Searching \ for '[EE] 9600 baud FSK modem IC?' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=9600+baud+fsk+modem
Search entire site for: '9600 baud FSK modem IC?'.

Exact match. Not showing close matches.
PICList Thread
'[EE] 9600 baud FSK modem IC?'
2010\11\15@201320 by Vitaliy

face
flavicon
face
Need one for a quick proof of concept. Does one exist, or is the only recourse to use a PIC and write custom firmware?

2010\11\15@201932 by enkitec

picon face
On 15-Nov-10 23:12, Vitaliy wrote:
> Need one for a quick proof of concept. Does one exist, or is the only
> recourse to use a PIC and write custom firmware?
>

    Try MX589 or FX589.

    Mark Jorda

2010\11\15@204634 by Vitaliy

face
flavicon
face
enkitec@gmail.com wrote:
>> Need one for a quick proof of concept. Does one exist, or is the only
>> recourse to use a PIC and write custom firmware?
>>
>
>     Try MX589 or FX589.

Sorry to be asking a stupid question, but what is the difference b/w GMSK and FSK? The output waveforms look the same (to my untrained eye), and Google wasn't too helpful.

Vitaliy

2010\11\15@212937 by Michael Watterson

face picon face
 On 16/11/2010 01:46, Vitaliy wrote:
> spam_OUTenkitecTakeThisOuTspamgmail.com wrote:
>>> Need one for a quick proof of concept. Does one exist, or is the only
>>> recourse to use a PIC and write custom firmware?
>>>
>>      Try MX589 or FX589.
> Sorry to be asking a stupid question, but what is the difference b/w GMSK
> and FSK? The output waveforms look the same (to my untrained eye), and
> Google wasn't too helpful.
>
> Vitaliy
>




Similar to difference between Phase Modulation and Frequency Modulation.

A real FSK @ 9600bps modem might be unusual. Though there are 9,600 baud FM packet Radio systems running. Not sure if 12.5KHz or 25kHz channels.


Historically FSK even used two oscillators (RTTY), switching between them. Historically you made the Shift bigger than needed to ease demodulation, which limited speed on a phone line to 1200 /75 using FSK  1200/1200 and higher use phase modulation. 9600 baud on a phone line using PSK or FSK isn't possible. The 4800/9600 baud uses QAM

28K is Trellis encoded. 33k is using compression, it's really only 28k, that's actually the highest speed analogue modem for end to end PSTN.

56K is really a form of ISDN. It only works into a gateway at a local exchange, never end to end over the network. The uplink is still 28K analogue (33k with compression)


GMSK uses a single oscillator modulated to minimise the bandwidth

2010\11\16@054540 by alan.b.pearce

face picon face
> Need one for a quick proof of concept. Does one exist, or is the only
> recourse to use a PIC and write custom firmware?

IIRC Microchip have a downloadable modem DSP library for a dsPic.


-- Scanned by iCritical.

2010\11\16@113810 by Herbert Graf

picon face
On Tue, 2010-11-16 at 02:29 +0000, Michael Watterson wrote:
> 28K is Trellis encoded. 33k is using compression, it's really only 28k,
> that's actually the highest speed analogue modem for end to end PSTN.

I'm not sure where you get that.

28.8k is 3200baud, while 33.6k is 3429 baud, there is no "compression"
involved to get 33.6k.

The max theoretical bandwidth from analog end to analog end is about
35kbit/s.

The fact that one end is digital (i.e. an ISP) allows for V.90 and V.92
which are capable of 56/33.6 and 56/48 respectively.

https://secure.wikimedia.org/wikipedia/en/wiki/Modem

TTYL

2010\11\16@132303 by Michael Watterson

face picon face
 On 16/11/2010 16:39, Herbert Graf wrote:
> On Tue, 2010-11-16 at 02:29 +0000, Michael Watterson wrote:
>> 28K is Trellis encoded. 33k is using compression, it's really only 28k,
>> that's actually the highest speed analogue modem for end to end PSTN.
> I'm not sure where you get that.
>
> 28.8k is 3200baud, while 33.6k is 3429 baud, there is no "compression"
> involved to get 33.6k.
>
> The max theoretical bandwidth from analog end to analog end is about
> 35kbit/s.
>
It's a long while since I had one (28k/33k) Modem. I don't think Wikipedia existed then. Maybe I presumed the 33k mode was compression as I never saw that speed!

