Jose P. of Intel has said that new Radio Access Network (RAN) will be about 90% –to-95% of CAPEX. He does not break that down in terms of base stations and subscriber equipment, because it varies depending on the cell size, i.e. macro cells vs pico…
Dear,
Our company name is Modacom Co., Ltd from Korea.
Modacom is a Korean company specializing in developing WIMAX devices of USB dongle, ExpressCard, Indoor CPE and Embedded Modules for mobile devices.
Now we are supplying commercial WiMAX products to KT (Korea ), UQ (Japan ), China, Delta(Azerbaijan) and other region.
We have the biggest Market share of Wimax Wave2 market in Korea, Japan, China, Azerjaiban.
If you do not mind, could you tell me ur email adress, I like to send you the detail product information by mail.
dear
hope u will b fine....i m working on a project in which we have to deliver triple play service i.e voice, video and data on wimax to our customers using redline equipment of 3.4 GHz(AN100U and AN100UX)....kindly help me regarding configuration of service classes,service flows and classifiers in detail....i will b thankful.
I agree with you. I faced similar problems of BER when Guard Band was reduced. In fact I recently queried from the Base Station vendor for 'out of band emission values'. For 3.5 GHz they did provide me a table mentioning Maximum Spurious Emissions at different bandwidths. But this BS is SC (Single Carrier) and when it comes to multi-carrier I believe the real challenge is to reduce ISI, the guard interval for reducing multipath effects becomes critical. With sub-channelization of the overall bandwidth we are more concerned with CCI rather than ACI, as all sectors of BS share sub-channels (which comprised of All SC)
For instance for a Hybrid placement of sub-channels the inner radius of the cells share all SC while the outer edges are provided with distinct sub-channels so that they do not interfere with neighboring cells .Different permutation base is used to minimize Co-sub channel interference.
The frequency scheme I discussed was for PUSC-All SC that means irrespective of the three frequency points (5 MHz each) the total 15 MHZ is considered for sub-channelization.
I think few things are getting mixed here, let me re-define my understanding and than re-ask the question,
Nominal Channel Bandwidth is the band assigned by the local regulatory authority. Used Bandwidth is the bandwidth which is physically occupied by WiMAX signal in frequency domain. The used bandwidth must be smaller than the nominal BW.
Guard Intervals is a particular ratio of the useful time inserted in OFDM symbol to collect multipath information, The ratio is called guard period (typical values are G 1/4 1/8 1/16 or 1/32) the absolute time is called cyclic prefix. We are unconcerned of this parameter at this moment.
Guard subcarriers are the outer carriers, which are not used for transmission. (see the attached image)
Now, my question is that even though the regulator has not defined explicit guard bands between the two operators nominal bandwidth, the used bandwidth must not interfere due to existent edge NULL (or Guard) sub-carriers. But still we allocate certain bandwidth as guards. As in my last explanation of 1,3,3 frequency plan 1.5 MHz was kept reserve at the beginning/ending of operator's band. Is the band occupied by NULL sub-carriers insufficient to avoid inter-operator interference?
Secondly for the ACI (Adjacent Channel Interference) part for a particular operator's band, aren't the guard bands enough to avoid ACI for frequency distribution like F1=3524, F2=3527.5 and F3=3531 (for channel size for 3.5 MHz) at same BTS with a total of 10.5 MHz allocation.
Referring to your example,
Rima wrote:
Let's suppose that we have two adjacent channels at the 3.5 GHz band with a channel size of 3.5 MHz (for instance the channel 3524 MHz & the channel 3527.5 MHz (central frequency)) could they be used by two near face-to-face sectors ??
For 3.5 MHz of channel size we have FFT size of 256, with 56 NULL carriers, with each sub-carrier consuming 15.625 KHz, the total band at edges is 875 KHz (437.5 KHz at each side), the next carrier will obviously be 0.875 MHz apart. What is the purpose of Guard Bands if such an allocation is not possible?
Reference to the Guard Band discussion, Guard bands in FDD makes sense, but in TDD if the allocated channel size itself has Guard sub-carriers (or Null Sub-carries at edges) than why we need to allocate additional Guard bands ? Is it there to have additional protection against the unwanted emissions that are out-of-band ?
I have seen a frequency plan that has 21 MHz of total bandwidth, a reuse scheme of 1,3,3 with 5 MHz channel and the rest 6 MHz is equally divided (1.5 MHz each) as guard bands. Although the NULL sub-carriers were already providing 1 MHz (0.5 + 0.5) of separation between the two adjacent channels, that makes the total guard band of 2.5 MHz.
o yeah.. that is cool.. so there is a lot i can ask u on the network :)....actually in malaysia we just started. nothing much i can tell u on the network. like me.. im working on the trial sites. so there is no life network still in malaysia.. for other wimax operator in malaysia yes... they do have but only for not more than 25 sites. hehe.. so ots to early to tell on the performance wise.. so means that for tunisia theres already more than 500 sites on air to cover 5 big city.. is it?