of spatial information, this means that the beamforming algorithm cannot distinguish (spatially) different signals. In this case, appropriately apply "constructive combination" to interfering signals with AoA similar to the desired signal. In addition, since the beamformer cannot achieve the spatial distinction between these signals anyway, this seems to be a suitable option. It should be noted that the optimal weight for constructive combination may already be available, because the optimal weight can be calculated for the received desired signal (ie, the second frequency interval). This means that if for example the lack of spatial separation logic indicator L for the next first neighboring interference signal_lsIf it is "True", the optimal weight of the desired signal (second frequency interval) should also be applied to the received first adjacent interference signal (first frequency interval) to obtain a constructive combination. The same considerations apply to indicators L,_s, That is, if L,_sIf it is "True", the optimal weight of the desired signal (second frequency interval) should also be applied to the received upper first adjacent interference signal (third frequency interval) to obtain a constructive combination. Therefore, if both indicators are "true", the best weight (used in the second frequency region)

Between) is used for the lower first adjacent interference signal (first frequency interval) and for the upper first adjacent interference signal (third frequency interval). In this case all signals are combined constructively.

[0200] In another embodiment, the optimal weight for combination can also be "adjusted" with the information of the covariance matrix of the first frequency interval (or the third frequency interval). For example, the weight applied to the combined second frequency interval may be an "adjusted/weighted" version of the best weight for the second frequency interval. The adjustment can be based on information from the spatial covariance matrix of the first (or third) frequency interval. In this way, the first weight determination block can determine the adjusted second weight value by modifying the second weight value based on the spatial covariance matrix of the information in the plurality of antenna signals corresponding to the first frequency interval; and The weighting value is set to the adjusted second weighting value. Similarly, the third weight determination block may set the third weight value based on the corresponding signal associated with the third frequency interval.

[0201] Therefore, due to the application of constructive combination method, if the two lack of spatial separation logic indicator [L_1S, U] is "true", then no self-zeroing occurs when the signals are too close in space. However, it should be noted that the result of this constructive combination method is that if two logical indicators [L_ls, L,_sIf any one of] is "true", the suppression of the first neighboring interference signal is no longer possible. On the other hand, this should be expected because the beamformer does not have spatial information in this case to compare the received desired signal with the received lower (L_lsIs "true"), on (L,_sIs "true") or the lower 1st adjacent interference signal and the upper 1st adjacent interference signal (L_1S, L_1SBoth are "true") separately.

[0202] Apply weighted value

[0203] As noted above, the method of FIG. 6 includes three correction paths: a first correction path 657, a second correction path 659, and a third correction path 661.

[0204] The first correction path 657 is used to combine the first frequency interval of each of the plurality of antenna signals consistent with the first weight value. As discussed above, the first weighting value is set to constructively or destructively combine the first frequency interval of each of the multiple antenna signals.

[0205] The second correction path 659 is used to constructively combine the second frequency interval of each of the plurality of antenna signals consistent with the second weight value. As discussed above, the second weighting value is set to constructively combine the second frequency interval.

[0206] The third correction path 661 is used to combine the third frequency interval of each of the plurality of antenna signals consistent with the third weight value. As discussed above, the third weighting value is set to constructively or destructively combine the third frequency interval.

[0207] In this example, the first weight value, the second weight value, and the third weight value each include a plurality of antenna weight values, and each of the plurality of antenna signals has an antenna weight value. Each of the antenna weights may be a complex number.

[0208] In the first correction path 657, the first step 686 of frequency shift +200 kHz is performed, so that the center of the first frequency interval of each antenna signal is located at 0 Hz. Then in step 692, the frequency-shifted signal provided by step 686 is low-pass filtered (LPF) through a filter with a cutoff frequency of 100kHz, centered at about 0Hz (that is, from -100kHz to 100kHz), where the finite impulse response (FIR ) In this example, there are 32 taps. As a result, the output signal from the filtering at step 692 will include the entire first frequency interval. These output signals can be considered as antenna signals in the first frequency range.

