Transmission challenges and solutions for all digital AM IBOC NAB 2014

Transcript

1 TRANSMISSION CHALLENGES AND SOLUTIONS FOR ALL-DIGITAL AM IBOC BRIAN W. WALKER Nautel Limited Hackett’s Cove, Nova Scotia, Canada Nautel NX50 50 kW AM transmitter. The IBOC signals ABSTRACT were supplied by an Exporter+ and Exgine combination into the transmitter using a digital I/Q interface. Recent tests have been performed by the NAB to assess the viability of the all-digital AM IBOC mode, MA3. Hybrid AM transmission using MA1 has been common- 2 ORGANIZATION OF ALL-DIGITAL SIGNAL place for several years now, but the all-digital mode presents some unique challenges. The peak to average The all-digital AM In-Band On Channel (IBOC) signal is power ratio of the signal increases significantly com- a logical re-organization of the hybrid AM IBOC wave- pared to a station broadcasting an analog AM or hy- form that simulcasts the analog program along with the brid signal, requiring either a reduction in power or some digitally encoded audio on HD-1. The hybrid waveform form of peak control. A typical hybrid transmitter instal- is termed the MA1 mode, while the all-digital waveform lation would have the transmitter optimized for analog is termed the MA3 mode; both are fully specified in [1]. AM performance, but a different set of criteria are re- MA1 is comprised of 3 blocks of orthogonal frequency quired for the all-digital signal to optimize spectral per- division multiplexed (OFDM) carriers around the AM formance and MER. carrier; the primary, secondary and tertiary carriers each This paper will present an analysis of the MA3 signal with their own power levels as shown in Figure 1. The and spectral mask, along with the implications for pass- levels are reduced closer to the AM carrier such as to ing it successfully through both current and past genera- minimize bleed through from the digital carriers into the tion AM transmitters. Suggested power levels for a given analog AM transmission. Each block contains 25 indi- transmitter will be reviewed, along with how these could vidual carriers separated by 181.7 Hz, with a total of 156 be increased by reducing the amount of power in the AM carriers including the reference carriers. The higher pri- carrier. Finally, an innovative signal conditioning tech- mary carriers are 64-QAM (Quadrature Amplitude Mod- nique that reduces the envelope and phase frequency con- ulation) modulated and the secondary carriers are 16- tent will be presented, allowing the operator to choose QAM modulated. An example of 64-QAM used in MA3 a trade-off between MER and spectral performance into is shown in Figure 2. Tertiary carriers are reduced to difficult loads. QPSK (quadrature phase shift keying) modulation where the information is carried in the phase of the carriers 1 INTRODUCTION such as not to interfere with the AM signal content in the same space. With a symbol duration of 5.8 ms, this produces a raw data rate of around 92 kbps, which after There has been renewed interest in the all-digital IBOC forward error correction (FEC) leaves a throughput rate signal recently, with a few on-air tests occurring within of 36.8 kbps broken into a 20.2 kbps P1 logical chan- the past year. The all-digital signal potentially offers the nel (primary carriers) and a 16.2 kbps P3 logical channel advantage of much greater coverage compared to exist- (secondary and tertiary carriers). ing AM stations at similar power levels, whether it is compared against analog or hybrid. There is the poten- P1 carries the core audio information that entirely con- tial to either cover a wider audience, or to reduce power tains the audio signal content on its own. P3 contains levels and operating costs while serving the same mar- enhanced audio information that provides better audio kets. The recent adoption of modulation dependent car- definition to the P1 core audio stream. With good sig- rier level (MDCL) control in the United States would in- nal reception, a high quality audio stream is delivered dicate that there is considerable interest in saving power to the listener, while in marginal conditions the receiver to lower costs. This paper will examine some of the chal- can operate on the P1 core audio stream alone. This has lenges that are placed upon the transmitter to broadcast led to an IBOC signal configuration where select stations this signal, and explore some possible solutions. opted to mute the secondary and tertiary carriers in favor of their analog AM signal, reducing self-interference for All of the results in this paper were obtained using a

