Study on the Influential Factors of Noise Characteristics in Dense-Graded Asphalt Mixtures and Field Asphalt Pavements

Determining the influential factors of noise characteristics in dense-graded asphalt mixtures and field asphalt pavement is important in constructing highways that are both low noise and environmentally friendly. In this study, the effects of nominal maximum aggregate size, asphalt binder type, air void percentage, and the service life of pavement on the noise absorption characteristics of asphalt mixtures were first investigated in laboratory. Thereafter, tire/pavement noise measurements were conducted on different types of dense-graded asphalt pavements. The effects of the service lives of the pavements, the types of the pavements, driving speeds, and test temperatures on the noise levels of the pavements were also studied. The Zwicker method is used to calculate psychoacoustic parameters on the tire/pavement noise spectrum. The laboratory results indicate that reducing the nominal maximum aggregate size, using rubber asphalt, and increasing air void percentage as well as surface texture depth improve the sound absorption performance of asphalt mixtures. The field measurements show that laying down asphalt pavements with a shorter service life or larger texture depth, using rubber asphalt, reducing traffic speed, and increasing air temperature can reduce noise.


Introduction
Traffic noise is one of the primary contributors to sound pollution in a city [1,2]. It has been widely recognized that traffic noise can introduce issues in long-term psychosocial health and well-being [3]. In general, traffic noise can come from four different sources: engine noise, exhaust noise, aerodynamic noise, and tire/pavement interaction noise. Among them, tire/pavement interaction is the dominant noise source for properly maintained vehicles at speeds above 50 km/h [4][5][6][7]. erefore, knowing how to reduce tire/pavement noise plays an important role in controlling the problems associated with highway noise.
Many past studies have focused on optimizing the design and manufacture of quieter automobiles and tires [8,9], effectively constructing sound barriers [10,11], planting noise absorbent tree belts on both sides of a highway [12], and paving low-noise or porous pavement [13,14]. Nevertheless, it must be noted that the reduction in the amplitude from tire/pavement noise caused by the improved design and manufacture of a car or its tire has begun exhibiting diminishing returns. On the other hand, sound barriers and noise absorbing tree belts are limited in their potential application due to the impact they have on the surrounding road space as well their engineering costs. As a result, low-noise pavement has been developed and applied in many sites to reduce tire/pavement noise [15].
Porous asphalt pavement was used mostly as the lownoise pavement of choice over the past twenty years [16]. It had been reported that the reduction on the sound pressure level of noise is 3∼5 dB for porous asphalt pavement [17]. However, porous asphalt pavement suffers from a shorter service life when compared with dense asphalt mixes, and this consideration has discouraged its widespread application [18]. In fact, most of the asphalt pavement used in worldside highways is made of dense asphalt mixes. erefore, how to control and reduce the tire/pavement noise from dense-graded asphalt pavement has become a central question with important strategic implications for developing quieter environment in the future. Regardless, few studies have been conducted to investigate the noise reduction characteristics of dense-graded asphalt pavement.
In this study, the influential factors of the noise reduction characteristics in laboratory compacted densegraded asphalt mixtures are investigated first. An impedance tube experiment is conducted to analyze the sound characteristics of the mixtures. e effects of the nominal maximum aggregate size (NMAS), types of asphalt binders, and air void percentages on the noise reduction performance are studied. Here NMAS is measured specifically as one sieve size larger than the first sieve to retain more than 10 percent of the material. In addition, cored cylinder samples from three field pavements are also obtained to determine the effect of service life on noise reduction performance.
ereafter, the influential factors of noise reduction characteristics are investigated further in dense-graded field asphalt pavements. e effects of the types of asphalt binder, driving speeds, and test temperatures are studied. During field testing, the sound pressure level was used to characterize noise level. In addition, many researchers have focused on the quality of sound as well as the sound pressure level of noises with the aim of reducing the harm to human life from noise through improving sound quality.
is novel aim provides a new research angle for evaluating tire-pavement noise. With this in mind, this study employs the Zwicker method to calculate and analyze tire-pavement sound quality parameters (e.g., loudness, sharpness, harshness, and fluctuation) in several field pavements [19,20]. e findings of this study contribute to our understanding of the noise characteristics of dense-graded asphalt pavement, promoting the research on low-noise pavement and the improvement of tire/pavement noise control technology.

