Department of Informatics, University of Oslo
29 September 1995
Comments and corrections to
D.H. Johnson & D.E. Dudgeon,
“Array Signal Processing: Concepts and Techniques”
Chapter 2:
Figure 2.1 (p 9). In ultrasound and radar imaging the theta is usually the elevation and the
phi the azimuth. This is the opposite of the definitions in the figure text. There is however
not a really well-defined definition for these angles.
Page 17: The distance between consecutive zero-crossings is given by kr = π or
r = λ ⁄ 2.
Chapter 3:
In eq. 3.2 and the equation before it (p. 60), when compared to the Fourier transform in
eq. 2.16 (p. 40), the (2 pi)^3 and the (2 pi)^4 should be removed.
Page 87 line 5: replace F(k,t) by F(k,w)
In the example on p 93 and the formula for the wavenumber vector, the last entry should
have a 2 pi and read: 2 pi (n1 - n2 + 2n3).
On p 95 below the example, it is a little confusing to read that the number of distinct lags
is M(M-1)/2, when one can count a higher number in the examples on the following
pages. It is conceptually simpler to count the number of distinct lags as 1 + M(M-1),
including lag 0 and both positive and negative lags.
Chapter 4:
The magnitude in the nearfield shown in fig 4.13 (p 150) shows that the amplitude has its
maximum at the focal point. It is a well-established fact in for instance medical ultrasound
that the intensity maximum is always closer than the focal point. This is due to the r^-1 in
the array pattern (as in the eq. on top of page 149). Probably this factor has been omitted
in the simulations leading to this figure and perhaps also in fig 4.14 and 4.15 also.
Chapter 6:
There seems to be an extra dy in the second equation on p. 269.
Chapter 7:
The equation in the middle of page 352 should have s’ instead of s in the first term.