Methods and systems for compensating the frequency shift of filters for WLAN traffic

Abstract

A non-transitory, computer-readable storage medium (120; 506) having stored thereon executable computer program instructions that, when executed by one or more processors (504), cause the one or more processors (504) to perform operations comprising: Determining whether a frequency response of a filter (143) has shifted, wherein the shift in the frequency response comprises a shift in a stopband (220) of the filter (143) and the filter (143) is a component of a network device (100) for communicating data frames in a wireless network; in response to determining that the frequency response of the filter (143) has shifted, applying a digital predistortion to a signal input to the filter (143) to compensate for the shift in frequency response; Determining whether the digital pre-distortion has compensated for the shift in the frequency response of the filter (143); and in response to determining that the digital pre-distortion has not compensated for the shift in the frequency response of the filter (143), applying at least one PHY Protocol Data Unit (PPDU) scheduling scheme to the data frames transmitted by the network device (100).

Description

BackgroundThe Unlicensed National Information Infrastructure (U-NII) radio band is part of the unlicensed radio frequency spectrum used by IEEE 802.11 devices and wireless Internet Service Providers (ISPs) for Wi-Fi communication. Currently, U-NII allocates Wi-Fi channels in the 5 GHz band in four subbands: U-NII-1 (5.150 to 5.250 GHz), U-NII-2 (5.250 GHz to 5.725 GHz), U-NII-3 (5.725 GHz to 5.850 GHz), and U-NII-4 (5.850 GHz to 5.925 GHz).Recently, there has been efforts to use the 6 GHz band for Wi-Fi communication. For example, it has been proposed to assign Wi-Fi channels in the 6 GHz-U-NII radio band in four subbands: U-NII-5 (5.945 to 6.425 GHz), U-NII6 (6.425 GHz to 6.525 GHz), U-NII-7 (6.525 GHz to 6.875 GHz), and U-NII-8 (6.875 GHz to 7.125 GHz). Such channel assignment in the 6 GHz band would significantly increase the number of available channels for Wi-Fi communication, especially as the currently available Wi-Fi bands (e.g., 2.4 GHz, 5 GHz) are increasingly congested through use by ISPs and wireless local area networks (WLANs). In addition, such channel allocation in the 6 GHz band provides the possibility of higher aggregated throughput across all Wi-Fi bands.US 2017 / 0 077 945 A1 describes a converter comprising preferably: (1) a plurality of oversampling converters each processing a different frequency band and operating in parallel; (2) multirate (i.e. polyphase) delta-sigma modulators (preferably second order or higher); (3) multibit quantizers; (4) multibit-to-variable level signal converters, such as resistor ladder networks or current source networks; (5) adaptive nonlinear bit mapping to compensate for mismatches in the multibit-to-variable level signal converters (e.g. by mimicking such mismatches and then shifting the resulting noise to a frequency range in which it is filtered out by a corresponding bandpass (reconstruction) filter); (6) Multiband (e.g., programmable noise transfer function response) band-pass delta-sigma modulators; and / or (7) a digital predistortion linearizer (DPL) for cancelling noise and distortions caused by an analog signal band-pass (reconstruction) filterbank.DE 10 2021 126 873 B4 describes systems and techniques aimed at compensating the frequency response shift of the filter, including the attenuation compensation. The attenuation compensation may apply pre-distortion to balance the size of the attenuated resource units (RUs). Moreover, shifting the filter frequency response may require the use of PPDU (Phy Protocol Data Unit) scheduling schemes.Brief Description of the InventionA non-transitory computer readable storage medium according to claims 1 to 12 and a method according to claim 13 are disclosed.Brief Description of the DrawingsThe present disclosure will be described in detail in accordance with one or more different embodiments with reference to the following figures. The figures are for illustrative purposes only and are merely representative or exemplary embodiments. FIG. 1 is a block diagram of an example network device including aggregated radios (e.g., in the 5 GHz and 6 GHz Wi-Fi bands) and a filter frequency response shift compensation circuit, in accordance with some embodiments. FIG. 2A is a diagram illustrating attenuation of OFDMA resource units (OFDMA= Orthogonal Frequency-Division Multiple Access) in a pass band of a filter in the network device of FIG. 1 due to the temperature sensitivity of the filter according to some embodiments. FIG. 2B is a diagram illustrating application of the disclosed digital pre-distortion techniques to the affected OFDMA RUs in the passband shown in FIG. 2A, in accordance with some embodiments. FIG. 2C is a diagram illustrating application of the disclosed digital pre-distortion techniques to out-of-band (OOB) OFDMA RUs that may be transmitted due to the shift in the stop band shown in FIG. 2A, in accordance with some embodiments. FIG. 3 is a flow diagram illustrating a process for compensating for filter frequency response including digital pre-distortion and PHY protocol data unit (PPDU) scheduling schemes for handling attenuation of OFDMA RUs, in accordance with some embodiments. FIG. 4 is a flow diagram illustrating a process for compensating for filter frequency response including digital pre-distortion and PHY protocol data unit (PPDU) scheduling schemes for handling stopband shift, in accordance with some embodiments. FIG. 5 illustrates an example of a computer system that may be used in implementing filter frequency response compensation in connection with embodiments of the disclosed technology.The figures are not exhaustive and do not limit the present disclosure to the precise form disclosed.Detailed DescriptionThe access to WLAN channels (wireless local area network) is often effected via IEEE 802.11 protocols in accordance with the WiFi technology standard. Other devices may also access the same channels, e.g., Bluetooth. The radio frequency spectrum is of decisive importance for the infrastructure of wireless communication. By way of background, the legacy 802.11 protocol standards provide different radio frequency ranges for Wi-Fi communication, including: 900 MHz, 2.4 GHz, 3.6 GHz, 4.9 GHz, 5 GHz, 5.9 GHz, and 60 GHz. Each of the frequency ranges may be divided into a plurality of channels. These channels may be numbered within a band spaced 5 MHz apart (except in the 60 GHz band where they are spaced 2.16 GHz apart), where the number refers to the center frequency of the channel. Although the channels are numbered 5 MHz apart, the transmitters typically occupy at least 20 MHz and the standards allow channels to be interconnected to form wider channels for higher throughput. These are also numbered after the center frequency of the interconnected group.As more and more radio technologies compete for the spectrum within the Wi-Fi frequency ranges, cloned radios must employ various techniques that allow simultaneous operation (coexistence) to avoid interference issues. In some cases, aggregated radios may operate with different radio technologies whose spectra overlap (or may not). In other cases, aggregated radios may also operate with the same radio technology and their spectra may overlap (or may not). By employing a technique referred to herein as frequency domain coexistence (using collapsed radios), a network device may simultaneously operate both a 5 GHz radio with the 5 GHz-U-NII channel assignment and a 6 GHz radio with the proposed U-NII-6 GHz channel assignment. This technique of coexistence in the frequency domain has the advantage that multiple channels can be operated, each of which can lie on adjacent bands. In other words, unrestricted simultaneous operation in both bands can be achieved.Despite the advantages of coexistence in the frequency domain, aggregated radios typically require a substantial band gap between their respective spectra for their operation. This technique relies on the use of filters (analog / digital) to prevent the radios from interfering with each other. However, if the band gap is too narrow, which may be the case with collapsed radios operating in the 5 GHz and 6 GHz Wi-Fi bands, coexistence in the frequency domain may become very difficult.To solve the problems associated with frequency domain coexistence solution, the disclosed methods for compensating for frequency response shift of the filter may include applying pre-distortion to an affected channel to compensate for the size of the attenuated resource units (RUs) or orthogonal frequency division multiplexed subcarriers (OFDM). This approach is referred to below as "damping compensation". Moreover, the techniques for compensating for filter frequency response shift may include applying one or more PHY protocol data unit (PPDU) scheduling schemes in a manner that compensates for any effects of the shift on the performance of the wireless channels. For example, a PPDU scheduling scheme may include reducing the bandwidth of a channel by balancing (or discarding) attenuated RUs (or subcarriers) as a fail-safe (e.g., in cases where attenuation compensation does not successfully compensate attenuation). This approach is referred to as "PPDU planning" below. Some of the PPDU scheduling schemes are associated with performance tradeoffs where some throughput may be lost through the application of the scheme, while the rest of the channel bandwidth may still be used. As will be described in detail, the techniques for compensating for the frequency response shift of the filter may be implemented using various mechanisms. For example, the aspects of attenuation compensation may be implemented via dedicated digital predistortion circuits (DPD) or via PPDU scheduling. For example, a network device may be adapted to include a font end that is specifically configured to pass a pre-distorted signal through the filter to compensate for attenuation of some RUs in the passband, which helps to increase the performance of the RUs involved again. As will be described in detail, attenuation of RUs may be caused by the inadvertent shift in frequency response of a filter. As will be described in detail with reference to FIG. 2, inadvertent shifts in the frequency response of a filter may occur as a result of filters having the basic characteristic of temperature sensitive frequency responses (e.g., the frequency response of a filter shifts by X MHz with each temperature change from Y ° C. ). Accordingly, examples of the techniques and systems described herein may effectively mitigate this undesirable shift in the frequency response of the filter (without having to stabilize the temperature of the filter), which may potentially result in degradation of the channels in the Wi-Fi bands.For discussion purposes, the disclosed techniques for compensating for filter frequency response shift with respect to coexistence are described specifically with respect to 5 GHz and 6 GHz radios, e.g., the referenced radios 130 and 132 within the network device 100 illustrated in FIG. 1. However, it should be appreciated that the disclosed techniques for compensating for the frequency response shift of the filter are applicable to filters used for different types of radio technologies (and are not limited to the Wi-Fi technologies described herein) operating in different frequency spectra (and are not limited to 5 GHz and 6 GHz bands as described herein).FIG. 1 is a block diagram of an example network device 100 for continuous 5-GHz and 6-GHz operation and for implementing the disclosed filter frequency response shift compensation techniques. The network device 100 includes at least one processing resource 110 and at least one machine readable medium 120 including (e.g., encoded with) at least instructions 122 executable by the at least one processing resource 110 of the network device 100 to implement the functionalities described herein with respect to the instructions 122. FIG. 1 is provided to illustrate that the network device 100 is configured to include instructions 122, which in particular implement the functionality for the PPDU scheduling scheme 123 as disclosed herein. Accordingly, the network device 100 is programmed to perform the PPDU scheduling aspects of the filter frequency response shift compensation techniques. The specific operations of the disclosed PPDU scheduling scheme are discussed in more detail below with reference to FIG. 3.In the example of FIG. 1, the