Kamien (1992) "Patent Licensing" Handbook of Game Theory, vol.1, chapter 11

This survey focuses on game-theoretic analysis of patent licensing. In the paper, the interaction between a patentee and licensees is described in terms of a three-stage game. In the first stage, a patentee sells the patent by using some mechanism. Firms in the industry simultaneously decide their buying strategies in the next stage. And finally, patents are distributed according to the rule of a mechanism and market competition (Cournot competition) is realized.
The main focus on the paper is comparison between different mechanisms which are (1) auction (2) fixed fee licensing (3) royalty (4) hybrid of (2)&(3). The interesting result is that licensing auction yields higher revenue than fixed fee or royalty does. In some situations, however, the hybrid mechanism colled "chutzpah" mechanism yields even higher revenue than auction does.
The analyses introduced in this survey has several restrictions such as; no private information, identical firms, only considering profit maximization.
In the end, the paper concludes as follows.

It is often the case that a survey of a line of research is a signal of it having peaked. This is certainly not true for game-theoretic analysis of paten licensing.

I hope this comment is still true although it has been written in 15 year ago...

Interesting Papers on References

Jensen (1989) "Reputational spillovers, innovation, licensing and entry" IJIO, 10-2
Katz and Shapiro (1985) "On the licensing of innovation" Rand, 16
Katz and Shapiro (1986) "How to license intangible property" QJE, 101
Muto (1987) "Possibility of relicensing and patent protection" EER, 31
Reinganum (1989) "The timing of innovation: Research, development and diffusion" in Handbook of IO