Receive new posts as email.
This site operates as an independent editorial operation. Advertising, sponsorships, and other non-editorial materials represent the opinions and messages of their respective origins, and not of the site operator or JiWire, Inc.
Entire site and all contents except otherwise noted © Copyright 2001-2006 by Glenn Fleishman. Some images ©2006 Jupiterimages Corporation. All rights reserved. Please contact us for reprint rights. Linking is, of course, free and encouraged.
Wi-Fi Alliance works to settle issues preventing VoWLAN rollout: The trade group is, as usual, ahead of the IEEE: markets come before standards. The 802.11e task group which has been working for years on packet prioritization and related issues to ensure good VoIP calls and non-stuttering streaming audio and video over Wi-Fi is nearly done. The WMM (Wi-Fi Multimedia) spec from the Wi-Fi Alliance resembles WPA: it’s not the complete standard, but it’s a useful enough stub to push out into devices and certify.
The goal is to allow multiple handsets to work with a single Wi-Fi access point without overwhelming its capacity, and to have handsets that have enough juice in them to work for the very long periods of time expected from normal cordless phones.
This article notes that WMM hasn’t seen much uptake, despite its inclusion of packet prioritization, key for allowing voice data to supercede other data as needed for audio continuity. SpectraLink added it, but all VoWLAN and VoLAN vendors have already built their own quality of service (QoS) into their proprietary systems, so may have little need to provide early interoperability with an in-progress standard until its customers demand it.
The Register reports that LG will ship the first UMA handset: UMA uses a local link to connect to a voice signal to an existing cellular infrastructure. Many analysts expect cell operators to push UMA as a way to move load from cell towers to home and office Wi-Fi (and even wired) networks. LG’s LG-CL400 handset should be the first to hit the U.S. market, Andrew Orlowski reports.
The article says that BenQ’s UMA PocketPC handset is slowly coming to market, while Nokia and Symbian will work with Kineto Wireless to produce UMA devices.
The key difference between Internet telephony using VoIP and UMA is that UMA doesn’t transit voice data over the public Internet. UMA will probably require partnerships with service providers to be entirely effective.
South Korea’s second-largest broadband provider is testing Wi-Fi phones: The company will spend the next month testing the service with a commercial roll-out next year. Wi-FI phones make a heck of a lot more sense in South Korea, a smaller country than the U.S. with a more densely packed population and vastly more hotspots per capita.
Broadcom’s BCM1161 processor integrates array of common features: The single-chip VoWLAN processor includes USB, an LCD, and analog codecs for microphone and speaker connections. It supports a 2 megapixel digital camera, recording audio, the playback of audio and video clips, and 3-way and speaker phone calls. It doesn’t include Wi-Fi on board, which is hard to figure from the coverage and press release. Broadcom offers Wi-Fi in a single chip to complement this VoIP offering.
Broadcom is also offering a reference design for integrating their product with a Wi-Fi chipset into a fully functioning handset.
The eBay acquisition of Skype has led to more talk about Wi-Fi and VoIP: Don’t buy into the hype that cell handsets will integrate Skype and Wi-Fi so that a single phone will let you call at very cheap rates (or free) from anywhere a Wi-Fi signal penetrates. Ain’t going to happen short term; long term, the entire telephony infrastructure is changing and both Wi-Fi and Skype will be part of it.
There are three kinds of Wi-Fi + VoIP combinations: pure VoWLAN, which are IP phones used in an enterprise with a robust infrastructure in which the phone is essentially exactly like a wired, conventional phone—only better; roaming VoWLAN in which a phone can be switched to a mode in which it can be used at home or at work, but only on known networks; and hotspot VoIP, in which secret sauce is inserted into a phone’s software to allow it work at hotspots.
The first market is already well developed, and it’s grown like mad in recent years. The increased security now available for enterprise Wi-Fi networks means that objections to offering VoIP over the WLAN have evaporated. The calls can be kept secure at several layers, making the voice traffic as secure as that going over a wired PBX setup or the PSTN.
The second market is starting to emerge in which people will carry phones that have cell standards and Wi-Fi. The VoWLAN feature will allow a few networks to be characterized, and the infrastructure won’t be robust within the enterprise, but rather be outsourced. Calls are primarily made to and from the PSTN, if not exclusively.
The third market is what has people holding their breath. When you look at companies like T-Mobile, which has a very poor U.S. 3G plan compared to its much larger competitors, Wi-Fi hotspots become their only tool to expand customer base and revenue. Further, they’re the only domestic hotspot network to have rolled out 802.1X security, which allows unique logins and unique encryption keys to be assigned over the local wireless link to each user on the hotspot network.