[0209] Similarly, in the third correction path 661, the first step 690 of frequency shifting -200 kHz is performed, so that the center of the third frequency interval of each antenna signal is located at 0 Hz. Then in step 694, the frequency-shifted signal provided in step 690 is low-pass filtered (LPF) through a filter with a cut-off frequency of 100kHz, centered at about 0Hz (that is, from -100kHz to 100kHz), where the finite impulse response (FII ?) In this example, there are 32 taps. As a result, the output signal from the filtering in step 694 will include the entire third frequency interval. These output signals can be regarded as antenna signals in the second frequency range. [0210] In the second correction path 659, since the center of the second frequency interval of each antenna signal is already located at 0 Hz, no frequency shift step is required. In step 698, the antenna signal is low-passed through a filter with a cutoff frequency of 100kHz

Filtering (LPF), centered at about 0 Hz (that is, from -100 kHz to 100 kHz), where the finite impulse response (FIR) is 32 taps in this example. As a result, the output signal from the filtering at step 698 will include the entire second frequency interval. These output signals can be considered as antenna signals in the second frequency range.

[02111 The method of FIG. 6 includes a first weighting application step 618 that applies the first weighting value to the first frequency interval antenna signal, so as to combine the signals and provide a weighted first frequency interval signal. In this way, the first weighting value is applied to the information corresponding to the first frequency interval in the plurality of antenna signals in order to constructively or destructively combine the signals.

[0212] Similarly, the third weighting application step 646 applies a third weighting value (ffoptj) to the third frequency interval antenna signal in order to combine the signals and provide a weighted third frequency interval signal. In this way, the third weighting value is applied to the information corresponding to the third frequency interval among the multiple antenna signals in order to constructively or destructively combine the signals.

[0213] The second weighting application step 626 applies a third weighting value (".") to the second frequency interval antenna signal, so as to combine the signals and provide a weighted second frequency interval signal. In this way, the second weighting value is applied to the information corresponding to the second frequency interval among the multiple antenna signals so as to constructively combine the signals.

[0214] The signal combining step 634 then combines the weighted first frequency interval signal, the weighted second frequency interval signal, and the weighted third frequency interval signal. In this example, because the frequency shift is applied in the first correction path 657 and the third correction path 661, before combining the three signals, the corresponding inverse frequency shift is applied to the weighted first frequency interval signal and the weighted third frequency Each of the interval signals. The output of the signal combination step 634 is output signaling, which has an improved SINR and has a reduction in selective application of interference in the first frequency interval and the third frequency interval. The selective reduction of interference is applied according to whether the signaling in the first frequency interval and the third frequency interval originates from a similar direction as the signaling in the second frequency interval.

[0215] Additional example embodiments

[0216] We will now describe another example embodiment that utilizes two replacement/training signals in a first frequency interval and a third frequency interval. The two signals used for each frequency interval can be referred to as the outer sub-group and the inner sub-group of the frequency interval. [0217] The electronically steered beamforming method for maximizing the SINR discussed above can perform constructive (coherent) addition of the received desired signal. This type of beamforming with suppression and combination capabilities should then be able to make an effective choice between suppression and combination in terms of SNR, a so-called "suppression versus combination trade-off" procedure. The following discussion expands the program for AoA calculation through two additional parts about 100kHz wide (100kHz frequency interval) to obtain two additional training signals or substitute signals (representations). The two additional alternative signals are:

[0218] i) The internal subgroup of the first frequency interval (between -200kHz and -100kHz), which can be referred to as a "downmix signal", in which the expected IBOC signal is mixed with noise and the possible lower first Neighboring interference signal, and [0219] ii) The internal subgroup of the third frequency interval (between +100kHz and +200kHz), which can be called "upmix signal", where the expected IBOC signal is mixed with noise And possibly the first neighboring interference signal.

[0220] Similarly, the part of the first frequency interval between -300kHz and -200kHz is referred to as the outer subgroup of the first frequency interval or "lower first neighbor (FM) interference signal". The part of the third frequency interval between +200kHz and +300kHz is referred to as the external subgroup of the third frequency interval or "upper first adjacent (FM) interference signal".

[0221] Therefore, to calculate the AoA for the trade-off procedure proposed in the following description, for both the upper first neighboring interference source and the lower first neighboring interference source, we will use a total of five alternative signals, namely:

[0222] i) For the desired IBOC signal (between -100kHz and +100kHz),

[0223] ii) For the next first adjacent (FM) interference signal (between -300kHz and -200kHz),

[0224] iii) For the upper first neighbor (FM) interference signal (between +200kHz and +300kHz),

[0225] i_v) For downmix signals (between -200kHz and -100kHz), and

[0226] v) For upmix signals (between +100kHz and +200kHz).