Mix Design and Sample Preparation in Laboratory
Before investigating the factors that affect noise characteristics in dense-graded asphalt concrete mixtures, the mix design for the different kinds of asphalt concrete mixtures to be used in this study was established. e detailed test results can be found in Tables 1-3. e number AH90# represents the virgin asphalt binder with a penetration of 80∼100 (0.01 mm) following the ASTM D946/D946M-15 standard test methodology. Note that styrene-butadiene-styrene (SBS) and rubber are polymer materials used in the modification of asphalt binder. In total, nine types of dense-graded asphalt concrete mixtures, with five gradations and five types of asphalt binders, were selected. In the line of gradations, the first number shows the NMAS, while the second number denotes the percent of aggregates that pass through the 4.75 mm sieve and settle in the 0.075 mm sieve. For example, AC16-30 translates to the asphalt concrete that has a 16 mm NMAS with 30% of its aggregates passing through the 4.75 mm sieve and settling in the 0.075 mm sieve. e detailed gradations for the experimental asphalt mixtures are presented in Table 2. After determining the gradations, a mix design was conducted. e results can be found in Table 3.
In addition, cylindrical asphalt concrete samples with a height of 4 cm, cored from three field pavement sections, were also prepared in order to investigate the effects of service lives on the noise characteristics of asphalt concrete. e varying lengths of service life for the pavements are presented in Table 4.

Sound Absorption Performance of Asphalt Mixtures in Laboratory
In order to reduce noise level, materials with greater sound absorption efficiency can be used. Specifically, better sound absorption performance indicates a state when most sound is absorbed by a given material and less sound is reflected back at the observer, thereby significantly reducing the noise level around the sound absorptive material [21]. ere are two effective ways to characterize and measure the sound absorption performance of materials in a laboratory setting. e first is in using of a reverberation room, which has considerable costs and is typically only reserved for very particular studies. e second is by conducting an impedance tube experiment.
Past studies have shown that the impedance tube experiment is precise in its measures for the sound absorption properties of asphalt mixtures at lower frequencies [22]. e test picture and schematic plot of the impedance tube can be found in Figures 1 and 2. In this experiment, the test sample is mounted at one end of a straight, smooth, and closed impedance tube. During the measurement and calculation, the standing wave ratio method was selected to determine the sound absorption coefficients of materials. As indicated in Figure 2, the sample was put on one side of the impedance tube, while the loudspeaker on the other side produces an acoustic wave P i . e standing wave was formed due to the superposition of acoustic wave P i and the sample's reflected wave P r . en the pressure maximum (antinode) P max can be calculated when P i and P r are in phase, while the pressure minimum (node) P min can be obtained when P i and P r are out of phase. e definitions of P max and P min can be expressed in (1) and (2). In the equations, r is the reflectance. e standing wave ratio G and the sound absorption coefficient a can be then calculated as expressed in (3) and (4). To reduce the relevant noise level, a material with a higher sound absorption coefficient will be sought [23].
where P 0 is the amplitude of the acoustic wave produced by the loudspeaker and r is the reflectance which is defined to be P r /P i . For this study's impedance tube experiment, details regarding testing methodology can be found in ASTM C384-04 [24]. e sound absorption coefficient for each kind of asphalt mixture is measured at different test frequencies. In addition, the noise reduction coefficient is calculated by finding the arithmetic mean of the sound absorption coefficients at 250, 500, 1000, and 2000 Hz [25]. A noise reduction coefficient of 0 indicates perfect reflection, while a noise reduction coefficient of 1 indicates perfect absorption. It needs to be carefully noticed that the noise reduction coefficient is just a general idea of the noise reduction effectiveness of a particular material. In order to clearly understand the sound absorption properties of materials, the sound absorption coefficients over the test frequencies have also been provided and analyzed in this study.