2 in a RBM configuration. Along with a reduction in bit rate, this provides a low bandwidth signal configuration only occupying 10 kHz of bandwidth. Figure 1: Spectrum of the MA1 signal. Figure 3: Spectrum of the MA3 signal. Since the logical channel organization is similar to MA1, on-air bit rates remain comparable. It is likely not the addition of audio and data services that will drive adoption of MA3, but rather the improved IBOC cover- age resulting from the IBOC carrier power increase and improved signal robustness. 3 PEAK AND AVERAGE POWER IMPLICATIONS As noted in the previous section, the all-digital wave- form consists of a large analog carrier surrounded by OFDM carriers. The default specification for the MA3 signal calls for the primary carriers to be at -15 dBc and for the secondary and tertiary carriers to be at -30 dBc. Figure 2: Constellation and MER display for the secondary carrier group in MA3 mode. This waveform has some interesting properties that dif- fer from both a standard analog broadcast and a typical OFDM broadcast. In this configuration, approximately analog listeners. This is called the reduced bandwidth 38% of the transmitted power is allocated to the analog mode (RBM) since the digital bandwidth is reduced by carrier. This gives the receiver something to lock onto, the 16.2 kbps in the P3 partitions. and should help to improve the transmitter spectral per- formance. Unfortunately, this energy doesn’t carry any MA3 is a reorganization of MA1 for all-digital oper- information and reduces the potential peak digital power ation, as can be seen in Figure 3. The primary carriers out of the transmitter. There has been some discussion have been relocated to replace the analog signal entirely about reducing the level of this carrier, since even drop- and are increased by 15 dB in power. The secondary car- ping it by a few dB would provide significant power sav- riers have moved to the upper sideband and the tertiary ings. carriers moved to the lower sideband both are increased It should be noted that it is more appropriate to con- in power to -30 dBc. The overall broadcast signal band- sider the power level of the station referenced to the to- width is reduced from around 30 kHz to under 19 kHz, tal power from the transmitter, rather than the carrier with a total of 102 carriers. The same P1 and P3 channel power. In hybrid AM mode, an AM station would typi- organization is maintained with a modest increase in P3 cally maintain the carrier power level that it would have channel capacity to 20.2 kbps. Both P1 and P3 continue before implementing IBOC. This would not work for a to operate in core and enhanced mode. Just like in MA1 conversion to all-digital, since this would imply that for a mode, the secondary and tertiary carriers can be disabled