[0227] Parts i-iii are described above with reference to FIGS. 5 and 6. Therefore, we will focus on the iv and ν parts of the down-mix signal and the up-mix signal in the description of this embodiment.

[0228] We then propose to use AoA to determine (similar to the function described above with reference to Figures 5 and 6) the "spatial separation" between the desired signal and the first neighboring interference signal. However, in this embodiment, we extend this criterion based on the AoA of the down or up mixed signal by determining the "spatial distance" between the desired signal and the mixed signal. Finally, if there is "lack of spatial separation" between the desired signal and the first neighboring interference signal, or in this embodiment, if there is "lack of spatial distance" between the desired signal and the mixed signal, then we decide to combine constructively, similar to In the function of Figure 5 and Figure 6. Next we will discuss the generation of two additional mixed signals that we propose for the trade-off procedure.

[0229] FIG. 7 shows another method of operating the receiver system, which can be considered as an extension of the processing of FIG. 6. The features of FIG. 7 that have been described with reference to FIG. 6 will not be repeated here.

[0230] In addition to the three estimation paths 756, 758, 760 of FIG. 6, the method of FIG. 7 further includes a first internal estimation path 757 and a third internal estimation path 759. The first internal estimation path 757 is used to determine the AoA of the downmix signal (the internal subgroup of the first frequency interval between -200kHz and _100kHz), and is also used to facilitate the determination of the first weighting value. The third internal estimation path 759 is used to determine the AoA of the upmix signal (the internal subgroup of the third frequency interval between +100kHz and +200kHz), and is also used to facilitate the determination of the third weighting value.

[0231] Down and Up Mix Signal Generation

[0232] In a similar manner to the first estimation path 756 and the third estimation path 760 described with reference to FIG. 6, the downmix signal is frequency shifted by +150 kHz so that its center is at 0 kHz. In addition, the frequency of the upmix signal is shifted by -150kHz so that its center is at 0kHz. A 50kHz low-pass filter (LPF) is then used to filter each of these frequency-shifted signals at about 0Hz.

[0233] It can be seen from FIG. 3 that the up and down mixed signals include a mixture of the received desired (H)IBOC signal and the first neighboring interference signal. For the trade-off procedure we are interested in, we determine the AoA of these mixed signals to determine the combination or suppression of the lower or upper first adjacent interference signal. That is, whether to set a weighting value for constructive or destructive combination of the first frequency interval and the third frequency interval. However, before we can use the mixed signal AoA, we first need to calculate the steering vector of the mixed signal, which is the topic in the following chapter under the heading "Mixed signal steering vector calculation". [0234] Mixed signal steering vector calculation

[0235] In this chapter, the calculation of the steering vector of the down or up mixed signal will be introduced. The calculation of the steering vector of the mixed signal is the same procedure as described above for the outer subgroup of the first frequency interval and the third frequency interval. In addition, each of the steering vectors of the mixed signal is obtained by using internal subgroups of the first frequency interval and the third frequency interval.

[0236] The alternatives of the received downmix signal and the received upmix signal are passed to the so-called: "suppression pair combination trade-off" algorithm. This trade-off algorithm (which can be implemented through the "determine steering vector and AoA through MMSE" step 707) calculates two additional steering vectors, which are compared with the steering vector calculated in the example of FIG. 6. The two additional steering vectors are the steering vector of the received down-mix signal and the steering vector of the up-mix signal. As discussed above, the steering vector of the received signal has a one-to-one relationship with the angle of arrival (AoA) of the signal. Therefore, the steering vector of the mixed signal contains the necessary spatial information to compare the mixed signal in the spatial domain with the received spatial information of the desired IBOC signal. In addition, we will address specific comparison criteria for down-mix signals and for up-mix signals under the heading "Comparison of Mixed Signal Arrival Angles" below. As pointed out above, and in addition for mixed signals, through

By using the "principal component analysis" (pea) method, the so-called "estimation and packing" technique to solve the eigenvalue problem, the calculation of the steering vector can be almost instantaneous. Furthermore, as discussed above, the principal component is proportional to the required steering vector. In addition, also here, the proposed composite baseband "suppression pair combination trade-off" algorithm is fast, that is, it has low latency. In fact, the delay is only determined by the observations (samples) required to calculate the spatial covariance matrix. It should be noted that the above description of FIG. 6 involves calculating the steering vector of the received desired signal and the first neighboring interference signal using the pea method. In addition, for the embodiment of FIG. 7, a similar procedure is used to obtain the steering vectors of the mixed signal, and then these steering vectors (ie, AoA) are used to make an appropriate trade-off between suppression and combination.