3.1.
e Effects of Nominal Maximum Aggregate Sizes. Figure 3 presents the sound absorption coefficients of asphalt mixtures with different NMAS. As shown below, the noise reduction coefficients can be found in Table 5.
It can be seen in Table 5 that the noise reduction coefficients of the asphalt mixtures decrease with an increase in NMAS. However, the variations in sound absorption     Figure 1: Test picture of the impedance tube.

Shock and Vibration
coe cients for the three types of asphalt mixtures shift in di erent frequency ranges. When the test frequency is lower than 500 Hz, the asphalt mixture with 10 mm NMAS has a higher sound absorption coe cient when compared with the AC13 and AC16 mixtures. When the test frequency is between 500 Hz and 800 Hz, the asphalt mixture with 16 mm NMAS exhibits a higher sound absorption coe cient. In addition, the asphalt mixture with 13 mm NMAS is found to have a relatively higher sound absorption coe cient at the high range of frequencies. erefore, it can be deduced that each asphalt mixture has its own dominant range of frequencies at which a higher sound absorption coe cient can be expected. Some past studies have indicated that the tra c noise for passenger vehicles and heavy trucks is primarily found in the frequencies around 1000 Hz and ranging from 500 to 1000 Hz, respectively [26]. As such, the type of pavement asphalt mixture to be used would best be individually recommended for speci c highways with variable tra c conditions. For highways with more passenger vehicles, asphalt mixtures with a larger NMAS will have larger sound absorption coe cients and better noise reduction performance. As for highways with more heavy trucks, asphalt mixtures with a smaller NMAS are recommended.

e E ects of Asphalt Binder Types.
e e ect of the type of asphalt binder on the sound absorption performance of the asphalt mixture is analyzed next. e calculated sound absorption coe cients and noise reduction coe cients are presented in Figures 4 and 5.
It is interesting to note that, as seen in Figure 4, all the asphalt mixtures have a similar minimum sound absorption coe cient and similar sound absorption coe cients at the frequencies around 500 Hz.
is indicates that the densegraded asphalt mixtures exhibit poor noise reduction performance around the 500 Hz frequencies. In addition, the asphalt mixtures with di erent types of binders show di erent dominant frequency ranges from the viewpoint of noise reduction. For example, when the test frequencies are smaller than 500 Hz, the SBS modi ed asphalt exhibits comparably better noise reduction performance. However, the di erences found among the asphalt mixtures with di erent binders are limited. When the test frequencies range from 500 to 1000 Hz, the virgin asphalt binder (AH90) has a relatively better noise reduction performance. e dominant frequencies for rubber asphalt mixtures are those above than 1000 Hz.
From this, we can speci cally see that by adding more crumb rubber to the mixture, smaller sound absorption coe cients will be exhibited at frequencies lower than 1000 Hz while larger sound absorption coe cients will be exhibited at frequencies higher than 1000 Hz. e noise reduction coe cients can be also calculated and plotted thereafter, as seen in Figure 5. It is clear that SBS-modi ed asphalt and virgin asphalt binders have a similar noise reduction performance. Moreover, rubber-modi ed asphalt will have an even better noise reduction performance when compared with SBS-modi ed asphalt and virgin asphalt binders.
is phenomenon had been reported by many researchers before [27][28][29]. A possible reason for this phenomenon is due to the better deformability of rubber asphalt, which can absorb the vibration energy of tra c vehicles and then reduce the tire/pavement noise level. erefore, we can see that the type of asphalt binder can be tailored in order to reduce the noise level of a given  highway. For highways with more passenger vehicles, rubber asphalt binder is to be recommended. As for highways with more heavy truck tra c, SBS-modi ed asphalt and virgin asphalt binders will present a relatively better noise absorption performance.