3 50 kW transmitter it would need to deliver over 130 kW that for a peak-limited transmitter, the overall coverage RMS. The typical AM transmitter is rated for an RMS (which is largely governed by the power in the digital power corresponding to 100% tone modulation, or 150% carriers) could be increased by approximately 1.3 dB due of the carrier power level. With the example of the 50 kW to the removal of the analog carrier. For the first two rea- transmitter, that would imply that it should be rated for sons stated above, some carrier level is required, so the a maximum RMS power level of 75 kW. This turns out same analysis can be done for a 6 dB reduced carrier. to be an overly optimistic estimate of the power that can In this case, comparing against the standard signal, the be achieved from the transmitter. Despite the headroom power from the transmitter is reduced by about 1.5 dB, built into the power amplifiers to allow AM operation, and the peaks are similarly increased by 0.5 dB. This the peak power of the digital signal dictates the achiev- corresponds to a 1 dB increase in the overall coverage, able power level. which is worth considering if that carrier level would be sufficient. Modern AM transmitters are designed to be able to handle at least 125% positive peak modulation when broadcasting AM. In many cases, such as with the NX, the transmitter is tuned to allow 140% positive peaks. This implies that the headroom beyond the rated car- rier power level of the transmitter is 7.6 dB and this is the peak to average power ratio that the transmitter can sustain without clipping when operating at an equivalent RMS power level. Coincidentally, this is approximately the limit as to where the MA3 signal can be clipped and still meet the spectral mask, assuming no other transmit- ter nonlinearity. In reality, any transmitter will have diffi- culty sustaining this power level and meeting the current spectral mask without a method for reducing the spectral impact of clipping peaks. Figure 4: CCDF of the MA3 signal as specified (green), with a 6 dB reduction on the carrier (blue), and with no carrier (red). This plot shows the probability of exceeding a given peak level, as referenced 4 REDUCING THE ANALOG CARRIER to the signal RMS power. One point of discussion for the MA3 signal is how much It is true that with a typical AM transmitter architec- power should be devoted to the analog carrier. The car- ture that amplifies the envelope and RF drive signals sep- rier serves a few purposes. arately, there is usually some benefit to having a DC com- ponent to the envelope since it helps reduce the high fre- • The receiver can use the carrier to obtain a perfect quency content. In this case, the added carrier provides frequency and phase lock to the incoming signal. very little benefit because it simply isn’t large enough The receiver automatic gain control can use the car- • compared to the OFDM carriers. Increasing the level of rier to set the appropriate signal level. the carrier to help with spectral issues is impractical, so the transmitter must essentially be linear enough to pass a • The carrier may make the signal easier to transmit signal that has no carrier at all. If the transmitter already while meeting the spectral mask. has that linearity requirement placed upon it, meeting the spectral mask should be possible at any of the carriers There is an open question as to what carrier level is re- levels discussed here. When the discussed signals were quired to meet the first two purposes. With a number experimentally passed through an NX transmitter, there of receivers already in the field, the transmitted signal were several dB of clearance from the mask at all points, should be kept compatible if the all-digital standard is to with only minor differences in performance. succeed. It may be possible to reduce the carrier by 3 or even 6 dB and maintain compatibility with receivers, since finer equalization and level control uses reference 5 MEASURING POWER cells within the OFDM signal. It should be noted that measuring power with the MA3 As for making the signal easier to transmit, this may be signal must be done differently than with an analog AM a false assumption. In Figure 4, the complementary cu- signal. The typical AM site uses a current probe and av- mulative distribution function (CCDF) of the signal was eraging meter to determine the carrier power, since by plotted with the original carrier level, with a 6 dB re- averaging the envelope voltage the effect of modulation duction, and with no carrier at all. Removing the analog can be removed. This technique no longer works as ex- carrier entirely increases the absolute peaks relative to pected once the OFDM signal is introduced. The value the RMS by 0.8 dB, but it reduces the overall power out obtained by that type of measurement will not be equal of the transmitter by approximately 2.1 dB. This means

