I am looking for making a chip or a microchip, which can be traced with the help of the satellites. Means it is merely a chip and when I tune my satellite to a particular frequency I would get to know that this particular chip is present here and so. But it should be a chip only means no battery and all i.e. no other circuitry. I am thinking so if it is possible now and if not please suggest something similar to it so that I can use that in the similar way. Thanks
Firstly you seem to be under the common misunderstanding that the GPS satellites somehow calculate where a GPS receiver is. They don't. The satellites are transmit only systems (ignoring the command and control radio links with their control stations) and a GPS receiver is a receive only system.
As user114749 indicated RFID is the closest to what you want and even that doesn't quite fit the requirements, it still needs an antenna.
A 900MHz RFID tag doesn't actually transmit anything, it's not picking up anywhere near enough power to run a transmitter, it works by modulating the reflections of the readers signal bouncing off its antenna.
A good tag can have a read range of ~20 meters when running the transmitter at 1W with a moderately directional antenna. Radio follows the inverse square rule, if you double the range you need 4 times the power (a 6 dB increase). But with RFID since the signal has to reflect back to the transmitter, you lose 6 dB getting to the tag and another 6 dB getting back to the transmitter meaning 12 dB of losses for every doubling of the range.
The lowest possible practical orbit is around 350km (the fuel cost to maintain a stable orbit at this altitude is very high but it's possible). So when your satellite is directly over the tag you are trying to read we have log(350km/20m) / log(2) doublings of our 20 meter range. That works out as just over 14.
Which means we need 14 x 12 dB = 168 dB more power needed at the transmitter over our 1 W RFID reader.
So in order to pick up a tag that is directly under it we need our RFID reading satellite to broadcast with around 63,000 TW of power, about 2.5 times the total worlds power output.
To locate a tag you'd need a range of several times this (add an extra 0 or two to the power number), you can't fly directly over every point on the earth, and a fleet of satellites to give reasonable coverage. But those are relatively small issues in comparison to getting it to work with a single satellite to start with.
These numbers may be off by a little bit. In fact they could well be off by a lot. But unless I've gotten it wrong by a factor of a few million it's not even close to being practical.
I second Pisda's point, but lets get hypothetical here: RFID and related chip technology already uses a chip that is powered by the radio signal send to it. So in 'theory' you could use that to located the chip.
HOWEVER, you are asking about orbital detection of a chip. Over more then 100 kilometers (let me Google this for you: "distance to lowest satalite in orbit") comes up with a satellite at 705 km up in orbit. If I do not count all the interference caused by: other radio sources (on the same frequency), space radiation, static created by (atmospheric) magnetic fields etc etc. You still need to somehow send out radio waves that reach said chip. And lets keep that part more simple, use multiple coils and pick up charge from multiple frequencies. You are still going to TRY to send a signal from such a weak power source all the way back to your satellite. I'll not try to come up with a proper calculation as this is not really my field and I hope I have already demonstrated why this question is a bit silly.
I deem this endeavour unlikely with current technology unless there are something within quantum mechanics that could fix the distance problem. (I am thinking about quantum entanglement, but that field is still in early development and does not fix the direction problem. But that you could potentially fix by letting the chip locate its location with GPS signals and send the coordinates with the technology I just mentioned)
So allow me to propose something different: A tracer element like a radioactive material. Preferably one that radiates gamma radiation in larger quantities (material with low half life) and detect said radiation. Let me skip over the part that this creates a large health hazard :-) and tell you that there is already a net of satellite detectors that can sniff out radioactive radiation and I'm pretty sure they can determine some specific characteristics about said source to point out your tracer.