The link I provided doesnt seem to work for quite a few people....so here is the text of the pdf file. It is from Legg Mason looking at the competitive threats to Nextel's PTT service.
I tried to highlight the most important issues....if anyone wants a copy of the original PDF file, PM your e-mail address.
attach4.groups.yahoo.com
• Summary: Based on our research on vendors and potential carriers offering push-to-talk, we believe Nextel has a significant technology advantage over competitive products coming to the market in 2003. Most notably, we believe any competitive product will have a 6-10 second delay in the call setup time, which may seem like an eternity to many existing Nextel Direct Connect customers who have become accustomed to virtually instantaneous communication. Further, maintaining instantaneous communication once the call has been setup may require significant spectrum allocation from carriers that are already constrained. We recognize that these hurdles are not insurmountable and will not give Nextel an indefinite competitive advantage, but, over the near term, we believe the greatest risk to Nextel will be the headline risk when new products are introduced.
• We expect to see competitive push-to-talk products from Verizon Wireless, Sprint PCS, and Alltel in 2003 to
challenge Nextel's Direct Connect product.
• Call-setup latency, network management, and implementation will be large challenges for the carriers as they implement push-to-talk products and manage voice networks.
• Vendors are offering solutions that address the switch only, and may cause integration problems with the radio
frequency (RF) layer and handsets.
• No competitive push-to-talk product has been implemented by carriers in the United States, and, to date, most
products exist only in a test environment.
• The risk of an inferior product and undefined target market will reduce potential returns from carriers looking to implement a competitive push-to-talk product.
• We believe Nextel will improve its competitive position by introducing nationwide Direct Connect in the second half
of 2003.
Issue:
Core to our investment thesis for Nextel is the company maintaining its differentiation strategy with its
push-to-talk product -Direct Connect. With the rollout of packet data networks, CDMA operators and GSM operators now have the ability to rollout a push-to-talk product, and have made a number of statements about
their intent to have a competitive product in 2003. With those comments come various promises of product
performance and delivery dates. As this issue has been gaining some steam over the last few months, we
believe it was time to evaluate the issue focusing on technology, implementation, vendor issues, and carrier
issues.
What is push-to-talk?
Push-to-talk (PTT) is a two-way radio communication designed in 1992 utilizing Motorola's proprietary iDEN
technology using the brand name Direct Connect. It was targeted predominantly for the blue/gray collar
workforce as an effective way to manage a mobile workforce and communicate instantaneously on a private
one-to-one call or group call. Direct Connect has been very successful as Nextel has built a number of
successful business networks (construction, suppliers) that tailor to a single industry or interest group. These
customers boast the highest ARPU and lowest churn in the industry. In addition, Nextel has reported that
add-on units represent approximately one-third of their gross additions, further demonstrating the high level of
customer satisfaction.
To further enhance its product, Nextel is planning to expand Direct Connect from local calling areas to traveling
Direct Connect by the end of 2002 and, subsequently, nationwide services by mid-2003. We have maintained
for some time that Nextel's Direct Connect product is the only differentiator in wireless and is the key attribute
to the company maintaining its strong competitive position.
Carriers:
Dating back to the summer of 2001, there has been speculation regarding which carrier will offer a push-to-talk product and when that product will come to market. The speculation began because all of the carriers now
have the ability to offer a push-to talk product when they upgrade their networks to packet switching (2.5G)
from circuit switching (2G). Of the carriers that we have spoken with, we believe there is interest in the product
from the CDMA camp - Sprint PCS, Verizon Wireless, and Alltel, and potentially AT&T Wireless from the
GSM/TDMA world. With the exception of AT&T Wireless, all of the mentioned carriers have noted that with the
launch of their cdma2000 1x networks, they have the ability to launch a push-to-talk product and all expect to
have a competitive product in the market sometime during 2003.
Vendors:
The next logical question that investors have been asking is that if carriers do not have a Motorola, as Nextel
does to help develop the push-to-talk solution, then what solution/vendor will the carriers use? From our
analysis, we believe the carriers are predominantly focused on two vendors: Winphoria networks for the CDMA
solution and Sonim technologies for the GSM solution. As a part of our research, we spoke with both vendors
to gain clarity into the underpinnings of their product.
