page image

CHAPTER IV - Other Approaches for Improving Wireless Performance

In addition to the three strategies discussed above, there are others that can be used to improve performance, according to Dennis Roberson, Vice Provost for Research at the Illinois Institute of Technology. He noted that each offers attractive benefits, but also has limitations:

  • Offloading cellular traffic onto Wi-Fi. If all wireless data traffic had to be accommodated on cellular networks, user demand for connectivity would almost certainly have outstripped the system’s capacity. But fortunately, there has been a widely available alternative—Wi-Fi, which offers two main advantages: it typically supports higher connection speeds than cellular networks, and it is generally free. Since Wi-Fi uses unlicensed spectrum, there is no definitive count of the number of hotspots in operation, but there are certainly millions of them in homes, offices and public locations. In 2015, for the first time, more than half (51 percent) of all wireless data traffic was carried over Wi-Fi rather than cellular networks.

    Wi-Fi is not only widely deployed (Roberson noted that from a middle floor of the Sears Tower in Chicago, it is now possible to “see” some 6,000 Wi-Fi hotspots!), but its performance continues to improve through repeated technology upgrades. A substantial amount of spectrum has been allocated for unlicensed use, and many mobile devices, including smartphones, have the ability to switch from a cellular to a Wi-Fi network. Currently, portions of the spectrum where Wi-Fi operates—especially in the 2.4 GHz band—is congested, which limits the ability of Wi-Fi to offload more cellular traffic. However, more capacity will be added as 5 GHz Wi-Fi comes online, and the FCC is considering allocating additional spectrum for unlicensed uses such as Wi-Fi.
  • Spectrum Sharing. An emerging option for increasing capacity is the shared use of specific spectrum bands by different users. The 2012 report from the President’s Council of Advisors on Science and Technology (PCAST) recommended that the federal government take the lead in opening up portions of the spectrum that are under its control for sharing with non-governmental users. The report called on the government to identify 1,000 MHz of federally controlled spectrum to create “the first shared-use spectrum superhighway,” and described a number of technical approaches and management schemes for sharing that would enable the U.S. to “move spectrum access from scarcity to abundance.”

    Techniques for sharing spectrum range from the tried and true (e.g., separating transmitters by geography and/or frequency), to sensing (e.g., dynamic frequency selection technology that enables Wi-Fi to avoid government radars in the 5 GHz band) to versions of sharing where a transmitter is given directions from a central database to avoid interference (e.g., TV white spaces data base, LSA/ASA in the 2.3 GHz band), to combinations of the above.

    Some promising actions have already been taken: In May 2015, the FCC authorized the creation of a Citizens Broadband Radio Service (CBRS) that would enable multi-tiered sharing of up to 150 MHz of spectrum in the 3.5 GHz band, which has been used by Department of Defense radar systems and for non-DOD commercial satellite communications. And results from early testing of sharing schemes known as Licensed Shared Access (LSA) in Europe and Spectrum Access System (SAS) in the U.S. suggest the possibility of further increases in spectrum capacity for cellular services.

    Given the need to make more efficient use of spectrum, sharing technologies are likely to continue to develop in sophistication and capability.
  • Millimeter Wave (mmW) and Massive MIMO. The portion of the spectrum from 30 to 300 gigahertz is generally characterized as “extremely high frequency” or millimeter wave (mmW) bands. According to the FCC, these frequencies “have historically been considered unsuitable for mobile applications because ofpropagation losses…and the inability of mmW signals to propagate around obstacles. [However,] technological advances hold promise of unlocking the potential of using mmW bands for mobile uses.”

    In fact, the characteristics of mmW that had previously seemed unattractive—short transmission paths and high propagation losses—could usefully contribute to supporting high performance transmission in small cells that can accommodate larger numbers of users. In addition, Massive MIMO antennas (arrays made up of hundreds or thousands of small antennas) are well suited to mmW signals and can be easily adapted to fit conventional mobile devices. mmW is likely to be a key element of 5G to support applications requiring very high capacity.

    On the other hand, while mmW is evolving rapidly, the technology is still in its infancy and is currently expensive to deploy and use. Although mmW will play a prominent role in improving wireless performance, particularly in areas with concentrated high demand, it will not fully replace the need for additional low band spectrum for 5G services.
  • Bi-directional transmission. The ability to support full-duplex (bi-directional) operation on a single channel immediately doubles the data carrying capacity of an existing channel. The capacity has now been demonstrated for a radio to cancel out a high powered local transmit signal, enabling it to “see” a weak signal from a distant transmitter. The technique promises large gains in spectral efficiency, but the technology is relatively immature and still expensive. In addition it is unclear how broadly applicable this technique will be, for example, in environments where transmission is primarily in one direction (i.e., asymmetrical, as in the case of streaming video content).

Other potential approaches to expanding capacity and increasing performance include carrier aggregation, which combines multiple, smaller bands of spectrum into a single bigger band; hybrid services, such as LTE-U and LAA, that augment cellular services operating in licensed spectrum with unlicensed spectrum to improve performance; and updating policies that set limits on interference to reflect the use of new technologies to enable closer packing of wireless systems in terms of frequency, space or time. Other innovations such as Software Defined Networks (SDNs) and Network Function Virtualization (NFV) that operate at higher levels of the wireless telecom protocol stack hold promise of granting additional flexibility for 5G networks, enabling them to support advanced services and provide capacity on demand.

Looming Deficits
Despite the promise of these technical and regulatory approaches to improving spectral efficiency and expanding the availability of spectrum, mobile usage will have to overcome shortages of some key wireless resources. For example, a few years ago, the FCC projected that strong, wireless data growth would lead to a “spectrum deficit” as early as 2013. Another looming problem is a shortage of backhaul—the side of the network that connects wireless users (usually by a wired connection) to the core network. A 2013 study by the firm Strategy Analytics projected a potential shortfall of up to 16 petabytes in backhaul capacity.

Growth in wireless traffic has primarily been met by implementation of succeeding generations of technology, each substantially more capable than the previous generation. Two key performance metrics for these standards are the maximum data rate, which has moved from narrowband to broadband speed, and latency, the inherent delay between the time a signal enters and exits a network. Data rate determines the type of media that can be supported by a network, with rich media like video requiring broadband speeds (and lots of capacity). Latency is directly connected to the “naturalness” of communication between parties on a network, and also determines the responsiveness of online games. And it is critically important for applications related to such things as support for autonomous vehicles, which obviously depend on very quick response times to operate safely.

Wireless Network “Generations”

Generation

Technology Standard

Capabilities

Maximum data rate

Latency

Year introduced

1G

AMPS

Analog voice

<10 Kbps

1979 (Japan)
1983 (US)

2G

GSM, CDMA

Digital voice

64 Kbps

1991 (Finland)

2.5G

GPRS, EDGE

Digital voice + data

144 Kbps

600 ms

2001

3G

WCDMA, EvDO

Circuit switched data

2 Mbps

200 ms

2002
(S. Korea, US)

3.5G

HSDPA

Packet switched data

20 Mbps

170 ms

2008

4G

LTE

Wireless broadband

450 Mbps

<100 ms

2009 (Sweden)

5G

IMT-2020

Flexibility to address multiple use cases

> 1 Gbps

< 1 ms*

2020

* This is envisioned RAN latency, not e2e

Share On