I gave away my 56K modem last year. . It spent a lot of time only doing 19K :(  Though when I had a Wireless Phone for a while it did 49k. Then the Wireless company went bust and I had to go back to copper pair. Not used it since Nov 2005 for Internet when I got in a 10.6GHz DOCSIS over Microwave link. Cheaper than phone line and gives 8M down/1M up + 2 x real phone numbers using QOS VLAN to the ISPs own PSTN gateway. The QOS and codec is good enough that my Fax works on it.



> The fact that one end is digital (i.e. an ISP) allows for V.90 and V.92
> which are capable of 56/33.6 and 56/48 respectively.
>
That bit I'm clear on.
> https://secure.wikimedia.org/wikipedia/en/wiki/Modem
>
> TTYL
>

2010\11\16@133517 by Charles Craft

picon face
Do you have a link to your wireless ISP?
What's the distance between your antenna and their end?

On 11/16/2010 1:22 PM, Michael Watterson wrote:
>
> I gave away my 56K modem last year. . It spent a lot of time only doing
> 19K :(  Though when I had a Wireless Phone for a while it did 49k. Then
> the Wireless company went bust and I had to go back to copper pair. Not
> used it since Nov 2005 for Internet when I got in a 10.6GHz DOCSIS over
> Microwave link. Cheaper than phone line and gives 8M down/1M up + 2 x
> real phone numbers using QOS VLAN to the ISPs own PSTN gateway. The QOS
> and codec is good enough that my Fax works on it.
>    
>

2010\11\16@141539 by Dave Lagzdin

picon face
On 16 November 2010 11:39, Herbert Graf <.....hkgrafKILLspamspam@spam@gmail.com> wrote:

> The fact that one end is digital (i.e. an ISP) allows for V.90 and V.92
> which are capable of 56/33.6 and 56/48 respectively.

A little trouble with that wording for me, help me with "end"?
You likely have an analogue loaded loop for both directions of
transmission up to the
nearest CO, then likely a PCM channel bank card or newer equivalent
aggregating into
a high speed digital channel in the backhaul to the next CO.

AFAIR- unlike us, Europe embraced ISDN and maybe 2B+Q over non-loaded
loops is the norm so 48K+ is "digital" to them?

2010\11\16@142157 by Herbert Graf

picon face
On Tue, 2010-11-16 at 14:15 -0500, Dave Lagzdin wrote:
> On 16 November 2010 11:39, Herbert Graf <hkgrafspamKILLspamgmail.com> wrote:
>
> > The fact that one end is digital (i.e. an ISP) allows for V.90 and V.92
> > which are capable of 56/33.6 and 56/48 respectively.
>
> A little trouble with that wording for me, help me with "end"?

Analog phone networks are actually digital networks with analog "ends".

The links between pretty much everything is digital, save the link from
your CO to you.

So, if both you and the number you are calling have an analog "end",
i.e. you have an analog link to your CO and they have an analog link to
their CO, then V34 (33.6k ideally) is the most you're going to get.

If OTOH the number you are calling is connected digitally to their CO
then your modem can take advantage of V90/V92 and get up to 53.3K (in
NA) down.

TTYL

2010\11\16@145955 by Michael Watterson

face picon face
 On 16/11/2010 18:35, Charles Craft wrote:
> Do you have a link to your wireless ISP?
> What's the distance between your antenna and their end?
12.5km

http://www.digiweb.ie
> On 11/16/2010 1:22 PM, Michael Watterson wrote:
>> I gave away my 56K modem last year. . It spent a lot of time only doing
>> 19K :(  Though when I had a Wireless Phone for a while it did 49k. Then
>> the Wireless company went bust and I had to go back to copper pair. Not
>> used it since Nov 2005 for Internet when I got in a 10.6GHz DOCSIS over
>> Microwave link. Cheaper than phone line and gives 8M down/1M up + 2 x
>> real phone numbers using QOS VLAN to the ISPs own PSTN gateway. The QOS
>> and codec is good enough that my Fax works on it.
>>
>>

2010\11\16@163711 by Charles Craft

picon face
On 11/16/2010 2:59 PM, Michael Watterson wrote:
>    On 16/11/2010 18:35, Charles Craft wrote:
>    
>> Do you have a link to your wireless ISP?
>> What's the distance between your antenna and their end?
>>      
> 12.5km
>
> http://www.digiweb.ie
>    
That's pretty sweet - "The antenna is small and unobtrusive, effectively a 12cm square plastic box."

