dslreports logo
 story category
The Challenge of the Last Mile

Delivering broadband communications to users is a complicated infrastructure problem. One of the important aspects of this is referred to “The Last Mile” which is the final hookup of the residence to the larger backbone of the network. While there are several solutions, they all must be balanced against considerations including the economics, the ease of deployment as well as the reliability.

Click for full size
While this was a traditional problem that dates back to copper phone lines and voice communications, with increasing broadband speeds it has become an even more critical issue, with attempts to find more modern solutions than traditional copper wires.

This all gets compounded even further in rural areas with the users spread further apart by geography, making the hookups even more costly.

The time honored solution to the last mile challenge was copper wiring. The original form was phone wiring for plain old telephone service (POTS). ISDN (integrated services digital network) was developed back in 1984, and provided speeds of 128 Kbps on both the download and the upload, but it is of historical interest at this point.

Subsequently DSL was patented in 1988 and was able to double the throughput of ISDN to 256 Kbps. DSL is a current last mile solution, and used by some telcos including AT&T U-verse and Verizon DSL. Note that this last mile solution is based on the decaying copper phone network, that is in need of investment in many areas. Also, DSL is limited in that it must be within proximity from the physical plant.

The cable providers use copper wiring as well. However, it is not telephone wiring, but rather their coaxial cable that was already in place to transmit video. By using the various DOCSIS specifications, the coaxial cable can be used to transmit simultaneously both data and voice.

With cable dominating the broadband in the US, coax for the last mile is a popular option. However, there are issues with the signal to noise ratio, with limits to how far the data can be transmitted due to attenuation of the signal in the wire. This is addressed with amplifiers in the coax, but increases the complexity and the power needs of the network.

Enter Fiber to the Home

With the above limitations for using a copper wire for last mile technology, the next solution was to use optical fiber, and to bring it directly to the end user’s residence. This is termed FTTH- fiber to the home, or FTTP- fiber to the premises. This approach is used by Verizon Fios, Google Fiber, and other smaller fiber broadband providers.

The current approach for last mile fiber is G-PON- gigabit passive optical networking. Optical fiber avoids some of the inherent limitations of copper wiring, such as susceptibility to interference, and also may offer greater speeds in the long term.

However, it is quite expensive, and requires signal boosters to maintain the laser signal. Each residence also requires an ONT- optical network terminal. These are larger, and require installation that is more expensive typically than a cable or DSL modem. Be that as it may, Altice recently announced a plan for FTTP, partially driven by electricity costs, as well as speed increases and less maintenance long term.

With the difficulties of running either copper wire or optical fiber, carriers are looking towards wireless solutions. This is also driven by the existing providers not allowing access to the utility poles, which too often ends up embroiled in legal action.

Google has been a prototypical example of the issues with the last mile issue. While hardly an upstart company at this point, they still had to trench fiber to deliver internet access to residences- a method that proved costly in less populated areas.

The Industry's Shift to Wireless

More recently, Google has been testing wireless delivery, using the 3.5 GHz spectrum, and putting transmitters on street lamps. This portion of the spectrum was previously used for Citizen’s Band radio, and was released this year to be used for wireless broadband applications. Google Fiber has currently slowed down their rollouts as they decide if they want to switch to wireless for their last mile.

However, Google Fiber may not be the only wireless solution. Upstart Starry, that so far has only released a wireless router, made some significant promises of 1 Gbps high speed internet access (although more recently they have toned this down to 200-300 Mbps), with the last mile offered via wireless transmission. Their plan was to keep the costs down, by transmitting it via wavelengths of 37 to 40 GHz- in the millimeter wireless band. Starry also claims to maintain affordability with their wireless last mile potentially costing as low as $25 per customer. Verizon Fios is also turning to “Fixed 5G” as their last mile solution for their Boston rollout.

While Google and Starry have their respective wireless last mile solutions in development, in rural areas wireless is already used for the last mile to deliver broadband. This solution gets used when the ISP is unwilling, or at least it is economically disadvantageous for them to run wire to the residence. This is a fixed wireless solution, with an external antenna on the residence, and the transmission on the 2.4 GHz or 5 GHz frequency.

Finally, in the most isolated areas, where there is virtually no infrastructure, satellite technology gets turned to as a last mile solution. Satellite has historically offered slow speeds, at expensive prices, causing users to turn to other solutions for their residences. Satellite has been used for internet on passenger airplane service. With the expenses of satellites, Google has turned to balloons, and Facebook to solar drones as potential ways to provide a signal from the sky.

The challenge of delivering the internet for the last mile remains, and ISP’s are constantly balancing the economic factors against the reliability and ease of deployment. It is intriguing to see the multiple number of options to address this issue. Feel free to discuss your thoughts and experiences on the last mile problem.

This article was contributed by the DSLReports.com community. If you'd like to receive payment for writing content like this for our front page, please drop us a line.

Most recommended from 79 comments



TestBoy
Premium Member
join:2009-10-13
Irmo, SC

21 recommendations

TestBoy

Premium Member

It isn't the tech..it's greed, bureaucracy, collusion, and corruption

We hear the money argument thrown around constantly but it simply does not hold water.
Pulling fiber sis a license to print money.

An example... the IFITL system pulled by a few telcos to some areas.
They put that stuff in the ground decades ago and guess what? they simply change the hardware and reuse the fiber.
What was 10 megs is now multi gigs.

Yes, it's expensive. Costs WILL BE RECOUPED AND THEN SOME.
They just want the costs recouped next quarter and that's not going to happen.

It's got very little to do with cost. It's greed, bureaucracy, collusion, and corruption.
The cost is an issue because of pure greed. Nothing more.
Fiber in the ground is a license to print money.. for the next half a century or more.

Anonec1aa
@antiqueexport.net

8 recommendations

Anonec1aa

Anon

Nice summary of broadband options and how they may develop

Thanks for a nice summary of broadband options and how they may develop to solve last mile connectivity. It looks like 1 single solution is not the answer as some suggest(FTTP only), but a basket of solutions depending on geography and costs.
dstratton
join:2005-06-15
Shelbyville, KY

3 recommendations

dstratton

Member

A correction about 3.5ghz

"...Google has been testing wireless delivery, using the 3.5 GHz spectrum, and putting transmitters on street lamps. This portion of the spectrum was previously used for Citizen s Band radio..."

Er, no. CB radio is 28mhz, not 3500mhz/3.5ghz. And 28mhz is not capable of the kind of bandwidth described.