Are we running out of wireless spectrum?
Recently I read with surprise an article in The New York Times' business section predicting doom and gloom for the U.S. wireless communications industry. According to the article, several studies submitted to the Federal Communications Commission by the Department of Defense and nonprofit and healthcare organizations indicate that it would be costly and almost impossible to relocate spectrum to accommodate new user demand for broadband wireless. The article implies that we are running out of spectrum; therefore, broadband wireless may be a dead issue.
I find this hard to believe. In wireline communications we have exceeded theoretical bounds (30K bit/sec) to drive copper wire to megabit speeds and will soon achieve gigabit speeds for short distances. I have no doubt that the wireless world will see technology and signal theory overcome the limitations of spectrum allocation.
Currently the mobile wireless world is trying to decide whether to forklift-upgrade to 3G technology or add-on-upgrade to 2.5G and then forklift to 4G. The bandwidth limitations of 3G are becoming self-evident, and the cost of a license plus the upgrade may not be economically feasible based on return on investment. However, with 4G it will be possible to achieve 2M bit/sec or greater bandwidth per mobile user in the same spectrum in use today.
While all the spectrum attention has been placed on mobile wireless, fixed wireless is also becoming a communications industry hot topic. The fixed wireless industry currently relies on expensive local multipoint distribution services technology, which uses licensed spectrum in the access WAN and IEEE 802.11 with frequency hopping to deliver speeds of 1M to 2M bit/sec to the wireless LAN.
Recent developments have produced speeds of up to 11M bit/sec (IEEE 802.11b), with the potential to get to 40M bit/sec (IEEE 802.11a) and eventually 100M bit/sec (IEEE 802.11a extended). These approaches use the unlicensed 2.4- and 5-GHz spectrums. Currently the 2.4-GHz spectrum is used for Bluetooth-embedded system communications and wireless home or office telephones, so there is potential for interference problems.
Two other fixed wireless standards -- IEEE 802.16 for broadband wireless access and IEEE 802.15 for wireless personal-area networks (WPAN) -- are changing the landscape of bandwidth delivery within current spectrum boundaries. WAN local-loop access using IEEE 802.16 in the unlicensed 2.4- and 5.8-GHz spectrum will deliver access speeds of 10M bit/sec or greater per user as a last-mile technology in competition with wireline xDSL. The low implementation cost of this standard, coupled with vendor use of IP and quality of service (QoS), make it a compelling option for wireless access for consumers and small office/home office users in the near term.
The WPAN using IEEE 802.15 is another story when it comes to spectrum utilization. Currently WPANs use Bluetooth at 2.4 GHz within a short-distance radius. However, they will use a new wireless communications technology called Ultra Wideband, currently in the approval process at the FCC. Ultra Wideband uses high-frequency microwave pulses with low-power intensity as a communications mechanism. Ultra Wideband transmissions cross all spectrum layers without interference, and as a security side benefit all transmissions are self-encrypted. Today, the distance cannot be great due to its low power, but per-user bandwidth speeds of 40M to 100M bit/sec can be demonstrated in the laboratory.
Ultra Wideband is a transmission technology that is not intimately tied to a media access control layer; therefore it has the potential of being applicable to IEEE 802.11 and 802.16 with new coding and signal processing research. This lower-layer technology, coupled with the efficiencies of higher-layer IP and the guarantees of QoS, could meet user bandwidth demands over the LAN, WPAN and access WAN.
Ever-advancing research and technology is freeing us from the spectrum conundrum and its political side effects. We may be running out of allocated wireless spectrum, but not the ability to deliver broadband communications capability to a range of users throughout the U.S.