Our carrier is 900 Mhz and I've got this beast on the roof. Better than the lag with satellite. :-)
  sites.google.com/site/bubbasnmp2/157_5716_a.JPG
(the bottom antenna with the hardware cloth behind it is for OTA HD TV.)

{Quote hidden}

>

2010\11\16@174216 by Sean Breheny

face picon face
On Tue, Nov 16, 2010 at 11:39 AM, Herbert Graf <.....hkgrafKILLspamspam.....gmail.com> wrote:
> The max theoretical bandwidth from analog end to analog end is about
> 35kbit/s.

That sounds right to me as a practical limit, but not a theoretical
limit. Where did you get that from? If it were a purely analog line
then the throughput would be governed by the bandwidth (fixed at about
8kHz) and the signal to noise ratio (which could always be improved
and so does not impose a theoretical limit). That leaves the
digitization step as the theoretical limit and that is 64 kbit/sec.

Sea

2010\11\16@174302 by Sean Breheny

face picon face
Correction to my own post: bandwidth is fixed at about 3kHz. 8kHz is
the sampling rate for the digitization.

On Tue, Nov 16, 2010 at 5:42 PM, Sean Breheny <EraseMEshb7spam_OUTspamTakeThisOuTcornell.edu> wrote:
{Quote hidden}

2010\11\16@174659 by Olin Lathrop

face picon face
Sean Breheny wrote:
> If it were a purely analog line
> then the throughput would be governed by the bandwidth (fixed at about
> 8kHz) and the signal to noise ratio (which could always be improved
> and so does not impose a theoretical limit).

But it's not a purely analog line and it does have inherent signal to noise
limits.  The ends may be analog, but everything in between is digital.  Even
if the analog ends are perfect, the digitization sets a upper limit on the
signal to noise ratio.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000

2010\11\16@175945 by Sean Breheny

face picon face
I understand that, I was just wondering where the 35kbit/sec number
comes from. It sounds like a rule of thumb and not a theoretical
limit. If it were a theoretical limit, it should apply in a lab-type
setting, too, and yet it would be fairly easy to inject a signal into
a telephone codec which got any arbitrary sequence of codes out of the
ADC, which could then deliver a data rate of 64 kbits/sec. Even with a
kilometer of wire in front of the codec, it could be done in the lab
by doing system identification on the wire to determine the transfer
function and then pre-distorting the signal to compensate.

In fact, this must actually be possible (not quite 64kbps) in a
real-world situation on a high-quality phone line since 56 kbps modems
do work on at least one analog end.

Sean


On Tue, Nov 16, 2010 at 5:47 PM, Olin Lathrop <@spam@olin_piclistKILLspamspamembedinc.com> wrote:
{Quote hidden}

>

2010\11\16@191244 by Dave Tweed

face
flavicon
face
Sean Breheny wrote:
> I understand that, I was just wondering where the 35kbit/sec number
> comes from. It sounds like a rule of thumb and not a theoretical
> limit. If it were a theoretical limit, it should apply in a lab-type
> setting, too, and yet it would be fairly easy to inject a signal into
> a telephone codec which got any arbitrary sequence of codes out of the
> ADC, which could then deliver a data rate of 64 kbits/sec. Even with a
> kilometer of wire in front of the codec, it could be done in the lab
> by doing system identification on the wire to determine the transfer
> function and then pre-distorting the signal to compensate.
>
> In fact, this must actually be possible (not quite 64kbps) in a
> real-world situation on a high-quality phone line since 56 kbps modems
> do work on at least one analog end.

56k works in the ISP-to-end user direction because the end user's modem
uses a 14- to 16-bit ADC to measure the voltage coming down the line, and
with enough training, it can syncrhonize itself precisely to the 8-kHz
sampling rate of the digital part of the system, and it can distinguish
about 128 unique voltage levels in each sample interval. Note that because
of u-law companding, these voltage levels are not linearly distributed,
which is why you need such a high-resolution ADC. It isn't 256 levels
becuase in the u-law system there are pairs of codes that are
indistinguishable, and furthermore, the phone company does some in-band
signalling by robbing (overwriting) one of the bits every now and then.
If both ends of the link are analog, then you have to treat the digital
part in the middle as a simple noisy channel, because there's no way to
syncrhonize to the sampling or calibrate to the telco's actual codec
voltage levels. As a result, you have an available bandwidth of about
3400-3700 Hz, and an effective S/N ratio of about 40-50 dB -- but keep in
mind that this is quantization noise in a companding codec, not simple AWGN..
That's why you end up with a limit of about 35 kbps in this mode -- 9 to 10
bits per baud at 3200-3400 baud.

