Campus Cellular Telephone Access Project Executive Summary

September 30, 2008

Introduction

Background

Many University of Michigan, Ann Arbor campus, buildings have poor cellular telephone coverage. This project would improve cellular coverage so it is good to excellent in all major U-M owned Ann Arbor campus buildings.

There has been interest in improving cellular coverage for productivity and general service purposes for several years. Recently—improving personal safety—has been identified as an additional key value of ubiquitous, readily available cellular service.

Improving personal safety has become a campus concern at many levels. In addition to events at other universities, an event at the U-M College of Engineering (CoE) made it clear that personal safety is impacted by an individual's ability to call for assistance. The sheer number of cell phones could be beneficial in an emergency situation. However, this benefit cannot, at present, be realized because in many buildings the reception is inadequate as was the case in the CoE building.

Several efforts by various campus departments to improve cellular service have been initiated in recent years. All efforts, thus far, have required the university to guarantee handset/cellular contract purchases from the single carrier improving its service. This model does not serve the security purpose because it is not a multi-carrier solution. Additionally, many single-carrier improvement projects have not been as successful as had been anticipated.

Carrier buy-in to a campus cellular improvement plan is required because the frequencies are licensed and paid for by the carriers. This means another entity, such as the university, cannot rebroadcast those frequencies, even within their own buildings, without carrier approval. Carrier business models and costs have not made it attractive to provide pervasive in-building campus cellular coverage in the past, but it appears this is changing. The university can help to shape the business model evolution.

This project proposes a model for implementing cellular coverage improvements that will provide coverage across campus at a relatively low cost.

Purpose

The purpose of this document is to: 1) clarify the allocation of resources, including personnel that would be required by departments; 2) set campus expectations relative to the effort and the results; and 3) propose a coordinated effort for the implementation so that carrier frequency interference does not occur and costs are controlled.

Executive Summary

General Overview

This project is being managed by ITCom, focusing on improving cellular coverage in the majority of campus buildings excluding the hospitals, medical campus, and Northwood apartments. A single organization is managing the project because of the nature of radio frequencies together with the physical infrastructure required. For example, in many cases a single exterior antenna will impact cellular service in multiple buildings. Control of the signal is critical to the quality of the service.

The goal is to implement a multi-carrier solution. Planning for Health System space can occur in parallel, but would be managed by MCIT. The total scope of the ITCom project are the approximately eighteen million square feet in the approximately 200 buildings with more than 10,000 square feet each. These measures typically exclude unoccupied basements, tunnels and mechanical areas. These figures are for planning purposes, and the final amount of space and number of buildings may vary by +/- 5%. (NOTE: although excluded from the measurement numbers, it is expected that there will be improved cellular coverage in the majority of basement and mechanical areas.)

Effort to Date

Effort to date has included: researching technologies for deployment, working in-depth with carriers to understand their business models and drivers, talking to peer universities about their activities, testing approximately 900,000 square feet to assess general quality and solution options, putting in place the mechanisms to test in a production environment a promising new technology that would expedite deployment on the rest of the campus, and initiating the testing of all campus buildings.

Testing in four north campus CoE buildings of an indoor-only solution, indoor/outdoor hybrid, and a promising new antenna technology, found that all tested options are viable. The new technology option, an antenna installed in air ducts, has significant promise in terms of expedience in deployment, being non-obtrusive to operations, and being more cost-effective than traditional in-building technologies.

Next Steps and Timing for Overall Project

The next steps will be to: assess the new technology performance for multiple carriers in a group of five buildings that account for about 981,000 square feet, under actual user conditions; assess the overall cost and the business model proposed by the vendors in terms of acceptance and feasibility of success with carriers; test the approximately 18M square feet of campus space for baseline quality measurements; produce carrier-specific data as well as summary data for campus dissemination resulting from the testing; draft a conceptual design for improvement on the campus; secure funding as necessary, and implement.

Engineering for the technology test buildings is currently underway. Installation is expected to begin after October 10, when the provider will be chosen. The requested turn-up date is December 10, subsequent to which there will be approximately 30 days of testing. The system will remain in place for at least 90 days allowing for testing and an assessment of user comments. After that time we plan to make it permanent, assuming it performs to expectations.

Information obtained from discussions with campus units will be used with engineering requirements and constraints to develop a specific campus building priority list. This effort will be done concurrently with the antenna technology test in five CoE buildings.