e E ects of Air Void
Percentages. e e ect of air void percentage on the sound absorption performance of the asphalt mixtures is investigated next. e measured results are presented in Figures 6 and 7.
It is clear that the sound absorption coe cients and noise reduction coe cients of the asphalt mixtures presented in Figures 6 and 7 increase consistently with an increase in air void percentage. Note that similar conclusions have been presented and calibrated in past studies on the topic [14][15][16]. A larger air void percentage means that the increased porosity in the material can be used predictably to absorb sound, resulting in a greater performance in noise reduction. e variations of sound absorption coe cients seen in Figure 6 also tell us the respective peak frequencies for each of these asphalt mixtures.
e SBS-AC10-10 and SBS-AC10-20 asphalt mixtures will exhibit greater noise reduction performance at a frequency around 800 Hz, while SBS-AC10-30 will exhibit a peak sound absorption coe cient at a frequency around 1000 Hz. However, the SBS-AC10-10 and SBS-AC10-20 asphalt mixtures (i.e., those with higher air void percentages) will have greater sound absorption performance on the entire range of tested frequencies when compared with the SBS-AC10-30 mixture. erefore, higher levels of air void percentage can be recommended for dense-graded asphalt mixtures in order to reduce noise levels.   3.4. e E ects of Service Lives. As mentioned above, cylindrical asphalt concrete samples with a height of 4 cm were cored from three eld pavement sections. e variations in their sound absorption coe cients along with their test frequencies can be found in Figure 8.
As the cored samples are characterized by relatively smoother sides when compared with the laboratory compacted asphalt mixtures, the variations in the sound absorption coe cients for cored asphalt mixture samples are di erent. It is clear that the cored asphalt mixture samples from the old pavement exhibit poorer sound absorption performance when compared with the pavements that have a shorter service life. A possible reason for this is that the surface texture of old pavement is simply worn out. Table 6 indicates the texture depths of the cored sample surfaces. A clear decrease in texture depth consistent with an increase in service life is observed.
In order to analyze the e ects of texture depth on the sound absorption coe cients, the surface texture of the cored asphalt samples were cut with a saw. e sound absorption coe cient was measured again thereafter. e comparison of test results is presented in Figure 9. A small texture depth corresponds to a lower sound absorption coe cient. erefore, it can be concluded that for densegraded asphalt pavement with a long service life, some kind of construction technology (e.g., milling and grooving which can increase the texture depth of pavement) can be used to control and reduce tire/pavement noise levels. To summarize, reducing the NMAS, using rubber asphalt, and increasing the air void percentage as well as the surface texture depth can all work to improve the sound absorption performance of materials while reducing tire/pavement noise levels.

Statistical Analysis of the Test Results.
In order to analyze the in uential factors of noise characteristics for the laboratory compacted and cored asphalt concrete samples, statistical analyses are conducted. As to the di erent side surface characteristics, the laboratory compacted and cored asphalt concrete samples are analyzed, respectively. e stepwise regression analysis method is selected to analyze the signi cance of the in uential factors. For this statistical analysis, the sound absorption coe cient was selected as the dependent variable. e test frequency, viscosity of asphalt binder at 175°C (which can di erentiate the mechanical properties of asphalt binder and represent di erent types of asphalt), NMAS, and air void percentage were selected as the independent variables for the laboratory compacted asphalt sample, expressed as X1, X2, X3, and X4. e results presented in Table 7 show that the test frequency and air void percentage play the most important role in determining noise absorption performance.
A similar statistical analysis is conducted on the cored asphalt concrete samples. e test frequency, service life, and surface texture depth were selected as the independent variables, expressed as X1, X2, and X3. e results presented in Table 8 show that all the three factors play an important role in determining the noise absorption performance of a mixture.

