Antenna array suitable for localizing transmitter bearings

Question

I'm working with radio tagged animals. Fish tagged with 150 MHz tags. I'm interested in localizing these tag positions.

Ideally, I'd like to build a few stations that can return bearing estimates of the tag location. I was wondering how I could do this.

Initially (and knowing very little about radio), I thought 2 orthogonally oriented Yagi antennas could make such a station. Knowing the radiation patterns of the antennas, the difference in received levels could produce a bearing estimate. However, I'm also aware that closely placed Yagi antennas will interfere with each other, so that idea sounds unfeasible.

I was hoping to get some advice on how I could go about building a station that returns bearings of a radio transmitter. Perhaps there are more suitable antennas for this purpose, or perhaps there are alternative solutions than using antenna arrays.

additional info based on comments so far

Here im specifically working with river fish in shallow (fresh) water. In deep water, we use acoustic transmitters and do time difference of arrival positioning. Stations would be above water on the river banks, stationary (unattended), and distances to the fish would be max 200 meters away, so im interested in small spatial scales here.

I already write models for the underwater acoustic transmitters... But in shallow water (a meter or less), this tech works poorly. Radio transmitters are already widely used in fishes (tracking river movements)... But they are traditionally just tracked with handheld yagi antenna.

The radio tags used give out short pulses.. around 200ms long every few seconds.

Answer 1

Being out in the open, Time Difference of Arrival (TDOA) is pretty easy.

A single station with two antennas placed less about a quarter wavelength apart with some pin diodes and an oscillator will produce a tone proportional to the angle and a null when the signal is perpendicular.

I built one of these back in college to find rockets using this guide on radio direction finding, and it works really well.

One improvement to make on the simple two antenna version is when the signal is perpendicular, the system can not detect whether the signal is directly in front or in back. There are several methods to over come this, including making another set of antennas that are perpendicular to the first set.

Answer 2

Look up 'doppler direction finding'. It's an established technology, and used by the popular 'Lojack' vehicle recovery system. If I remember, a simple, stationary, antenna array can give a bearing without referencing complex antenna patterns. Multiple antenna arrays in different areas can triangulate the transmitter. If you have one array, make it mobile, and 'home in' on the signal directly.

Answer 3

Look up how the radio hams do 'Foxhunting' which is basically a competition to locate a small intermittently operating transmitter.

The antennas they use for the 2M band should scale well enough for the purpose.

Usually a dipole (Look for the null, it is shaper then the maxima) or yagi, steered by the operator.

There is another approach, you can do 'doppler DF' with 4 verticals electronically switched into a FM receiver, this can be made to give you the classic spy movie circle of leds indicating relative bearing, but multipath can be a real issue.