Costs/Business Model

In the industry and among academic institutions there are numerous funding models from 100% building-owner funded, to some carrier contribution, to carrier contribution based upon increased sales, to 100% carrier funded. In addition, there is a "3rd-party neutral host" business model whereby an entity that is not a carrier builds a system and then charges the carriers to use it. Although this 3rd-party model had some measured success several years ago, it appears to be losing favor with both building owners and cellular carriers. Carriers have indicated that they are willing to fund the design and deployment of the equipment on the U-M Ann Arbor campus. However, actual commitments have not yet been received.

The cost for testing existing coverage on the campus is approximately $150,000, with some of that anticipated to be recovered from the carriers. The data from this study will be used to design the cellular infrastructure.

The cost for the pilot in five CoE buildings, approximately 981,000 square feet, is expected to be under $300,000. We have requested that the provider(s) of the antenna technology for the test buildings quote a price just for the test, which could be zero. We have also requested that they propose a price and business model that is acceptable to the carriers for a permanent installation in those buildings as well as a model for the entire campus. It is believed that this will clarify future costs to the university for the wider, full-campus implementation.

Due to the business model options, the hard-dollar cost to the university for a campus-wide solution can range from $1M to approximately $7-8M. Even if the carriers fund 100% of the hard costs, there will be staff time required for integration with interbuilding fiber, collocation planning, building user coordination, campus communication and other activities.

With this document, ITCom is communicating its need for, and requesting approval for support from campus units. To plan and implement the project we will need access to building space, an understanding of key priority areas, special concerns, user coordination, etc. In addition, should outdoor nodes be required, building mounting assets would also be required. If outdoor nodes are part of a final design, ITCom would work with the campus Exterior Elements Design Review Committee (EEDR) to gain approval for their placement. Finally, it would be most advantageous to have a single point of contact identified for each building.

Risks and Contingencies

1. One key challenge will be to get the carriers focused and active in engineering and implementing a solution. They are typically slow to act.

2. We do not believe that this project will, by itself, impact the quantity of landline phones used by staff and faculty. However, with reliable cellular service in University Housing residence hall students will probably use the landlines in the rooms even less and their cell phones will be able to more reliably communicate an emergency. Due to student pressure and a possible Housing interest in redirecting funds, landlines in the rooms might be eliminated. If this occurs, the fixed costs for the landline service would be redistributed over the remaining campus landline users, increasing their costs.

3. This project may lead Public Safety and Plant to evaluate their radio communications. This robust cellular network together with the increasing capability of mobile devices would provide a cost effective, state-of-the-art solution that may meet many of their needs.

4. It is critical that there be only one or two U-M points of contact for the carriers—ITCom for the campus and MCIT for the hospitals. If there are multiple distributed antenna systems (DAS) being implemented, the frequency interference could be a significant technical problem. Some existing cellular coverage infrastructures in buildings may need to be decommissioned.

5. The technology test is based on using a buildings' HVAC system as a waveguide for the cellular frequencies. The trial will prove whether or not that technology works. There are, of course, many buildings that do not have a forced air HVAC system. In those instances a more expensive but more tested infrastructure will be used.

   In parallel, we will have an immediate implementation option for the trial buildings as well as a proposed plan for the remainder of the campus.

Standards of Performance

For the pilot, ITCom has specified performance requirements for numerous transmission quality measures for voice and data to be available in 95% of the space being tested. For capacity, it is specified that in the most demanding time periods (e.g., an emergency, game day) the capacity of the system should be adequate to provide <2% dropped calls. It is specified that E911 service be consistent with applicable regulatory requirements. These same parameters are expected for any permanent installation.

The campus community should be aware that the above specifications are target requirements, and may not be met immediately upon implementation. For example, if the new technology solution proves not viable in the actual environment on campus, we could revert to an outdoor-only solution. That solution provides quality service in about 70% of the target pilot space, not 95% as provided by the new technology. Thus additional retro-fitting would be required.

For design purposes our target is to improve cellular coverage in 95% of the occupied space. Pragmatically, general industry standards are that approximately 90% of the space is covered. Although the technology to be used and the building structure will ultimately determine what spaces cannot be covered, the uncovered space would likely be the tunnels, inner hallways or stairwells, possibly restrooms, and some mechanical rooms. Lastly, space may be improved in "chunks," i.e., there may be a vast improvement in 60% of the space in a building, and then some months later, an improvement in the remaining space.

Recommendations

Since the campus is not committing to major cell phone contract purchases, the university is being viewed by cellular carriers as a "public venue" much like an airport. With a coordinated effort on the campus, we will be able to leverage the campus and the constituency of the campus community in negotiating with the carriers to improve their service. It is recommended that ITCom, with active participation of the departments, manage the planning and deployment of technology to improve cellular performance in all campus buildings as presented above. A trial in the College of Engineering is proceeding. This trial will provide information to more thoroughly propose a campus wide plan.