4 to either the carrier or the RMS power out of the trans- 3. Frequency response on the envelope path (or phase path) mitter. To illustrate the point, Table 1 lists the readings that would be obtained for three different signals. The 4. Peak clipping in the power amplifier AM modulation used for the analog case is the same as Most AM transmitters are designed using some form that on the MA1 signal. The averaging meter works well of envelope elimination and restoration (EER) amplifier. for analog mode, results in a slight error for MA1, and The principle in this architecture is that the envelope gives an error of almost 20% when using it for MA3. The voltage is developed using nonlinear amplification and is error would only increase if the analog carrier were re- mixed with the RF drive in the final stage power ampli- duced. When in MA3 mode the NX is configured to use fier to produce the desired waveform. The main advan- the RMS power for monitoring, protection, and power tage of this type of transmitter over linear amplification control, while MA1 mode is referenced to AM carrier is efficiency. The downside can be that the magnitude power as for standard AM broadcasting. This can be ob- and phase signals that are passed through the amplifier served in Figure 5, which shows an NX50 operating in have much wider bandwidth than the RF signal for all MA3. cases except analog AM. The resulting envelope spec- RMS Carrier Signal Averaging trum can be observed in Figure 6. Note that the original meter 20 kHz bandwidth on the RF signal has been expanded Analog AM 1 1 1.05 to 100 kHz on the magnitude; similarly, there is an ex- MA1 with Analog 1 1.11 1.02 pansion to 150 kHz on the phase. If the amplifier is not 1 MA3 2.62 2.11 capable of passing these signals without significant dis- tortion, then the transmitter will have difficulty with the Table 1: Carrier, RMS, and averaging power meter readings in vari- spectral mask. Some of these concepts were discussed in ous modes, each referenced to the analog carrier in that mode. a previous paper[2] with respect to the MA1 signal. MA 3 Magnitude Spectrum − 0 10 − 20 − 30 − 40 − − 50 − 60 70 − Relative magnitude (dBc) − 80 90 − − 100 30 40 50 60 70 80 90 100 0 10 20 Frequency (kHz) Figure 6: MA3 magnitude spectrum. 7 AM-AM CORRECTION Figure 5: The MA3 signal passing through an NX50 transmitter. The amplifiers in the transmitter will tend to have some Note that the power level displayed in the top banner is RMS, while degree of amplitude distortion. The distortion is typi- the sidebar shows the actual carrier power. cally low enough that it does not cause issue with analog transmission, but it benefits from being corrected for bet- ter audio specifications as well as digital transmission. 6 TRANSMITTER NONLINEARITY CORRECTION The actual mechanism for applying this type of correc- tion in the NX transmitter is by digitally applying a look- The transmitter would not have any difficulty with meet- up table (LUT) to the magnitude signal before generat- ing the spectral mask if it were not for some inherent ing the pulse duration modulation (PDM) signals for the nonlinearities in the amplification. There are four princi- amplifiers. It is adapted by comparing the desired sig- pal sources of spectral regrowth: nal against the measured output voltage of the transmit- ter, after appropriate delay-matching and filtering. The 1. Amplitude nonlinearity, otherwise known as AM- correction that would usually be applied on a transmit- AM distortion ter is accomplished with a 1 kHz tone, and assumes that 2. Incidental phase modulation caused by envelope the amplitude distortion is relatively constant with fre- variation, also know as AM-PM distortion quency. This simplification holds true at low frequency,

5 but as the frequency increases the response of the modu- of the RF output is dependent on the magnitude of the envelope voltage. This effect is certainly present when lator filter comes into play. broadcasting in analog, and a suitable correction can be The transmitter AM-AM correction was adapted us- applied to remove it. A relatively simple method involves ing several different frequencies as shown in Figure 7. using a lookup table to find the opposite phase shift at a At 1 kHz, the modulator filter has very little response given amplitude and applying it to cancel the distortion. and so its impedance can be modelled as a short. As An example of such a lookup table is shown in Figure 8. the frequency increases, the source impedance added by It can be seen from the characteristic in the graph that the the modulator filter increases, and an interesting effect correction required is quite significant at low amplitudes, happens. The PA tends to behave as a lower resistance and drops off to almost nothing above the nominal carrier value at very low amplitudes, which then has a tendency level. to counteract the modulator distortion once the source impedance increases. As a result, less AM-AM correc- PM correction on NX50 at 1 MHz AM − tion is required at higher modulating frequencies. This 45 implies that while the curve determined at 1 kHz, which 40 reflects the low frequency characteristic of the ampli- 35 30 fier, help with AM performance it is not ideal for digi- 25 tal, where the modulating frequencies tend to be much 20 higher. 15 10 − AM Correction with Training Frequency AM phase correction (degrees) 5 1 0 0.9 5 − 40 20 45 25 15 0 5 10 30 35 50 0.8 signal level (% full scale) 0.7 0.6 Figure 8: Example AM-PM correction curves from NX50 operating gain correction 0.5 at 1 MHz, using 1 kHz training frequency. 0.4 1khz 3khz 0.3 5khz When operating with an all-digital signal, this type of 7khz 0.2 AM-PM distortion can result in poor spectral compliance 30 40 45 50 0 25 20 15 10 5 35 signal level (% full scale) if left uncorrected. This is considerably worse than if the same transmitter were transmitting analog. It may be counter intuitive, since there would obviously also be Figure 7: AM-AM correction curves at 1, 3, 5, and 7 kHz. troughs on an analog signal that would result in phase modulation, but the high frequency envelope content in The principal problem with running the AM-AM cor- the troughs of a digital signal in turn causes high fre- rection at higher frequencies, or on an actual signal, is quency phase modulation. This has far more of a ten- that there tends to be a frequency response on the resul- dency to make the transmitter exceed the spectral mask, tant signal due to the modulator filter. The requirement especially at frequencies farther away from the carrier. for delay matching on the desired signal vs. the mea- In contrast, the incidental phase modulation of an analog sured output also becomes much stricter, since otherwise transmitter tends to fall closer to the carrier and be within a correction update may be applied to the wrong index the spectral emissions mask. in the look-up table. This can cause instability in the precorrection algorithm, and will result in the transmit- ter emitting unwanted spectral content because of noise 9 ENVELOPE EQUALIZATION and poor convergence in the LUT. In order to help alle- viate this issue, each update to the lookup table can be In an AM transmitter using an EER architecture, with a applied as a weighted average to a region of the look-up modulator followed by a RF amplifier, there is often a table. This has the effect of filtering the update, and helps bandwidth limitation on the envelope path. This is im- tremendously with the stability of the algorithm. Using posed by the combination of the PDM process and the this technique to run the adaptation on the actual signal reconstruction filter. The PDM frequency should be kept allows for the most appropriate correction to be used, and low to minimize AM-AM distortion and to boost the ef- does not interrupt the on-air signal. ficiency. At the same time, the modulator filter must attenuate any harmonics of the PDM that might pass through the amplifier and cause unwanted spectral emis- 8 AM-PM CORRECTION sions. These goals are at odds with the desire to keep the envelope bandwidth as wide as possible to maximize Another common distortion found on AM transmitters transmitter linearity. is AM-PM distortion. When this occurs, the phase shift

