Hotspot Project for a Modern Market (data, voice and video)
Bidemi A. Olaiya
I need to setup a solution for a modern market that will take care of voice, video and data. The idea starts from hotspot but not limited to hotspot. On the Hotspot infrastructure, we will need to do advertisement for the shops that are connected to the hotspot service.... Does anyone have an idea on how better we can make this fly? I need to build a Proposal for the management.
My thought: I intend to use mesh potato since there is no readily available means of shop-to-shop communication in the whole modern shop.
Again, It's a modern market (Shopping mall) of about 2,000 units of shops and with about 4floors in height. Very massive..... What do you think? Any deeper idea? Counting on you guys....
I hope to read from everyone who has ideas. I appreciate the group a lot.
Bidemi A. Olaiya
.
T Gillett
Bidemi A. Olaiya
.
--
You received this message because you are subscribed to the Google Groups "Village Telco Development Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to village-telco-...@googlegroups.com.
To post to this group, send email to village-...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/village-telco-dev/414D6D5B-A309-4454-BA0F-C54EAF6D180D%40ampersandhitech.com.
For more options, visit https://groups.google.com/d/optout.
Dirk van der Walt
Bidemi A. Olaiya
--
You received this message because you are subscribed to the Google Groups "Village Telco Development Community" group.
To unsubscribe from this group and stop receiving emails from it, send an email to village-telco-...@googlegroups.com.
To post to this group, send email to village-...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/village-telco-dev/963bcdef-d3ee-44ed-9338-0043b168a110%40googlegroups.com.
Dave Duchesneau
Bidemi,
> If there are 2000 units…
I agree with Dirk that you may need a proven commercial solution. To manage at that scale you there needs to be a lot of stability, and a means to manage each node (including updates) without always having to visit it. I’m not sure how affordable Ruckus is, but it is certainly commercial gear.
If you’d like to look into something that shares its roots with VillageTelco, you might want to look into Open-Mesh (http://www.open-mesh.com/), which is similarly based on open-source batman/adv (http://www.open-mesh.org/), but is a distinctly commercial offering. The Open-Mesh OM-2P family of mesh nodes look quite a bit like the new VillageTelco MP02 nodes, but do not have any FXS port options. You may or may not like the fact that the Open-Mesh nodes are managed via a free cloud-based service, but you also can choose your own RADIUS-based services if you’d rather, for example.
Just in case it sounds like I’m advertising Open-Mesh nodes, let me just say that I’m not. I’m also not bashing VillageTelco’s MP02 nodes. Moving forward, I would very much like to see MP02 nodes accepted as first-class citizens on the free Open-Mesh cloud, so that deployments could mix and match node types as needed, while being able to take advantage of a ready-to-go management solution. Likewise, Open-Mesh nodes should be able to be used in VillageTelco deployments (they already can be, if connected back-to-back via Ethernet).
I’ve personally deployed many dozens of Open-Mesh nodes over several years at multiple not-for-profit campuses and community hubs. Both the product line and the behind-the-scenes software have evolved significantly and are quite easy to set up. With only a MAC address from a node’s label (or from a manifest), I can give an Open-Mesh node to someone to plug in, and then do everything else remotely.
All current Open-Mesh nodes have a built-in hardware watchdog to force a reboot and avoid extended node lock-ups. To me, that’s mandatory for any network larger than the one in your own home. Although Open-Mesh previously had dual-band nodes, they currently don’t. From the specs I’ve heard on the upcoming dual-band hardware, it will open up another set of options for high-density networks. They may be priced a bit higher, because Open-Mesh doesn’t fool around with advertising phony power levels, so the hardware will have to be priced higher to cover a second band with a higher-power radio.
All the best,
Dave Duchesneau
paulc...@gmail.com
Bidemi A. Olaiya
I have looked through the site you recommended and i find the equipments very interesting in look but unfortunately, none of their contact page and form is working. Do you mind asking them to relate with me through mail?
Thank you.
Bidemi
To view this discussion on the web visit https://groups.google.com/d/msgid/village-telco-dev/e471db46-1ce9-4b4a-850c-fea05c4cce90%40googlegroups.com.
Dave Duchesneau
Bidemi,
>> If there are 2000 units…
Ø For such a large installation i would be looking at Altai beamforming radios
Ø http://www.altaitechnologies.com/
The Altai gear appears to be well-designed, but in a completely different pricing category. Unlike the Open-Mesh (http://www.open-mesh.com/) gear I referred to in my earlier post, which is on par pricewise with VillageTelco gear (typically under $100 USD per node), the multi-beam Altai gear is two orders of magnitude more expensive (roughly $3400 USD for a single Altai A8n base station, with four sector antennas).