-- Dave Twee

2010\11\16@191250 by William \Chops\ Westfield

face picon face
> 9600 baud FSK modem IC?
>
So *what* are we talking about here?

AFAIK, a 9600bps FSK modem over normal phone lines is impossible.  The  9600bps modems used much more exotic modulation schemes (as is being  discussed.)  IIRC, regular FSK topped out at about 700bps, and all  modems 1200bps and faster are more exotic...

But that doesn't mean you couldn't have a 9600bps FSK modem over a  link with inherently higher bandwidth than a phone line (say, a radio  link.)

Was the original poster looking for a 9600bps FSK modem, or a 9600bps  phone-line modem?

BillW

2010\11\16@200343 by Michael Watterson

face picon face
 On 16/11/2010 21:37, Charles Craft wrote:
> On 11/16/2010 2:59 PM, Michael Watterson wrote:
>>     On 16/11/2010 18:35, Charles Craft wrote:
>>
>>> Do you have a link to your wireless ISP?
>>> What's the distance between your antenna and their end?
>>>
>> 12.5km
>>
>> http://www.digiweb.ie
>>
> That's pretty sweet - "The antenna is small and unobtrusive, effectively
> a 12cm square plastic box."

My latency is about 20ms to 30ms, typically 22 to Irish sites.

It saves the usual cost of oven reference for the 20MHz DOCSIS upstream to 10.6GHz conversion by using a single 9.7GHz approx L.O. in an PLL,
The Base station has a pilot at 10.252GHz. A 495MHz approx pilot receiver locks the Microwave L.O. used for downconversion.

The indoor Cable modem is thus about 485MHz (there are 4 x 8MHz channels) Downstream and about 20MHz upstream. This is mixed to about 820MHz and filtered in 2 x SAW filters and mixed the the single 9.7GHz approx L.O.  The LO. is spilt with wilkinson stripline to TX mixer and then the other 1/2 split for PLL prescaler and RX down mixer.  It uses a PIC to control the two LMX series synth / PLL (the 800MHz 1st IF for TX and the 9.7GHz syncronised to the off air 10.252GHz beacon).

Obviously the Basestation uses Oven reference and up converts the CMTS cable downstream to feed a 30W rated 10.25GHz approx PA at about 3W. And Downconverts the 10.6GHz from domestic units to the CMTS IF in frequency..

So Cable Broadband, with only about 3m to 20m coax at each house and full duplex X band microwave to headend. Some sites they have 3 x 28MHz FDD allocations. In theory if you have LOS and put a dish instead of the pair of PCB aerials (H &V for TX and RX, or vice versa), the system could work to 180km (160km is the DOCSIS max plant length, but that is coax, Radio goes further in same time).

So newer VSAT is changing from Mix of DVB-S + DVB-RCS and proprietary systems to a satellite version of DOCSIS. Wildblue in USA and Tooway in Europe.

More to Come
http://www.techtir.ie/blog/watty/ka-sat-countdown

DOCSIS over wireless is more efficient than Wimax and lower latency. But no good for Mobility. Though oddly it can support Nomadic, (1min to 3min sector re-connect)

DOCSIS is also used at 700MHz and 800MHz. (Caribbean)? and can use ANY bandwidth from about 2MHz to 10MHz per channel downstream by changing IF filters and and Symbol Rate. 6MHz & 8MHz are of course standard US and Euro channels used on all the Microwave DOCSIS, but Arriss does 2MHz filtered Modems (you can use adjoining channels!) for the UHF DOCSIS they sell branded WiDox :)  Many Microwave bands available. DOCSIS standard implementations use TDMA assigned slots by base CMTS for upstream in 0.8, 1.6, 3.2 or 6.4MHz channels. QPSK to 32QAM possible.