network device 100 may perform any network data transfer operations, including, but not limited to, switching, routing, bridging, or a combination thereof. In some examples, the network device 100 may include a wireless access point (WAP). In the examples described herein, a "WAP" generally refers to reception points for any known or suitable wireless access technology that may become known later. In particular, the term WAP is not intended to be limited to WAPs conforming to the IEEE 802.11 standards. A WAP is typically an electronic device that allows wireless devices to connect to a wired network via various communication standards. A WAP may include all necessary hardware components to carry out the inventions disclosed herein, including, but not limited to: processors, memory, display devices, input devices, communication devices, etc. One skilled in the art will appreciate that the network device 100 may be any suitable type of network device from one or more suitable manufacturers.In the example of FIG. 1, the network device 100 includes a first radio 130 and a second radio 140. In some examples, both the first radio 130 and the second radio 140 may operate in one or more frequency bands conforming to one or more IEEE standards (e.g., 802.11ax). In some examples, the first radio 130 may operate with one or more channels in the 5 GHz band. For example, the first radio 130 may operate on one or more channels in subbands U-NII-1, U-NII-2, U-NII-3, and U-NII-4. In some examples, the second radio 140 may operate with one or more channels in the 6 GHz band. For example, the second radio 140 may operate on one or more channels in the proposed subbands U-NII-5, U-NII-6, UNII-7, and U-NII-8. It will be appreciated by one of ordinary skill in the art that the first radio 130 and the second radio 140 may operate in any suitable frequency band and may conform to any suitable type of wireless communication standards known today or later developed (e.g., in accordance with future and / or emerging Wi-Fi standards). Although FIG. 1 shows the network device 100 having two radios, those skilled in the art will understand that the network device 100 may include four, eight, or any other suitable number of radios.In the example of FIG. 1, the network device includes a first antenna 139 and a second antenna 149. In some examples, each of the first antenna 139 and the second antenna 149 may transmit and / or receive directional signals, omni-directional signals, or a combination thereof. In the examples described herein, a "directional" signal refers to a signal that radiates more strongly in one or more directions compared to one or more other directions along an azimuth plane (i.e., horizontal plane), while a "non-directional" signal refers to a signal that evenly radiates in all directions along an azimuth plane. In some examples, both the first antenna 139 and the second antenna 149 may be a phased array antenna. In the examples described herein, a "phased array antenna" refers to a group of antennas that can generate a directional signal that can be electronically controlled to point in different directions without requiring the antennas to be moved. In such examples, a phased array antenna may include a group of directional and / or omni-directional antennas that may focus RF energy in particular spatial directions. Those skilled in the art will understand that the first antenna 139 and the second antenna 149 may comprise any suitable type of antenna(s) known today or later developed. Although FIG. 1 shows the network device 100 having two antennas, it will be understood by those skilled in the art that the network device 100 may include four, eight, or any other suitable number of antennas.As described above, a radio operating in the 6 GHz band, e.g., the second radio 140, may offer a great potential for increasing channel availability and throughput for Wi-Fi communication. However, as mentioned above, the assignment of Wi-Fi channels in the 6 GHz band could have a narrow band gap between Wi-Fi channels in the 5 GHz band, which could lead to coexistence issues with existing 5 GHz Wi-Fi channels. In operation, for example, the network device 110 may simultaneously use the first radio 130 for operation in the 5 GHz band using the 5 GHz-U-NII channel assignment and the second radio 140 for operation in the 6 GHz band using the proposed U-NII-6 GHz channel assignment. However, coexistence in the frequency domain needs to be efficiently realized in order for the first radio 130 and the second radio 140 to take advantage of cooperation without causing interference and interference in certain Wi-Fi channels due to the narrow band gap between the 5 GHz and 6 GHz bands.In particular, when the network device 100 receives a signal in a Wi-Fi channel at or near an upper limit of the 5 GHz band while generating another signal in a Wi-Fi channel at or near a lower limit of the 6 GHz band (or alternatively, when the network device 100 receives a signal in a Wi-Fi channel at or near a lower limit of the 6 GHz band while generating another signal in a Wi-Fi channel at or near an upper limit of the 5 GHz band), the received signal may experience interference and interference from the generated signal, resulting in desensitization (e.g., Shorter range) and degradation of the signal quality (e.g., lower throughput) of the received signal. The coexistence problems may become worse if the passband of the spurious bands is very wide (e.g., >500 MHz) because the filter design becomes very complicated. To achieve frequency domain coexistence, some existing RF filter technologies require a large transition bandwidth (e.g., 250 MHz or more) between the pass filters for 5 GHz and 6 GHz bands to achieve a minimum DB suppression (e.g., at least 50 dB) required to mitigate such interference and interference. As an improvement over these existing RF filter technologies, the network device 100 is clearly designed to adequately address problems of coexistence of 5-GHz and 6-GHz Wi-Fi channels, even if only a narrow transition bandwidth (e.g., 50 or 110 MHz) between Wi-Fi channels is available at or near the boundaries of the 5-GHz and 6-GHz bands. According to embodiments, the network device 100 includes front ends 135, 145 with an improved design configured to: 1) ensure proper frequency-domain coexistence between the respective spectra of the referenced first radio 130 and second radio 140; and 2) provide feedback to the respective radio 130, 140 to apply appropriate pre-distortion to compensate for a detected frequency response shift. As seen in FIG. 1, each of the radios 130, 140 has its own front end 135, 145, respectively. In the example, radio 130 uses front end 135 and radio 140 uses front end 145. FIG. 1 shows an example configuration for the internal circuit of front end 145 for illustrative purposes only. However, it should be appreciated that the internal front side circuit 135 may have the same or substantially similar configuration.FIG. 1 shows the example configuration for the internal circuit of the front end 145 (illustrated by the dashed box) with regard to its connection to the second radio device 140 and the second antenna 149. In particular, the filters 143 of the front end 145 operate finely tuned and precisely, especially for the narrow subband, and can achieve very high rejection in the narrow band gap. However, there is a trade-off for using such sophisticated filters 143 (e.g., narrowband filters) to prevent the first radio 130 and the second radio 140 from interfering with each other in a manner that maintains frequency-domain coexistence for narrow band gaps (e.g., 5 GHz and 6 GHz Wi-Fi bands). Narrowband filters, such as filter 143, typically have a filtering characteristic that is very sensitive to temperature changes. For example, the edges of the passband of filters 143 may shift in the range between 3 MHz and 5 MHz (left or right) in the frequency spectrum as the temperature changes (e.g., from hot / high to cold / low). This shift in the frequency response of the filters can attenuate some RUs or OFDM subcarriers in the range of 6 dB-10 dB and thus ultimately lead to high EVM problems (error vector magnitude). As a result, an entire affected channel may become completely unusable (due to degradation). For example, although coexistence is supported, attenuation in the pass band of the filters 143 (due to the frequency response shift due to temperature sensitivity) may result in complete loss of channels having a width of 20 MHz, 40 MHz, 80 MHz, 160 MHz, or 320 MHz, which further affects the overall performance of the network device 100. To solve these problems associated with the temperature-induced frequency response shift that may occur at filters 143, front end 145 is further improved to provide feedback to support the digital predistortion functions in first radio 130 and second radio 140 to achieve attenuation compensation.FIG. 2A shows a diagram 200 illustrating an example of attenuation in a pass band that may occur with narrowband filters (such as filters 143 shown in FIG. 1 ) when temperature changes during operation. As described above, narrowband filters are typically prone to response shifts due to temperature variations. According to embodiments, the narrowband filters are specifically designed for 160 MHz channels at the top of the 5 GHz band and at the bottom of the 6 GHz band. However, significant shifts in filter response can render these channels unusable for full bandwidth operation. By way of background, the temperature coefficient of frequency (TCF) is the basic property of filters that measures the shift of the filter response over frequency. The TCF may be expressed in parts per million per degree Celsius (PPM / ° C). A TCF of 25 PPM / ° C at 6000 MHz means, for example, that the response of a filter can shift by 0.15 MHz with each temperature change. An example: The nominal operating temperature range of filters may range from -40°C to 100°C with a nominal operating temperature of about 20°C. If the filter is designed to be centered at 20°C, a temperature variation of 60°C may occur on both sides, which may result in a shift of the filter response in the frequency spectrum by 9 MHz. Filters now commonly using high quality technologies such as SAW (surface acoustic wave), BAW (bulk acoustic wave) and DR (dielectric cavity resonance) all have TCFs in the range of 20-40 PPM / °C. This means that the response of these filters can shift by 7-15 MHz over temperature at 6000 MHz. In particular, the graph 200 illustrates two conditions that may result in a shift in response of a filter based on temperature sensitivity (or TCF): 1) attenuation in the passband (indicated by oval 215) and 2) shift in the stopband (indicated by oval 220).FIG. 2A shows a scenario in which the narrowband filter (shown in FIG. 1 ) may experience a temperature rise of 27° C. For example, when the network device (shown in FIG. 1 ) is operating, the various electrical components may emit heat, which in turn heats the environment inside the device. Since the filter is an internal component of the network device, it may also heat up. Additionally, as mentioned above, the filter may have a TCF of 25 PPM / °C at 6000 MHz. The diagram 200 shows an intrinsic pass band 205 of the filter or the pass band related to the intended response of the filter (e.g., rated operation). Moreover, the graph 200 shows a passband 210 (indicated by the dashed line) affected by the temperature sensitivity, in which the response of the filter has been shifted due to the temperature increase and the temperature sensitivity of the filter (e.g., deviation from the intended response after the design). As can be seen, the diagram 200 shows that the frequency response of the filter has shifted to the left by about 4 MHz in the frequency spectrum (with respect to the position of the actual passband 205) due to the temperature rise (+27° C.) or the passband 210 affected by the temperature sensitivity. In general, this 4 MHz shift in the temperature sensitive passband 210 has caused attenuation on the right side of the plot 200 (indicated by oval 215) and the shift of the stopband (indicated by oval 220) on the left side of the plot 200. In the passband 210 affected by temperature sensitivity, the OFDMA RUs 230 have experienced attenuation. This is illustrated by the reduction in the amplitude of the signal or the "truncation" of the two EVUs 230 to the far right of the diagram 200 as compared to the remaining OFDMA VEUs 225 of the diagram 200. In particular, the diagram 200 shows that the two RUs 230 concerned have an attenuation of up