|
[Submitter's First Name] Reza [Submitter's Last Name] Arefi [Membership Status: (M)ember; (O)bserver; (N)either] M [Starting Page #] Pg.14 [Starting Line #] 35 [(T)echnical for Content-Related Material; (E)ditorial for typos,
T [Detailed Description of Proposed Insertion, Deletion, Change] Replace the text with the following. Recommendation 1 : Adopt a "6 dB below receiver thermal noise in the victim receiver criterion " as being a value of interference from the interfering operators, which is "acceptable." The document recommends this value in recognition of the fact that it is not practical to insist upon an "interference-free "environment. Having once adopted this value, there are some important consequences: Each operator acknowledges that he is willing to accept a 1 dB degradation in his receiver sensitivity from the operators. Depending upon the particular deployment environment, an operator may have a cumulative –6 dB contribution from multiple CoCh and AdjCh operators. Each operator should include design. margin in his system which is capable of simultaneously accepting the compound effect of interference from all other relevant operators, at the -6 dB level. The design margin in (b)above should be included preemptively at initial deployment, even if the operator in question is the first to deploy in a region and is not experiencing interference. All parties should recognize that, in predicting signal levels, which result in the -6 dB interference value, it is difficult to be precise in including the aggregating effect of multiple terminals, the effect of uncorrelated rain, etc. The actual degradation in performance and the value of signal level below receiver noise in the victim receiver, need to be further studied in order to assure that high performance, high availability, BWA networks can be deployed with sufficient operational flexibility. [Reason for Edit] WCA Engineering Committee The -1 dB degradation in performance (receiver sensitivity) from each possible interferer is not acceptable for the following reasons: Given that there is the potential for numerous interferers that could affect victim system performance, it will become punitive to account for several dB of degradation into system link margins. The 6 dB below thermal noise floor criterion from each interfering operator is not acceptable for the following reason:
As can be seen, the effect of adding multiple interferers would have a notable impact on the receiver performance. This cumulative effect should be taken into consideration. [Submitter's First Name] Reza [Submitter's Last Name] Arefi [Membership Status: (M)ember; (O)bserver; (N)either] M [Starting Page #] 15 [Starting Line #] 11 [(T)echnical for Content-Related Material; (E)ditorial for typos,
T [Detailed Description of Proposed Insertion, Deletion, Change] Replace the text with the following. Recommendation 2 : [Each operator should take the initiative to collaborate with other known operators prior to initial deployment and at every relevant system modification. This recommendation should be followed even if an operator is the first to actually deploy in a region.](Refer to Recommendation 4 comments). To encourage this behavior, the document introduces the concept of using specific received interference signal level (dBm) values to "trigger " different levels of initiatives taken by an operator to give notificationto other operators. If power spectral flux density values (psfd)are specified as trigger values, a translation methodology is utrilized (as given in Annex YYY) to convert the received signal levels into psfd values. The specific trigger values and their application to the two deployment scenarios are discussed in Recommendations 5 and 6 below and in Section 7.In some regulatory environments, the fact that the "triggers " were properly analyzed and that the proper cooperative initiative was made can be used as evidence of operating in good faith to promote coexistence. [Reason for Edit] WCA Engineering Committee Use of power spectral flux density (psfd) as a trigger mechanism is not desirable. A much simpler, more effective method is for the operator to state received signal level, in dBm, along the border between adjacent license areas for co-channel frequencies. This method removes the element of determining bandwidths occupied, antenna gains, etc., and provides, along with the radio transmitter’s location and antenna orientation, everything the carrier on the other side of the boundary needs to know to determine if there will be any potential interference issues. Reasonable received signal levels can be determined based on the noise floor and then the impact of multiple interferers on the increase of that noise floor. Moreover, most of the computer software tools on the market today that model RF Fixed Broadband Wireless Access systems provide information in terms of simple received signal level. [Submitter's First Name] Reza [Submitter's Last Name] Arefi [Membership Status: (M)ember; (O)bserver; (N)either] M
[Starting Page #] 15 [Starting Line #] 21 [(T)echnical for Content-Related Material; (E)ditorial for typos,
T [Detailed Description of Proposed Insertion, Deletion, Change] Replace the text with the following. Recommendation 3 : Each operator should design and deploy his own system for the maximum amount of frequency reuse The logic behind this Recommendation is that the same techniques of base station site selection,antenna pattern management and emission control that must be employed to facilitate aggressive frequency reuse within a system will contribute to its coexistence with other systems. Recommendations 9,10 and 11 below and in Section 6 provide recommended minimum antenna patterns, spectral masks and maximum EIRP from the vantage point of coexistence. These do not, however, guarantee coexistence. Even the most dense frequency reuse system does not guarantee coexistence. However, starting from a foundation of a "better " engineered system can facilitate the later resolution of coexistence issues. Coexistence requirements will need to be carefully balanced with the operational and performance flexibility requirements of BWA networks.