[0237] Also here, regarding FIG. 6, the calculation standard of the steering vector in the trade-off algorithm is the minimization of the mean square error (MMSE), that is, the Wiener-Hopf standard. In addition, from the sample-based spatial covariance matrix R for the down-mix signal u and the up-mix signal ν respectively_uuAnd R_vvCalculate the steering vector sv_uThe procedure for SVv and SVv is similar to the procedure described above for the desired signal and for the first neighboring interference signal. Therefore, for the additional mixed signals proposed in this embodiment, we directly show the calculated steering vectors (represented by their optimal weights), which gives:

[0238] U; _i( ··* PI for downmix signals,

[0239] Two for the up-mix signal, equation 34

[0240]

[0241] In the case of a sample-based spatial covariance matrix;

<img file="CN107370499A_D0006.tif" />

For downmix signals,

[0242]

[0243] For the up-mix signal, the equation³⁵ Where Mki is a stream of sample vectors used for the substitute of the received downmix signal, and is a stream of sample vectors used for the substitute of the received upmix signal. Also here, the eigenvectors are calculated by solving the quadratic eigenfunctions of the two replacement signals in the two additional frequency intervals. As a result, we obtain two additional steering vectors^ We will use these two additional steering vectors^ to calculate the AoA of these received mixed signals. These AoAs are then compared to determine whether there is a lack of spatial distance from the desired (H) IBOC signal received. This mixed signal AoA comparison procedure will be described in the next chapter.

[0244] Mixed signal arrival angle comparison

[0245] In this section, we will introduce comparison criteria to make an appropriate trade-off between suppression and combination in terms of SNR, for example, for the beamforming method as described with reference to FIG. 6.

[0246] This trade-off algorithm is based on the following availability:

[0247] i) The steering vector of the mixed signal (internal subgroup of the first frequency interval and the third frequency interval) calculated as explained in the previous section,

[0248] ii) The received steering vector of the desired signal (second frequency interval), and

[0249] iii) The steering vectors of the first adjacent interference signal (the outer subgroup of the first frequency interval and the third frequency interval), both of which are described above with reference to FIG. 6.

[0250] The determination criterion uses the AoA information obtained from these steering vectors. More precisely, the determination is based on the difference between the received desired signal and the received AoA of the lower or upper first adjacent interference signal, where the difference is the same as the received desired signal and the received lower or The difference combination between the AoA of the upmix signal. In the rest of this chapter we will

Explain the comparison and the corresponding criteria.

[0251] In a manner similar to the manner in FIG. 6, the AoA of the received down-mix signal and the received up-mix signal can be obtained from the steering vectors of the received down-mix signal and the received up-mix signal respectively (for both ULA of each antenna), as follows:

<img file="CN107370499A_D0007.tif" />

Where for the downmix signal, lsv......,d

[0252] φ._{ι (} shV^J

[0253] ₇.&lt;_:. TM j where fast-for up-mixed signals, equation 36

[0254] Now we propose to use the obtained AoA to discriminate whether the received down-mix signal and the up-mix signal have sufficient spatial distance from the received desired signal. However, the criterion for sufficient spatial distance between the mixed signal and the desired signal also depends on the spatial separation between the desired signal and the first adjacent interference signal. It should be noted that the lower or upper mixed signal is actually a mixture of the desired signal and the lower or upper first adjacent interference signal. If the spatial distance between the mixed signal and the desired signal is less than α% of the spatial separation between the first adjacent interference signal and the desired signal, where α &lt;1, then the lower or upper mixed signal and the desired signal There is the so-called "lack of spatial distance". The criterion for determining whether there is a lack of spatial distance between the received downmix signal and the received desired signal is given by the following formula:. / Wrong-(p_s\ &lt;σ_ο|φ_?·-Φ_&amp;\

