执法小组委员会建议2017年11月

Contents Subcommittee Members..............................................................................................................................2Questions from NTIA.....................................................................................................................................2Introduction..................................................................................................................................................2Findings.........................................................................................................................................................2Recommendations........................................................................................................................................5Addendum I‒Status of Enforcement Operations, Current Policy Framework, and Previous Work inAutomated Enforcement..............................................................................................................................9AddendumII‒Questions for SAS Entities Regarding Automated Enforcement......................................11Figure 1. Straw Man Enforcement Cycle Model........................................................................................13 Subcommittee Members MembersincludedMary Brown, Mark Crosby, Dale Hatfield, Paul Kolodsky, Mark McHenry,JaniceObuchowski,Rick Reaser,Dennis Roberson,Andrew Roy,Mariam Sorond, BryanTramont, Jennifer Warren and Bob Weller.NTIA advisors were Bruce Jacobs, David Reed,andYang Weng. Questionsfrom NTIA 1.What options do you see for making enforcement more robust, including by increasingautomation to prevent interference, and to identify and respond to interference when itdoes occur in the near or longer term?2.What are the principal technical and operational options for enabling automatedenforcement, at both the network and device levels, and how would you addresscybersecurity and privacy requirements?Please consider, among others, optionsrelated to: station IDs; data cloud/fog architectures; and crowd-sourcing.3.What options for automated enforcement are unique to the development anddeployment of 5G technologies/applications?4.What steps do you recommend the Federal Government, specifically NTIA, take toimplement automatedenforcement processes? What steps will the private sector needto take? Please consider steps relating to technical, process and policy issues, includingpotential operator-to-operator coordination approaches. Introduction The subcommittee did not have sufficient time to address every question initsentirety. We didnot address the aspects of automated enforcement addressing cybersecurity requirements,steps that the private sector can take,oroperator-to-operator coordination approaches.Webelieve that there is a possibility to address those questions in future CSMAC groups.Inresponse to the above questions, theEnforcement Subcommitteehasidentifiedfive(5)findingsandis proposing five (5)recommendationsasNTIAaction items. Findings 1.Mostcompanies hire consultants to locate and identify interference, though somelarger companies (e.g., wireless carriers) have internal teams. Current framework limits the ability of consultants to precisely locate and identify sources due to privacy andaccess issues. 2.Sources of interference with regard to developing automated enforcement broadlydivide into three types (see also AddendumI): a.Intra-system (self-interference).Interference due to mechanisms entirely within thecontrol of the system operator. This interference is often due to poor systemimplementation practices, errors in equipment configuration, and less-commonly,malfunctioning equipment. b.Proximate (nearby) interference.1 Interference affecting just one site or user.Thisinterference is often the most challenging to resolve, as it may not be observableusing external equipment and there are often multiple sources.For example, theexquisite sensitivity of tower-top antennas with low-noise amplifiers often meansthat the interference is not “visible” to external observers and so location of thesource becomes difficult. Proximate interference also includes cases where anequipment issue may impact multiple sites or users, but the interference is localizedto individual sites. c.Widespread interference.Interference affecting multiple sites or users.Thisinterference is generally easy to identify because it is usually strong, but there maybe challenges in identifying the precise source(s) due to access or privacy issues, forexample. 3.“Automated” systems are relatively primitive or limited to specific standards-compliantsystems (e.g.,cellular PCIs, WiFi SSID/MAC addresses,Structured packet Detection,Preamble Detection, IEEE 802.11y identifier including a geo tag,etc.).There aresignificant challenges that these systems must overcome.(See FCC TAC statement ofwork for next generation enforcement architecture.2)Currently, there are basically fourcategories ofautomated