Also, I do not believe that the Altai gear has “beamforming” capability at all, as the term is conventionally defined. Altai’s marketing material uses various phrases which are commonly used in conjunction with beamforming, and are thus suggestive of beamforming. However, I have found nothing in any of Altai’s literature or on their web site which indicates that they actually claim to support beamforming (the shaping of an RF signal to increase gain in a particular direction, by means of electronically adjust the signal strength and phase of an RF signal to each antenna element in an array of elements), nor is there any technical indication that beamforming is in use. Rather, Altai appears to use multiple colocated fixed-pattern antennas, with each antenna connected to one or more radios (radios on different frequencies can share the same antenna by means of adequate filtering, such as by using a diplexer that combines two bands into one RF path). Altai seems strong on RF filtering, which allows it to co-exist with traditional cellular tower gear by reducing out-of-band emissions and rejecting inbound cellular signals. Altai’s major strength is probably the integration of multiple antennas with minimal cabling. Here’s an excerpt from a February 2011 press release (http://www.prnewswire.com/news-releases/altai-technologies-all-in-one-integrated-a8-i-super-wifi-base-station-launching-at-2011-gsma-mobile-world-congress-115877419.html), emphasis mine:
HONG KONG, Feb. 11, 2011 /PRNewswire-Asia/ -- Riding on A8's patented long range wireless technology, Altai Technologies today announces it will officially unveil its carrier grade A8-i Super WiFi Base Station at the 2011 GSMA Mobile World Congress in Barcelona, Spain (Feb 14-17).
Being the latest addition to its product portfolio of Super WiFi Solutions, the A8-i is an all-in-one multi-beam antenna array super WiFi base station that integrates the base unit and antennas into one panel. Utilizing smart antenna technologies and patented signal processing algorithm, the all-in-one A8-i provides the industry's best 360 degree signal coverage per base station and eliminates the RF cabling. Equipped with multiple antennas in a single enclosure, the A8-i provides up to 3 times the range and 10 times the per site coverage as traditional access points.
Altai’s technology evolved later to incorporate 802.11n (which is based on MIMO), and ultimately led to their 8x8 MIMO marketing material. Here’s a quote from page 2 of one of Altai’s 2013 fact sheets (http://www.altaitechnologies.com/wp-content/uploads/2013/10/Altai-Fact-Sheet-131008.pdf), emphasis mine:
Super Coverage
Altai has developed an advanced smart antenna technology that is part of its Super WiFi base station family and boosts the performance of any standard WiFi client. Altai’s patented smart antenna relies on a uniquely structured 8-element antenna array and a proprietary signal processing algorithm:
· The user signal information is received through all the 8 antenna elements in the array
· The smart antenna technology processes the signal in real time and optimize the SINR of each WiFi client
· The smart antenna technology obtains angular and polarization diversities – based on its 8-antenna array and multiple radios
The powerful 8-antenna array also empowers the 8x8 MIMO signal processer [sp] that is part of Altai’s base stations, providing further signal gain in both the uplink and downlink.
Super WiFi capacity Altai uses two different technologies concurrently, increasing its WiFi base station capacity by a factor of 10:
· 8x8:2 MIMO, providing much higher antenna diversity gain and strong enough signal for multiplying capacity
· Altai’s AirFi technology – a throughput optimization algorithm implemented at the WiFi layer, increasing its efficiency and capacity
Super Interference immunity
One of the benefits of Altai’s smart antenna technology with its narrow beam directed toward the user is that it inherently attenuates interference from all directions that are not within the beam directed to the WiFi client.
· It selects the channel with the least interference at all times
· The focused antenna beam attenuates interference coming from all other directions
To me, Altai’s technology is not at all what it appears to be, unless you read deeply. In particular, Altai appears to have a sophisticated antenna diversity system, which is very, very different than MIMO, and was developed and deployed prior to the existence of MIMO. Now that MIMO is available cheaply (for 2x2 MIMO anyway), Altai has adapted their 8-antenna (diversity) array to 2x2 MIMO radios and called the whole thing 8x8 MIMO (but sometimes also 8x8:2 MIMO, which is telling).