Noise Characteristics for the Dense-Graded
Field Asphalt Pavement

Test Pavement Conditions and Test Description.
ree pavements with the same AC13 asphalt concrete layers, but varying service lives, were selected rst to analyze noise levels. Details can be found in Tables 4 and 6. In addition, in order to   analyze the influential factors of both tire/pavement sound quality and noise level, three types of dense-graded asphalt concrete pavement samples containing SBS-AC13, AR-AC13, and SBS-SMA13 located in Beijing were found. SMA here denotes stone mastic asphalt, one specific type of densegraded asphalt concrete characterized by a larger surface texture depth. ese sample pavements were exposed to a natural traffic environment and aged 3∼6 years without any overhaul. e pavement conditions are presented in Table 9.
Five driving speeds (40 km/h, 50 km/h, 60 km/h, 70 km/h, and 80 km/h) were measured to study the effect of vehicle speed on noise level and noise quality. ree test temperatures (30°C, 20°C, and 0°C) were also selected for the analysis. During field measurements, the tire/pavement noise spectrum was measured using a tire-pavement noise detection trailer according to the close proximity method [30]. e test tires and microphones were placed in a closed system installed in the test trailer. During testing, the trailer moves forward at some specified speed. Four microphones installed next to the test tire are then used to collect the continuous sound pressure level from the tire/pavement noise emitted at the given speed. Examples of the tested tires and their respective measuring devices can be found in Figure 10.
To avoid introducing external variables, only one specific type of tire was used: the Dunlop GRANDTREK AT20 265/65R17 112S. e numbers and letters denote that the tire has a 265 mm normal width cross section, a 65% height to width aspect ratio, and a 17 inch wheel diameter.

e Method to Calculate the Sound Pressure Level of Noise Spectrum and Sound Quality Parameters.
Tire-pavement noise data were collected and sorted using the four channel data processing system developed by the Beijing Shengwang Commercial Company (BSWA). During data collection and treatment, an A-weighting curve was utilized to define the various international standards generally used for measuring sound levels. Note that this method is widely used for environmental noise measurement. Specifically, a 1/3 octave A-weighted sound pressure level was adopted for this study. As to the evaluation index for sound quality, loudness, sharpness, harshness, and fluctuation were selected, as explained above.
Calculating the sound quality parameters was conducted using the Zwicker method, which is based primarily on the 1/3 octave spectrum. e audible noise frequency band is divided into 24 critical bands (on the Bark scale), as seen in Table 10.
Loudness (N) is a perceptual measure in which people judge the intensity of a sound, and it is dependent upon their hearing. It is a subjective feeling of both sound volume and intensity. Here, it is calculated by integrating the characteristic value of loudness at each critical band. e formula for calculating the specific value of loudness (N) and the excitation level of critical bandwidth is presented as follows: where E TQ is the excitation of absolute audible threshold in quiet conditions, E 0 is the excitation under reference sound intensity, E is the corresponding excitation of calculated sound, and N ′ (z) is the loudness at the specified frequency band number z. e variables E TQ , E 0 , and E can be obtained in field data collection. e total loudness (N) can be calculated by integrating N ′ on bands 0∼24 using the following formula: Sharpness (S) is a parameter that describes the proportion of high frequency components in the sound spectrum. e variable reflects the cacophony of sound signals.
e calculation formula is shown as follows: e weighted item g(z) is expressed by (8). e variable z represents band number.
Harshness (H) denotes the psychoacoustic parameter, which describes the modulation amplitude and modulation frequency distribution of the sound signal. e calculation formula is presented as follows: where ΔL E is the variation of excitation level, f mod is the modulation frequency (kHz), N ′ max is the specified maximum loudness in a critical band, and N ′ min is the specified minimum loudness in a critical band. Fluctuation (F) describes the level of sensitivity in the human ear to slowly modulated sound, reflecting the subjective feelings affected by fluctuation degree of sound loudness. e formula for calculating fluctuation is shown as follows: where f o is the modulated fundamental frequency of 4 Hz.  Road-Section 2, and Fengxiang Street) were selected to study the effects of service lives on tire/pavement noise pressure levels. e results are plotted in Figure 11.
It is clear that noise pressure level increases corresponds with an increase in the service life of the pavement and driving speed.
is variation in field tire/pavement noise level is similar to that for the cored asphalt samples (presented in Figure 8). In considering the similar gradation of the asphalt mixtures but variable service lives and surface texture depths, best practices for noise reduction regarding dense-graded asphalt pavement can be recommended. As noted above, some kind of technology (e.g., milling and grooving) can be utilized to reduce the field tire/pavement noise pressure level for these ends.