[0255] I: Square-Hang formula 37'' in potential \φ_η -φ"two-- Φ.νϊ

[0256] The calculation of J is described above with reference to FIG. 6, and the criterion for determining whether there is a lack of spatial distance between the received up-mix signal and the received desired signal is given by the following formula: ^\ &lt;Yi/ ^m

[0257] 'l false sample> «,,Ιφ/ Kaorusun Discussion

[0258] where the calculation of W is described above with reference to FIG. 6, and where K,, α_ν1} is a variable that can be set to an appropriate value (for example, 0.45 for both). These values can be obtained through other methods such as "field testing" or (6) the spatial modeling of the IBOC transmission scene.

[0259] In addition, through the first simulation with the captured IBOC signal outside the scene, it was found that AoA, h^y may have considerable fluctuations. Therefore, it is proposed that N of N observations_uAverage these AoAs on the, Nv block (through "running average"), which is obtained for the downmix signal:

<td></td><td>Λ· «Ιφζ-</td><td colspan="2">-φ^Ι &lt;σ</td><td>Ψί-</td><td>-⁽p_sl</td><td></td>

<td>[0260]</td><td>E_u = &lt;_：</td><td></td><td></td><td></td><td></td><td>Equation 39</td>

<td></td><td>'False«-</td><td></td><td>&gt; a</td><td>Ψι ~</td><td></td><td></td>

<td>[0261]</td><td>And for the upmix signal:</td><td></td><td></td><td></td><td></td><td></td>

<td></td><td>ί ft- «\φ_ν -</td><td>άΙ</td><td>&lt;</td><td>\Ψΐ'</td><td></td><td></td>

<td>[0262]</td><td>ΛΤΓ.·... ΙΊΊ!·.'·. y.·</td><td></td><td></td><td></td><td></td><td>Equation 40</td>

<td></td><td>,False ί W ··</td><td>'^fPs\</td><td></td><td></td><td></td><td></td>

<td>[0263]</td><td colspan="2">The running average AoA value is given by:</td><td></td><td></td><td></td><td></td>

[0264] φ ~ Α-ί-V. h! For the downmix signal,

[0265] 77·^4-/For the up-mix signal, equation 41 ?,; :: ί

[0266] where {N_U, N_V} Is a variable that can be set to an appropriate value (for example, set to 4 for both). These values can be obtained through other methods such as "field testing" or (6) the spatial modeling of the IBOC transmission scene. It should be noted that due to this additional averaging, the delay increases in proportion to the number of M,Nv} blocks of N observations. Logical indicator [Eu, E_v] Used to decide which procedure to follow to make the appropriate SNR trade-off between suppression and combination. The procedure followed is the last part of this embodiment that suppresses the combination trade-off algorithm, and will be discussed in the following sections.

[0267] Suppressing the trade-off of combination

[0268] Finally, the last part of the suppression-to-combination trade-off algorithm actually makes an appropriate choice between suppression and combination. Logical indicator [Eu, E_v] Indicates whether the spatial distance is lacking, and through this information, the method decides whether to perform suppression (destructive combination) or combination (constructive combination) to obtain the best SNR for the digitally modulated signal.

[0269] In FIG. 7, the step 751 for determining the weighted value for maximizing the SINR is based on the same three frequency interval matrices 776, 778, 780 described above with reference to FIG. 6.

