Wi-Fi将死?实在太夸张了

预测是件很有意思的事。一半的预测都是错的，特别是在技术领域。比尔·盖茨有一项很著名的预测，那就是没有人会开发出32位操作系统。1943年，IBM董事长托马斯·沃森说：“我觉得全球计算机市场的规模可能是5台。”另外，谁能忘得了iPhone问世时微软首席执行官史蒂夫·鲍尔默的预测呢：“iPhone的市场份额根本不可能有多大。”

“预测未来很容易；难的是预测正确。”这句谚语是有道理的。因此，最近出现的Wi-Fi将消亡的预测引发怀疑并不意外。

无限数据方案、各式各样的LTE-U、市民宽带无线电服务（CBRS）以及5G的出现都被急火火地视为潜在的Wi-Fi杀手。然而，尽管在快速发展的无线通信领域任何事情都有可能，但借用马克·吐温的一句话，说Wi-Fi已经死了，是极度的夸大其词。

2006年以来，思科的年度可视网络指数（VNI）一直是预测蜂窝和Wi-Fi网络流量的黄金指标，其准确性获得了普遍认可，而且被视为研究无线通信行业的全球标准。

最新的VNI预计，2021年全球公共Wi-Fi热点总量将是2016年的6倍。单北美地区就会有大幅增长，到2021年公共Wi-Fi热点数量将增长3倍，达到9370万个。此外，市场研究机构Markets and Markets预计，作为服务，到2021年Wi-Fi市场规模的年均复合增长率为38%。

如果说Wi-Fi是即将被一颗巨大小行星毁了的恐龙，那么所有主要行业分析师就都遗漏了这个问题。

实际情况是仅凭LTE不可能应付流媒体和共享等带宽密集型活动产生的海量数据。在机场和体育馆等人群摩肩接踵的高密度地区尤其如此。这样的环境下手机信号很差，难以有效保持带宽，这就让Wi-Fi成了可以跟LTE抗衡的较可靠解决方案。

在如今的重要地点，蜂窝和Wi-Fi网络均并行存在。电信运营商用Wi-Fi来满足巨大的移动数据需求。Sprint目前正在将用户的LTE流量转移给Wi-Fi，预计将有更多通信公司布其后尘。思科的2017年VNI预计，到2021年，64%的智能手机流量和72%的平板电脑流量将转移到Wi-Fi网络。

市场咨询机构ABI Research认为，物联网开始崭露头角也会给Wi-Fi带来有利影响。

预计到2025年，物联网每年将产生11.1万亿美元的经济影响。这可是万亿美元。Wi-Fi是物联网设备的生命线。爱立信最近公布的数据显示，到2020年蜂窝网络的覆盖面可能不到北美物联网的15%。绝大多数物联网都可能在非授权类通信技术的支持之下。目前，Wi-Fi连接着家庭中的所有设备——Alexa语言助手、电灯、Sonos音箱和门铃，所有大家能想到的东西。今后，在制造、交通运输和公用事业等领域，Wi-Fi还可能成为所有使用物联网的企业做出变革的必须手段。

和传统LTE相比，Wi-Fi的基础设施成本较低，延迟较短，吞吐量较大，可以服务于所有电信运营商的终端，而且较容易部署。这让Wi-Fi成了室内必备网络，无论是在家里，在写字楼，还是在工厂。

无限移动数据方案将宣告Wi-Fi之死的预期从1995年就开始了。3Com创始人罗伯特·梅特卡夫当时说：“我觉得互联网很快就会出现大爆发，但会在1996年出现灾难性崩溃。”还记得几年前吗？那时蜂窝网络不限流量，但消费者还是选择用Wi-Fi。此外，数以亿计的设备都没有LTE连接，它们都需要Wi-Fi。

Wi-Fi在5G和CBRS等技术领域也有一席之地。

5G是4G的同胞兄弟。作为最新移动通信标准，它更年轻、更快、更强大。虽然预计到2020年以后才会出现，但5G的速度将远远超过4G，理论下载速度可达到每秒1万Mb。那么大家就会问了，5G已经那么快了，我们为什么还需要Wi-Fi呢？

5G将基于多个频段的累加。无线通信行业基本上已经同意把LTE、Wi-Fi以及其他未授权类通道都融合在一起。如果授权类（LTE）和未授权类（Wi-Fi）频段不能共存，5G从经济上讲就不可行。换句话说，5G并不侧重于哪种技术，它是一个整合型网络，通过更多的频段来吞吐更多数据。

