I've been involved in 5G preparations and roll-out for over a year now. It seems to be going rather well for our Telco friends. Verizon launched in it's 10th US city and EE are planning to be in 16 UK cities before the end of the year.
Of course, the 5G won't cover the cities, only very small parts of the cities - typically a few square miles in and around the financial and retail centre of each. The 5G signal also doesn't penetrate walls, trees, buses, or anything reasonable solid that might stand between you and the small 5G antenna. Unless you are in the retail store of the carrier whose 5G smartphone you have, it isn't going to work.
The 28GHz high frequency bands used to achieve the fastest 5G speeds simply don't allow for much more than a few hundred metres between your phone and the cell tower, in a straight line, on a clear day. If you throw in a street corner, some foliage, and a few other people using the signal, then that drops to not-very-far-at-all.
So the Telcos need to put out 1,000s more small 5G radio cells, either on existing posts or their own new posts. And to give you the speed of network to support that gigabit per second speed, all of those poles need to have fibre-optic connection back into the Telco core network. That sounds expensive, because it is. Estimated at 3x to 5x more expensive than rolling out 4G.
So that's why only city centres, and business parks are likely to see 5G anytime soon. The problem of patchy mobile coverage, not-spots and woeful data speeds whenever between cities will worsen.
When asked, the Telcos will tell us that 5G can also work at lower frequencies. 3.5 GHz and even 700MHz have been auctioned for Telco use of 5G in UK and Europe. But despite the glitzy 5G moniker, running it at 3.5 GHz has all the downsides of your home WiFi signal, with very few benefits. 5G radio does allow you to squeeze 15% more data through than 4G LTE. Not exactly a quantum leap. 700MHz will at least give you a few miles of coverage area, even if when you do get online, you might be underwhelmed by the speed.
Should those of us not resident within a mile of London's Oxford Street or Downtown Manhattan despair ? Well, not entirely. The smartphone experience is driven by the tasks you try to achieve, and where you try to achieve them.
The good news is that we all mostly stopped worrying about whether emails and attachments would get to our boss from the train home. If it didn't go immediately, it would go as we slopped for stations in built-up areas. And that non-real-time aspect was key.
We can happily download a Netflix episode in the time it takes to leave Waterloo station, safe in the knowledge that the black-spots through Vauxhall and Sunbury wouldn't affect your viewing. And that is the point. Our real-time expectations are buffered because clever companies that had to deal with patchy coverage have optimised the user experience by using the network when it was available, and allowing brief periods of network disappearance to not affect the fun.
Not every use case works that way. Watching a game live on your smartphone would be great, if you could rely on the network. If you can't then it will be a frustrating experience. Maybe Golf would be okay, but football (soccer), rugby, or anything where missing a few seconds can ruin the experience, needs a real-time uninterrupted feed. Video-conferencing also needs that. Not that many people want to engage in person-to-person facetime-ing whilst on a train, or in the street. But content and audio sharing, as part of a web-conference, for management updates, training programs and workshop sessions is how much of the service industry works today. That can't be done with patchy service, regardless of how fast the connections is when it is up.
The poster-child use-cases for 5G are just that. Aspirational posters about what the future promises. They are about as realistic as the 1960s posters of flying cars and 3 day working weeks. Remote surgery, self-driving cities, minority report-style human computer interaction. None of these have gone past prototype stage. In the meantime the (non-5G) Internet of Things (IoT) is gradually gaining momentum, with networks of sensors and connected objects providing homeowners and companies with key data streams that help them run their lives, homes or businesses. Almost none of these need 5G. Most, in fact, need a cut-down forms of 4G, stripped back to just good coverage and quite low data rates.
Although not as news-worthy as augmented reality headsets and drone deliveries, these IoT applications are likely to have more impact on our daily lives than 5G will, for the next 5 years. Every time you are given an hour slot for an amazon delivery, every time the bus stop screen tells you when the next bus is due, every time the signposts tell you which car parks are full and then the car park tells you which floor has spaces. That's the Internet of Things, and none of it needs 5G.