to 6 dB by the RUs 225 in the passband 210. Large attenuation in the RUs, such as the 3 dB-6 dB attenuation shown in the affected RUs 230 of FIG. 2, may be potentially catastrophic and degrade the reliability and quality of the channels. For example, a 26-tone RU of RUs 225 may be approximately -2 MHz wide and have 1024 QAM modulated subcarriers carrying approximately 260 bits. Without the use of the disclosed DPD technique to compensate for this shift, the RUs 225 could experience significant attenuation at high temperatures, ultimately resulting in a high EVM.In addition, the diagram 200 shows a shift of the ratchet band (indicated by the oval 220). This shift may possibly result in desensitizing problems in a simultaneously operated adjacent channel. The effect of such a shift in the stopband may result in a reduced range of the channel involved. For example, if the 160 MHz channel 15 filter (in UNII-5) experiences such a shift in response as shown in FIG. 2A, this may possibly result in a desensitizing of the 160 MHz channel 163 (in UNII-3 / 4). That is, if channels 163 and 15 are operating simultaneously on a WAP, this problem will decrease the range of operation of channel 163 for the clients. In some embodiments, the front-end circuitry and PPDU scheduling techniques as disclosed herein may be applied in a manner that specifically compensates for the notch band shift that may also occur as a result of the temperature sensitivity of the filter. A method which is used in particular for compensating this unintentional displacement of the stop band (in the frequency response of the narrowband filter) is illustrated in FIG. 4. In addition to compensating for "stopband shift", these techniques also help meet very stringent requirements for the out-of-band (OOB) stop in the gap between bands UNII-4 and UNII-5.Returning to FIG. 1, the front ends 135, 145 may be configured to generate a feedback signal that enables the radios 130, 140 to compensate for attenuation experienced by the RU (as shown in FIG. 2A ). In general, the internal circuitry for the front ends 135, 145 extends the linear gain range of the power amplifier (PA) 141 by pre-distortion of the baseband signal. The front ends 135, 145 may be designed according to an algorithm related to a digital pre-distortion scheme trained on the phase and magnitude distortion of a power amplifier (PA) output near compression. The algorithm may be subjected to preprocessing in which the algorithm is trained on the behavior of the PA. Thereafter, during processing, the DPD algorithm may determine a pre-distortion that may be applied to the baseband signal such that the signal is shaped as a linearly amplified signal after undergoing the PA distortion (RF gain) at the PA output. In particular, the front ends 135, 145 may be configured to add an appropriate amount of distortion to a signal required to compensate for attenuation of the filter response. That is, the front ends 135, 145 distort a signal prior to filtering, thereby amplifying the signal so as to cancel any attenuation (learned by the algorithm) of the filter response.In the example configuration, the front end 145 is shown to internally include a first coupler 142 (also referred to herein as a prefilter coupler) connected in series with an output of the PA 141. In addition, the front end 145 includes a second coupler 144 (also referred to herein as post-filter coupler). The second coupler 144 is connected to the output of the filters 143 and, in particular, is positioned to receive the response of the filters 143. The second coupler 144 returns the RU attenuation via a feedback line 146 to the radio 140, which then performs DPD. In the configuration shown, a type of "feedback loop" is thus formed between the filters 143, the post-filter coupler 144 and the radio device 140. The signal received by the second coupler 144 in response to the filters 143 is fed back to the radio 140. In an alternative embodiment where the first coupler 144 is not present, DPD may still be performed using a predetermined attenuation value based on a known temperature shift. In this case, the purpose of coupler 142 would be to perform closed loop power control (CLPC) of the transmit power and would not play a role in active DPD feedback.Accordingly, the DPD algorithm may determine whether the signal is experiencing attenuation (with respect to an intrinsic response) based on the feedback signal from the second coupler, and then that distortion needs to be applied. The first radio 140 may then be used to pre-distort the signal before it enters the filters 143 to compensate for attenuation. For example, the first radio 140 may amplify the signal by an amount that compensates for the attenuation and compensates for the shift in response of the filters 143.Moreover, it should be understood that the internal circuit of front end 145 shown in FIG. 1 is not to be considered limiting and alternative configurations may be used to implement front end 145 in accordance with embodiments. For example, as mentioned above, coupler 142 may be used for RF CLPL, which is generally used in APs. According to this embodiment, a predetermined attenuation value based on a known temperature shift may be used. Accordingly, in this alternative configuration, the second coupler 144 may not be required (or present) in the front end circuit 145.FIG. 2B relates to the scenario illustrated in FIG. 2A. In particular, FIG. 2B shows that the front ends may be used to enable each radio to pre-distort a signal such that the amplitude of the affected RUs 255 is increased by 6 dB. Accordingly, the subcarriers of the last two RUs 255 are shifted up by 6 dB in baseband or amplified by the radios before the signal enters their respective filters. Thus, the pre-distortion applied by the radios sufficiently amplifies the RUs 255 to compensate for the expected attenuation (at extreme temperatures) of the filter response that shifts the RUs down by 6 dB. For example, the graph 250 illustrates that when the pre-distorted signal is filtered by the filters 143 (indicated by the downward arrow), the additional attenuation in the last two RUs 260 (due to the temperature dependent frequency response shift of the filter) is cancelled by the higher amplitude of the pre-distorted RU 255 as input to the filters 143. As can be seen in the diagram 250, the last two RUs 260 (which are affected by the filter frequency response shift) have the same amplitude as the remaining EVUs 225 (which are not affected by the filter frequency response shift) in the passband 210. Thus, the radios may maintain a low EVM of the attenuation-affected RUs, thereby maintaining signal quality while the referenced radios 130, 140 are simultaneously operating at 5 GHz and 6 GHz. It should be appreciated that the disclosed DPD techniques ensure that only the RVUs involved are distorted or otherwise enhanced for DPD. Thus, there are no effects on the RUs that are not affected by the attenuation in the passband, which would result in the corresponding channels being degraded by the DPD techniques. For example, the unaffected RUs on the left side of the passband are not distorted (see FIG. 2B ). An example of a method for applying DPD and PPDU planning, in particular for compensating the attenuation of the OFDMA EUs (as indicated by the oval 215 in FIG. 2A ), is illustrated in FIG. 3. In addition, the first coupler 142 may be used to extend the linear gain of the PA 141.In some embodiments, the disclosed DPD and PPDU scheduling techniques may also be used to compensate for other conditions resulting from the temperature sensitivity frequency response shift of the narrowband filters 143. As explained above with reference to FIG. 2A, the left side of the passband shows a shift in the stopband (indicated by oval 220). According to embodiments, the disclosed front ends 135, 145 (used to implement DPD) and the PPDU scheduling scheme 123 may be tuned to compensate for the offset of the stopband. As already mentioned, FIG. 4 shows an example of a method for applying DPD and PPDU planning, in particular for compensating the displacement of the stopband (characterized by the oval 220).FIG. 2C relates to the scenario illustrated in FIG. 2A. In detail, FIG. 2C shows that the front ends can be used to enable each radio to pre-distort a signal so that the amplitude of all out-of-band (OOB) RUs 270 is reduced to compensate for the shift in stop band. FIG. 2C illustrates that RUs 270 may be out of band of the actual passband 205 of the developed filter. However, the diagram 275 illustrates that these OOB RUs 270 may continue to be emitted due to the shift of the stopband 210 (e.g., because the RUs 270 are included in the pass band that has experienced a shift). As seen in diagram 270, the use of pre-distortion aggressively reduces the amplitude of the OOB RUs 270 now within the shifted pass band 210. With such a low amplitude (relative to the other RUs 276), these OOB RUs 270 are substantially negligible and have less potential to substantially affect transmissions of a concurrently operating channel outside the designated stop band. Thus, this DPD technique compensates for the shift of the stopband. Moreover, as the temperature decreases, the filters 143 may have an opposite frequency response shift due to temperature sensitivity. For example, if the filters 143 cool strongly, the effect may be described as a shift of the pass band to the right with respect to the actual pass band (e.g., the opposite direction of the pass band shift shown in FIG. 2A ). Accordingly, the front ends 135, 145 (which are employed for implementing DPD) and the PPDU scheduling scheme 123 may be configured to compensate for the opposite frequency response offset (e.g., the passband right offset), which may be more appropriate when the network device 100 is likely to be in a low temperature operating environment. In some embodiments, the front ends 135, 145 may be configured such that the radios perform pre-distortion (aggressive amplification) of the affected RUs on the left side of the channel bandwidth for these cases of opposite frequency response shift (e.g., passband right shift).The function of the filters 143 will now be described. As already indicated, the filters 143 are uniquely designed for improved operation that will meet the referenced radios 130, 140, such that frequency domain coexistence is maintained and interference is mitigated.In the example of FIG. 1, the instructions 122 of the network device 100 may be configured to receive a first signal in a 5 GHz band or a 6 GHz band. In some examples, the first signal may be received by the first antenna 139 of the network device 100. In the example of FIG. 1, the instructions 122 of the network device 100 may be configured to generate a second signal in the other of bands 5 GHz and 6 GHz. In some examples, the second signal may be generated by the second radio 140 of the network device 100, where the second radio 140 operates in the other of the 5 GHz and 6 GHz bands.In the example of FIG. 1, the instructions 122 of the network device may be configured to select at least one of the filters 143 applied to the first or second signal in the 5 GHz band. For example, the filters 143 may be composed of a plurality of filters, a second filter transmits a higher frequency band than a first filter in the 5 GHz band, and the second filter is narrower than the first filter. In some examples, the first or second filter may be selected by the front end 145. In some examples, the first or second filter may be selected in response to instructions received by the network device 100 from a controller. For example, the instructions the network device 100 receives from the controller may indicate a channel in the 5 GHz band corresponding to one of the first or second filters, and the first or second filter may be selected in response to the indicated channel.In some examples, the first filter of the filters 143 may pass a frequency band in the range of 5150 to 5735 MHz, and the second filter may pass a frequency band in the range of 5735 to 5895 MHz. One skilled in the art will understand that each of the filters 143 may pass different frequency bands in the 5 GHz band and that each frequency band may correspond to a plurality of Wi-Fi channels.In some examples, the first filter may have a center frequency (i.e., a frequency at the center of the frequency band that the filter transmits) and a stop frequency (i.e., a frequency at a boundary of the frequency band that the filter transmits) such that: 2%≤| ((stop frequency - center frequency)) / ((center frequency))|*100≤ 10%. In some