3 ms·
In order of your responses: ``3'' : "Work well" for what? I was specifically addressing comms, as that's where the problem currently exists. There are too many
by iheartmemcache 10y ago
In order of your responses:
``3'' : "Work well" for what? I was specifically addressing comms, as that's where the problem currently exists. There are too many concurrent protocols/specs for me to keep up with LoRA (I'm only familiar with the 3GPP standards i.e. the NarrowBand IOT analog to LoRA, for obvious reasons), but I did some reading[1]. It has some awesome features like low-power consumption that allows long range, but again, you're not going to be streaming Youtube videos via that protocol. Their site officially says[2] their solution "is a network ideal for Internet of Things (IoT), metering, security, asset tracking, and machine-to-machine (M2M) applications." So while a great solution for, say, plopping down a few thousand soil hydrometers, it's not going to help with the over-saturation of the 4G networks.
``1'': Spectrum efficiency varies as a function of carrier wave length[3 for a brief overview] There's an information theory limit as to how much data you can effectively transport (throughput, not latency) limited by the Shannon-Hartley limit. (Or maybe the Nyquist something-something...forgive my ignorance on the topic, I really should have went to grad school or at least taken more physics classes.) To offer more bandwidth you can do plenty of things without resorting to mmWave, i.e. picking up rights to a carrier wave when Auction 73 occurred, or use OFDM amongst the ISM bands
Which brings me to (``2''- at least, as I understand it). What providers are trying to do is use the standard UWB/OFDM that already has well established, cheap commodity hardware behind it (think: a Cisco Aironet equivalent). The poster at the start of the thread said >> 2. Users have shown they aren't willing to pay more wireless. And for most things consumers do like VR and 360 video etc, wifi is fine and cheaper.
I was simply saying that any new technology will be bundled into the next revision of phones. It's not going to cost the consumer any more than usual, because (at least in the US) the 24-month sign+subsidized phone practice is commonplace enough that the next-gen technologies will be bundled in at the same rate.
I'm right there with you on mmWave - if you're looking to replace 4G with that, well, probably not viable at the moment as people still tune transmission lines using machined components (expensive, can't really use powdered metal bulk solutions when you need .1 thou tolerances..) that are designed filter cavities like this[4] in one offs (even more expensive without economies of scale) by generally those with graduate degrees. I haven't heard any ballpark quotes for 5G pricing from the usual suppliers.. who's quoting what, and from where?
[1] https://www.ietf.org/mail-archive/web/lp-wan/current/pdfbVG486gGVD.pdf https://www.ietf.org/mail-archive/web/lp-wan/current/pdfbVG4...
[2] http://www.semtech.com/wireless-rf/lora.html http://www.semtech.com/wireless-rf/lora.html
[3] https://en.wikipedia.org/wiki/Channel_capacity https://en.wikipedia.org/wiki/Channel_capacity
[4] http://www.rfsworld.com/stayconnected/userfiles/stayconnected/issueimages/bilder/0308/WiMAX-filter_1_RZ.jpg http://www.rfsworld.com/stayconnected/userfiles/stayconnecte...
- petra 10y ago3. With regards to the IOT - most apps fits the LoRa capabilities. What's left is probably cameras(but for them WiFi mesh combined with backhaul/wired access would be fine, mostly. There's might be a place for drones, but i haven't though about that yet). 1. Auction 73 is just 100MHZ, not big enough to matter. Also how much bandwidth could you get from "OFDM amongst the ISM bands" , after discounting bands that are reserved due to politics,or for future use by government/military etc ? why aren't they being auctioned ? 2. Why would carriers bundle an expensive technology that doesn't promise them more money long term(after everybody deploys it, which will happen if someone does) ? they have simple ways to signal each other to do nothing. Also assuming this tech offers similar spectral efficiency, if all it buy is more bandwidth, by deploying more cells, just deploy more 4G cells - and save yourself money/risk/etc. Can't find the quote for the 5G price, but i think it was in the context of mmWave, just some discussion, nothing formal. But why would 5G(non mmWave) be cheaper ? companies have to justify r&d, more newer patents, start from lower volumes, complexity will probably grow, etc. Also a side question: is there a way to build around mmWave unlicensed bands to create noise floor similar to using it as a licensed band ? maybe with phase array antennas(assuming no other interfere but the network) ? or maybe a different legal regime that gives neighborhoods control of their spectrum ? because if so, and if we can solve the mmWave tech(big if), it would look like the future of wireless.