Below we summarize our conclusions:
Winphoria
The company develops a commercially available next-generation packet-based Mobile Switching Center that
increases the capacity of wireless voice networks, and supports both circuit- and packet-switched networks for
2G, 3G, and all-IP wireless technologies. Winphoria is a private venture-capital-funded company based in
Tewksbury, MA, founded in March 2000.
What is their push-to-talk product?
The company offers a product called Global Instant Rendezvous (GIR). It supports PTT connections on
handsets with the Winphoria client software. The product uses a push-to-talk server-based solution and resides
on a mobile switch. The solution works on both cdma2000 1x and GPRS networks, but we believe much of their early success has been with the CDMA community. The PTT server controls call setup and controls similar to Nextel's DAP (Digital Application Processor). It operates on any packet network and works the following way. First, the signaling access, or the protocol used to communicate with the server, invites users to a one-to-one or group call by sending a session initiation protocol (SIP) invite. SIP is a signaling protocol that works over any IP-based device that sends out an invite to a user from the server to join a push-to-talk session. SIP has been rapidly adopted as a Voice over Internet Protocol (VoIP) standard as a method of merging voice and data networks. Once users respond to the SIP invite, a PTT session is established.
GIR specifications
include:
1. Group call register - keeps track of who is registered and can handle up to one million members.
2. Call processing - issues a SIP invite to initiate a message.
3. Conversion Real-time-protocol (RTP) where speech is converted to RTP and travels over packet
data-switched network to server.
4. Works on cdma2000 1x and GPRS networks.
5. Estimated costs for deployment including servers is in the millions based on the number of subs.
6. Unlimited number of people for conferences.
Users have the ability to access saved group lists and can leave messages if the other party is not available.
Similar to what we have heard from other vendors, call set-up will take 6-10 seconds and Winphoria's solution
will manage the switching layer only, leaving the RF and handset solution up to the carrier to resolve. The
company noted that it does not make the software in the handsets, but that it partnered with unnamed handset
vendors and expects to have commercial availability of handsets by 4Q03. While we have seen the form factor only on PDA's rather than phones, we believe that it will use a push-to-talk button similar to Nextel. As stated above, we believe Winphoria has engaged Sprint PCS and may also supply the push-to-talk solution for Verizon Wireless and Alltel.
Sonim
Sonim Technologies is a leading provider of IP-based instant communication solutions optimized for GPRS, 3G
and 802.11 wireless data networks. The company is private and was founded in late 1999 by former
employees of Intel, TI and Nortel. The company focuses on the GPRS carriers, but noted that its solution is
air-interface independent.
What is their push-to-talk product?
Sonim's push to-talk solution combines mobile instant messaging and SMS with a voice interface. According to
Sonim, the service leverages scaleable client/server architecture, with a client embedded in the device and a
network-based server platform delivering the push-to-talk services. Similar to Winphoria, the solution provides
SIP-based call control and presence management (for group calls). Sonim features include:
1. Contact list that may contain both individuals and groups and provides presence information to see what
users within a contact list are available for a push-to-talk session.
2. Messaging capabilities, which allows users who are not available for a a push-to-talk session to send a
push-to-talk message (non-real time).
3. Voice memo service that allows users to capture impromptu information.
In the U.S., the Sonim believes T-Mobile is the most aggressive carrier in coming up with a PTT solution,
followed by Cingular and AT&T Wireless and expects an offering in U.S. and Europe by the end of 2003.
Push-To-Talk Challenges:
With any new technology, there are challenges that need to be addressed to allow the product interoperability
with current products in the marketplace. In speaking with Nextel, it has taken them ten years to get Direct
Connect to where it is today with instant call set-up and efficient use of network resources, even with a network
that was designed to handle push-to-talk first and interconnect second. With that being said, we outline below
what we believe to be the most significant challenges carriers will face in competing with Nextel's Direct
Connect product.