[Reason for Edit] WCA Engineering Committee Although this Recommendation may generally considered to be good engineering practice, it would be unduly burdensome for a carrier to be forced to adopt a requirement to use frequencies in a uniform manner across the frequency band. This requirement can become an operational constraint that is unacceptable. Further, these types of operational specifications are outside the scope of a document that is only expected to deal with coexistence criteria, in the form of protection criteria and interference power limits that will trigger coordination. Moreover, this requirement or practice ignores the current demand for adaptive broadband systems (either TDD or FDD), which would require dynamic adjustments of transmitted bandwidths across the full spectrum. [Submitter's First Name] Reza [Submitter's Last Name] Arefi [Membership Status: (M)ember; (O)bserver; (N)either] M
[Starting Page #] 16 [Starting Line #] 1 [(T)echnical for Content-Related Material; (E)ditorial for typos,
T [Detailed Description of Proposed Insertion, Deletion, Change] Replace the text with the following. Recommendation 5 : No coordination is needed in any direction if the transmitter is greater than 16 km from either the service area boundary or the neighbor ’s boundary (if known)in that direction.
[Reason for Edit] WCA Engineering Committee The 60 kilometer (km) distance from the service boundary is unnecessarily excessive and recommends a much smaller distance, e.g., 16 km as determined by FCC 47 CFR § 101.103. The 60 km criterion seems to have been derived from some arbitrary calculations involving the radio horizon, and does not have any relationship to achievable distances from the service boundary. The 16 km criterion is entirely satisfactory. This recommendation is also inconsistent with the analyses in the Annex. The radio horizon approach may apply to systems operating at lower frequencies. Further, the radio horizon approach assumes hub-hub omni-directional operation, which is also inconsistent. (Note: Annex information needs to be updated to reflect the 16 km criterion) [Submitter's First Name] Reza [Submitter's Last Name] Arefi [Membership Status: (M)ember; (O)bserver; (N)either] M
[Starting Page #] 16 [Starting Line #] 5 [(T)echnical for Content-Related Material; (E)ditorial for typos,
T [Detailed Description of Proposed Insertion, Deletion, Change] Replace the text with the following. Recommendation 6: Recommendation 2 above introduced the concept of using interference signal levels (dBm) and/or power spectral flux density "triggers " as a stimulus for an operator to take certain initiatives to collaborate with his neighbor. The coordination trigger values (see Annex B)of –127 dBW/MHz/m2 (24,26,28GHz bands) and –127 dBW/MHz/m2 (38,42GHz bands)are employed in this document, in the initiative procedure described in Recommendation 7 below. These values were derived as that power spectral flux density values which, if present at an average base station antenna and average receiver, would result in approximately the – -6 dB interference value cited in Recommendation 1.It should be emphasized that the trigger values are useful only as thresholds for taking certain actions with other operators; they do not make an absolute statement as to whether there is, or is not, interference potential. Several administrations have permitted significant deployment of point-to-point links as well as point-to-multipoint systems, with psfd trigger levels of -127 dBW/MHz/m2at 38 GHz band. [Reason for Edit] WCA Engineering Committee An acceptable level of interference power at the service boundary is given under "Specific Comments." The psfd approach is the wrong approach. Nevertheless, additional comments are provided on why even the recommended psfd levels are not acceptable. The 38-42 GHz coordination trigger value of –111 dBW/MHz/m2 is not supported in Annex B or anywhere else in the document. Annex B presents a range of values from -105 dBW/MHz/m2 to –135 dBW/MHz/m2, but the trigger value in this Recommendation is not found. It is also appropriate to take into consideration the psfd values considered by the ITU-R, and by the ITU WRC-2000, where worst case values of close to –120 dBW/MHz/m2 at elevation angles above 25 degrees have been used for interference from satellite systems into BWA/HDFS systems in the 38 GHz band. The psfd limits at lower elevation angles below 25 degrees drop to 137 dBW/m2/MHz at 5 degrees elevation angle. The psfd limits for intersystem coexistence must be consistent. ITU WRC-2000 states that the pfd limit for GSO satellites is –127 dBW/MHz/m2 (maximum) and –139 dBW/MHz/m2 (nominal) at the worst-case scenario of angles less than 5 degrees. The maximum GSO limit of –127 dBW/MHz/m2 is the number we feel should be used for a psfd limit. Please see notes above under Recommendation 2 regarding reasons why pfd should not be used. Also, the NSMA has been delegated by the FCC to develop inter-system coordination criteria, and acceptable interference power levels for co-frequency and adjacent channel operations. The IEEE 802.16 recommendations are inconsistent with the work going on within the NSMA on this issue. Coordination with NSMA, at a minimum, is strongly recommended, if not required. [Submitter's First Name] Reza [Submitter's Last Name] Arefi [Membership Status: (M)ember; (O)bserver; (N)either] M