6 Using a compensation filter on the magnitude signal tion should be considered after all precorrection options can help extend the bandwidth by boosting those fre- are exhausted, since it will degrade the modulation er- quencies that are attenuated by the modulator reconstruc- ror ratio (MER) slightly. It has been incorporated in the tion filter. This compensation filter also equalizes group NX series for wide bandwidth Digital Radio Mondiale delay, which is often just as if not more important than (DRM), but is equally applicable to the MA3 signal with correcting for amplitude variation. The challenge often the appropriate configuration[3]. The principle used is lies in finding the ideal filter for the system. On the similar to some peak to average power ratio (PAPR) re- NX series, a modified version of the least mean squares duction schemes, where the problematic portion of the (LMS) algorithm is used to adapt the filter from the mea- signal is modified slightly to reduce its impact. With a sured voltage sample. PAPR reduction the problem areas are peaks; with the typical AM transmitter, the most challenging portions of One method that has been used with success to adapt the signal are in the troughs, for a few reasons. the filter is simply to run the transmitter in AM mode with wide band (>70 kHz) noise as the modulation The envelope signal has the highest frequency con- • source. This has the advantage of being very simple to tent in the troughs when operating with a digital sig- implement, it tends not to be affected by any other non- nal. linear effects, and it gives a good characterization of the modulator response since there is known frequency con- • The AM-AM distortion is most pronounced at low tent. The disadvantages are that it can be affected by amplitudes, and the variation with frequency means the RF filter response and a broadcaster could not real- that it cannot be perfectly cancelled. istically run a signal like this into the antenna, since it would cause interference with several other stations. The AM-PM distortion on the signal is largest at • To optimize the transmitter for the all-digital signal, low amplitude where the current from the PA is low. the equalization needs to be adapted with the transmit- This can typically be corrected extremely well, but ted signal, ideally into the antenna. This allows for the errors in the other corrections will cause phase er- unique characteristics of the load to be taken into ac- rors since the predicted amplitude would not match count, has the advantage of not driving the RF load at fre- the voltage across the RF bridge. quencies that are not of interest, and allows on-air adap- Bandwidth limitations on the antenna can translate • tation. The challenge with using the transmitted signal is to either increasing or decreasing impedance being that the frequency content will not be flat for the adap- presented by the RF PA with frequency. This may tation. Using this filtering algorithm, this type of signal result in a requirement for the modulator to apply a adapts the filter fairly well where there is significant fre- negative voltage to the RF PA in order to achieve the quency content, but not where there is less energy. This desired output, which is not possible without the ad- is insufficient to help with the spectral regrowth, and in dition of considerable complexity to the amplifier. many cases can make it worse than having no equalizer at all. There are a few essential stages to the algorithm, with The envelope signal can be altered to allow adaptation several methods being possible for each step. by using a shaping filter to flatten the envelope frequency content. This is only used in the adaptation algorithm, • Identify the problematic sections of the signal not on the transmitted signal. The implementation on the NX series measures the signal going through the trans- Flag and extend the identified sections • mitter to dynamically design the shaping filter to use. Replace each section with a lower frequency transi- • The overall effect of this process is that the equalization tion (as measured with the magnitude and phase) filter will adapt more closely to the high-frequency re- gions in the trough. The same portions of the signal tend • Low pass filter the resulting signal to ensure it stays to be responsible for most spectral regrowth, so this gives within the original RF bandwidth the best results. The sections of the signal requiring correction can be identified by high-pass filtering the magnitude and phase 10 SIGNAL ENVELOPE CONDITIONING signals, and looking for the output to cross a threshold. An alternate approach to meeting the spectral mask re- Alternatively, the magnitude signal can be examined for quirements for MA3 is to make the signal easier to pass falling below a threshold, although this could flag sec- through the transmitter, rather than attempting to correct tions of the signal that are not high frequency and thus all nonlinearity. In a real system where the antenna may have less requirement for modification. The final method not present an ideal load, or with a previous generation that can be used is to examine the second derivative for transmitter not designed for the high bandwidth require- exceeding a threshold. Regardless of how the sections ments of an all-digital signal, this may be the only way of the signal are identified, the next step is to determine to meet the mask. The approach described in this sec- what to do about them.