[0270] In addition to the function of FIG. 6, in this example we also pass two logical indicators [Eu, E_v] Introduce the possible extension of the beamformer, these two logical indicators [Eu, E_v] Indicates whether there is a lack of spatial distance between the received mixed signal and the received desired signal. It should be noted that the lack of spatial distance means that the AoA of either or both of the desired signal and the mixed signal have a "close enough" value, that is, they are spatially "close enough" to indicate the lack of spatial distance. The lack of spatial distance can lead to determining that the beamforming algorithm should not attempt to suppress the first neighboring interference signal. Therefore, since the mixed signal information in this case indicates that the first neighboring interference signal is not, for example, destructive, that is, it is more or less "harmless". If it is actually the "harmless" first neighboring interference signal, it is appropriate to apply "constructive combination" on the first neighboring interference signal. Therefore, if the lower or upper mixed signal has an AoA similar to the desired signal, it is appropriate to use constructive combination for the lower or upper first adjacent interference signal (the first frequency interval and the third frequency interval). It should be noted that the optimal weights for constructive combinations may already be available because they can be calculated for the desired signal received. Therefore, this completely means that if for example the lack of a spatial distance logical indicator E for the downmix signal_uIf it is "true", the optimal weight of the desired signal should also be applied to the received lower first adjacent interference signal to obtain a constructive combination in the first frequency interval and in the second frequency interval. For indicator Ε_νAlso consider, that is, if Ε_νIf it is "true", the optimal weight of the desired signal should also be applied to the received upper first adjacent interference signal to obtain a constructive combination in the third frequency interval and the second frequency interval. Therefore, if both indicators are "true", the best weight is used for the lower first neighboring interference signal and for the upper first neighboring interference signal. In this case, in the three correction paths, constructive combinations are applied in all three frequency intervals, that is, in the first frequency interval, the second frequency interval, and the third frequency interval.

[0271] This can be implemented as a logical OR function between logical variables L and E through step 755. That is, if L_lsEither or Eu is true, then the first weighting value (Wopt, ,) is set as a value for constructive combination. If L,_sEither or Eν is true, then the third weighting value (wopt.j) is set as a value for constructive combination.

[0272] In a manner similar to the manner described above with reference to FIG. 6, in another embodiment, the optimal weight for combination may also be the first frequency interval (or the third frequency interval) covariance matrix Information is "adjusted." For example, they can

"Adjustment" based on the covariance matrix used for the internal and/or external first frequency interval (and/or similar for the third frequency interval).

[0273] Therefore, due to the application of the constructive combination method, if the spatial distance logical indicator [Eu, E_vIf any one of] is "true", then when the mixed signal is spatially "close enough" to the desired (H)IB0C signal, unnecessary suppression does not occur. It should be noted, however, that the result of this constructive combination method is that if the logical indicator [E_u, Ε_νIf any one of] is "true", the suppression of the "harmless" first neighboring interference signal is also impossible. On the other hand, because the beamformer has "no reason" to suppress the received "harmless" first neighboring interference signal (E_uIs "true"), the received "harmless" first neighboring interference signal (Ε_νIs "true"), or the received "harmless" lower first adjacent interference signal and upper first adjacent interference signal (Eu, E_vBoth are "true"), which should be acceptable. In addition, this should also avoid "unnecessary attempts" to suppress the lower second neighboring interference signal or the upper second neighboring interference signal.

[0274] FIG. 8 shows a receiver system 800 representing a generalized version of the processing of FIG. 7 for one of the first neighboring interference signals (first frequency interval). It should be understood that the system 800 of FIG. 8 can be extended so that it can also handle another first neighboring interference signal (third frequency interval). Fig. 8 shows a structure similar to the structure of Fig. 5. The features that have been described with reference to FIG. 5 will not be repeated here.

[0275] The receiver system 800 includes a first AoA block 806, which in this example includes an inner first AoA block 806a and an outer first AoA block 806b. The inner first AoA block 806a determines the inner first AoA ((p_u) 808a. The outer first AoA block 806b determines the outer first AθΑ(5H) 808E associated with the outer subgroup of the first frequency interval.

[0276] The receiver system 800 includes a second AoA block 810 that determines a second AoA 812 in the same manner as in FIG. 5.

[0277] The first weight determination block 814 performs subsequent processing to determine whether to set a first weight value for constructively or destructively combining information corresponding to a first frequency interval of a plurality of antenna signals:

[0278] To assist "self-zero prevention" in the first frequency interval (using L_lsOr U, as described above with reference to Figure 6):

[0279] Refer to the outer first angle difference (|(pr(p_s|) is determined as the outer first angle of arrival ((pJSOSb and the second angle of arrival ((p_s) The difference between 812.

[0280] Refer to if the outer first angle difference (|(Mp_s|) Less than the external threshold (Δ φ _is),then

[0281] The first weighting value (wopt.x) 816 is set as a value for constructively combining the information of the first frequency interval corresponding to the plurality of antenna signals. This is based on the first interference signal (in the first frequency interval) that is considered to be too close to the desired signal (in the second frequency interval) in space, so that any destructive combination has the desired attenuation in the first frequency interval The risk of digital sidebands.

