Complex modulation/demodulation -

[flong]

I am looking at a problem in CDMA relating to demodulation of the CDMA pilot signal. I would appreciate if someone could show me the error of my ways as it would appear to me that either my understanding of complex modulation/demodulation is incorrect (or the originator of the code that I am looking is incorrect).....

My understanding is as follows;

The CDMA2000 pilot is generated using Walsh code 0. Thus the I and Q data streams are all "zero" data. The output of the complex modulator is basically the sum and difference of the spreading Pn and Pq pseudo random sequences (as defined in the CDMA2000 specifications) i.e.

I = Pn - Pq
Q= Pn + Pq

Assuming that 0 represents a '1' and 1 represents a -1 then the encoding (after normalisation) should be as follows;

Pn Pq I Q Output
1 -1 1 0 cos
1 1 0 1 sin
-1 -1 0 -1 -sin
-1 1 -1 0 cos

Thus the output of the modulator will be

I * Cos(2*pi*fc* t) + Q * Sin(2*pi*fc* t).
In essence one outputs EITHER + OR - Cos OR + OR - Sin

In an ideal world the receiver output (baseband, after mixing) will be the original I and Q sequence.

If there is a locally generated pilot sequence in the receiver, then, as part of the process of locking onto the pilot, the incoming sequence is multiplied by the locally generated pilot in the following manner

I' = I*Pn + Q*Pq
Q' = Q*Pn - I * Pq

[charan langton]

Are you sure the pilot uses a 0 Walsh code. That does not seem correct. There have to be fairly good transitions for pilot to be useful.

[flong]

Thanks for your response Charan.

Yes, Walsh code zero is used for the Forward pilot in CDMA2000 ref. 3GPP2 C.S0010-C v2.0.

BUT

the I and Q channels (in this case "0" data") are spread with the PN sequences derived form teh generators defined below

2.1.3.1.17.1 Spreading Rate 1
2 The PN sequences shall be based upon the following characteristic polynomials:
3 PI(x) = x15 + x13 + x9 + x8 + x7 + x5 + 1
4 (for the in-phase (I) sequence)
5 and
6 PQ(x) = x15 + x12 + x11+ x10 + x6 + x5 + x4 + x3 + 1
7 (for the quadrature-phase (Q) sequence).

so you will certianly get plenty of transitions on both I & Q

That said the main question behind my initial question is to whether I have interpretted correctly the process of complex modulation and demodulation.

hope you can still help me on that one.

Regards
Finbarr

[charan langton]

No, your ideas look correct.

Charan