Since you simply need a full duplex transverter for HF & UHF to an FDD band, it's easy to put DOCSIS on a new band. Digiweb has about  60K subscribers on 5Mbps to 12Mbps  (64QAM downlink) packages. There is also an Irish ISP with 3.6GHz FDD DOCSIS (16QAM for longer range)

DOCSIS 3.0 isn't so interesting for Wireless as it's bonding entire channels. More Hybrid Fibre Cable operators.

The Average DSL speed in Ireland is < 3Mbps
The Average Cable speed is over 10Mbps (set to rise to over 30MBps)
The Average Wireless DOCSIS is over 6Mbps

My speed test just now on 8Mbps down, 1Mbps up package
http://www.speedtest.net/result/1033367017.png

Cable Modem Stats  (12.5km microwave)
*RX  Power*  0 dBmV    Frequency 485MHz
*Signal to Noise Ratio*        29 dB        *Modulation*        QAM64

*Shared Upstream Data Rate*        5120 Ksym/sec        *Modulation*        QPSK

*Power *44 dBmV   Logical Channel 1   (I think about 20MHz)

Real Downstream is about 10.248GHz, Upstream about 10.590GHz
The PA transistor is regular MAR6 size package. mWs of power... Aerial panel gain is about 17dBi I think....
N.F. of RX  is about 2.5dB!

Ordinary RG6 with 12V 600mA DC inserted connects cable modem to "Radio". The Radio is under 200 Euro I think. Cable Modem with two phone jacks (2 x ATA) is under 50 Euro.

Irish Line rental for copper is 26 euro a line. Before you add calls or DSL.   My package is about 40 Euro including two "lines" with real numbers. I have DECT phones and fax.
QOS, almost zero packet loss and low jitter and ISP has own SIP servers and PSTN gateway means even Fax works. Or indeed the dialup modem built into my laptop :-)

2010\11\16@201420 by Michael Watterson

face picon face
 On 16/11/2010 22:42, Sean Breheny wrote:
> On Tue, Nov 16, 2010 at 11:39 AM, Herbert Graf<KILLspamhkgrafKILLspamspamgmail.com>  wrote:
>> The max theoretical bandwidth from analog end to analog end is about
>> 35kbit/s.
> That sounds right to me as a practical limit, but not a theoretical
> limit. Where did you get that from? If it were a purely analog line
> then the throughput would be governed by the bandwidth (fixed at about
> 8kHz) and the signal to noise ratio (which could always be improved
> and so does not impose a theoretical limit). That leaves the
> digitization step as the theoretical limit and that is 64 kbit/sec.
>
> Sean
Sample rate is 8KHz
So practical bandwidth is < 4KHz, hence the 3200 symbol rate or  3429 symbol rate limits. SNR, uLaw/ALaw and 8 bit companded resolution limit the bitrate (bits per symbol)

The V90/V92 mode by pass the digitisation with uLaw or Alaw and 8kHz sampling. Either for Downstream only or Downstream and Upstream.

If V90/V92 had been designed for Europe only, it would have been 64k Max rather than 56k Max

Actual ISDN in Europe, entry level is always been 2 x B + D = 2 x 64 +16 = 144kbps  Some Telcos did offer 9,600bps X.25 over the D (signalling channel) as the user data only uses the B channels.

2010\11\16@202411 by Michael Watterson

face picon face
 On 16/11/2010 22:59, Sean Breheny wrote:
> I understand that, I was just wondering where the 35kbit/sec number
> comes from. It sounds like a rule of thumb and not a theoretical
> limit. If it were a theoretical limit, it should apply in a lab-type
> setting, too, and yet it would be fairly easy to inject a signal into
> a telephone codec which got any arbitrary sequence of codes out of the
> ADC, which could then deliver a data rate of 64 kbits/sec. Even with a
> kilometer of wire in front of the codec, it could be done in the lab
> by doing system identification on the wire to determine the transfer
> function and then pre-distorting the signal to compensate.
>
> In fact, this must actually be possible (not quite 64kbps) in a
> real-world situation on a high-quality phone line since 56 kbps modems
> do work on at least one analog end.
>
> Sean

The practical limit is 33kbps because most telco do not implement full V92. Its 33k analogue for upstream (max) and the pseudo 56k ISDN (to suit USA ISDN) on downstream only. Hence end to end rather than Internet you are limited to top analogue upstream of each end.

You can get 64K end to end (or 128k). You have a modem called and ISDN TA and it does all the stuff you suggest. Except they designed it for error free 144k.