examples, the second filter may pass a frequency band that is greater than an integer multiple of the channel bandwidth of a channel in the 5 GHz band. For example, the channel may be a channel having the largest channel bandwidth among a plurality of channels in the 5 GHz band. In some examples, the second filter may have a center frequency and a stop frequency such that: | ((stop frequency - center frequency)) / ((center frequency)) | *100 ≤ 2%. As already mentioned, these filters 143 can be implemented as narrowband filters, specifically designed for 160 MHz channels at the top of the 5 GHz band and at the bottom of the 6 GHz band, emphasizing how important it is to adequately compensate for shifts in filter response that may otherwise render these channels unusable.FIG. 3 is a flow diagram illustrating a process 300 that includes both the disclosed DPD techniques and PPDU scheduling. As described above with reference to FIG. 1, the sophisticated filters used to maintain frequency domain coexistence in the 5 GHz band and the 6 GHz band may be susceptible to temperature-related frequency response shifts. The PPDU planning is another approach (in addition to DPD) that can be used to correct or otherwise compensate for the attenuation of passband RUs that can be caused by this shift. In some embodiments, the PPDU scheduling aspects of process 300 are deployed as a form of fail-safe, for example, when the radios apply DPD in a manner that is unable to provide sufficient compensation in cases of severe attenuation. The process 300 may be implemented by a network device, such as an AP, that includes aggregated radios operating simultaneously in the 5 GHz band and the 6 GHz band, as shown in FIG. 1. Moreover, the various operations of process 300 may be implemented by a processor executing instructions stored on the network device, the radios of the network device, or a combination thereof.As shown in FIG. 3, the process 300 begins with operation 305. Next, in operation 310, the PPDU scheduling process 300 may determine whether the filter frequency response has shifted due to temperature changes. The detection of a temperature-induced frequency response shift (shown as "yes") may serve as an indication that some RUs are experiencing some attenuation, typically the RUs at the edges of the channel width. In some embodiments, there may be a known correspondence between the amount of temperature shift and the amount of attenuation. For example, a DPD algorithm may learn that a temperature shift of 27° C. at the filter (for a filter with a TCF of 25 ppm / ° C. at 6000 MHz) corresponds to an attenuation of 6 dB. As a result, the DPD algorithm can determine the amount of pre-distortion that must be applied to appropriately compensate for this attenuation. Thereafter, the PPDU scheduling process 300 may proceed to operation 315 to perform DPD techniques for preamplification of the affected RUs. The DPD techniques may be implemented via the radios (including frontends) described in detail above with reference to FIG. 1.If operation 310 determines that there is no shift in the filter response (shown as "no"), the process 300 may proceed to operation 311 and end the process 300 or return to operation 305 to restart the process in an iterative manner.Continuing with operation 320, process 300 may perform a test to determine whether the RUs are still attenuated even after DPD techniques are employed. If attenuation is still present in the affected RUs (e.g., at the ends of the channel width) even after the pre-distortion of the subcarriers, this may result in errors at the receiver while decoding the symbols corresponding to these subcarriers. Decoding errors can result in subframe errors in the PPDU. That is, some MPDUs within the PPDU may not be correctly received by the receiver (AP in case of uplink (UL) traffic or the client(s) in case of downlink (DL) traffic). For OFDMA frames, the impact could be greater for stations using such edge RUs. If it is determined that the affected RUs are still attenuated, the process 300 may proceed with PPDU scheduling as a secondary approach (or fail-safe) to correct the shift in filter response that was not fully corrected by the DPD techniques.Otherwise, if it is determined in operation 310 that there is no attenuation in the RUs, this determination indicates that the DPD techniques of the previous operation 315 have successfully compensated for the frequency response shift caused by temperature sensitivity. Thus, the process 300 may proceed to operation 321 and end the process 300 or return to operation 305 to iteratively restart the process.Generally, operation 325 may be considered the beginning of the PPDU scheduling functions of process 300. The disclosed PPDU scheduling techniques may be configured to apply a particular scheme based on the direction of traffic (e.g., whether sent or received by the AP) and the type of PPDU used for the traffic (whether it is a single user (SU) or multi-user (MU)). By adapting the scheme specifically for the traffic to be transmitted, the PPDU scheduling aspects of the process 300 may be used to optimize performance and ensure optimal use of the entire width of the channel. A particular scheme may be applied for: 1) downlink traffic, for PPDUs transmitted by the AP; and 2) uplink traffic, for PPDUs received by the AP. Thus, at operation 325, a check is made to determine whether the AP is transmitting frames or is receiving frames to distinguish between DL traffic and UL traffic. The UL traffic may be either SU or trigger-based (TB). The TB PPDUs may use either UL MUMIMO or UL MU-OFDMA, or both together.If operation 325 determines that the AP receives the frames, then the traffic is considered UL traffic. For UL traffic, the process 300 continues to operation 335 to begin with the PPDU scheduling scheme adapted for UL traffic (the right branch of the flowchart). If operation 325 determines that the AP is transmitting frames, the traffic is considered DL traffic and process 300 continues with operation 330 to apply the PPDU scheduling scheme for the DL traffic (the left branch of the flowchart).In the case where DL traffic is determined in operation 325 (shown as TX (DL)), the process 300 may then proceed to operation 326. In operation 326, a check is made to see if the transmission, including DL traffic, is full bandwidth or sub-bandwidth. If the transmission is full bandwidth, the process 300 continues to operation 330. A full bandwidth transmission is, for example, a SU or a full bandwidth MU-MIMO. If during the test operation 326, it is determined that transmission is at a sub-bandwidth, the process 300 proceeds to operation 327. A sub-bandwidth transmission may include MU-OFDMA or MU-MIMO with sub-bandwidth.By determining that transmission is only partial bandwidth, the PPDU scheduling scheme can use the available bandwidth. In particular, at operation 327, the affected RUs may be assigned to clients already having sufficient SNR. If there are no clients with sufficient SNR, the transmission rate (TX) may be set to a lower value and / or the transmit power for the affected RUs may be increased in operation 327. Next, in operation 328, it is checked whether the performance has improved due to the actions in the previous operation 327. If it is determined in operation 328 that performance has been improved appropriately (shown as "Yes"), the process 300 may continue with operation 329 to end the process 300 or return to operation 305 to iteratively restart the process. However, if operation 328 determines that performance has not been improved (shown as "No") after completion of the previous operation 327, then the process 300 continues with operation 341.In operation 341, transmit beamforming (TX) is performed at least for the affected RU, such that the antenna weights are matched to filter attenuation. After the TX beamforming is performed in operation 341, another check is performed in operation 342 to determine whether performance has been improved as a result of the actions in the previous operation 341. If operation 342 determines that performance has been improved appropriately (shown as "Yes"), then the process 300 may proceed to operation 343 to end the process 300 or return to operation 305 to restart the process in an iterative manner. However, if operation 342 determines that performance has not been improved (shown as "no") after completion of the previous operation 341, then the process 300 continues with operation 344.Subsequently, at operation 344, RUs may be assigned to one or more clients whose SNR is high enough to reliably detect the preamble of the transmission. These assigned RUs originate from a subchannel containing the RUs involved. After the RUs have been allocated in operation 344, another check is made in operation 345 to determine if performance has improved. If it is determined in operation 345 that performance has been improved appropriately (shown as "Yes"), the process 300 may proceed to operation 346 to end the process 300 or return to operation 305 to iteratively restart the process. However, if operation 345 determines that performance has not been improved by previous operation 344 (shown as "NO"), then process 300 continues with operation 347.In operation 347, a check is made to see if the preamble puncturing is enabled. If preamble puncturing is enabled, this technique may be applied accordingly by the PPDU scheduling scheme in operation 348. The preamble puncturing technique allows the AP to avoid transmissions on one or more secondary subchannels and provides a mechanism that allows the receiver not to use the punctured subchannel for reception. Since the preamble puncturing applies only to secondary subchannels, operation 348 may also consist in the ZB first ensuring that its primary subchannel is not at the beginning or end of the channel, since the subchannels at the beginning / end of the channel are those containing the attenuated RUs, and changing the primary subchannel using standard techniques if it is the edge subchannel.Specifically, at operation 348, preamble puncturing is performed (e.g., by the AP) such that the subchannel to which the attenuated RUs belong is punctured. The preamble puncturing is carried out for MU PPDUs according to the 802.11ax standard by programming the field HE-SIG-A1 and specifying the RUs to be punctured by programming the sub-field RU assignment in the field HE-SIG-B. In MU-MIMO, the preamble puncturing can be implemented by programming the Split Subchannel bitmap subfield of the NDP Advertisement frame so that the clients only provide feedback for the non-punctured subchannels. By using preamble puncturing, transmissions on the attenuated subchannels are avoided, thereby reducing bandwidth and transmitting DL traffic in a manner that compensates for attenuation.If the preamble puncturing is not enabled, the process 300 may not use the preamble puncturing in transmission. Therefore, the process 300 does not perform preamble puncturing and instead proceeds to operation 390 where another PPDU scheduling technique is applied. Operation 390 is discussed in more detail below.Returning to operation 326, if the transmission is determined to utilize full bandwidth, process 300 loops through another branch of the flow beginning at operation 330. For DL traffic, the applied PPDU scheduling scheme may mitigate attenuation by increasing the signal-to-noise ratio (SNR) and / or reducing the physical transmission rate of the traffic, which would enable the use of a more robust modulation scheme. As shown, operation 330 includes adjusting the transmission rate to a lower value and / or increasing the transmission power to increase the SNR for all RUs or subcarriers. In other words, at operation 330, the transmit power for the PPDU is increased to increase the net SNR, which may result in better decoding of the symbols. Alternatively or additionally to increasing transmit power, operation 330 may include decreasing the modulation and coding rate (MCS) used for transmission by the AP. By reducing the transmission rate, a more robust modulation scheme can ensure that subframe errors (which would otherwise be caused by attenuation) are avoided.Increasing the transmit power within the required limits may allow the system to avoid reducing the MCS rate to a substantially lower value and thus potentially achieve optimum performance. The increased SNR and / or the reduced transmission rate may be applied for both HE SU PPDUs and HEMOU PPDUs (OFDMA and MIMO). In the case of MU OFDMA, the techniques in operation 330 may be limited only to the edge EUs (or subcarriers) being