Of course, the 5G won't cover the cities, only very small parts of the cities - typically a few square miles in and around the financial and retail centre of each. The 5G signal also doesn't penetrate walls, trees, buses, or anything reasonable solid that might stand between you and the small 5G antenna. Unless you are in the retail store of the carrier whose 5G smartphone you have, it isn't going to work.
The 28GHz high frequency bands used to achieve the fastest 5G speeds simply don't allow for much more than a few hundred metres between your phone and the cell tower, in a straight line, on a clear day. If you throw in a street corner, some foliage, and a few other people using the signal, then that drops to not-very-far-at-all.
So the Telcos need to put out 1,000s more small 5G radio cells, either on existing posts or their own new posts. And to give you the speed of network to support that gigabit per second speed, all of those poles need to have fibre-optic connection back into the Telco core network. That sounds expensive, because it is. Estimated at 3x to 5x more expensive than rolling out 4G.
So that's why only city centres, and business parks are likely to see 5G anytime soon. The problem of patchy mobile coverage, not-spots and woeful data speeds whenever between cities will worsen.
When asked, the Telcos will tell us that 5G can also work at lower frequencies. 3.5 GHz and even 700MHz have been auctioned for Telco use of 5G in UK and Europe. But despite the glitzy 5G moniker, running it at 3.5 GHz has all the downsides of your home WiFi signal, with very few benefits. 5G radio does allow you to squeeze 15% more data through than 4G LTE. Not exactly a quantum leap. 700MHz will at least give you a few miles of coverage area, even if when you do get online, you might be underwhelmed by the speed.
Should those of us not resident within a mile of London's Oxford Street or Downtown Manhattan despair ? Well, not entirely. The smartphone experience is driven by the tasks you try to achieve, and where you try to achieve them.
The good news is that we all mostly stopped worrying about whether emails and attachments would get to our boss from the train home. If it didn't go immediately, it would go as we slopped for stations in built-up areas. And that non-real-time aspect was key.
We can happily download a Netflix episode in the time it takes to leave Waterloo station, safe in the knowledge that the black-spots through Vauxhall and Sunbury wouldn't affect your viewing. And that is the point. Our real-time expectations are buffered because clever companies that had to deal with patchy coverage have optimised the user experience by using the network when it was available, and allowing brief periods of network disappearance to not affect the fun.
Not every use case works that way. Watching a game live on your smartphone would be great, if you could rely on the network. If you can't then it will be a frustrating experience. Maybe Golf would be okay, but football (soccer), rugby, or anything where missing a few seconds can ruin the experience, needs a real-time uninterrupted feed. Video-conferencing also needs that. Not that many people want to engage in person-to-person facetime-ing whilst on a train, or in the street. But content and audio sharing, as part of a web-conference, for management updates, training programs and workshop sessions is how much of the service industry works today. That can't be done with patchy service, regardless of how fast the connections is when it is up.
The poster-child use-cases for 5G are just that. Aspirational posters about what the future promises. They are about as realistic as the 1960s posters of flying cars and 3 day working weeks. Remote surgery, self-driving cities, minority report-style human computer interaction. None of these have gone past prototype stage. In the meantime the (non-5G) Internet of Things (IoT) is gradually gaining momentum, with networks of sensors and connected objects providing homeowners and companies with key data streams that help them run their lives, homes or businesses. Almost none of these need 5G. Most, in fact, need a cut-down forms of 4G, stripped back to just good coverage and quite low data rates.
Although not as news-worthy as augmented reality headsets and drone deliveries, these IoT applications are likely to have more impact on our daily lives than 5G will, for the next 5 years. Every time you are given an hour slot for an amazon delivery, every time the bus stop screen tells you when the next bus is due, every time the signposts tell you which car parks are full and then the car park tells you which floor has spaces. That's the Internet of Things, and none of it needs 5G.
No comments:
Post a Comment