Call setup
This is the most prevalent issue in the marketplace, and also the biggest advantage to Nextel. Call setup is the
time it takes from when a customer pushes the PTT button and speaks, to when the other party receives the voice packets and hears the voice. In Nextel's world, this set-up is 1 second or less, compared to approximately 6-10 seconds with all other competitive products. Vendors have told us that the six-second delay consists of: 1/2 second for channel assignment, 1 second for service connect, 1/2 second for RLC synch., and 3 seconds for PPP. What this means is that to set up a group or individual Direct Connect call, users will have to wait the approximate length of a landline call to connect. In response to extended call set-up time, Winphoria noted that the majority of call set-up and latency problems are a function of the CDMA 1xRTTT network technology, rather than solutions being brought by Winphoria or other vendors. We believe this issue is significant, as CDMA handsets need to be woken up, a process by which the handset requests a data session and is assigned network resources before it transmits or receives data packets. Any change to this process to shorten the call set-up process would need to go through the CDMA standards body and could take significant
time to resolve. We note that Winphoria has tried to solve this problem by getting its handsets to start making the network wakeup call as soon as the user begins the PTT session, as well as voice buffering, or recording the voice of the PTT user, before the session is initiated. It may be premature to form a conclusion on these network work-arounds, but we can conclude that call set-up is a significant advantage to Nextel.
The vendors have noted that once call-setup is established, communication is instantaneous. However, that would require carriers to extend the "hang time" allocated for a call. Hang time is the amount of time the network will allow a group or individual call to use a voice channel before it requires the initiation of another call set-up. This is important because Nextel's call set-up is fast enough that they only need to allocate approximately 7 seconds for hang time, while other carriers will have to allocate at least 10-20 seconds so their call set-up can be completed. In practical terms, it determines the length of time allowed between voice packets being sent. For example, if two users are using the PTT function and one user takes 10-15 seconds to locate a document for the conversation, the call may be lost if the hang time is not over 15 seconds. This would
exacerbate the call-setup issue currently facing the vendors as it would require multiple setups. The other
potential problem, is that if carriers substantially increase the hang-time, they are tying up valuable network
resources in the form of voice channels for longer periods of time. This turns PTT into a capacity drain for other
carriers versus a capacity benefit as it is for Nextel, requiring half the voice channels as interconnect service.
We believe this is an essential feature in Nextel's service as customers have become accustomed to fast call setup, and other carriers would essentially ask Nextel customers to take a step back in efficiency. This would be the equivalent of getting consumers to buy a slower speed computer processor after they have been
working on their high-speed Pentium computers for years. The exact reason for the fast call setup is not
completely known to the outside world, but we do understand a component in the fast call setup is the sleep
mode of the handset. When handsets are not in use, they fall into the sleep mode, and, depending on the
network that sleep mode could be light or deep sleep. For Nextel's network, they do not allow its handsets to
fall into a "deep sleep mode," by having handsets that periodically send a page to the cell site noting they are
still active. By doing this, the call set up process can be completed faster. The downside to this strategy is that
it drains battery levels and spectrum usage, but it allows the phones to stay in an active mode. We believe this
will be the largest hurdle for competitive products to overcome. No vendor has been able to solve this problem
and it would require wholesale changes within the CDMA standards world to change the sleep-cycle mode of
the network. As such, we believe that call set up will be the largest hurdle for a competitor and a major
deterrent in taking away market share.
Network Issues
Another potential problem for the vendor solutions that will be implemented by the carriers is that many of the
current solutions only integrate with one-third of the core network elements. The three parts of the network, the
Radio Frequency (RF) layer, switch, and handsets need to speak to each other to maximize PTT efficiencies.
The RF layer, or the Radio Access Network (RAN), controls the transmission and reception of radio signals,
while the switching network provides switching and transport for traffic coming to and from the RAN. The
problem with solutions for the switch only means the carrier will have to integrate with other core network
elements on its own.