attenuated. In DL OFDMA transmissions, the attenuated RUs may be assigned to clients known to have a higher SNR. For example, lowering 40 dB SNR to 34 dB SNR may still result in an adequate MCS for successful transmission of the data traffic. Therefore, the selected client may use this attenuated RU based on the SNR and the number of bytes. For example, the client may have a short packet of 100 bytes and 40 dB SNR, such that using a lower MCS would not substantially degrade its performance.Thereafter, the process 300 may proceed to operation 340, which checks whether performance has been improved by increasing the SNR and / or decreasing the transmission rate in the previous operation 330. If it is determined in operation 340 that the performance has been improved (shown as "yes"), this may serve as an indication that the attenuation has been successfully corrected. As a result, the process 300 may proceed to operation 331 to end the process 300 or return to operation 305 to iteratively restart the process.Otherwise, if operation 340 determines that performance has not been improved (shown as "no"), the process 300 may apply a different PPDU scheduling technique to the DL traffic, namely transmit beamforming (TXBF), in operation 360. In operation 360, transmit beamforming is applied to the antenna weights to adapt to filter attenuation. The filter attenuation can be considered similar to frequency selective fading (with the difference that the attenuation occurs on all transmit-receive paths (Tx-Rx) and we do not get diversity gain at the receiver to alleviate it). Thus, transmit beamforming may be applied to all transmissions (including SU) from the AP, as the probing process of transmit beamforming provides feedback from the client that involves the effect of attenuation for the Tx-Rx paths. The antenna weights obtained as a result of the beamforming process adapt to the filter attenuation, thus improving the reception of the symbols on the attenuated RUs. This technique can be applied to both HE SU PPDUs and full or partial bandwidth HE MU MIMO PPDUs. Transmit beamforming may also be applied in cases where both OFDMA and MIMO are used in combination.Next, the process 300 may proceed to operation 365, which checks whether performance has been improved by transmit beamforming in the previous operation 360. If it is determined in operation 365 that the performance has been improved (shown as "yes"), this may serve as an indication that the attenuation has been successfully corrected. As a result, the process 300 may proceed to operation 361 to end the process 300 or return to operation 305 to iteratively restart the process.Alternatively, if operation 365 determines that performance has not been improved (shown as "no"), process 300 may apply another PPDU scheduling technique to the DL traffic. As disclosed, the PPDU scheduling techniques may be further tuned to apply a particular scheme to SU or MU frames in DL traffic. Accordingly, in operation 385, it is determined whether the transmission is SU or MU (e.g., OFDMA or MIMO). Based on this determination, process 300 may use either pre-amble punching (at MURframes) or a decrease in bandwidth for transmission (at SUframes).If the DL traffic is detected as SU (shown as "SU") at operation 385, the process 300 proceeds to operation 390. In operation 390, the bandwidth for the transmission of the SU frames in the DL traffic is reduced. Operation 390 may include dynamically reducing the bandwidth of the PPDU to a width less than the client's channel width. According to some embodiments, the bandwidth may be reduced to the primary half of the channel width.On the other hand, if MU frames are detected in operation 385 (shown as "MU"), the process 300 proceeds to operation 395. In operation 395, a check is made to determine whether the preamble puncturing is actually enabled before applying this technique to the DL traffic. For MU transmissions from the AP on the DL, preamble puncturing, which is a feature of the 802.11ax standard, may be used. The determination at operation 395 may be based on examining the settings of a Forced Sounding Support bit in the HE MAC capabilities of the client and / or the Forced Preamble RX field of the HE PHY capabilities of the client. If the preamble puncturing is not enabled (represented as "no"), as determined by operation 395, then the preamble puncturing cannot be used. As a result, process 300 may proceed to operation 390, which dynamically reduces the bandwidth of the PPDU to a width less than the client's channel width (similar to bandwidth reduction at SU frames). The process 300 may then proceed to operation 391 to end the process 300 or return to operation 305 to iteratively restart the process.Subsequently, after the bandwidth reduction for the DL traffic is performed as the last PPDU scheduling technique, the process 300 may be ended with operation 397.Referring to the case that UL traffic is determined in operation 325, the process 300 may then proceed to operation 335. The UL traffic may be either not trigger based, such as SU, or trigger based (TB) MU. In most cases, HE TB PPDUs may use either UL MU MIMO or UL MU OFDMA, or both in combination. As already indicated, the disclosed PPDU scheduling techniques may be tuned such that a particular scheme may be applied to trigger-based (e.g., MU) frames and another scheme may be applied to non-trigger-based (e.g., SU frames), thereby optimizing the PPDU scheduling scheme for the type of traffic being transmitted. Thus, operation 335 is a check that determines whether the UL traffic is non-trigger-based (e.g., SU) frames or trigger-based (e.g., MU) frames. If operation 335 determines that the UL traffic is not trigger-based frames, process 300 may limit bandwidth as a PPDU scheduling scheme. Alternatively, process 300 may increase SNR and / or reduce transmission rate when operation 335 determines trigger-based frames in UL traffic.If operation 335 detects trigger-based frames in UL traffic (shown as "TB"), process 300 proceeds to operation 350, where the SNR is increased and / or the transmission rate is reduced. As seen, operation 350 may include assigning the attenuated RU to clients with higher SNR. In the case of trigger frames, operation 350 may include programming a higher value in the UL target RSSI sub-field and / or a lower rate in the UL HE MCS sub-field for the higher SNR clients. The increase in SNR and / or the decrease in transmission rate for TB MU PPDUs may be achieved in operation 330 by programming a higher value for the UL target RSSI subfield in the basic trigger frame. Alternatively or additionally, operation 350 may also include programming a lower rate in the UL HE MCS subfield of the basic trigger frame. The aim is to achieve the same final effect of SNR and rate as explained above for the case of DL traffic.Next, the process 300 may proceed to operation 375, which checks whether performance has been improved by increasing SNR and / or decreasing transmission rate in the previous operation 350. If it is determined in operation 375 that the performance has been improved (shown as "yes"), this may serve as an indication that the attenuation has been successfully corrected. As a result, the process 300 may proceed to operation 376 to end the process 300 or return to operation 305 to iteratively restart the process.Otherwise, if it is determined in operation 375 that the performance has not been improved (shown as "no"), this may be an indication that the PPDU scheduling scheme applied in operation 350 has not compensated for attenuation and the performance continues to be negatively affected. Thus, the process 300 continues to operation 370 to apply another PPDU scheduling scheme to the UL traffic to remedy the attenuation.At operation 377, clients are assigned RUs that have a sufficiently high SNR to reliably detect the preamble of the transmission. These RUs are allocated in particular from the sub-channel containing the RUs concerned.Again, in operation 378, the process 300 checks whether performance has been improved. If it is determined in operation 378 that the performance has been improved (shown as "yes"), this may serve as an indication that the attenuation has been successfully corrected by the actions of operation 377. As a result, the process 300 may proceed to operation 379 to end the process 300 or return to operation 305 to iteratively restart the process.Otherwise, if it is determined in operation 378 that the performance has not been improved (shown as "no"), this may be an indication that the PPDU scheduling scheme applied in operation 350 has not compensated for attenuation and the performance continues to be negatively affected. Thus, the process 300 continues with operation 380 to apply another PPDU scheduling scheme to the UL traffic to remedy the attenuation, namely the bandwidth reduction.Thereafter, at operation 380, the process 300 may avoid using the attenuated RUs by blocking or otherwise dropping them completely. For example, the attenuated RUs may be assigned a reserved value of A1D12in the target frame (corresponding to the attenuated RUs), or the UL bandwidth subfield may be programmed to use less bandwidth than the channel width, both effectively dropping the attenuated RUs. In other words, if programming the rate and UL RSSI of the TB MU frames in UL traffic does not contribute to reducing the PER and improving performance (operation 350), operation 380 may employ a more brachial approach that limits the use of the attenuated RUs. In some embodiments, programming the AID16 subfield includes programming the subfield for the RU in the base trigger to a reserved value (e.g., any value between 2008-2044 or 2047-4094) that does not correspond to the association ID (AID) of one of the connected clients. Alternatively, operation 380 may be to restrict the bandwidth for the MU PPDU in UL traffic to a smaller width than the client's channel width by programming a lower value into the UL BW subfield of the basic trigger frame. The restriction of the bandwidth can be considered a brute force method, since this approach reduces the overall bandwidth of the UL MU PPDU, but is advantageous if the reservation of the RUs by programming reserved AID16 values is not possible for reasons of interoperability. The process 300 may then proceed from operation 380 to operation 397, thereby completing the process 300.Returning now to the case where non-trigger-based PPDUs in UL traffic were determined in operation 335, process 300 proceeds to operation 345. In operation 345, an impact of the UL rate adjustment from the clients may be evaluated. For example, for HE SU PPDUs, the AP monitors the subframe packet error rate (PER) and the rates used by the client side rate matching algorithm to determine if it selects an appropriate transmission rate and / or transmit power to reduce symbol errors due to attenuated RUs.Next, the process 300 may proceed to operation 355, which checks whether performance has been improved based on the evaluated UL rate adjustment in the previous operation 345. If it is determined in operation 355 that the performance has been improved (shown as "yes"), this may serve as an indication that the attenuation has been successfully corrected. As a result, the process 300 may proceed to operation 346 to end the process 300 or return to operation 305 to iteratively restart the process.Otherwise, if operation 355 determines that the performance has not been improved (shown as "no"), this may be an indication that the attenuation has not been properly compensated for and remains a negative impact on the performance. Thus, the process 300 continues to operation 370 to apply another PPDU scheduling scheme to the UL traffic to remedy the attenuation.Subsequently, at operation 370, an operating mode message may be used to reduce the operating channel width to half, such that the attenuated RUs are avoided. For example, the SU PPDUs of the UL traffic can be limited to a bandwidth that is less than the channel width by transmitting the Operating Mode Notification (OMN) action frame that limits the bandwidth for all PPDUs from this time on. As a result, the RUs are discarded outside the matched bandwidth and assuming that the RUs involved are in these edge bandwidths, thereby mitigating the negative effects of attenuation. Subsequently, the process 300 may be ended with operation 397.It will be appreciated that the aforementioned PPDU scheduling techniques as described in FIG. 3 are either applicable on a per-PPDU basis or can be reversed at any time without the clients losing connection or experiencing traffic