e Effects of Asphalt Pavements Types and Driving
Speeds.
e same noise measurements were conducted on the three pavements containing SBS-AC13, AR-AC13, and SBS-SMA13 asphalt concrete mixtures. As noted above, noise quality and noise pressure levels were measured and calculated for the three pavements. e results are presented in Figures 12-16. Note that AC, SMA, and AR denote the SBS-AC13, SBS-SMA13, and AR-AC13 mixtures, respectively, in Figures 12-16.
With an increase in driving speed (ranging 50 km/ h∼80 km/h), the tire-pavement noise sharpness for the SBS-AC13 pavement exhibits small variation. However, the tire-pavement noise sharpness for the SBS-SMA13 and AR-AC13 pavements exhibits peak values at speeds of around 70 km/h. is observation may be attributed to the similar surface texture depths for the SBS-SMA13 and AR-AC13 pavements. With this consideration, a tra c speed limit can be proposed based on the variation in noise sharpness so as to avoid the excessive noise pollution. To sum up, the sharpness of the SBS-AC13 pavement is lower than that of the SBS-SMA13 and AR-AC13 pavements. is means that for the SBS-AC13 pavement, the high frequency component in the noise spectrum is relatively small and exhibits only slight variation when speed is changed. erefore, it is likely that a larger surface texture depth has a negative in uence on the variation of tire/pavement noise.
Harshness and uctuation are used to describe the feelings conjured in humans when noise signals vary. e Shock and Vibration 9 variation of harshness and uctuation seen in Figures 15 and  16 are generally similar. e general trend is that the harshness and uctuation both decrease with the increase of driving speed. Speci cally, the AR-AC13 pavement exhibits a relatively larger harshness and uctuation at most test speeds. is may be due to the more elastic properties found in the rubber asphalt pavement. A larger elastic deformation is to be expected when tra c loads are applied to the AR-AC13 pavement, inducing an increase in both harshness and uctuation. When evaluating the sound quality of tire/pavement noise, the noise pressure level, loudness, and sharpness are generally used to form an objective index for better representing sound characteristics. erefore, it can be concluded from Figures 12-14 that a more e cient solution for reducing the tire/pavement noise of dense-graded asphalt pavement highways is by paving them with rubber asphalts.

e E ects of Test Temperatures.
Tire/pavement noise was also measured at di erent air temperatures. One of the objectives behind this study is to investigate the variation in noise pressure levels at di erent seasons. During the eld measurements included in this study, only two kinds of asphalt pavements (containing AR-AC13 pavement and SBS-SMA13 pavement) were selected. A speed of 50 km/h was decided upon as the test's driving speed. Measurements were taken, and the noise spectrum was plotted (see .
A peak noise level value at around 1000 Hz can be found for the AR-AC13 and SBS-SMA13 pavements at di erent test temperatures. Considering the dominant frequency of noise caused by passenger vehicles, the passenger vehicle can be regarded as the primary source of tire/pavement noise in light of this nding. In addition, the comparisons of noise levels between the AR-AC13 and SBS-SMA13 pavements are different at various frequency ranges. When the test frequency is smaller than 500 Hz, the AR-AC13 pavement exhibits greater noise when compared to the SBS-SMA13 pavement. When the frequency is greater than 500 Hz, it is the SBS-SMA13 pavement that produces the greater noise. erefore, in considering the di erent dominant noise frequencies for passenger vehicles and heavy trucks [26], the AR-AC13 pavement is recommended for paving highways predicted to experience large volumes of passenger vehicle tra c. As for highways that will see more heavy trucks, the SBS-SMA13 pavement would be more suitable for controlling tire/pavement noise levels.
Speci cally, the e ects of the test temperatures on the noise emitted by the AR-AC13 and SBS-SMA13 pavements can be found in Figures 20 and 21.
It can be seen from Figures 20 and 21 that although lower test temperatures can induce higher levels of tire/pavement noise, the di erences in the noise levels at di erent temperatures are not signi cant. Speci cally, the di erence is only about 2∼3 dB for temperatures between 0°C and 20°C. erefore, the noise issues in colder regions or seasons are more urgent than for warmer climates. is phenomenon can be attributed to the brittle property of asphalt mixtures in cold temperature, which works to reduce their sound absorption performance.