然而，5G系统开始部署时，下一代Wi-Fi，也就是802.11ax也将可以投入市场。11ax的加持将极大提升Wi-Fi网络的效率。更具体地说，11ax将解决此前Wi-Fi技术缺乏上行链路调度的问题，人们一直强调，这种能力是未授权类频段使用LTE网络后形成的3GPP生态系统的主要好处之一。也就是说，随着802.11ax的推出，3GPP和IEEE 802.11 MAC两种技术标准的差异就会消失。

CBRS取代Wi-Fi的说法也会让人产生被误导的预期。这两项技术相互补充。可以把CBRS视为使用3.5GHz频段的Wi-Fi，而Wi-Fi使用的是2.4 GHz和5 GHz频段。基于LTE的CBRS解决方案可实现户外最大覆盖，这就有了经济性，而Wi-Fi仍将适于企业市场和家庭网络。值得注意的是，在现有频段部署CBRS的做法主要可用于美国市场，而且其成本要比Wi-Fi高得多，因为CBRS需要用新的LTE设备全面重建基础设施。

客观地说，Wi-Fi确实有缺点。公共网络带来安全顾虑，使用公共网络所需的用户名/密码登录过程也不让人满意。Passpoint等新技术提供了最高级别的公共Wi-Fi加密，解决了令人讨厌的登录问题，而且不需要用户干预就可在LTE和Wi-Fi之间来回切换。

1966年，《时代》杂志做了一次著名预测：“虽然完全可行，但远程购物会彻底失败。”虽然用了40年才起步，但“远程购物”现在已经站住了脚。Wi-Fi的情况也是如此。更明确地说，在如今的通信领域和明天的融合世界，授权类和未授权类频段都需要一席之地，而且最为确定的是，Wi-Fi将是其中的一部分。就像LTE不断发展一样，Wi-Fi也将不断进步，从而支撑起一个无缝连接的世界。

Wi-Fi不会死。Wi-Fi万岁。（财富中文网）

译者：Charlie

审稿：夏林

德里克·彼得森是移动网络连接服务提供商Boingo Wireless首席技术官以及科罗拉多理工大学客座教授。

Predictions are a funny business. Odds are you’re going to be wrong half the time—especially when it comes to technology. Bill Gates famously predicted nobody would ever make a 32-bit operating system. Thomas Watson, chairman of IBM, said in 1943: "I think there is a world market for maybe five computers.” And who can forget former Microsoft CEO Steve Ballmer’s prediction when the iPhone launched: “There’s no chance that the iPhone is going to get any significant market share.”

There’s a reason for the adage “Predicting the future is easy; getting it right is the hard part.” So it comes as no surprise that skepticism abounds surrounding the recent predictions that Wi-Fi will go the way of the Dodo.

Unlimited data plans, LTE-U with all its flavors, Citizens Broadband Radio Service (CBRS) and the rollout of 5G have all been trotted out as potential Wi-Fi killers. And while anything is within the realm of possibility when it comes to the fast-moving wireless space, to borrow a line from Mark Twain, the reports of Wi-Fi’s death have been greatly exaggerated.

Since 2006, Cisco’s annual Visual Networking Index (VNI) forecast has been the gold standard for cellular and Wi-Fi traffic projections. It’s widely respected for its accuracy and is considered the global standard for research in the wireless industry. As Cisco’s says, it's projections have been accurate within 10% over the last 11 years.

Cisco’s latest VNI predicts that globally, total public Wi-Fi hotspots will grow 6-fold from 2016 to 2021. In North America alone, the growth will still be significant, spiking 4-fold to 93.7 million by 2021. In addition, Markets and Markets has pinpointed a CAGR of 38% for the Wi-Fi as a Service market size by 2021.

If Wi-Fi is a dinosaur about to be destroyed by a giant asteroid, every major industry analyst missed the memo.

The reality is it is not possible for LTE alone to handle the onslaught of data caused by bandwidth-intensive activities like streaming and sharing. This is especially true in high-density locations, such as airports and stadiums, where people are sitting “hip to hip." Cell signals struggle in these environments to effectively transport enough bandwidth, making Wi-Fi a more reliable solution that can complement LTE.