[Starting Page #] 16 [Starting Line #] 19 [(T)echnical for Content-Related Material; (E)ditorial for typos,
T [Detailed Description of Proposed Insertion, Deletion, Change] Replace the text with the following. Recommendation 7 : Apply the "triggers " of Recommendations 5 and 6 prior to deployment and prior to each relevant system modification. Should the trigger values be exceeded, then the operator should try to modify the deployment to meet the trigger, and failing which the operator should coordinate with the affected operator. [Reason for Edit] WCA Engineering Committee The coordination procedures are based on Recommendations 5 and 6, which have already been commented on above. The psfd values that trigger coordination need to be carefully selected, which does not seem to be the case here. [Submitter's First Name] Reza [Submitter's Last Name] Arefi [Membership Status: (M)ember; (O)bserver; (N)either] M
[Starting Page #] 16 [Starting Line #] 25 [(T)echnical for Content-Related Material; (E)ditorial for typos,
T [Detailed Description of Proposed Insertion, Deletion, Change] Replace the text with the following. Recommendation 8 : For same area /adjacent channel interference cases, deployment will usually benefit by having one guard channel between nearby transmitters. Where the transmissions are of different bandwidth, the guard channel could be equal to the wider channel. Where administrations do not require guard channels, the affected operators may reach agreement on how the guard channel is apportioned between them. However, setting aside a full or portion of a guard channel is not a requirement, as long as the emission mask requirement at the band edge is met., Careful and intelligent frequency planning and/or use of orthogonal polarization will significantly alleviate the need for this guard channel. [Reason for Edit] WCA Engineering Committee Spectrum is a valuable resource. Therefore, any requirement that freezes the use of spectrum by an Operator by requiring him to set aside a large piece of guard band spectrum, will cause undue harm. Once again, the current document goes beyond its intended scope. If the purpose is to provide protection to operators in an adjacent band/co-coverage environment, it needs to be done by specifying a level of out-of-band emission, as function of frequency from the band-edge. It is up to the operators to implement their in-band frequency plan to achieve the specified out-of-band limit. Adjacent channel interference issues should be dealt with through good frequency planning and engineering practices. From a business perspective, operators must have the flexibility to utilize 100% of the spectrum, as long as they conform to the out-of-band emission rules. [Submitter's First Name] Reza [Submitter's Last Name] Arefi [Membership Status: (M)ember; (O)bserver; (N)either] M
[Starting Page #] 16 [Starting Line #] 33 [(T)echnical for Content-Related Material; (E)ditorial for typos,
T [Detailed Description of Proposed Insertion, Deletion, Change] Replace the text with the following. Recommendation 9 : Utilize antennas for the base station and subscriber terminals at least as good as shown in Section 6.2.The coexistence simulations which led to the Recommendations contained herein revealed that a significant part of coexistence problems are the result of main-beam interference. The side lobe levels of the Base Station antennas are of a significant, but secondary influence. The sidelobe levels of the subscriber antenna are of tertiary importance. In the context of coexistence, therefore, antennas, such as those presented in Section 6.2 are sufficient. It should be emphasized that utilizing antennas with sidelobe (and polarization)performance better than the minimum will not degrade the coexistence performance and, in fact, are an effective mitigation technique for specific instances. In many cases, intra-system considerations may place higher demands on antenna performance than those required for inter-system coordination. [Reason for Edit] WCA Engineering Committee Clarification.