interruptions.In some embodiments, not shown in FIG. 3, a PPDU scheduling technique may include using edge RUs only for traffic belonging to access categories (ACs) that may tolerate higher error rates and thus are able to tolerate the degradation associated with attenuated RUs. For example, edge RUs can be assigned only to clients that have traffic on lower QoS ACs, such as best effort (BE) and background (BK) ACs. In addition, the edge RUs can be avoided for clients that transmit traffic from ACs with higher QoS, such as voice (VO) and / or video (VI) ACs. In some embodiments, these PPDU scheduling techniques may be dynamically turned on or off based on the frequency shift, and may be terminated when it is determined that the frequency shift caused by temperature sensitivity is no longer determined during operation of the filters (e.g., when the filters again reach their nominal operating temperature).Although the DPD and PPDU scheduling techniques described above are applicable to all channel bandwidths to compensate for filter performance, the effects on 160 MHz channels are significant from the point of view of Wi-Fi-6E. As already indicated, 160 MHz channels (as compared to lower bandwidths) are transmitted at high transmit power and can potentially assign broader tone RUs to clients with high throughput requirements. In some cases, the TCF characteristic of filters may shift the filter response in a range of 4 MHz-6 MHz at extreme temperatures. This could severely limit the operation of 160 MHz channels and thus force lower throughput performance by using lower bandwidth channels. Such performance problems would be more apparent with the new Wi-Fi standards, such as the 802.11be standard, which allows 320 MHz bandwidths. Thus, the advantages realized by the techniques and systems disclosed herein will alleviate these problems and have even greater utility in such Wi-Fi systems while providing flexible channel allocation and density.Referring now to FIG. 4, a process 400 is illustrated that implements DPD and PPDU planning to address the shift of the stopband in particular. Generally, the process 400 attempts to limit or eliminate OOB emissions when the AP is transmitting. However, when the AP receives, process 400 ensures that reception is not obstructed by transmission in the adjacent channels. Therefore, process 400 ultimately checks whether the OOB transmissions on the transmit branch (TX) are sufficiently reduced (e.g., operations 453- 475), as opposed to checking whether performance on the receive branch (RX) has been improved (e.g., operations 430- 496).The process 400 begins with operation 405. Next, in operation 410, the PPDU scheduling process 400 may determine whether the filter frequency response has shifted due to temperature changes. The detection of a temperature-related frequency response shift (shown as "yes") may serve as an indication that some RUs are experiencing some attenuation, typically the RUs at the edges of the channel width. Thereafter, the PPDU scheduling process 400 may proceed to operation 415 to perform DPD techniques to attenuate the OOB emissions and thereby remedy the shift of the stopband. The DPD techniques may be implemented via the radios (including frontends) described in detail above with reference to FIG. 1.Alternatively, process 400 may proceed to operation 411 and end process 400 or return to operation 405 to iteratively restart the process if operation 410 determines that there is no shift in the response of the filter (shown as "No").Continuing with operation 420, process 400 may perform a test to determine whether the OOB RUs are sufficiently attenuated after DPD techniques are employed. If the OOB RUs were not adequately attenuated by the DPD techniques, there is still the possibility of problems such as desensitizing in a co-operating adjacent channel. Thus, if it is determined that the OOB RUs are not attenuated sufficiently to compensate for the shift in the stopband, the process 400 may proceed with PPDU scheduling as a secondary approach (or fail-safe) to correct for the shift in filter response that was not fully corrected by the DPD techniques.Otherwise, if operation 410 determines that the OOB RUs have been properly attenuated, this determination indicates that the DPD techniques of the previous operation 415 successfully compensated for the frequency response shift caused by temperature sensitivity. Thus, process 400 may proceed to operation 421 and end process 400 or return to operation 405 to iteratively restart the process.Generally, operation 425 may be considered the beginning of the PPDU scheduling functions of process 400. The disclosed PPDU scheduling techniques may be configured to apply a particular scheme based on the direction of traffic (e.g., whether sent or received by the AP). By adapting the scheme specifically for the traffic being transmitted, the PPDU scheduling aspects of the process 400 may be used for power optimization. A particular scheme may be applied for: 1) downlink traffic, for PPDUs transmitted by the AP; and 2) uplink traffic, for PPDUs received by the AP. As already indicated, the process 400 generally aims to determine whether the OOB transmissions are sufficiently reduced during the downlink (e.g., operations 453- 475), as opposed to checking whether performance is improved during the uplink (e.g., operations 430- 496). Thus, in operation 425, a check is made as to whether the AP is transmitting or receiving frames to distinguish between DL traffic and UL traffic. The UL traffic may be either SU or trigger-based (TB). The TB PPDUs may use either UL MU MIMO or UL MU OFDMA, or both together.If operation 425 determines that the AP receives the frames, then the traffic is considered UL traffic. For UL traffic, process 400 continues with operation 430 to begin the PPDU scheduling scheme adapted for UL traffic (the right branch of the flowchart). If operation 425 determines that the AP is transmitting frames, the traffic is considered DL traffic and process 400 continues with operation 435 to apply the PPDU scheduling scheme for the DL traffic (the left branch of the flowchart).In operation 435, a check is made to see if the transmission, including DL traffic, is full bandwidth or sub-bandwidth. If the transmission is full bandwidth, the process 400 proceeds to operation 440. A full bandwidth transmission is, for example, a SU or a full bandwidth MU-MIMO. If during test operation 435 it is determined that transmission is at a sub-bandwidth, process 400 proceeds to operation 451. A sub-bandwidth transmission may include MU-OFDMA or MU-MIMO with sub-bandwidth.If it is determined in operation 435 that the transmission is full bandwidth, the process 400 proceeds to operation 440. At operation 440, transmit power (TX) for the edge subcarrier / subcarriers may be decreased. In some cases, operation 440 may include a decrease in transmit power for the entire channel width. After lowering the transmit power in operation 440, process 400 proceeds to operation 445 to verify the effectiveness of this approach.Next, in operation 445, a check is made as to whether the OOB emissions have been sufficiently attenuated by the reduction in transmit power in the previous operation 440. If the OOB emissions are deemed not to be sufficiently attenuated (shown as "No") in operation 445, then there is not sufficient compensation for the shift in the stopband. Thus, in operation 450, the process 400 may perform another PPDU scheduling technique to attempt to compensate for the shift of the stopband. If it is determined in operation 445 that the OOB RUs have been properly attenuated (shown as "yes"), then this determination indicates that the measures of the previous operation 440 have successfully compensated for the shift of the stopband. Thus, process 400 may proceed to operation 446 and end process 400 or return to operation 405 to iteratively restart the process.At operation 450, the bandwidth for transmission is reduced to an unaffected portion of the channel width. At operation 450, it may be ensured that the primary subchannel is included in the unaffected portion of the channel. Subsequently, process 400 may proceed from operation 450 to operation 497, terminating process 400.Returning to operation 435, if the test reveals that transmission is at a sub-bandwidth, process 400 continues to operation 451. In operation 451, the transmit power is decreased, but only for the edge RUs.Thereafter, in operation 455, it is checked whether the OOB emissions have been sufficiently attenuated by the reduction of the transmission power in the previous operation 451. If the OOB emissions are deemed not to be sufficiently attenuated (shown as "No") at operation 455, then there is not sufficient compensation for the shift in the stopband. Thus, in operation 460, the process 400 may perform another PPDU scheduling technique to attempt to compensate for the shift of the stopband. If it is determined in operation 455 that the OOB RUs have been properly attenuated, this determination means that the previous operation 451 actions have successfully compensated for the shift of the stopband. Thus, process 400 may proceed to operation 456 and end process 400 or return to operation 405 to iteratively restart the process.Subsequently, at operation 460, clients may be assigned RUs whose SNR is high enough to reliably detect the preamble of the transmission even if the transmit power is further reduced. These assigned RUs originate from a subchannel containing the RUs involved. After the RUs are assigned in operation 360, another check is made in operation 465 to determine if the OOB emissions have been sufficiently attenuated. If operation 365 determines that the OOB emissions have been properly attenuated (shown as "Yes"), the process 400 may continue with operation 446 to end the process 400 or return to operation 405 to iteratively restart the process. However, if it is determined at operation 445 that the OOB emissions have not been properly attenuated (represented as "NO") by the previous operation 460, the process 400 proceeds to operation 470.In operation 470, a check is made to see if the preamble puncturing is enabled. If preamble puncturing is enabled, this technique may be applied accordingly by the PPDU scheduling scheme in operation 475. The preamble puncturing technique allows the AP to avoid transmissions on one or more secondary subchannels and provides a mechanism that allows the receiver not to use the punctured subchannel for reception. Since preamble puncturing applies only to secondary subchannels, at operation 475, the AP may first ensure that its primary subchannel is not at the beginning or end of the channel, as the subchannels at the beginning / end of the channel are those containing the attenuated RUs and alter the primary subchannel using standard techniques if it is the edge subchannel. After the preamble puncturing is performed by operation 475, the process 400 may continue with operation 497, thereby terminating the process 400.If the preamble puncturing is not enabled, the process 400 may not use the preamble puncturing in transmission. Therefore, the process 400 does not perform preamble puncturing and instead proceeds to operation 450, where another PPDU scheduling technique (i.e., reducing bandwidth to the unaffected portion of the channel) is applied.Returning to the case that UL traffic is determined in operation 425, process 400 may then proceed to operation 430. The UL traffic may be either non-trigger based (non-TB), such as SU, or trigger based (TB) MU. As already indicated, the disclosed PPDU scheduling techniques may be tuned such that a particular scheme is applied to trigger-based (e.g., MU) frames and another scheme may be applied to non-trigger-based (e.g., SU) frames, thereby optimizing the PPDU scheduling scheme for the type of traffic being transmitted. Thus, operation 430 is a check that determines whether the UL traffic is non-trigger-based (e.g., SU) frames or trigger-based (e.g., MU) frames. In general, if operation 430 does not determine trigger-based frames of UL traffic, process 400 may limit bandwidth as a PPDU scheduling scheme. Alternatively, process 400 may increase SNR and / or reduce transmission rate when operation 430 determines trigger-based frames in UL traffic.If operation 430 detects trigger-based frames in UL traffic (shown as "TB"), process 400 proceeds to operation 436, where the SNR is increased and / or the transmission rate is reduced. As seen, operation 436 may include assigning the attenuated RU to clients with higher SNR. In the case of trigger frames, operation 