Cellular and Wi-Fi are already deployed alongside each other in large venues today. Carriers leverage Wi-Fi to accommodate the tremendous demand for mobile data. Sprint currently offloads customer LTE traffic over to Wi-Fi and more carriers are expected to follow suit. Cisco’s 2017 VNI predicts the amount of traffic offloaded from smartphones to Wi-Fi will be 64% by 2021, and tablets at 72%.

Wi-Fi will also see a positive impact from the dawn of the Internet of Things (IoT), according to ABI Research.

By 2025, IoT is estimated to have an economic impact of $11.1 trillion a year. Trillion. Wi-Fi is the lifeblood for IoT devices, with recent data from Ericsson suggesting that cellular-based networks could be used to provide less than 15% of overall IoT connectivity in North America by 2020. The vast majority of IoT could be supported by unlicensed technologies. Today, Wi-Fi connects all the devices in a person’s home—Alexa, lightbulbs, Sonos, doorbells, you name it. Tomorrow Wi-Fi could be a go-to for revolutionizing the IoT-powered enterprise across manufacturing, factories, transportation, utilities and more.

Compared to traditional LTE, Wi-Fi has a lower cost of infrastructure, reduced latency, more throughput, can serve all endpoints respective of carrier affiliation, and is easier to deploy, making it the must have for indoor coverage, whether it’s for a home, office building or industrial plant.

Hearing predictions that unlimited mobile data plans will be the cause of Wi-Fi’s demise hearken back to 1995, when Robert Metcalfe, the founder of 3Com said, “I predict the Internet will soon go spectacularly supernova and in 1996 catastrophically collapse.” Remember a few years ago when cellular plans used to be unlimited—and consumers still turned to Wi-Fi? In addition, billions of devices don’t have LTE connectivity; they need Wi-Fi.

Wi-Fi also has a place in technologies like 5G and CBRS.

5G is 4G’s younger, faster, and stronger brother—the next and newest mobile wireless standard. While 5G isn't expected until 2020, it will be significantly faster than 4G, allowing for a theoretical download speed of 10,000 Mbps. So if you’re wondering if that is fast, why do we need Wi-Fi?

5G will be based on the unified aggregation of multiple bands. The wireless industry has largely accepted that this will include the convergence of LTE and Wi-Fi or other unlicensed channels. 5G is not economically practical without coexistence between licensed (LTE) and unlicensed (Wi-Fi) spectrum. In other words, 5G does not prefer one technology over another, but rather one consolidated network that handles more data with more spectrum.

However, by the time 5G systems start getting deployed, the next generation of Wi-Fi, 802.11ax, will be ready for market. 11ax enhancements will significantly increase the efficiency of Wi-Fi networks. More specifically, 11ax will address the lack of up-link scheduling in previous versions of Wi-Fi, a capability that has been highlighted as one of the primary benefits in the 3GPP ecosystem of LTE operation in unlicensed bands. This would suggest that the divergences between 3GPP and IEEE 802.11 MAC designs are set to diminish with the introduction of 802.11ax.

The argument of CBRS’ takeover of Wi-Fi also brings a misinformed perspective. The technologies are complementary. CBRS can be viewed as another band of Wi-Fi, operating in the 3.5 GHz unlicensed channel alongside Wi-Fi at 2.4 GHz and 5 GHz. LTE-based CBRS solutions can maximize outdoor coverage where it makes economic sense, whereas Wi-Fi will remain the suitor for connectivity in the enterprise market and at home. It’s worth noting that the deployment of CBRS with the current defined spectrum is primarily available for the U.S.market and that it comes at a much higher cost than Wi-Fi, as it will require facilities to overhaul infrastructures with new LTE equipment.

To be fair, Wi-Fi does have its shortcomings. Public networks bring to bear security concerns and an unsatisfactory log-in process of user names and passwords when using public networks. New technologies like Passpoint offer the highest level of public Wi-Fi encryption and solve tedious log-in requirements, switching devices back and forth from LTE to Wi-Fi without user intervention.

Back in 1966, Time Magazine famously predicted, "Remote shopping, while entirely feasible, will flop.” While it took another 40 years to really take off, “remote shopping” is here to stay. The same is true of Wi-Fi. To be clear, both licensed and unlicensed spectrum need a seat at the table in the connected world of today and the converged world of tomorrow. But Wi-Fi will most certainly be part of that equation. Even as LTE evolves, Wi-Fi too will evolve to power a seamlessly connected society.

Wi-Fi is not dead. Long live Wi-Fi.

Derek Peterson is chief technology officer of Boingo Wireless. He is also an adjunct professor at Colorado Technical University.