7 In order for the identified sections to be modified, the MER, it should have imperceptible impact on coverage, first step is to extend the flagged regions of the signal but it may help with a system needing to meet the mask. in the time domain. This allows a more gradual modifi- The advantage of this type of approach is that it does not cation to be made, reducing its spectral impact. It also require knowledge of the transmitter, and there are no serves to join together any isolated samples, and helps to RF samples required as there would be with precorrec- tion. The downside is that it has limited ability to over- create a region to be corrected. There are some additional come certain transmitter issues, and may only be useful checks performed within the implementation to ensure that a replacement region does not become too large, and for systems that are close to, but are not quite spectrally that the corrections do not become too frequent; other- compliant. wise the MER might be too severely impacted. The next stage has the purpose of replacing the sec- 11 CONCLUSION tion of the signal with a segment that is smooth in both I and Q as well as magnitude and phase. This requires The specification for the MA3 signal might be mislead- deliberate intervention to make sure the signal will not ing if the assumption is made that the carrier level would fall back to its original samples when the final low pass be the same as with MA1 or analog. Transitioning to all- filter is applied. The method used here determines a new digital implies that the licensed power level for a given midpoint for the section being replaced, then uses cubic station should be reconsidered, and coverage tests still spline interpolation to fill in the remaining samples. The need to be completed to determine appropriate power midpoint as determined in this algorithm is the mean of levels. The capabilities of the existing equipment in the the magnitude and phase of the samples entering and ex- field should be taken into account: many transmitters iting the replaced section. An example of a correction may only be capable of an RMS setpoint that is less than made by the algorithm can be seen in Figure 9. The sig- their current carrier power. The level of the analog carrier nal has been moved away from zero, which was causing in the signal should be reviewed; if possible, the 6 dB re- high frequency content on the magnitude and phase. duction shown in this paper should be considered, since it makes a significant difference in both the transmitter power capability and the RMS transmitted power. The other consideration for deploying MA3 will be the ability of the existing transmitters in the field to meet the 4 spectral mask. Many AM stations currently struggle to meet the MA1 mask, and the MA3 mask is significantly 2 more stringent. It is likely that many legacy transmit- ters will be unable to meet the mask unless it is relaxed 0 somewhat, or they employ some of the techniques de- Quadrature − 2 scribed in this paper. The NX transmitter used for these Original tests is not representative of many of the older systems Modified − 4 4 in the field that have been retrofitted for IBOC. The en- 2 velope conditioning technique may be of benefit to some 300 250 0 200 of these systems, since adding it to an existing transmit- 150 − 2 100 50 4 − ter is far less invasive than the changes required to allow phase − In Time precorrection. Figure 9: Example of a correction made by this algorithm. Note that References the signal has been moved away from the origin, but is still smooth TM in the I/Q plane. [1] HD Radio Air Interface Design Description - Layer 1 AM , iBiquity Digital Corporation, Finally, a low pass filter is applied to the signal to sup- http://www.nrscstandards.org/SG/NRSC-5- press any out of band content created during the interpo- C/1012sf.pdf, August 23, 2011, rev. F. lation process. The cutoff on this filter must be very ag- [2] B. W. Walker, “Easing the Transition to AM IBOC: gressive in order to avoid modifying the desired signal, Tools and Techniques to Help the Broadcaster,” but it does not require a significant amount of attenua- NAB Broadcast Engineering Conference Proceed- tion since the signal should not have too much content to ings , 2009. suppress. The resulting signal has lower frequency content on [3] B. W. Walker, “Signal Correction for Digital Audio the magnitude and phase signals than the original. This Broadcasting Using EER Amplifiers,” Master’s the- helps to reduce the effects of some of the possible trans- sis, Dalhousie University, Dec 2006. mitter nonlinearity described in this paper. Provided the algorithm is set in such a way that it has little impact on