436 may include programming a higher value in the UL target RSSI sub-field and / or a lower rate in the UL HE MCS sub-field for the higher SNR clients. The increase in SNR and / or the decrease in transmission rate for TB MU PPDUs may be achieved in operation 436 by programming a higher value for the UL target RSSI subfield in the basic trigger frame. Alternatively or additionally, operation 436 may also include programming a lower rate in the UL HE MCS subfield of the basic trigger frame.Next, the process 400 may proceed to operation 491, which checks whether performance has been improved by increasing SNR and / or decreasing transmission rate in the previous operation 436. If it is determined in operation 491 that the performance has been improved (shown as "yes"), this may serve as an indication that the attenuation has been successfully corrected. As a result, the process 400 may proceed to operation 492 to end the process 400, or return to operation 405 to restart the process in an iterative manner.Otherwise, if it is determined in operation 491 that the performance has not been improved (shown as "no"), this may be an indication that the PPDU scheduling scheme applied in operation 436 has not compensated for the shift of the stopband and the performance continues to be negatively affected. Thus, process 400 proceeds to operation 493 to apply another PPDU scheduling scheme to the UL traffic to remedy the attenuation.In operation 493, the clients are assigned RUs whose SNR is high enough to reliably detect the preamble of the transmission even with a further reduction in the transmission power (TX). These RUs are allocated in particular from the sub-channel containing the RUs concerned.Again, in operation 494, the process 400 checks whether performance has been improved. If it is determined in operation 494 that performance has been improved (shown as "yes"), this may serve as an indication that the shift in the stopband has been successfully corrected by the actions in operation 493. As a result, the process 400 may proceed to operation 495 to end the process 400, or return to operation 405 to iteratively restart the process.Otherwise, if operation 400 determines that performance has not been improved (shown as "no"), this may be an indication that the PPDU scheduling scheme applied in operation 493 has not compensated for the shift in the stopband and the performance continues to be negatively affected. Thus, the process 400 continues to operation 496 to apply another PPDU scheduling scheme to the UL traffic to remedy the attenuation, namely the bandwidth reduction.Thereafter, at operation 496, the process 400 may avoid using the attenuated RUs by blocking or otherwise dropping them completely. For example, the attenuated RUs may be assigned a reserved value of A1D12in the target frame (corresponding to the attenuated RUs), or the UL bandwidth sub-field may be programmed to use less bandwidth than the channel width, both effectively dropping the attenuated RUs. Process 400 may then proceed from operation 496 to operation 497, thereby completing process 400.Returning now to the case where non-trigger-based PPDUs in UL traffic are determined in operation 430, process 400 proceeds to operation 480. In operation 380, an impact of the UL rate adjustment may be evaluated by the clients. For example, for HE SU PPDUs, the AP monitors the subframe packet error rate (PER) and the rates used by the client side rate matching algorithm to determine if it selects an appropriate transmission rate and / or transmit power to reduce symbol errors due to attenuated RUs.Next, the process 400 may proceed to operation 485, which checks whether performance was improved based on the evaluated UL rate adjustment in the previous operation 480. If it is determined in operation 485 that the performance has been improved (shown as "yes"), this may serve as an indication that the attenuation has been successfully corrected. As a result, the process 400 may proceed to operation 486 to end the process 400 or return to operation 405 to iteratively restart the process.Otherwise, if it is determined in operation 485 that the performance has not been improved (shown as "no"), this may be an indication that the stopband shift has not been properly compensated for and still has a negative impact on the performance. Thus, the process 400 continues with operation 490 to apply another PPDU scheduling scheme to the UL traffic to remedy the attenuation.Subsequently, at operation 490, an operating mode message may be used to reduce the operating channel width so that the attenuated RUs are avoided. For example, the SU PPDUs of the UL traffic may be limited to a bandwidth that is less than the channel width by transmitting the Operating Mode Notification (OMN) action frame that limits the bandwidth for all PPDUs from this time on. As a result, the RUs outside the set bandwidth are discarded. Subsequently, the process 400 may end with operation 497.FIG. 5 shows a block diagram of an example computer system 500 in which the embodiments described herein may be implemented. The computer system 500 includes a bus 502 or other communication mechanism for communicating information, and one or more hardware processors 504 coupled to the bus 502 for processing information. The hardware processor(s) 504 may be, for example, one or more general purpose microprocessors.The computer system 500 also includes a main memory 506, such as random access memory (RAM), a cache, and / or other dynamic storage devices, coupled to the bus 502 for storing information and instructions to be executed by the processor 504. Main memory 506 may also be used to store temporary variables or other intermediate information during execution of instructions to be executed by processor 504. When such instructions are stored in storage media accessible by processor 504, computer system 500 becomes a special purpose machine adapted to perform the operations specified in the instructions.Computer system 500 also includes read only memory (ROM) 508 or other static storage device coupled to bus 502 for storing static information and instructions for processor 504. A storage device 510, e.g., a magnetic disk, optical disk, or USB stick (flash drive), etc., is provided and connected to bus 502 to store information and instructions.The computer system 500 may be connected via the bus 502 to a display 512, e.g., a liquid crystal display (LCD) (or touch screen), to display information to a computer user. An input device 514, including alphanumeric and other keys, is coupled to bus 502 for communicating information and command selections to processor 504. Another type of user input device is cursor control 516, such as a mouse, trackball, or cursor direction keys for communicating direction information and command selections to processor 504 and for controlling cursor movement on display 512. In some embodiments, the same directional information and command selections as cursor control may be implemented via receiving touches on a touch screen without cursor.Computer system 500 may include a user interface module for implementing a graphical user interface, which may be stored in a mass storage device as executable software code executed by the computing device(s). This and other modules may include, for example, components such as software components, object oriented software components, class components and task components, processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays, and variables.In general, the word "component", "engine", "system", "database", "data storage", and the like, as used herein, may refer to logic embodied in hardware or firmware, or to a collection of software instructions that may have entry and exit points and are written in a programming language such as Java, C, or C++. A software component may be compiled and linked into an executable program, installed in a dynamic link library, or written in an interpreted programming language such as BASIC, Perl, or Python. It will be appreciated that software components may be callable from other components or may be callable by themselves and / or may be invoked in response to detected events or interrupts. Software components configured for execution on computing devices may be provided on a computer readable medium, such as a compact disc, digital video disc, flash drive, magnetic disk, or other tangible medium, or as a digital download (and may be originally stored in a compressed or installable format that requires installation, decompression, or decryption prior to execution). Such software code may be partially or completely stored in a memory of the executing computing device for execution by the computing device. Software instructions may be embedded in firmware such as an EPROM. Moreover, the hardware components may consist of connected logic units such as gates and flip-flops and / or programmable units such as programmable gate arrays or processors.The computer system 500 may implement the techniques described herein using custom hard-wired logic, one or more ASICs or FPGAs, firmware, and / or program logic that, in combination with the computer system, makes or programs the computer system 500 a special-purpose machine. According to one embodiment, the techniques described herein are performed by computer system 500 in response to processor(s) 504, executing / executing one or more sequences of one or more instructions contained in main memory 506. Such instructions may be read into main memory 506 from another storage medium, such as storage device 510. Execution of the sequences of instructions contained in main memory 506 causes processor(s) 504 to perform the process steps described herein. In alternative embodiments, hardwired circuitry may be used in place of or in combination with software instructions.The term "non-transitory media" and similar terms as used herein refer to any media that stores data and / or instructions that cause operation of a machine in a particular manner. Such non-volatile media may include non-volatile media and / or volatile media. The non-volatile media includes, for example, optical or magnetic hard disks, such as storage device 510. The volatile media includes dynamic memory, such as main memory 506. Common forms of non-volatile media include, for example, floppy disks, flexible disks, hard disks, solid state drives, magnetic tapes or other magnetic data storage media, CD-ROMs, other optical data storage media, physical media with patterns of holes, RAM, PROM and EPROM, FLASH-EPROM, NVRAM, other memory chips or cartridges, and their networked versions.Non-transitory media are different from transmission media but may be used in conjunction with them. Transmission media participates in the transmission of information between non-transitory media. Transmission media includes, for example, coaxial cables, copper and fiber optic cables, including the wires making up bus 502. Transmission media can also occur in the form of sound or light waves, as are generated during data communication via radio and infrared.Computer system 500 also includes a communication interface 518 connected to bus 502. Network interface 518 establishes a two-way data communication link to one or more network links that are connected to one or more local area networks. The communication interface 518 may be, for example, an integrated services digital network (ISDN) card, a cable modem, a satellite modem, or a modem to establish a data communication connection to a corresponding type of telephone line. As another example, network interface 518 may be a local area network (LAN) card to establish a data communication link to a compatible LAN (or WAN component for communication with a WAN). Wireless connections may also be implemented. In each of these implementations, the network interface 518 sends and receives electrical, electromagnetic, or optical signals that transmit digital data streams having different types of information.A network connection typically allows data communication over one or more networks to other data devices. For example, a network connection may connect via a local area network to a host computer or to data devices operated by an Internet Service Provider (ISP). The ISP, in turn, provides data communication services over the world wide packet data communication network, commonly referred to today as the "Internet.". Both the local area network and the Internet use electrical, electromagnetic or optical signals that transmit digital data streams. The signals over the various networks and the signals on the network connection and over the communication