As a reminder, the Nextel/Motorola relationship works very well because Motorola manages all parts of the
network as well as mobility management. In Nextel's world, mobility management is the management of the
network by Motorola of Direct Connect and interconnect. Specifically, it determines for Direct Connect how
many cell-sites should be searched for a particular user or users and how long to keep a voice channel
available. These are the keys to mobility management and carriers will need to appropriately manage the
balance between Direct Connect and interconnect service to maintain service quality levels.
Vendors such as Winphoria and Sonim do not offer a full turnkey solution through all of the layers of the
network but rather a solution to be implemented at the switching layer. The carriers will be responsible for their
own mobility management, which may require a fairly steep learning curve and may affect the quality of the
interconnect network if voice channels are being allocated for Direct Connect usage.
Handsets
Handsets may be another area of concern for competitive push-to-talk products as the server will have to
arbitrate between multiple types of handsets. Specifically, Nextel's switches only need to recognize Motorola
handsets while Sprint PCS, Verizon, and Alltel have handsets in their portfolio from multiple handset makers.
The solution may be to sell a PTT-only phone that a server can recognize, but it will decrease the adoption time
for any competitive product as users will have to make an additional handset purchase, shortly after migrating
to new handsets for 1xRTT and GPRS networks. From the vendors we spoke with, Sonim was the only vendor
to indicate they have engaged 3 of top-6 handset providers to provide a handset solution. Based on normal
product development times, we would estimate that even if competitive PTT networks are ready for commercial
deployment in the first half of 2003, handsets would not be available in commercial quantities until the end of
2003 at the earliest. We do not know how the form factor would look or if the new handsets would have a PTT
button similar to Nextel as all of the demos from the vendors have been on PDAs.
We believe that regardless of the technical or handset issues, some customers can be bought with aggressive
promotions and free handsets and this will be a negative for Nextel, as they are losing a potential customer.
However, the flip side of the argument is that to the extent that Verizon, Sprint and Alltel increase awareness of
competitive products, they increase the overall pie. This potential market may not be available to Nextel today
and may make up for some of the loss of incremental customers from competitive offerings.
Business Networks and Interoperability
Direct Connect has been very successful because the company has built a number of successful business
networks (construction, suppliers) that tailor to a single industry or interest group. In the early days of Direct
Connect, business networks were formed so specific user groups could communicate. As Nextel expanded its
network to allow multiple fleets to communicate business networks became much more viable. Specifically, a
landscape architect could communicate with its subcontractors and its subcontractors could communicate with
its suppliers. The company was successful in linking up parts of the business cycle. This is relevant to other
carriers trying to compete with Nextel as they would have to convert all layers of a business network to a new
product before any part of the business network would switch. This may be challenging and costly if carriers
compete on price instead of product differentiation.
On a similar note, interoperability among existing Nextel Direct Connect customers and other carriers may be a
barrier to success. Unless Nextel opens its network to other carriers so all PTT customers have the ability to
communicate, we believe it will be very difficult to increase the overall pie. This is a challenge that has not been presented in the current cellular environment as a Verizon customer and Nextel customer could communicate
even though the carriers are using different technologies. In our view, it further isolates Nextel from its
competitors, which may be a doubled-edged sword. Specifically, it may prevent other carriers from having
success with Direct Connect, but limits Nextel options if another carrier is interested in combining the two
networks.
Target market
As carriers weigh the pros and cons about going after an established market with a technologically inferior
product, the choice of a target market will be a crucial decision. Specifically, vendors have mentioned to us that
carriers are experimenting with the idea of going after a different segment than Nextel's. We believe this to be
the best way to gain early market share as those users will not be as married to fast call set-up times as
existing customers may be. In our opinion, there is a market for PTT outside of the traditional construction,
manufacturing, and government world that carriers could target, but we struggle with the investment case. In
our view, carriers have become interested in PTT not because the service is available to them with packetized
networks, but rather that Nextel has consistently demonstrated that the PTT segment boasts the highest value
customers in the industry. However, that value may be derived by the differentiation in the product as well as
the target segment, which means that if carriers choose alternative segments such as the consumer market
without any product differentiation, then they will repeat the same mistake they have made in the voice market
today. This decision may ultimately become a non-starter for carriers evaluating the potential return from the
segment.