They’ve already scored a consumer electronics deal with Kodak, which will ship software later this year for their already-delayed EasyShare-One camera with Wi-Fi: the camera will be able to use 802.1X to log into T-Mobile’s network, minimizing network interface configuration.
Right now, most hotspots networks employ their own authentication schemes that allow a user to be connected to access. T-Mobile and iBahn are the only networks of scale that I know have 802.1X deployed. T-Mobile has it built into their own client; iBahn makes it available through standard 802.1X clients (found in Windows XP and Mac OS X 10.3 and 10.4).
Authentication remains the boring issue that has prevented the kind of growth (until lately) predicted for hotspot use. Authentication creates walled gardens for each user base. Companies like Boingo, iPass, and GoRemote have had to build client software that handles the authentication differences behind the scenes in order to aggregate the dozens of distinct hotspot networks in their system.
Boingo has experimented with VoIP over Wi-Fi hotspots through Vonage, but you had to be an existing Vonage subscriber, have a Vonage soft-phone subscription, and be a Boingo subscriber. The Boingo/Skype deal is essentially a pricing relationship that allows the use of Skype only on hotspots that Boingo aggregates for less money.
All these pieces to make VoHS (VoIP over Hot Spot?) work means that there’s a large software component and many intermediate pieces involved in the chain from voice conversation out to the Internet over an arbitrary hotspot.
Essentially, lack of roaming across all major networks at a reasonable price prevents VoIP over hot spot from becoming a near-term ubiquitous reality. Cell handsets aren’t easily upgradable, and it’s likely that the kind of software necessary to allow a handset with cellular and Wi-Fi support built in also will require frequent updates.
Which brings us to SK-EarthLink. The partnership between the largest telecom company in South Korea, and one that deploys some of the most advanced handsets and devices on their network in the world, and EarthLink bodes well for VoIP over hot spot. Because SK-EarthLink will be a MVNO (mobile virtual network operator), reselling access probably from Sprint PCS’s voice and 3G network, it will be in their interest to offer the best price with the most advanced features using the most conserved cost.
You can see the convergence. While Boingo is apparently just one of the suitors to be part of the SK-EarthLink deal—Boingo/EarthLink founder and SK-EarthLink CEO Sky Dayton has to recuse himself from business deals that involve Boingo—it’s likely to use its aggregated portfolio combined with existing VoIP-in-the-field research to make a good offer.
The handsets that SK-EarthLink offers could be the real portent of the future of this third VoIP over Wi-Fi market: the handsets could, one hopes, have cell voice, 3G data, and Wi-Fi built in, switching voice and data to the cheapest available network, whether one’s home, work, or hot spot. If SK-EarthLink can achieve that, they could represent the first real convergence in a way that benefits consumers—and threatens the foundation of the entire phone business. Ironic, then, that they’ll be buying service from Sprint to do it.
Techworld takes a look at a phone that can use the cheaper network: The CiceroPhone turns a Pocket PC device into a telephone, and it can route calls over a Wi-Fi network when it’s available, choosing cellular when not. The intent isn’t for hotspot use, but rather in making a single device roam across a user’s common working networks, such as home, remote offices, and possibly intra-office. Cicero claims its phone is first to dial out over both networks and to switch between them dynamically.
Peter Judge, author of the above article, also separately reviews the CiceroPhone, and finds it quite good and flexible, although voice quality needs improvement. The fact that any SIM can be inserted and you can configure which calls by phone number are routed over which network is a good glimpse of what the future could be like on a dedicated handset—if carriers and handset makers allow it.
Interesting outlook in Time Europe on how Nokia, Motorola, et al., are claiming their own position: Using VoIP and Wi-Fi handset makers can reduce their dependence on cellular operators. Customers want VoWLAN phones or at least phones that can use Wi-Fi as one option for transferring data and handling voice. In 2010, mobile revenue will be $550 billion—and the handset makers can reassign where that money goes.
Nokia says that next year, Wi-Fi will be a standard feature in its multimedia and business phones, the article notes, as well as midrange models. Motorola should be integrating Skype software with some phones next year. In-Stat’s Alan Nogee—a smart analyst—estimates a huge growth in handsets with Wi-Fi: 13.5 million in 2007, 52.8 million in 2008, and 136 million by 2010. The article notes this might be conservative and I agree: as with Wi-Fi growth curve, it either explodes or fails.