[Submitter's First Name] Reza [Submitter's Last Name] Arefi [Membership Status: (M)ember; (O)bserver; (N)either] M
[Starting Page #] 17 [Starting Line #] 1 [(T)echnical for Content-Related Material; (E)ditorial for typos,
T [Detailed Description of Proposed Insertion, Deletion, Change] Replace the text with the following. Recommendation 10 : The utility of emissions masks for controlling adjacent channel coexistence issues is strongly dependent upon the separation of the two emitters in space and in frequency. In the case where there is large spatial separation between emitters, the opportunity exists for an interfering emitter to be much closer to a receiver than the desired emitter. This unfavorable range differential can overwhelm even the best emissions mask. Likewise, emissions masks are most effective when at least 1 guard channel exists between allocations. The emissions mask presented in Section 6.1.4 is most appropriate for the case where there is one guard channel between allocations and a modest separation of emitters. For cases where there no guard band is provided, it is recommended that co-location of emitters be considered before trying to improve emission masks. For operating frequencies above 15 GHz, the FCC Technical Rules already contain an emission mask requirement. This mask is more than adequate for adjacent channel coexistence. [Reason for Edit] WCA Engineering Committee The emission mask that is recommended assumes at least one guard channel between allocations. The emission mask should show the required roll-off from the band-edge, and not assume the roll-off to begin from the inner-edge of the guard channel in the allocated spectrum. For operating frequencies above 15 GHz, the FCC Technical Rules already contain an emission mask requirement. The spectral mask presented here (IEEE 802.16) places a more stringent requirement on maximum attenuation (67 dB) at frequencies far removed from the operating band than does the FCC requirement, 47 CFR § 101.111 (56 dB). This could place undue burden on current operators who meet the FCC requirement, but not the requirement presented here. Further, the IEEE 802.16 Recommendation will require the roll-off to commence in-band at the beginning of the guard-band. Manufacturers who have built systems in compliance with the FCC Rules will have to reengineer their products to accommodate the more restrictive IEEE 802.16 criterion, if it is adopted. [Submitter's First Name] Reza [Submitter's Last Name] Arefi [Membership Status: (M)ember; (O)bserver; (N)either] M
[Starting Page #] 17 [Starting Line #] 13 [(T)echnical for Content-Related Material; (E)ditorial for typos,
T [Detailed Description of Proposed Insertion, Deletion, Change] Replace the text with the following. Recommendation 11 : Utilization of EIRP and Subscriber Power control in accordance with Section 6.1.1 and 6.1.2,respectively, can be of help in meeting the coexistence criterion. The interests of coexistence are served by reducing the amount of EIRP emitted by base station, subscriber and repeater terminals. [Reason for Edit] WCA Engineering Committee Operating at lower EIRP levels is a good engineering practice, and will be followed by BWA operators to varying extents depending on their service and business environments. Operation with low fade margins, in clear weather, is particularly important at frequencies above 25 GHz. Our main concern is that the group has once again gone beyond the intended scope by requiring universal operation at low EIRP limits. Operators should use whatever EIRP levels are needed to conduct its business and state what the resultant power levels are for the given transmitters at the service boundary. It must be noted that limiting EIRP levels is of primary concern to operators, and equipment vendors are being influenced to develop and implement downlink power control capabilities in addition to uplink capabilities. The objection is for requiring mandatory universal power limits. [Submitter's First Name] Reza [Submitter's Last Name] Arefi [Membership Status: (M)ember; (O)bserver; (N)either] M
[Starting Page #] 17 [Starting Line #] 20 [(T)echnical for Content-Related Material; (E)ditorial for typos,
T [Detailed Description of Proposed Insertion, Deletion, Change] Replace the text with the following. Recommendation 12 : It will not be necessary to engage in extensive calculations if the received interference signal level at the service boundary is specified in dBm. However, in order to reconcile with psfd values prescribed by several regulatory regimes, it is useful to translate the psfd values into signal levels (dBm). This translation methodology is provided in Annex YYY (to be developed). In conducting analyses to predict power spectral flux density, the following considerations may be taken into account:
-Clear air (no rain)plus relevant atmospheric absorption -Intervening terrain blockage
boundary should be the maximum value which occurs at some elevation point up to 500 m above local terrain elevation.
Clear sky propagation (maximum path length) conditions should be assumed. [Reason for Edit] WCA Engineering Committee First, no assumptions or cumbersome calculations will need to be made by the interfering operator in order to state received signal levels at the boundary. Second, no cumbersome calculations or assumptions will need to be made by the victim operator if these values are known at the boundary. The RF engineers for the operator will be very comfortable determining whether a given received interference signal level will present problems. There is no explanation as to why the psfd level should be calculated at an elevation 500 m above local terrain. It would appear that this is to represent an absolute worst case scenario, or is intended to simulate line-of-sight conditions whether or not they exist along the path. The issue would seem to be where the neighboring system might possibly have a base station that could receive the interference. The good judgment of the engineer is a better guide than an arbitrary calculation to 500 m [Submitter's First Name] Reza [Submitter's Last Name] Arefi [Membership Status: (M)ember; (O)bserver; (N)either] M
[Starting Page #] 17 [Starting Line #] 33 [(T)echnical for Content-Related Material; (E)ditorial for typos,
T [Detailed Description of Proposed Insertion, Deletion, Change] 4.2 Guidelines for geographical and frequency spacing Delete or significantly modify the model presented below. The need for guard bands should be eliminated. Spacing for acceptable performance is subjective.
[Reason for Edit] WCA Engineering Committee
The model will result in inaccurate results.
[Submitter's First Name] Reza [Submitter's Last Name] Arefi [Membership Status: (M)ember; (O)bserver; (N)either] M
[Starting Page #] 50 [Starting Line #] 15 [(T)echnical for Content-Related Material; (E)ditorial for typos,
T [Detailed Description of Proposed Insertion, Deletion, Change] In section 7.1.1 paragraph 3, replace the word "shall" with "should be" or restructure the sentence to incorporate this concept. [Reason for Edit] WCA Engineering Committee A recommended practice emphasizes the use of "Should", not "Shall". |