interface 518 that transmit the digital data to and from the computer system 500 are examples of transmission media.The computer system 500 may send messages and receive data including program code via the network(s), network connection, and communication interface 518. In the Internet example, a server could transmit requested code for an application program over the Internet, the ISP, the local area network, and the communication interface 518.The received code may be executed by the processor 504 upon receipt thereof and / or stored in the storage device 510 or other non-volatile memory for later execution.Each of the processes, methods, and algorithms described in the preceding paragraphs may be embodied in, and fully or partially automated by, code components executed by one or more computer systems or computer processors having computer hardware. The one or more computer systems or computer processors may also operate to support execution of the respective operations in a cloud computing environment or as a software as a service (SaaS). The processes and algorithms can be partially or fully implemented in application specific circuitry. The various features and methods described above may be used independently or combined in various ways. Various combinations and sub-combinations are intended to fall within the scope of this disclosure, and certain method or process blocks may be omitted in some implementations. The methods and processes described herein are also not limited to any particular order, and the blocks or states associated therewith may be performed in other suitable orders, in parallel, or in other ways. Blocks or states may be added to or removed from the disclosed examples. The execution of certain operations or processes may be distributed among computer systems or computer processors that are not only located in a single machine, but are distributed across a number of machines.As used herein, circuitry may be implemented in any form of hardware, software, or a combination thereof. For example, one or more processors, controllers, ASICs, PLAs, PALs, CPLDs, FPGAs, logic components, software routines, or other mechanisms may be implemented to form a circuit. In implementation, the various circuits described herein may be implemented as discrete circuits, or the described functions and features may be partially or totally shared among one or more circuits. Although various features or functional elements are individually described or claimed as separate circuits, these features and functions may be shared among one or more common circuits, and such description is not intended to imply or imply that separate circuits are required to implement these features or functions. When a circuit is implemented in whole or in part with software, this software may be implemented to operate on a computer or processing system capable of executing the functionality described with respect to it, such as computer system 500.As used herein, the term "or" may be understood in both the inclusive and exclusive sense. Moreover, the description of resources, acts, or structures in the singular is not to be understood as excluding the plural. Conditional terms such as "may", "could", "could" or "may", unless expressly stated otherwise or otherwise understood in the context, are generally to be understood such that certain embodiments include certain features, elements and / or steps, while other embodiments do not include these.The terms and expressions and their modifications used in this document are not to be understood as limiting, but rather as open-ended, unless expressly stated otherwise. Adjectives such as "conventional", "traditional", "normal", "standard", "known", and terms of similar meaning are not to be understood as limiting the described subject matter to a particular time period or to an available subject matter at a particular time, but should be understood as including conventional, traditional, normal, or standard technologies, which may be available or known now or at any time in the future. The presence of extending words and formulations such as "one or more", "at least", "but not limited to", or similar formulations in some instances is not to be understood as the narrower case is intended or required when such extending formulations are not present.

Claims

A non-transitory computer readable storage medium (120; 506) having stored thereon executable computer program instructions that, when executed by one or more processors (504), cause the one or more processors (504) to perform operations comprising: determining whether a frequency response of a filter (143) has shifted, wherein the shift in frequency response comprises a shift in a stop band (220) of the filter (143), and the filter (143) is a component of a network device (100) for communicating data frames in a wireless network; responsive to determining that the frequency response of the filter (143) has shifted, applying a digital pre-distortion to a signal input to the filter (143) to compensate for the shift in frequency response; determining whether the digital predistortion has compensated for the shift in frequency response of the filter (143); and responsive to determining that the digital predistortion has not compensated for the shift in frequency response of the filter (143), applying at least one PHY protocol data unit (PPDU) scheduling scheme to the data frames transmitted from the network device (100).The non-transitory computer readable storage medium (120; 506) of claim 1, wherein the shift in frequency response is associated with higher levels of out-of-band emissions from edge resource units in response to the shift in the stop band (220) of the filter (143), and wherein the digital pre-distortion applies attenuation to edge resource units to compensate for the shift in a stop band (220) of the filter (143).The non-transitory computer readable storage medium (120; 506) of claim 2, programmed to perform further operations comprising: determining whether the data frames transmitted by the network device (100) are associated with uplink or downlink traffic; in response to determining that the data frames transmitted by the network device (100) are associated with the downlink traffic, determining whether transmission of the downlink traffic is partial bandwidth or full bandwidth; in response to determining that transmission of the downlink traffic is full bandwidth, decreasing a transmission power of at least one of the edge subcarriers or the entire channel width; in response to reducing the transmit power of the edge subcarriers, determining whether the edge resource units are attenuated to compensate for the shift of a stop band (220); and in response to determining that the edge resource units are not attenuated to compensate for the shift of a stop band (220), reducing the bandwidth for transmitting the downlink traffic to an unaffected portion of a channel width.The non-transitory computer readable storage medium (120; 506) of claim 3, programmed to perform further operations comprising: in response to determining that transmission of the downlink traffic is occurring at sub-bandwidth, decreasing transmission power of the edge resource units; in response to decreasing transmission power of the edge resource units, determining whether the edge resource units are attenuated to compensate for the shift of a stop band (220); and in response to determining that the edge resource units are not attenuated to compensate for the shift of a stop band (220), allocating the other resource units from a sub-channel including the edge resource units to clients, wherein the allocated resource units are determined to have a signal-to-noise ratio (SNR) such that a transmission power is reduced and a preamble remains recognizable from transmission of the downlink traffic.The non-transitory computer readable storage medium (120; 506) of claim 4, programmed to perform further operations comprising: responsive to allocating resource units from the subchannel containing the edge resource units to clients, determining whether the edge resource units are attenuated to compensate for the shift of a stopband (220); responsive to determining that the edge resource units are not attenuated to compensate for the shift of a stopband (220), determining whether preamble puncturing is enabled; and responsive to determining that the preamble puncturing is enabled, puncturing a subchannel containing the edge resource units.The non-transitory computer readable storage medium (120; 506) of claim 5, wherein puncturing the subchannel comprises: determining whether a primary subchannel is the subchannel to be punctured; and in response to determining that the primary subchannel is the subchannel to be punctured, activating a subchannel other than a new primary subchannel prior to puncturing.The non-transitory computer readable storage medium (120; 506) of claim 3, programmed to perform further operations comprising: responsive to determining that the data frames transmitted by the network device (100) are associated with the uplink traffic, determining whether the uplink traffic comprises trigger-based PPDUs or non-trigger-based PPDUs; and responsive to determining that the uplink traffic comprises trigger-based PPDUs, assigning the edge resource units to clients associated with higher signal-to-noise ratios (SNRs); in response to assigning the edge resource units to clients associated with higher SNRs, determining whether the edge resource units are attenuated to compensate for the shift of a stop band (220); and in response to determining that the edge resource units are not attenuated to compensate for the shift of a stop band (220), assigning the other resource units from a sub-channel containing the edge resource units to clients, wherein the assigned resource units are determined to have a signal-to-noise ratio (SNR) such that a reduced transmission power and a preamble remain recognizable from the transmission of the downlink traffic.The non-transitory computer readable storage medium (120; 506) of claim 7, programmed to perform further operations comprising: in response to assigning the edge resource units to clients associated with higher SNRs, determining whether the edge resource units are attenuated to compensate for the shift of the stop band (220); and in response to determining that the edge resource units are not attenuated to compensate for the shift of a stop band (220), reducing a bandwidth of the trigger-based PPDU to a width less than a channel width, or dropping edge resource units to compensate for the shift of the stop band (220).The non-transitory computer readable storage medium (120; 506) of claim 7, programmed to perform further operations comprising: responsive to determining that uplink traffic does not comprise trigger-based PPDUs, evaluating an impact of uplink rate adaptation of clients; responsive to evaluating the impact of uplink rate adaptation of clients, determining whether the edge resource units are attenuated to compensate for the shift of the stop band (220).The non-transitory computer readable storage medium (120; 506) of claim 9, programmed to perform further operations comprising: in response to determining that the edge resource units are not attenuated to compensate for the shift of the stop band (220), reducing a width of an operating channel.The non-transitory computer readable storage medium (120; 506) of claim 10, wherein reducing the width of the operating channel avoids a sub-channel containing the edge resource units.The non-transitory computer readable storage medium (120; 506) of claim 11, wherein reducing the width of the operating channel comprises an op-mode notification.A method comprising: determining whether a frequency response of a filter (143) has shifted, wherein the filter (143) is a component of a network device (100) for communicating data frames in a wireless network, and the shift in frequency response comprises a shift in a stop band (220) of the filter (143); responsive to determining that the frequency response of the filter (143) has shifted, applying a digital pre-distortion to a signal input to the filter (143) to compensate for the shift in frequency response; determining whether the digital pre-distortion has compensated for the shift in frequency response of the filter (143); and responsive to determining that the digital pre-distortion has not compensated for the shift in frequency response of the filter (143), applying at least one PHY protocol data unit (PPDU) scheduling scheme to the data frames transmitted by the network device (100).

Images