Customer Survey - The 5 Cs
As part of our research, we surveyed several Nextel enterprise users in the construction industry to determine
the utility of Direct Connect. Our conclusions are as follows:
1) Call Setup - Instantaneous call setup is very valuable, as many Direct Connect sessions last only a few
seconds in length. The average Direct Connect sessions last 40 seconds, so waiting 6-10 seconds for the call
setup is material.
2) Communication Speed - Given their affinity for the fast call setup, it came as no surprise that Direct Connect
users also valued the sub-1 second ongoing communication in a Direct Connect session. As stated previously,
achieving this from other carriers will likely require additional spectrum.
3) Connectivity - The respondents unanimously agreed that the inherent value in the product comes from all
Nextel users being able to Direct Connect with each other. They went on to state that other carriers attempting
to compete would have a very difficult time penetrating the construction industry unless Nextel allowed
interconnection to its handsets. Basically, if your customers and suppliers both use Direct Connect, then it will be very difficult for you to switch to another carrier if Nextel does not allow interconnection, which we believe it will not.
4) Cost - Most users agreed that if a carrier offered significantly cheaper pricing (i.e. 25% discount) they would
look at it, but that switching would require the competing wireless carrier to offer very cheap or free phones as
well due to the large embedded base. While cost is important, none of the users we talked to said they would
switch at any cost if the call setup was 6-10 seconds and the phones wouldn't communicate with Nextel
phones.
5) Control - Surprisingly, one of the users surveyed mentioned that the Nextel phones are used as a control
mechanism for employees. Specifically, the particular company's phones are only set up to use Direct Connect
and can't make cellular calls. This proved to be beneficial from both cost and production points of view.
Nextel Path
With so much focus on the potential PTT products coming to the market, investors are forgetting that Nextel is
not maintaining the status quo in an effort to hold on to its market share. Nextel is on track for its nationwide
deployment of its Direct Connect product with phase one to occur in 4Q02, and to be completed by mid-year
2003. This is not news as the company has disclosed this timeline for some time, but rather an affirmation of
the success of the trials and the naming of the specific markets where the implementation will begin. The
expansion outside of local markets will increase its differentiation and usefulness to its core customer base
(over 94% of NXTL customers use Direct Connect) and it will allow the company to introduce the product to a
new segment of the market. Phase one will begin in 4Q02 in Boston and NY and allow customers to use Direct
Connect in the markets in which they travel. Several other cities across the country will follow Boston and NY in
early 2003. By mid-year 2003, after completion of phase one, Direct Connect will be extended nationwide,
allowing customers to contact other NXTL customers across the country. As a reminder, over 150 million Direct
Connect calls are made daily (50 billion in 2001), and all phones sold by Nextel have the Direct Connect
functionality. With the rollout of nationwide Direct Connect in mid-2003, we expect Nextel to introduce a tiered
pricing strategy, which we estimate will be in the $10- to $15-per-month range. We anticipate the early
adopters will be nationwide salesforces and the construction/manufacturing industries with multiple locations.
In addition to nationwide Direct Connect, on January 10 of this year, Nextel, Motorola, and Qualcomm announced plans to develop a Direct Connect product for global wireless operators using CDMA platforms. Specifically, Nextel and Qualcomm have entered into an exclusive eight-year licensing agreement for development and deployment of Qualcomm's QChat software. This new product will enable Nextel to maintain its product differentiation in the U.S., by securing the most credible threat to Direct Connect, the QChat software. Motorola will integrate the infrastructure solution for use in CDMA networks to ensure the interoperability necessary for Nextel's Direct Connect. The beta version of Q-Chat is scheduled for delivery in the first quarter of 2003 with the final version in 2Q03. However, this version of Q-chat needs to operate on cdma2000 1x release A. As a reminder, current 1x networks are Release 0 and we do not expect Release A to be implemented in 1H03. Because of the required network upgrade, we do not believe that there will be a commercial product until 2004.