However, I believe Altai is playing fast and loose with their claim of 8x8 MIMO (multiple input, multiple output) capability. If I’m correct, Altai’s claims might be considered false advertising in many legal jurisdictions. Although Altai’s A8n family has 8 RF paths (for internal or external antennas, depending on the model), it appears to be little more than multiple 2x2 MIMO radio nodes operating on different frequencies (there are several indicators of this, including the maximum bandwidth of 300 Mbps and 150 Mbps per stream, which is precisely a 2x2 MIMO constraint for 802.11n, whereas a 3x3 MIMO would enable 450 Mbps maximum, etc.). Also, note that, under the standard definition of MIMO, all the MIMO streams in an NxN designation are operating on the SAME frequency, whereas Altai’s 8 streams operate on different frequencies (i.e., different channels in each of the 2.4 GHz or 5 GHz bands). Altai’s congestion avoidance technique seems to split the congestion over multiple channels (just as you can do with mesh nodes on multiple frequencies, but connected to the same switch). Altai’s diversity antenna scheme uses some combination of spatial, pattern, and/or polarization diversity, but definitely NOT adaptive arrays or AESAs (active electronically scanned arrays) that manipulate phase shifters and attenuators at the face of each radiating site (which is essential for beamforming). Unfortunately, in the time I made available, I could not locate any patents for any of Altai’s “patented” technology. Furthermore, the only patent I could find relating to 8x8 MIMO belongs to Qualcomm, Inc. (http://www.freepatentsonline.com/y2010/0208781.html).
One Altai model in particular, the A8-Ein, has a unit which integrates an “all-in-one multi-beam antenna array” into one panel, featuring “high 11n throughput capacity up to 300 Mbps data rate” and “Multi-beam 8x8:2 MIMO Smart Antenna Technology to provide superior signal strength and link budget” (http://neotech.asia/wireless-networks/altai-super-wifi/, under A8n series tab). Altai’s non-standard designation of “8x8:2 MIMO” here (emphasis mine) does not have a meaning that correlates to 8x8 MIMO under its usual NxN definition, where N=8 (N transmitting “output” antennas communicating with N receiving “input” antennas, where each of the N chains contributes to the total throughput). Under 802.11n, when N=2 (2x2 MIMO), one can expect a maximum speed of 150 Mbps via each chain, so with two chains, the maximum total is 300 Mbps. Likewise, when N=3 (3x3 MIMO), one could expect a maximum of 450 Mbps via the three chains, all on the same frequency. With Altai’s 802.11n 8x8:2 MIMO gear, the maximum throughput of is not 8 x 150 = 1200 Mbps, but is instead only 300 Mbps, the same as 2x2 MIMO. When you look at Altai’s 2013 fact sheet (cited earlier), the specification for the A8n on page 6 indicates a transmit power of “27 dBm (24 dBm/chain),” which makes it very clear that there it has only two MIMO chains, not eight. Since every 3 dBm represents a doubling of power, adding two 24 dBm chains together yields a maximum power of 27 dBm (each chain contributes half the power). If there were really eight chains (8x8 MIMO), each chain would contribute only one-eighth of the power, so that the total (with antennas) would remain within the legal EIRP (effective isotropic radiated power).
To me, the major appeal of the Altai A8n or A8-Ein families is the RF filtering that appears to have been integrated into each unit. For a DIY installation (something like Altai, but without using Altai gear), in order to co-locate antennas on multiple frequencies in the same band with reduced interference, you would typically need excellent channel-specific filtering (to reduce interference from co-located transmitters on adjacent or nearby frequencies, even with non-overlapping channels) Each channel-specific filter typically costs about as much as each node, or in some cases, more (see http://search.l-com.com/search?keywords=filter). With multiple frequencies in the same band transmitting simultaneously, such channel-specific filtering may be required for each RF input/output, even when the antennas are not pointed at each other.
Bidemi, my original intent was to provide you some additional avenues of exploration. I had not heard of the Altai gear, so I’m glad Paul brought it up, although that pretty much forced me to investigate further. I wasn’t going to address the details of a large installation, due to lack of time, and that hasn’t changed. However, now that I’ve looked into the Altai gear, I would encourage you to look also at Ubiquiti gear (http://www.ubnt.com/), which is well proven in many contexts, and potentially much less expensive and more flexible than the Altai gear.