Before DSL was available, the ISDN was very popular if you could afford it and get it. Eventually they lowered the line rental to same as analogue and you simply paid twice as much if you used 128K instead of 64k.

NO Flat rate! Still no Flat rate! cents per minute to use the Internet. Even today, if they can supply POTS or  ISDN but not DSL, there is no true flat rate, it's effectively 10 times the cost of DSL for same amount of typical time (which is MUCH less data!)

2010\11\16@202718 by Michael Watterson

face picon face
 On 17/11/2010 00:12, William "Chops" Westfield wrote:
>> 9600 baud FSK modem IC?
>>
> So *what* are we talking about here?
>
> AFAIK, a 9600bps FSK modem over normal phone lines is impossible.  The
> 9600bps modems used much more exotic modulation schemes (as is being
> discussed.)  IIRC, regular FSK topped out at about 700bps, and all
> modems 1200bps and faster are more exotic...
>
> But that doesn't mean you couldn't have a 9600bps FSK modem over a
> link with inherently higher bandwidth than a phone line (say, a radio
> link.)
>
> Was the original poster looking for a 9600bps FSK modem, or a 9600bps
> phone-line modem?
>
> BillW
>
This was what I wondered. As 1200 bps down and 75bps up is fastest FSK I ever saw!

Typically FSK at 9600bps would want 40KHz or more. PSK is near 20kHz

2010\11\17@004156 by Sean Breheny

face picon face
OK, I understand now. In a way, 35kbits/sec is a theoretical limit,
but it is rather removed from first principles because it depends on a
lot of details about how the system is implemented. Thank you for the
excellent, detailed description.

Sean


On Tue, Nov 16, 2010 at 7:12 PM, Dave Tweed <RemoveMEpicTakeThisOuTspamdtweed.com> wrote:
{Quote hidden}

>

2010\11\17@095255 by Herbert Graf

picon face
On Tue, 2010-11-16 at 17:42 -0500, Sean Breheny wrote:
> On Tue, Nov 16, 2010 at 11:39 AM, Herbert Graf <spamBeGonehkgrafspamBeGonespamgmail.com> wrote:
> > The max theoretical bandwidth from analog end to analog end is about
> > 35kbit/s.
>
> That sounds right to me as a practical limit, but not a theoretical
> limit. Where did you get that from? If it were a purely analog line
> then the throughput would be governed by the bandwidth (fixed at about
> 8kHz) and the signal to noise ratio (which could always be improved
> and so does not impose a theoretical limit). That leaves the
> digitization step as the theoretical limit and that is 64 kbit/sec.

>From the wiki page I linked.

As for the figure, I didn't check the numbers, but it did seem
reasonable. FWIW the bandwidth on an analog POTS line isn't anywhere
near 8kHz. It's more like 3kHz (the exact figure depends on the
carrier). The 8kHz you mention is I believe the sample rate used by the
ADCs.

TTYL

2010\11\17@110035 by Sean Breheny

face picon face
Yes, I goofed on the bandwidth and I sent a correction message right
after my initial message.

Sean


On Wed, Nov 17, 2010 at 9:53 AM, Herbert Graf <TakeThisOuThkgrafEraseMEspamspam_OUTgmail.com> wrote:
> As for the figure, I didn't check the numbers, but it did seem
> reasonable. FWIW the bandwidth on an analog POTS line isn't anywhere
> near 8kHz. It's more like 3kHz (the exact figure depends on the
> carrier). The 8kHz you mention is I believe the sample rate used by the
> ADCs.
>
> TTYL
>
>

2010\11\25@010121 by Vitaliy

face
flavicon
face
William "Chops" Westfield wrote:
> But that doesn't mean you couldn't have a 9600bps FSK modem over a
> link with inherently higher bandwidth than a phone line (say, a radio
> link.)
>
> Was the original poster looking for a 9600bps FSK modem, or a 9600bps
> phone-line modem?

Not a phone-line modem. The channel bandwidth is about 20 kHz.