We believe this deal is very advantageous to Nextel in that it locks down QChat from being marketed to any
domestic CDMA carrier without Nextel's consent. As we have mentioned above, Sprint PCS, Verizon Wireless
or any CDMA carrier in the U.S. interested in push-to-talk functionality will have to either find an alternative to
QChat and Direct Connect or negotiate with Nextel to obtain licensing rights, which we do not believe will come
cheap. Also, the agreement reconnects Nextel with the CDMA migration path, which, we believe, the company
would have chosen if their financial position were cleaner a few years ago. In addition, many investors have
expressed concern over Nextel losing its product differentiation with the rollout of 3G 1X networks that can offer a voice over IP alternative product such as QChat. We now believe Nextel has increased its stranglehold on
the mobile workforce space by eliminating a reasonable facsimile to Direct Connect and forced its competitors
to turn to vendors that do not have a track record with push-to-talk functionality. This agreement puts together
the best intellectual property in the CDMA world Qualcomm, with the intellectual property in the push-to-talk
world.
Competitive products are coming to the market, but we believe most of the risk to Nextel will be the headline
risk associated with a launch and we believe any sell-off related to competitive PTT products will create an
attractive investment opportunity. If the challenges laid out above come to fruition, then, over time, Nextel will
have a difficult decision to make. Specifically, they could choose to license the technology in the U.S. to the
other carriers similar to Qualcomm in the CDMA world, and get a piece of every PTT sale, while giving up its
differentiation and becoming another wireless carrier competing on price. Or they could choose the alternative,
and continue to go at it alone and maintain its differentiation until a carrier comes up with a technology solution.
In our view, we are not convinced that other carriers are making PTT a number-one priority, like Nextel. They
all face significant challenges in front of them with network migrations, increasing margins and free cash flow,
and even potential consolidation. We believe that new product launches from competitive carriers will not
change their focus to PTT and, as such, will give Nextel a distinct competitive advantage throughout 2003.
Appendix:
Direct Connect Technology Primer
Even though we believe the description of a Direct Connect call becomes fairly technical, we thought it would
be helpful to investors to walk through the stages of a Direct Connect call. Below we highlight the 9 steps
during a typical Direct Connect call:
1. User #1 initiates a Direct Connect call by dialing number(s) or selecting from a list of names, then pressing
and holding the push-to-talk (PTT) button on the side of the phone.
2. Phone #1 communicates with the nearest cell site, identifying the sending and targeted user's information.
3. Site #1 forwards the call request to the Metro Packet Switch (MPS), which is simply a Frame Relay WAN
connecting cell sites, Dispatch Application Processor (DAPs), and Packet Duplicators (PDs).
4. The call request reaches the DAP. The DAP is a database containing user information, the fleet (or group of
units) they belong to, whether or not they subscribe to dispatch service, and who they are capable of calling.
The DAP knows what units are on, and where they are on the network.
5. The DAP sends a page out to the sites where it thinks the target (user #2) is located. Unit #2 hears the page,
and notifies cell site #2 that it is ready to accept a call. Site #2 sends the acceptance message through the
MPS to the DAP.
6. The DAP now assigns a packet duplicator to the call, and gives unit #1 permission to start transmitting voice.
Sites #1 and #2 assign RF traffic to the call and unit #1 notifies the user with a chirping sound to go ahead and
begin talking. The process up to this point is less than 1 second to complete the initial call setup.
7. Voice (in the form of digital packets) leaves unit #1 and travels to cell site #1, through the MPS, to the
assigned packet duplicator.
8. The PD makes a copy of the voice packet, and sends it to cell site #2, where it is sent to unit #2.
9. When user #1 lets go of the PTT button, user #2 may talk (by pressing the PTT button). Voice packets are
sent from phone to site, to the PD, and back to another site as required.
Source: Nextel Communications
Company Description
Nextel Communications is the leading provider of enhanced specialized mobile radio services (ESMR) in North
America, directly serving more than 9 million digital subscribers. Nextel and Nextel Partners, Inc. currently
serve 100 of the top-100 U.S. markets covering approximately 196 million POPs.
Nextel Communications, Inc. |