Related documents

Uranium 2014: Resources, Production and Demand

Uranium 2014: Resources, Production and Demand

A Joint Report by the OECD Nuclear Energy Agency and the International Atomic Energy Agency Uranium 2014: Resources, Production and Demand U ranium 2014: Resources, Production and Demand N E A

More info »
E:\PUBLAW\PUBL031.115

E:\PUBLAW\PUBL031.115

131 STAT. 135 PUBLIC LAW 115–31—MAY 5, 2017 * Public Law 115–31 115th Congress An Act Making appropriations for the fiscal year ending September 30, 2017, and for May 5, 2017 other purposes. [H.R. 244...

More info »
WEF GlobalInformationTechnology Report 2014

WEF GlobalInformationTechnology Report 2014

Insight Report The Global Information Technology Report 2014 Rewards and Risks of Big Data Beñat Bilbao-Osorio, Soumitra Dutta, and Bruno Lanvin, Editors

More info »
pisa 2012 results volume I

pisa 2012 results volume I

PISA 2012 Results: What Students Know and Can Do tICS, themA StuDent PeRfoRmAnCe In mA ReADIng AnD SCIenCe Volume I rogramme for ssessment A tudent S nternational I P

More info »
HS Full Reduced

HS Full Reduced

State(s) of Head Start

More info »
CityNT2019TentRoll 1

CityNT2019TentRoll 1

STATE OF NEW YORK 2 0 1 9 T E N T A T I V E A S S E S S M E N T R O L L PAGE 1 VALUATION DATE-JUL 01, 2018 COUNTY - Niagara T A X A B L E SECTION OF THE ROLL - 1 CITY - North Tonawanda TAX MAP NUMBER ...

More info »
Microsoft Word   Standard Sewer & Water Main Specification July 1,2014 CompleteBook Print Format

Microsoft Word Standard Sewer & Water Main Specification July 1,2014 CompleteBook Print Format

THE CITY OF NEW YORK DEPARTMENT OF ENVIRONMENTAL PROTECTION BUREAU OF WATER AND SEWER OPERATIONS STANDARD SEWER AND WATER MAIN SPECIFICATIONS JULY 1, 2014 Patalano Revised: July 1, 2014; Purnima Dhari...