Statistical Analysis of Field Test Results.
A similar stepwise regression statistical analysis is conducted to investigate the in uential factors of noise levels for densegraded eld asphalt pavement [31]. e noise pressure level functions as the regression's dependent variable, while driving speed, service life, air void percentage, texture depth, the viscosity of the asphalt binder at 175°C, and measurement temperature function as independent variables (expressed as X1, X2, X3, X4, X5, and X6, resp.). Table 11 presents the analysis results.
Driving speed, service life, asphalt type, and measurement temperature noticeably a ect the noise level of densegraded eld asphalt pavement. e statistical insigni cance exhibited for air void percentage, texture depth, and type of asphalt binder can be attributed to the small quantity of selected representative samples.

Conclusions and Recommendations
is study has investigated the in uential factors of noise characteristics in dense-graded asphalt mixtures and eld asphalt pavements. e paper's ndings can be used to guide noise control design and policy for the most widely used dense-graded asphalt mixtures and pavements. Speci cally, the following conclusions and recommendations are o ered in light of the ndings:   (1) For highways with greater passenger vehicle traffic, asphalt mixtures with a larger NMAS that use rubber asphalt binders will have superior noise reduction performance. As for highways with greater heavy truck traffic, asphalt mixtures with a smaller NMAS, SBS modifications, and virgin asphalt binders are recommended for reducing tire/pavement noise. e comparisons of noise reduction coefficients conducted in this study indicate that asphalt mixtures with a smaller NMAS and rubber asphalt exhibit superior noise reduction performance.
(2) Larger air void percentages in dense-graded asphalt mixtures exhibit superior sound absorption and noise reduction performance. erefore, in order to reduce the noise level, increasing air void percentage is recommended for dense-graded asphalt mixtures so long as the durability of the mixtures can be retained. (3) Pavements with a longer service life exhibit inferior noise absorption performance and large noise levels, which has been shown by laboratory test results of the cored asphalt concrete as well as by field noise measurements. e surface texture depth of pavement was found to play a key role in determining the noise reduction characteristics. Improvements in milling and grooving technology can be utilized to reduce the field tire/pavement noise level for densegraded asphalt pavement therein. (4) Field measurement results indicate that the reasonable control of traffic speed limits and the use of rubber asphalt can help to reduce tire/pavement noise and noise loudness for dense-graded asphalt pavement. (5) Field measurement results also indicate that AR-AC13 pavement can emit smaller levels of noise for highways with a large volume of passenger vehicles, while SBS-SMA13 pavement is more suitable for controlling the tire/pavement noise levels for highways with more heavy truck traffic. is recommendation derives from both field testing as well as laboratory testing. (6) e tire/pavement noise level in dense-graded asphalt pavement increases when the air temperature gets cooler. erefore, stricter noise control policies should be considered for locales that experience lower temperatures in general.
In this study, it has been found that the effects of asphalt pavement types and driving speeds on the specific sound quality parameters of sharpness, harshness, and fluctuation are still not clear. Future research is needed to analyze these subjective parameters further.
Data Availability e test data in lab and in field used to support the findings of this study are included within the article.

Conflicts of Interest
e authors declare that there are no conflicts of interest regarding the publication of this paper.