Without attempting 8x8 MIMO, but striving to emulate Altai’s 2X2 MIMO-per-channel setup, you could use three Ubiquiti airMAX Rocket M2 radios (http://www.ubnt.com/airmax#airMaxHardware), each on its own channel (1, 6, or 11), and each having two channel-specific filters (one filter per MIMO stream, since each of the node two antennas are on the same channel). The maximum throughput would be 300 Mbps per 2x2 MIMO channel, but to achieve anywhere close to this rate would require three separate 2x2 MIMO antennas pointed in different directions (each on a different channel). This over-simplified setup would cost under $100 for each radio (http://www.streakwave.com/Itemdesc.asp?ic=RocketM2), around $100 for each filter (http://search.l-com.com/search/keywords-filter--res_per_page-100), and around $100 for each 2x2 MIMO panel antenna (http://search.l-com.com/search?display_type=&keywords=mimo+panel+antenna), for a total of around $500 to $600 for each of the three channels. Rather than use such an approach for regular clients, however, I would use it ONLY for links to matching radios which provide backhaul or long-link connections to local meshes (e.g., VillageTelco, Open-Mesh, etc.). With the Ubiquiti gear (and some others), you can reduce the channel width and do other things to increase the range and reduce interference. Of course, you can introduce other bands.
If you want to extend the example by adding 5 GHz radios into the mix, you can do it without having to buy more panel antennas, lightning arrestors, etc. Several antenna models are not only MIMO capable (cross-polarized dual antenna ports), but also dual-band on those same ports (see http://www.l-com.com/wireless-antenna-24-49-58-ghz-13-15-dbi-cross-polarized-flat-panel-mimo-antenna-n-female-connector, which offers 13 dBi gain and a 43-deg beam on 2.4 GHz, with 15 dBi gain and a 25-deg beam on 5 GHz). To use both the 2.4 GH and 5 GHz bands with high-power Ubiquiti radios, you would need a diplexer (http://search.l-com.com/search?keywords=diplexer) for each dual-band antenna port. Merging a 2.4 GHz Rocket M2 and a 5 GHz Rocket M5 onto the same dual-band antenna would require two diplexers, each Rocket M node has two antenna ports. Each diplexer would combine one of the 2.4 GHz ports with one of the 5 GHz ports, with the combined signal going to the dual-band antenna. If channel-specific filtering is also used, those filters need to be between the Rocket M node and the diplexer. Ubiquiti Rocket M series nodes can be used off-the-shelf (such as for long-distance links or Internet backhaul), or flashed with VillageTelco’s SECN v2 software (for layer 2 mesh operation). With multiple co-located nodes, you can mix and match long-distance and local-mesh (SECN) functionality by hardwiring the nodes together as needed.
At lower power levels (e.g., 20 dBm), you can also choose VillageTelco-compatible dual-band radio gear that already has built-in diplexers such as the TP-Link WDR3600 (http://www.amazon.com/TP-LINK-TL-WDR3600-Wireless-Gigabit-300Mbps/dp/B008RV51EE/ref=sr_1_1?ie=UTF8&qid=1397503312&sr=8-1&keywords=tp-link+wdr3600), which (thanks to Terry Gillett) can be flashed with SECN v2. The WDR3600 has two antenna ports, each of which can directly drive a port on the aforementioned dual-band 2x2 MIMO panel antennas (with suitable RP-SMA to N-type adapters, lightning arresters, etc. A cheap-but-effective solution would be to use a TP-Link WDR3600 as with a dual-band panel antenna as dual-band backhaul or long-link solution communicating with one or more distant units, all flashed with SECN v2 and operating on both bands concurrently to take advantage of interface alternating (http://www.open-mesh.org/projects/batman-adv/wiki/Multi-link-optimize). To further improve the range, the channel width can be reduced from 20 MHz (default) to 10 MHz or 5 Mhz, which will provide stronger connections at the expense of maximum theoretical data rate (cutting the width by half also reduces the maximum theoretical rate by half, but may not reduce the actual rate, since the connection may be much better). This setup may provide a reliable-but-cheap long distance link, while also minimizing the power required for solar/battery-backed operation (a dual-radio WDR3600 under full load consumes around 4 to 5 watts, which compares favorably to the Altai 8n, which is specified at 30 watts. Ideally, one or more local mesh nodes (e.g., MP02) operating on a non-overlapping 2.4 GHz channel would then be plugged into any of the WDR3600’s four LAN ports (as long as such mesh nodes are behind the panel antenna attached to the TP-Link WDR3600, they will enjoy a front-to-back isolation of 25 dB from the panel antenna, without any additional channel filtering). I won’t get into all the possible configuration details or possibilities, but this should provide a sense of Altai alternatives that are available now and can be tailored at will.
I apologize for the length of this post, and for jumping around a bit. I wish I had time to edit it down and organize it better.
To view this discussion on the web visit https://groups.google.com/d/msgid/village-telco-dev/e471db46-1ce9-4b4a-850c-fea05c4cce90%40googlegroups.com.