More info »
a i4787e

a i4787e

2015 ISSN 2412-5474 nimal genetic resource diversity underpins the supply livestock products and A services across a wide range of production environments. It promotes resilience and serves as a basis...

More info »
The State of Sustainability Initiatives Review 2014: Standards and the Green Economy

The State of Sustainability Initiatives Review 2014: Standards and the Green Economy

The State of Sustainability Initiatives Review Standards and the 2014 Green Economy Jason Potts, Matthew Lynch, Ann Wilkings, Gabriel Huppé, Maxine Cunningham, Vivek Voora

More info »
Microsoft Word   147400

Microsoft Word 147400

Federal Communications Commission FCC 15 -24 the Before Federal Communications Commission Washington, D.C. 20554 In the Matter of ) ) GN Docket No. 14 -28 Protecting and Promoting the Open Internet ) ...

More info »
a i5199e

a i5199e

Status of the Main Report World’s Soil Resources © FAO | Giuseppe Bizzarri INTERGOVERNMENTAL INTERGOVERNMENTAL TECHNICAL PANEL ON SOILS TECHNICAL PANEL ON SOILS

More info »
625137

625137

2018-19 Nebraska All-Sports Record Book - Nebraska Communications Office -

More info »
StateoftheClimate2015 lowres

StateoftheClimate2015 lowres

STATE OF THE CLIMATE I N 2015 Special Supplement to the Bulletin of the American Meteorological Society Vol. 97, No. 8, August 2016

More info »
LaneEtAlPrivacyBigDataAndThePublicGood

LaneEtAlPrivacyBigDataAndThePublicGood

This is a prel version of the book Privacy, Big Data , and the Public Good : iminary er, and for Lane, Victoria Stodden, Stefan Bend nt, ed. Julia Helen Niss enbaum Frameworks Engageme Unive rsity Pre...

More info »
Thriving on Our Changing Planet: A Decadal Strategy for Earth Observation from Space

Thriving on Our Changing Planet: A Decadal Strategy for Earth Observation from Space

TIONAL ACADEMIES PRESS THE NA This PDF is available at http://nap.edu/24938 SHARE     Thriving on Our Changing Planet: A Decadal Strategy for Earth Observation from Space DET AILS 700 pages | 8.5 ...

More info »
At the Dawn of Belt and Road: China in the Developing World

At the Dawn of Belt and Road: China in the Developing World

At the Dawn of Belt and Road China in the Developing World Andrew Scobell, Bonny Lin, Howard J. Shatz, Michael Johnson, Larry Hanauer, Michael S. Chase, Astrid Stuth Cevallos, Ivan W. Rasmussen, Arthu...

More info »
ECI Historical Listing    Volume III   March 2019

ECI Historical Listing Volume III March 2019

Employment Cost Index National Compensation Survey – Volume III Historical Listing April www.bls.gov/ect 2019 (December 2005=100) 2019 -March Current Dollar, March 2001 The estimates from 2001 to 2005...

More info »
Draft Environmental Impact Statement for the Safer Affordable Fuel Efficient (SAFE) Vehicles Rule for Model Year 2021 2026 Passenger Cars and Light Trucks

Draft Environmental Impact Statement for the Safer Affordable Fuel Efficient (SAFE) Vehicles Rule for Model Year 2021 2026 Passenger Cars and Light Trucks

The Safer Affordable Fuel-Efficient (SAFE) Vehicles Rule for Model Year 2021–2026 Passenger Cars and Light Trucks Draft Environmental Impact Statement July 2018 Docket No. NHTSA-2017-0069

More info »
Fourth National Report on Human Exposure to Environmental Chemicals Update

Fourth National Report on Human Exposure to Environmental Chemicals Update

201 8 Fourth National Report on Human Exposure to Environmental Chemicals U pdated Tables, March 2018 , Volume One

More info »