Given the number of responses and not a single "hit", I shall assume that a 9600 baud pure-FSK modem IC does not exist. :)

Vitaliy

2010\11\25@023137 by RussellMc

face picon face
> Not a phone-line modem. The channel bandwidth is about 20 kHz.
>
> Given the number of responses and not a single "hit", I shall assume that a
> 9600 baud pure-FSK modem IC does not exist. :)

This may be useful

       http://www.amsat.org/amsat/articles/kd2bd/9k6modem/

As may this

  http://www.symek.com/g/fsk9601.html

Useful background

   http://www.avrfreaks.net/index.php?name=PNphpBB2&file=printview&t=53752&start=0


              Russel

2010\11\25@032916 by Michael Watterson

face picon face
 On 25/11/2010 05:34, Vitaliy wrote:
> William "Chops" Westfield wrote:
>> But that doesn't mean you couldn't have a 9600bps FSK modem over a
>> link with inherently higher bandwidth than a phone line (say, a radio
>> link.)
>>
>> Was the original poster looking for a 9600bps FSK modem, or a 9600bps
>> phone-line modem?
> Not a phone-line modem. The channel bandwidth is about 20 kHz.
>
> Given the number of responses and not a single "hit", I shall assume that a
> 9600 baud pure-FSK modem IC does not exist. :)
>
> Vitaliy
>
Yes, I'd assume QPSK at very least, no FSK.

summary background info I used for a talk & demo of packet Radio about 6 years ago
http://www.wattystuff.net/amateur/packet/index.htm

2010\11\25@060855 by William \Chops\ Westfield

face picon face

On Nov 24, 2010, at 9:34 PM, Vitaliy wrote:

> I shall assume that a
> 9600 baud pure-FSK modem IC does not exist. :)

Don't some of the low cost wireless transceivers claim to do 9600bps  FSK ?

Doing it in 20k of bandwidth seems like it would be pretty tricky,  though.
On phones we got 1/5th of supposed total bandwidths (600bps out of  3000Hz
bandwidth?)

BillW

2010\11\25@072052 by Michael Watterson

face picon face
 On 25/11/2010 11:08, William "Chops" Westfield wrote:
> On Nov 24, 2010, at 9:34 PM, Vitaliy wrote:
>
>> I shall assume that a
>> 9600 baud pure-FSK modem IC does not exist. :)
> Don't some of the low cost wireless transceivers claim to do 9600bps
> FSK ?
>
ASK/OOK on 433MHz and other ISM bands up to 9,600 using Manchester Encoding is feasible. The chips have no modem as such. The RX is a superhet and dataslicer.

Analog Devices and Maxim have some parts that carry FSK 9,600. But it's just the RF part to /from baseband (i.e. Superhet receiver). Not the actual "modem".

There are a couple of modules that have modem too..

http://in.rsdelivers.com/product/micrel/micrf620z-tr/433mhz-ism-band-transceiver-module/0382652.aspx

I have something like this that does 9,600 or maybe 19,200 full duplex maybe Micrel?
There is no baseband. UHF only.. may not really be FSK, but QAM / QPSK

The chips these are based on are often adaptable for 350MHz to 950MHz, but no other frequency.

> Doing it in 20k of bandwidth seems like it would be pretty tricky,
> though.
> On phones we got 1/5th of supposed total bandwidths (600bps out of
> 3000Hz
> bandwidth?)
>
> BillW
>

2010\11\25@095243 by John Coppens

flavicon
face
On Wed, 24 Nov 2010 22:34:35 -0700
"Vitaliy" <RemoveMEpiclistspamTakeThisOuTmaksimov.org> wrote:

> Given the number of responses and not a single "hit", I shall assume that a
> 9600 baud pure-FSK modem IC does not exist. :)

I've never seen FSK at 9k6, but if it really has to be that, and it has
to be baseband, I guess the old VCO/PLL pair could do the job (2 ICs
though, _and_ analog) IIRC XR2206 and XR2211 or so.

But I'd rather sacrifice the FSK, and go to some of those PSK systems
Russell indicated.

Joh

2010\11\25@190120 by William \Chops\ Westfield

face picon face

On Nov 25, 2010, at 4:20 AM, Michael Watterson wrote:

>> Don't some of the low cost wireless transceivers claim to do  
>> 9600bps FSK ?
>>
> ASK/OOK on 433MHz and other ISM bands up to 9,600 using Manchester
> Encoding is feasible. The chips have no modem as such.

I was thinking of modules like these:

http://lynx-dev.com/index.php?page=shop.browse&category_id=106&option=com_virtuemart&Itemid=2

(which definitely claim to be FSK...)