[0282] In order to help "prevent unnecessary suppression" (use E_uOr E_v, As described above with reference to Figure 7):

[0283] Refer to the internal first angle difference C|(p_u-(P_s|) is determined to be the inner first angle C(p_u) 808a and the second angle of arrival ((p_s) The difference between 812;

[0284] If the inner first angle difference (]^（pu-^（ps|) is less than the outer first angle difference (f(pi-(p_s|) the predetermined proportion (a_u), that is, the AoA of the downmix signal is closer to the expected signal than the first neighboring interference signal (this can mean that the internal first AoA is close to the expected signal, and therefore because it is a similar signal or at least does not have the same Signal any spatial difference, and the combination should be performed), then:

[0285] The first weighting value (W_0pU) 816 is set to constructively combine the first frequency corresponding to multiple antenna signals

The value of the information of the rate interval. This is based on the fact that the signals in the inner subgroup (between -200 kHz and -100 kHz) of the first frequency interval are spatially closer to the desired digital sideband than the first interference signal in the first frequency interval. Therefore, because the downmix signal AoA is close to the desired signal, it is not considered "harmful polluting", or the first neighbor has similar AoA. Nonetheless, apply combinations because CIC can lead to a "stronger" signal. Recall that the AoA of the second frequency interval between (-100kHz and +100kHz) will be the same as the AoA of the digital sideband in the internal subgroup of the first frequency interval (between -200kHz and -100kHz).

[0286] If the inner first angle difference (|(p_u-(P_sl) A predetermined ratio greater than the difference of the outer first angle (a_u), then:

[0287] The first weighting value 816 is set to a value for destructively combining the information of the first frequency interval corresponding to the plurality of antenna signals. This is based on the fact that the signals in the internal subgroup (between -200 kHz and -100 kHz) of the first frequency interval are spatially closer to the first interference signal in the first frequency interval than the expected digital sideband.

[0288] Unless a specific order is clearly stated, the instructions and/or flowchart steps in the above figures can be executed in any order. In addition, those skilled in the art will recognize that although an example instruction set/method has been discussed, the materials in this specification can be combined in a variety of ways to produce other examples, and should be described in detail herein within the context provided To understand.

[0289] In some example embodiments, the instruction set/method steps described above are implemented as functions and software instructions embodied as an executable instruction set, and these executable instructions are programmed and controlled in a computer or with the executable instructions. On the machine. Such instructions are loaded for execution on a processor (for example, one or more CPUs). The term processor includes microprocessors, microcontrollers, processor modules or subsystems (including one or more microprocessors or microcontrollers), or other control or computing devices. The processor can refer to a single component or to multiple components.

[0290] In other examples, the instruction set/method described herein and the data and instructions associated therewith are stored in a corresponding storage device, which is implemented as one or more non-transitory machines or computer-readable or computer-usable Storage media. Such computer-readable or computer-usable storage media are considered to be part of an article (or article). An article or article can refer to any manufactured single component or multiple components. Non-transitory machine or computer usable media or media as defined herein do not include signals, but such media or media can receive and process information from signals and/or other temporary media.

[0291] Example embodiments of the materials discussed herein may be implemented in whole or in part via a network, a computer, or a data-based device and/or service. These can include clouds, the Internet, intranets, mobile devices, desktop computers, processors, look-up tables, microcontrollers, consumer devices, infrastructure, or other enabling devices and services. As can be used herein and in the claims, the following non-exclusive definitions are provided.

[0292] In one example, one or more of the instructions or steps discussed herein are automated. The term automation or automatic (and similar variations) means the use of computers and/or mechanical/electrical devices to control the operation of equipment, systems, and/or processes without the need for human intervention, observation, effort, and/or decision-making.

[0293] It should be understood that any components referred to as coupled can be directly or indirectly coupled or connected. In the case of indirect coupling, additional components may be located between the two components that are said to be coupled.

[0294] In this article, example embodiments have been presented in terms of selected detail sets. However, those of ordinary skill in the art will understand that many other example embodiments including different selected sets of these details can be practiced. It is hoped that the appended claims cover all possible example embodiments.