BillW

2010\11\26@142723 by Vitaliy

face
flavicon
face
Michael Watterson wrote:
> ASK/OOK on 433MHz and other ISM bands up to 9,600 using Manchester
> Encoding is feasible. The chips have no modem as such. The RX is a
> superhet and dataslicer.
>
> Analog Devices and Maxim have some parts that carry FSK 9,600. But it's
> just the RF part to /from baseband (i.e. Superhet receiver). Not the
> actual "modem".
>
> There are a couple of modules that have modem too..
>
> http://in.rsdelivers.com/product/micrel/micrf620z-tr/433mhz-ism-band-transceiver-module/0382652.aspx

What currency is P.O.A.?

I was really looking for a modem to utilize an existing channel, but this is interesting nonetheless.

Vitaliy

2010\11\26@143704 by Vitaliy

face
flavicon
face
John Coppens wrote:
>> Given the number of responses and not a single "hit", I shall assume that
>> a
>> 9600 baud pure-FSK modem IC does not exist. :)
>
> I've never seen FSK at 9k6, but if it really has to be that, and it has
> to be baseband, I guess the old VCO/PLL pair could do the job (2 ICs
> though, _and_ analog) IIRC XR2206 and XR2211 or so.

This may be exactly what I was looking for... reading the datasheets now, thanks!


> But I'd rather sacrifice the FSK, and go to some of those PSK systems
> Russell indicated.

PSK appears to be more expensive than FSK.

Vitaliy

2010\11\26@193654 by John Coppens

flavicon
face
On Fri, 26 Nov 2010 12:35:50 -0700
"Vitaliy" <piclistEraseMEspam.....maksimov.org> wrote:

> > I've never seen FSK at 9k6, but if it really has to be that, and it has
> > to be baseband, I guess the old VCO/PLL pair could do the job (2 ICs
> > though, _and_ analog) IIRC XR2206 and XR2211 or so.
>
> This may be exactly what I was looking for... reading the datasheets now,
> thanks!

You're welcome... You might be helped with some other ICs from that
era, like the NE565 (or LM565), and some even used the CD4046 to decode
FSK. And don't discard using an FM chip to decode FSK. I once used the
TDA120 to do that for data and Fax signals (analog fax).

Luck,
Joh

2010\11\26@215839 by Vitaliy

face
flavicon
face
John Coppens wrote:
>> > I've never seen FSK at 9k6, but if it really has to be that, and it has
>> > to be baseband, I guess the old VCO/PLL pair could do the job (2 ICs
>> > though, _and_ analog) IIRC XR2206 and XR2211 or so.
>>
>> This may be exactly what I was looking for... reading the datasheets now,
>> thanks!
>
> You're welcome... You might be helped with some other ICs from that
> era, like the NE565 (or LM565), and some even used the CD4046 to decode
> FSK. And don't discard using an FM chip to decode FSK. I once used the
> TDA120 to do that for data and Fax signals (analog fax).

I decided to read read up on the theory in Miller's "Modern Electronic Communication", and guess what I found... a circuit of a Full-duplex modem based on the XR-2206 and XR-2211. :-)

Vitaliy

2010\11\27@052844 by Michael Watterson

face picon face
 On 27/11/2010 02:58, Vitaliy wrote:
> John Coppens wrote:
>>>> I've never seen FSK at 9k6, but if it really has to be that, and it has
>>>> to be baseband, I guess the old VCO/PLL pair could do the job (2 ICs
>>>> though, _and_ analog) IIRC XR2206 and XR2211 or so.
>>> This may be exactly what I was looking for... reading the datasheets now,
>>> thanks!
>> You're welcome... You might be helped with some other ICs from that
>> era, like the NE565 (or LM565), and some even used the CD4046 to decode
>> FSK. And don't discard using an FM chip to decode FSK. I once used the
>> TDA120 to do that for data and Fax signals (analog fax).
> I decided to read read up on the theory in Miller's "Modern Electronic
> Communication", and guess what I found... a circuit of a Full-duplex modem
> based on the XR-2206 and XR-2211. :-)
>
> Vitaliy
>
I like happy endings.
I've also seen RTTY decoder done with a pair of NE567s
FM receiver ICs, can work, but some might have internal small C and thus have a minimum frequency. I've used a low noise cascode amp that is 100MHz to 4GHz with lower limit due to internal capacitor

More... (looser matching)
- Last day of these posts
- In 2010 , 2011 only
- Today
- New search...